« The culture that is French, a continuing series | Main | The age-performance profile in baseball »
Where do our beliefs come from?
We all like to think that our beliefs come from rational thinking, deep experience and good judgment. But suppose that you had to predict someone else's beliefs, let's say their beliefs about taxes, welfare, regulation....economic policy of all kind. Let's put some money on it, the better your predictions the more money you make.
I will give you one piece of information to improve your predictions. Either I will tell you whether the person whose beliefs you must predict is an economist or a biologist or I will tell you whether the person whose beliefs you must predict is American or French. Which piece of information do you want?
What does this say about where beliefs come from?
Addendum: Suppose I asked you instead to predict the types of arguments that the person will use to justify their beliefs. Now which piece of information do you want? What is the role of education in determining beliefs?
Posted by Alex Tabarrok on December 14, 2007 at 07:43 AM in Economics, Philosophy, Religion | Permalink
Comments
Nationality, noting that in some countries a small elite believes something totally different from the average citizen. Haiti would be an example there, though a Haitian biologist and economist would still be pretty close in orientation.
Posted by: Tyler Cowen at Dec 14, 2007 8:01:25 AM
the obvious (at least to me) answer would be that nationality predicts beliefs, with a similar but not as concrete division along educational fields lines, but types of arguments/justifications will be determined by education history, with an almost irrelevant correlation to nationality.
In fact, I'd be surprised that someone would think otherwise...
What would also be interesting is whether people would choose religion, gender, and/or income level as determining facts, and what order they would be selected in for which arguments.
Posted by: shawn at Dec 14, 2007 8:28:58 AM
Beliefs of all kinds is simply too vague. Without specific examples I have no idea what you mean. Beliefs about food would probably be more effected by nationality. In general, are we only looking at people who are either economists or biologists? Surely having an advanced degree of any type will have a larger effect on beleifs than nationality compared to the average member of the population. Beliefs about proton pumps would definitely be more effected by being a biologist. I don't have a strong opinion about whether what they call an "economist" in France belongs to the category we Americans would call an "economist" rather than, say, a "sociologist", but I would expect a great deal more consensus among American and French "neoclassical economists", which is almost what "economist" means to me, then I would between American and French citizens and neoclassical economists of the same nationality.
Posted by: michael vassar at Dec 14, 2007 8:32:15 AM
Most of that increased consensus regarding economic and policy matters might come from economic education, but outside of economic matters it would simply be from IQ, socio-economic status, and years of education. Sadly, not much from the math tools, which they should do more with than prove their cleverness, but usually don't.
Posted by: michael vassar at Dec 14, 2007 8:35:27 AM
Maybe occupation (biologist or economist) doesn't say anything about how we choose our beliefs. Instead, maybe it indicates that people who hold certain beliefs are more inclined toward some disciplines than others.
Posted by: Frank at Dec 14, 2007 8:57:17 AM
Nationality, but only of the "_______ -American" variety.
I would be kidding myself to think I could accurately predict the answer from a member of the general public in any particular country besides the one in which I was raised.
We presume that we know other people, but we really know very little. Step into another culture and we know next to nil.
Posted by: Pearl Alexander at Dec 14, 2007 9:06:04 AM
I just wish everyone would read Ayn Rand, because that's where most beliefs SHOULD come from, and out society would be so much better if we took her prescriptions to heart.
Posted by: Mike at Dec 14, 2007 9:29:57 AM
Nationality, for certain.
For me, it all boils down to culture, and even (higher) education is just a function of culture. Each culture carries with it basic and unquestioned assumptions about how the world works and should work, which tend to be better predictors of beliefs (and the arguments for them) than a diploma.
Posted by: Jeff H. at Dec 14, 2007 9:31:06 AM
I know you are giving the example to make your question concrete, but it oversimplifies. The origins of an individual's beliefs are very complex. Whatever are the "most important" experiences in her life are likely to be the major determinants. For some people these are religious experiences, for some educational/intellectual experiences, for some personal tragedies, etc. However beliefs are also the result of a cumulative process I think. One's current beliefs are a function of recent experiences, past beliefs and "rational thinking, deep experience and good judgment".
Therefore when you try to boil it down to a choice between nationality and occupation (1) this oversimplifies the possible answers and (2) oversimplifies the character of belief formation. Further the answer to your question would be different if you chose two different countries and occupations (say France/England and artist/investment banker).
The second question you ask about justifying beliefs may have a simpler answer. More often than not this is probably most directly related to a person's occupation or training.
Posted by: Ben at Dec 14, 2007 9:53:19 AM
You should not want your beliefs to be determined by either profession or nationality. Your confidence could reasonably vary with what you learned more about. But your expectations should not so vary.
Posted by: Robin Hanson at Dec 14, 2007 9:55:20 AM
But people's beliefs do vary by profession and nationality (among other things), Robin. In fact, a post on MR last week showed that people's beliefs about how to treat their parents vary widely across countries. Maybe we have different definitions of "beliefs" in mind...
Posted by: Ben at Dec 14, 2007 10:03:41 AM
False dichotomy. Economists and biologists are both more scietific than the mainstream asnd would tend to view things in those terms. A better comparison would be "American or French" versus "economist or social worker."
Posted by: at Dec 14, 2007 10:07:30 AM
Robin, forgive me if you've already discussed this somewhere, but what are your thoughts about living in a different culture as a means of overcoming bias?
In my experience, integrating into a different culture virtually forces one to confront basic assumption about one's thinking--the result for me was to at least recognize more of my biases, if not exactly overcome all of them.
Posted by: Jeff H. at Dec 14, 2007 10:08:53 AM
But Tyler, in economics as in Haiti a small elite "believes something totally different from the average citizen"!
Alex, your question is interesting--hugely important, I'd say--but seems slightly ill-posed.
Thinking scientifically about belief formation and taxes or belief formation and Big Government to the extent of prediction with small posterior errors involves I think different methods--different types of arguments--than does thinking scientifically about belief formation in one person who with some chance error may be a Frenchman or a biologist. The sample sizes are vastly different, for one thing, introducing more heterogeneity and co-variates of belief in the former, where experimental error may be weighed against random error.
On the point I argue in The Cult of Statistical Significance (with Deirdre McCloskey, forthcoming Jan. 2008) that most statistical scientists--in almost every field-- have gotten all confused about their theories of belief in statistics.
The problems with statistical significance in economics you are familiar with (though Ziliak and McCloskey show in the new book how much worse the problem is in some parts of medicine and in most of pharmacology and clinical psychiatry, where the damage done is immediately palpable. Think Vioxx.)
The widespread inattention of scientists to oomph--to the stuff worthy of belief--we call "the sizeless stare of statistical significance."
A second major and even more widespread problem is the fallacy of the transposed conditional. In Fisherian statistics beliefs get tested in a null procedure. But this is not hypothesis testing--belief testing--at all. In Fisher's procedure which is to say in 95% of today's statistical science publications, the probability of the data, assuming a true null, is what you get. But what you conclude is "the probability of the null hypothesis, assuming the data are true."
This is the fallacy of the transposed conditional.
And it is easy to see in Bayes' theorem that the conventional test of significance entails an obvious logical fallacy:
P(H, d) = P(d, H)*P(H)/P(D)
or posterior probability equals the likelihood function (Fisher) times the prior probability.
Today's significance testers lop off the prior probability--what Bayesians and other people with common sense and a feel for profit have ever since "Student" and Harold Jeffreys referred to as "experimental" or "scientific" belief. Where does the prior belief come from? Depends on the field of inquiry. In experimental brewing, for example, your prior is formed by establishing aesthetic and chemical and other kinds of industrial standards for the beer, and then experimenting like hell on them until you've reduced your real experimental error. Now each time a new ale or stout comes along, you can test your prior beliefs about ales and stouts directly against the new ("Fisherian" if you will) evidence.
In observational sciences such as most of economics the problem of belief formation is the same.
So today's scientist neither tests nor estimates: he testimates. Testimation is the unhappy marriage of the sizeless stare of statistical significance to the fallacy of the transposed conditional, and it costs us jobs, justice, ecology, and even lives.
I'm sorry for the long post. Quickly then I'll respond to Alex's final question, concerning what types of arguments will be used: 28 valid topics, 4 master tropes, 1 tetrad, 1 pentad, several metaphors, and countless apostrophes!
Thanks for the question and a great blog. Best to you all.
Steve
Posted by: Steve Ziliak at Dec 14, 2007 10:41:46 AM
At 74, my experience is deeper than I like. My judgement is, in relative terms, not too bad. I am addicted to rational thinking.
More than 60 years ago, I used to be asked in an American classroom to stand up and make a plaedge including the words "I believe ...". It started me asking what that word believe meant. I still have not pinned that meaning down.
That said, I saw the idea of believing being sucessfully propagated in the classrooms of the USA. However, the only "beliefs" that practically all Americans share are in "life, liberty , property, and the pursuit of happiness." One joyous thing about the USA is that the rest of the beliefs of individual citizens are unpredictable. It is the only country where it would be unsurprising to find a creationist astronomer (a species that the Head of the Papal Observatory says does not exist).
So I would choose American or French. This is because a great many of the French would deny that they hold "creances"; though by Heaven they have biases!
Posted by: Diversity at Dec 14, 2007 10:46:00 AM
The comparison--nationality versus academic training--is supposed to highlight the irrational sources of belief. What it really highlights is that different cultures value different things, and therefore have different beliefs about what policies are better because they weight the trade-offs in a different way. This is only irrational if you believe that economics has no normative component to it, or is uninformed by human values. The diversity of opinions these very issues--taxes, welfare, regulation--among smart economists should be enough to disabuse that notion.
Posted by: Ken at Dec 14, 2007 10:59:18 AM
Steve,
Any economist worth anything should concede that you and Deirdre McCloskey have an important point to make. But advertising your new book on someone else's blog under a post only tangentially related to the topic of the book turns people off and makes you seem even more shrill and less scientific.
Posted by: Ben at Dec 14, 2007 11:00:30 AM
Alex, can you give some examples of things you believed and then did a mental double-check saying to yourself "Do I believe this for good reasons or just because of my nationality/profession?"?
Posted by: TGGP at Dec 14, 2007 11:24:51 AM
only one? eye color.
Posted by: Jason at Dec 14, 2007 11:27:19 AM
only one? eye color.
Posted by: Jason at Dec 14, 2007 11:27:33 AM
The first line cracked me up. Perhaps belief in the rationality of one's beliefs is itself an overprotected belief.
Posted by: sally at Dec 14, 2007 12:14:19 PM
If nationality was that important, there wouldn't be any national politics. I mean one could predict what kind of arguments one would make because that's how debates are framed but not necessarily one's conclusions.
and there are other factors.. Age could be relevant as generations often have a different worldview. And class of origin may be more important than the fact that 2 people have the same profession (and i'm not using class only in the upper/middle/lower sense). And education is, in a lot of different ways. For instance I doubt an US-educated haitian economist and a France-educated haitian economist would articulate their arguments the same way.
and then there's experience. Tyler's travels partially explains some of the minor differences of POV he has with other libertarians, doesnt it ?
Posted by: nu at Dec 14, 2007 12:34:30 PM
Robin; beliefs should influence your choice of profession, and sometimes even choice of nationality. Influence just should not flow in the opposite direction.
Steve has the right to be shrill on this. We're good rationalists here. There are few pieces of info that are critically important to understanding the world, and few fora frequented by many good rationalists. Let all the former echo through all the later. That said, Overcoming Bias might be an even more appropriate place to post.
Posted by: michael vassar at Dec 14, 2007 12:36:55 PM
"From an Enlightenment or Positivist point of view, which is Hume's point of view, and mine, there is simply no avoiding the conclusion that the human race is mad. There are scarcely any human beings who do not have some lunatic beliefs or other to which they attach great importance. People are mostly sane enough, of course, in the affairs of common life: the getting of food, shelter, and so on. But the moment they attempt any depth or generality of thought, they go mad almost infallibly. The vast majority adopt the local religious madness, as naturally as they adopt the local dress. But the more powerful minds will, equally infallibly, fall into the worship of some intelligent and dangerous lunatic, such as Plato, or Augustine, or Comte, or Hegel, or Marx."
David Stove, The Plato Cult, 1991
"Smart people believe wierd things because they are skilled at defending beliefs they arrived at for nonsmart reasons."
Michael Shermer, Scientific American: Smart People Believe Weird Things Sept. 2002
"One cannot overstate the childishness of the ideas that feed and stir the masses. Real ideas must as a rule be simplified to the level of a child's understanding if they are to arouse the masses to historic actions. A childish illusion, fixed in the minds of all children born in a certain decade and hammered home for four years, can easily reappear as a deadly serious political ideology twenty years later."
Sebastian Haffner, "Defying Hitler" pg. 17
The much more interesting question is how somebody's beliefs could change.
Posted by: Mike Huben at Dec 14, 2007 1:19:20 PM
Fun posting and comments.
My 2 cents: Aren't there many different kinds of belief? For example: "I believe this pencil will drop to the floor if it rolls over the edge of this desk" vs. "I belive in reincarnation" vs. "I believe my boss will have a fit if I don't tell her what she wants to hear." So if you want to talk about where beliefs come from, wouldn't it be helpful to separate out the various kinds of belief we might be talking about?
Hey, another two cents. You write: "We all like to think that our beliefs come from rational thinking, deep experience and good judgment." Am I really the only exception to this among MR readers? I mean, yes, I do want to believe that my belief that this pencil will fall if it rolls over the edge of the desk is a sensible one. But where Bigger Thoughts and Bigger Convictions go? God only knows where they come from, and that's OK with me.
But maybe this is a diff between economists and civilians?
Posted by: MIchael Blowhard at Dec 14, 2007 1:39:34 PM
