That is entirely possible. However, in that case, I would expect that other people would argue social-democratic positions well (assuming we hold that social-democratic positions have the same prior probability as those of any other ideology of equivalent complexity), and receive upvotes for it. Instead, I just saw an overwhelmingly neoliberal consensus in which I was actually one of the two or three people explaining or advocating left-wing positions at all.
Think of the Talmud’s old heuristic for a criminal court: a clear majority ruling is reliable, but a unanimous or nearly unanimous ruling indicates a failure to consider alternatives.
Now, admittedly, neoliberal positions appear often appealingly simple, even when counterintuitive. The problem is that they appear simple because the complexity is hiding in unexamined assumptions, assumptions often concealed in neat little parables like “money, markets, and businesses arise as a larger-scale elaboration of primitive barter relations”. These parables are simple and sound plausible, so we give them very large priors. Problem is, they are also complete ahistorical, and only sound simple for anthropic reasons (that is: any theory about history which neatly leads to us will sound simpler than one that leads to some alternative present, even if real history was in fact more complicated and our real present less genuinely probable).
So overall, it seems that for LessWrong, any non-neoliberal position (ie: position based on refuting those parables) is going to have a larger inferential distance and take a nasty complexity penalty compared to simply accepting the parables and not going looking for historical evidence. This may be a fault of anthropic bias, or even possibly a fault of Bayesian thinking itself (ie: large priors lead to very-confident belief even in the absence of definite evidence).
Now, admittedly, neoliberal positions appear often appealingly simple, even when counterintuitive. The problem is that they appear simple because the complexity is hiding in unexamined assumptions, assumptions often concealed in neat little parables like “money, markets, and businesses arise as a larger-scale elaboration of primitive barter relations”. These parables are simple and sound plausible, so we give them very large priors. Problem is, they are also complete ahistorical, and only sound simple for anthropic reasons (that is: any theory about history which neatly leads to us will sound simpler than one that leads to some alternative present, even if real history was in fact more complicated and our real present less genuinely probable).
This particular example doesn’t seem troublesome to me, because I’m comfortable with the idea of bartering for debt. That is, my neighbor gives me a cow, and now I owe him one- then I defend his home from raiders, and give him a chicken, and then we’re even. A tinker comes to town, and I trade him a pot of alcohol for a knife because there’s no real trust of future exchanges, and so on. Coinage eventually makes it much easier to keep track of these things, because then we don’t have my neighbor’s subjective estimate of how much I owe him versus my subjective estimate of how much I owe my neighbor, we can count pieces of silver.
Now, suppose I’m explaining to a child how markets work. There are simply less moving pieces to tell it as “twenty chickens for a cow” than “a cow now for something roughly proportional to the value of the cow in the future,” and so that’s the explanation I’ll use, but the theory still works for what actually happened. (Indeed, no doubt you can explain the preference for debt over immediate bartering as having lower frictional costs for transactions.)
In general, it’s important to keep “this is an illustrative example” separate from “this is how it happened,” which I don’t know if various neoliberals have done. Adam Smith, for example, claims that barter would be impractical, and thus people immediately moved to currency, which was sometimes things like cattle but generally something metal.
I would expect that other people would argue social-democratic positions well
In this particular thread or on LW in general?
In the particular thread, it’s likely that such people didn’t have time or inclination to argue, or maybe just missed this whole thing altogether. On LW in general, I don’t know—I haven’t seen enough to form an opinion.
In any case the survey results do not support your thesis that LW is dominated by neoliberals.
but a unanimous or nearly unanimous ruling indicates a failure to consider alternatives.
Haven’t seen much unanimity on sociopolitical issues here.
On the other hand there is that guy Bayes… hmm… what did you say about unanimity? :-D
Problem is, they are also complete ahistorical, and only sound simple for anthropic reasons
Graeber’s views are not quite mainstream consensus ones. And, as you say, *any* historical narrative will sound simple for anthropic reasons—it’s not something specific to neo-liberalism.
Not sure what you are proposing as an alternative to historical narratives leading to what actually happened. Basing theories of reality on counterfactuals doesn’t sound like a good idea to me.
In any case the survey results do not support your thesis that LW is dominated by neoliberals.
The survey results are out? Neat!
Not sure what you are proposing as an alternative to historical narratives leading to what actually happened. Basing theories of reality on counterfactuals doesn’t sound like a good idea to me.
I’m not saying we should base theories on counterfactuals. I’m saying that we should account for anthropic bias when giving out complexity penalties. The real path reality took to produce us is often more complicated than the idealized or imagined path.
Graeber’s view are not quite mainstream consensus ones.
The question is: are they non-mainstream in economics, anthropology, or both? I wouldn’t trust him to make any economic predictions, but if he tells me that the story of barter is false, I’m going to note that his training, employment, and social proof are as an academic anthropologist working with pre-industrial tribal cultures.
At minimum, it does seem like many anthropologists see Graeber’s work as much more tied into his politics than things even often are in that field, and that’s a field that has serious issues with that as a whole.
That is entirely possible. However, in that case, I would expect that other people would argue social-democratic positions well (assuming we hold that social-democratic positions have the same prior probability as those of any other ideology of equivalent complexity), and receive upvotes for it. Instead, I just saw an overwhelmingly neoliberal consensus in which I was actually one of the two or three people explaining or advocating left-wing positions at all.
Think of the Talmud’s old heuristic for a criminal court: a clear majority ruling is reliable, but a unanimous or nearly unanimous ruling indicates a failure to consider alternatives.
Now, admittedly, neoliberal positions appear often appealingly simple, even when counterintuitive. The problem is that they appear simple because the complexity is hiding in unexamined assumptions, assumptions often concealed in neat little parables like “money, markets, and businesses arise as a larger-scale elaboration of primitive barter relations”. These parables are simple and sound plausible, so we give them very large priors. Problem is, they are also complete ahistorical, and only sound simple for anthropic reasons (that is: any theory about history which neatly leads to us will sound simpler than one that leads to some alternative present, even if real history was in fact more complicated and our real present less genuinely probable).
So overall, it seems that for LessWrong, any non-neoliberal position (ie: position based on refuting those parables) is going to have a larger inferential distance and take a nasty complexity penalty compared to simply accepting the parables and not going looking for historical evidence. This may be a fault of anthropic bias, or even possibly a fault of Bayesian thinking itself (ie: large priors lead to very-confident belief even in the absence of definite evidence).
This particular example doesn’t seem troublesome to me, because I’m comfortable with the idea of bartering for debt. That is, my neighbor gives me a cow, and now I owe him one- then I defend his home from raiders, and give him a chicken, and then we’re even. A tinker comes to town, and I trade him a pot of alcohol for a knife because there’s no real trust of future exchanges, and so on. Coinage eventually makes it much easier to keep track of these things, because then we don’t have my neighbor’s subjective estimate of how much I owe him versus my subjective estimate of how much I owe my neighbor, we can count pieces of silver.
Now, suppose I’m explaining to a child how markets work. There are simply less moving pieces to tell it as “twenty chickens for a cow” than “a cow now for something roughly proportional to the value of the cow in the future,” and so that’s the explanation I’ll use, but the theory still works for what actually happened. (Indeed, no doubt you can explain the preference for debt over immediate bartering as having lower frictional costs for transactions.)
In general, it’s important to keep “this is an illustrative example” separate from “this is how it happened,” which I don’t know if various neoliberals have done. Adam Smith, for example, claims that barter would be impractical, and thus people immediately moved to currency, which was sometimes things like cattle but generally something metal.
In this particular thread or on LW in general?
In the particular thread, it’s likely that such people didn’t have time or inclination to argue, or maybe just missed this whole thing altogether. On LW in general, I don’t know—I haven’t seen enough to form an opinion.
In any case the survey results do not support your thesis that LW is dominated by neoliberals.
Haven’t seen much unanimity on sociopolitical issues here.
On the other hand there is that guy Bayes… hmm… what did you say about unanimity? :-D
Graeber’s views are not quite mainstream consensus ones. And, as you say, *any* historical narrative will sound simple for anthropic reasons—it’s not something specific to neo-liberalism.
Not sure what you are proposing as an alternative to historical narratives leading to what actually happened. Basing theories of reality on counterfactuals doesn’t sound like a good idea to me.
The survey results are out? Neat!
I’m not saying we should base theories on counterfactuals. I’m saying that we should account for anthropic bias when giving out complexity penalties. The real path reality took to produce us is often more complicated than the idealized or imagined path.
The question is: are they non-mainstream in economics, anthropology, or both? I wouldn’t trust him to make any economic predictions, but if he tells me that the story of barter is false, I’m going to note that his training, employment, and social proof are as an academic anthropologist working with pre-industrial tribal cultures.
Previous years’ survey results: 2012, 2011, 2009. The 2013 survey is currently ongoing.
How would that work?
I am not sure what the mainstream consensus in anthropology looks like, but I have the impression that Graeber’s research is quite controversial.
At minimum, it does seem like many anthropologists see Graeber’s work as much more tied into his politics than things even often are in that field, and that’s a field that has serious issues with that as a whole.