The Ideological Turing Test

Post Reply New Topic

We've had ideas like this before on WP.

http://econlog.econlib.org/archives/201 ... gical.html

The idea put forward here is simple: can you simulate an ideology that isn't your ideology? How well can you do it?

Yeah, been there done that. I learned long ago, I can't simulate my ACTUAL ideology enough to convince anybody that even I believe it.

I can do a very good imitation of a senior professor of the old school, though.

Krugman is so full of himself.

I don't think this is too hard anyone that actually discusses things with people they disagree with as a opposed calling them names. I'd say Austrians probably learn more about Keynesians than vice versa just because you have to seek out an Austrian argument on things while Keynesian ideas are the "mainstream". Most people in general have no idea what either mean tho.

As is every other human on the planet.

This is purely anecdotal, but I have observed this sort of mind-blindness among conservatives. They seem unable to even conceive of anyone who could disagree with them. I spent over an hour debating a conservative Christian friend of mine, and never did manage to get him to understand that it was possible for two people to disagree on the meaning of a given passage of Scripture.


Possibly true, but Krugman's claim was about liberals and conservatives- not liberals and right-libertarians. Conservatism is certainly mainstream enough that we can assume most liberals have at least heard of it.

I'd be willing to give the test a shot. Go ahead and assign me any ideology within reason. We have attempted this exercise in the past on the forum, through AG's "Argue for a position you disagree with" threads. Recall that the attempts of our most die-hard conservative poster in that thread were comically feeble.

you never attend a liberal cocktail party in san francisco and expect to hear positions argued for. Perhaps its declasse. What you do hear is "_______(social issue X) is unnacceptable in the world we live in, especially with all the wealth we have" and other statements of faith in the cause. There tends to be far more disagreement and views fully fleshed out at conservative events Ive attended.

We have a great many liberal intellectuals on WP but I don't think there are many who can at least regurgitate the very best right-wing scholarship. Sarah Palin, Glenn Beck, or Rush Limbaugh? they are like straw men and not a problem to quote. If its not a statement of faith, it is an assessment of the motives of the opposition or highlighting the positions/statements of glenn beck to knock down.

the average non-intellectual liberal only knows liberal positions and constructed conservative strawmen to knock down. When its apples-to-apples, it's dead even.

It is the mind. It is not just the Conservative [and I have had that happen myself] not just the Liberal [like certain relatives who assume EVERYBODY not obviously a redneck must be in agreement, not just the bigot, not just the animal righter, not just the mainstream linguist.

Some minds actually see only one way - not just as the only right way, the only IMAGINABLE way. And conversing with them is very hard, unless you are the same in which case it is either a symposium or a fight.

I have been running Inuyasha's account for the past 3 weeks does that count?

I see you point but I don't think it is dead even (obviously or I would be conservative)
I think that you have made a great argument of how the conservative media has undermined intellectual conservatism.
Intellectual conservatism has been undermined on a different front as well the ideological think tanks are either
libertarian(Koch) funded like the Cato institute or funded by fundies like the American Heritage Foundation.
It is really hard for a classical conservative thinker to find work without compromising his values and intellectual freedom.
and since Nozickian politics are essentially state of nature, invisible hand voodoo I think Krugman could explain them just fine.

I personally love how Bryan Caplan exaggerates his outsider status, how "out of the norm" right-libertarian economists supposedly are. I'd personally think your average economist would have a very hard time simulating the arguments of a sophisticated proponent of second generation rent controls or of a Post-Keynesian rejection of equilibirum analysis a lot more than simulating a poor old minority right-libertarian economist (although, to be fair, some Austrians also reject static equilibirum analysis).

Regardless, the whole test doesn't really tell you much except whose position is more "on the fringe", as people on the fringe of a discipline obviously have a lot more experience arguing against the mainstream and hence learning the nuances/unwritten premises of any given ideology. An economist who opposes free trade would probably do a lot better than even libertarians at "simulating the other's position".



And would attending a conservative country club lead to anything better? Although, to be honest, I've found that people in general dislike sophisticated ideological discussions in social settings.



What the hell counts as "Keynesian"? If you mean that your average New Keynesian, conservative or liberal, couldn't explain an Austrian well, then I suppose that may be true. But I bet your average conservative New Keyensian or even Austrian would be pretty poor at explaining Post-Keynesianism compared to the Post-Keynesian's ability to explain Austrianism/conservative New Keynesianism.

Last edited by Master_Pedant on 23 Jun 2011, 6:23 pm, edited 1 time in total.

Well... given that's pretty strongly libertarian, he is very out of the norm compared to Krugman.

Umm...... those are very specific positions, M_P. The Turing Test idea really appears to be much broader than "EXACT POLICY POINT X!! !", but rather that a person can simulate an overall ideology, not that a person can defend highly specified positions. Krugman isn't being called to simulate Bryan Caplan's position on all things, only the broadly defined ideology that Caplan holds himself to.


Look, comparing the Post-Keynesian to the Austrian is going to be difficult. The two groups don't talk much.

In any case, a major issue is still going to be that it's more likely that this is really a matter of a broadly defined ideology. For instance, Caplan in a later post simulates conservatism, but he doesn't go into all of the various arguments or details, and he isn't so narrowly focused on the specifics of an economic school of thought.

Do I think your criticism is entirely wrong? No. Do I think you're going too far? Absolutely.

The same "logic" used to apply to broader ideologies tends to equally apply to specific positions. Regardless, it seems that Caplan's whole post was about narrowing Krugman's inital, general claim about broader conservatives/saltwater trained economists down to right-libertarians. As a matter of fact, there appears to be some field-specific information Caplan thinks would be essential in this test, as he does specify:



Later, Caplan even gets more specific:



Personally, aside from to what a precise, detailed, narrow degree he thinks this test applies, it also seems to indicate some cluelessness. While an extremely high percentage of economists agree with free trade, there are heterodox positions and those economists who hold those heterodox postions may actually be better at explaining the logic of free trade than your average proponent.



Do you mean, in general, they don't reach a wider audience through their writings or that they don't talk with each other that much? If the later, that's generally true, but there has been some mutual acknowledgement:

http://www.informaworld.com/smpp/conten ... a791283942 < == Post-Keynesian author acknowledging Austrians on uncertainty
http://mises.org/journals/qjae/pdf/qjae2_4_5.pdf <=== Austrian Author acknowledges Post-Keynesians on uncertainty

Except this alteration isn't that significant. Economists generally tend to be socially liberal.


Um.... no, he really didn't say that field specific information is central. He said "We need to compare people to others in their category". "We need to baseline people to the knowledge of their field". He didn't say "Krugman's gotta be able to talk about Austrian economics", instead he stated "It doesn't inform us of anything if Krugman's not able to explain Nozickian political philosophy, but it would be impressive if he could"


He's not being as narrow and sticklerish as you are at all. Even further, I don't see anything about cluelessness. It may be true that critics are better at explaining things. I don't think this is centered on the issue though.


I clearly meant the latter. I *KNOW* there's been mutual acknowledgment. Hell, GLS Shackle and Ludwig Lachmann had a good long-standing relationship, and they are both figures with some significance to their respective schools of thought. There's even an interview with GLS Shackle on the mises.org website. That doesn't mean that the groups talk much, and that's all I really meant.

The whole style and thought processes conservatives use to come to their conclusions are generally quite different and less technical.



A lot of my comments on Austrian economics where directed a Jacoby, who brought the subject up, but the whole aspect of similar "the other side's beliefs" in one's function as an economist entails knowing "the other side's" economics. As a very extreme right-libertarian, that would probably indicate neoclassical monetarism if not Austrianism for Bryan Caplan's "side".



Are you and I reading the exact same post? It seems to be exactly what he's talking about - knowing more about the "other side" being a proxy for increased objectivity or thoughtfulness and the categorical conclusion that right-leaning, free market economists would be generally better at it.



In general, I'd agree that your statement holds.

I am not sure of your point. Conservatives usually use libertarian economists when promoting their ideas about markets.


He's not a macroeconomist, so he might not be focused on that issue as much. Caplan is a political economist and also a labor economist.


The idea of it being a "Turing test" isn't that you just test for knowledge, only that a person can simulate something with reasonable proficiency. Knowledge helps, but in the same sense that knowing language allows a computer to speak English. Knowing all of the specifics isn't necessary, but knowing enough to meaningfully engage the idea is important. You do have a point on how strongly minority beliefs have a real advantage, BUT, potentially not one that clouds Caplan's.

In any case, I am not actually JUST reading that post. Caplan and other authors on the blogosphere have engaged the idea of the ideological turing test. Caplan's own simulation was actually NOT very technical in its outlook, meaning that if we interpret Caplan by Caplan, we have to say that any comments on technicality have to be moderated by his example. http://econlog.econlib.org/archives/201 ... vativ.html (Note: His example comes across TOO libertarian to take seriously, I think)

I still think the style and depth of right-libertarians would be different and better than that of conservative supply-siders.



The whole thesis of the post seems to be "Isn't Krugman acting so high and mighty without merit??! ! I bet us libertarian economists, who are in the minority, could represent leftwing Keynesians who control the Ivy establishment a lot better than he could represent us!! !" That seems to require some specific info.



Urgh, it still seems that Bryan Caplan is arguing for pulling off a deception that could fool professional, ideologically oriented, economists into thinking they're dealing with one of their own. That still seems to imply specific info and in the comments section a poster notes that Caplan really isn't asking for a true Turning test.



Perhaps that's the problem. I haven't read the other post, so perhaps Caplan gave a misimpression when stating what he was suggesting and claiming.

Okay, I've read that other post. I don't see it as that applicable to the post we're debating here. There, it seemed just a sort of "what would I write if I wanted to win over sympathetic, general libertarians to the conservative side". In the other post, it was "could I fool an educated audience into thinking I was a leftwing, Keynsian economist better than leftwing Keynesians could fool people into thinkin they were Free Market economists". Those seem two completely different topics.