« Ways that sheep can die | Main | The Return of the Zombie »
Are there reasons to be dogmatic?
That's another request; the exact wording was "The five best reasons to think you should be more dogmatic about the economic beliefs you are not dogmatic about."
I'll give one reason, namely that, somewhat counterintuitively, dogmatism can further the generation of new ideas. Yes, this does require a special meaning of the word dogmatic. I'm not talking about a grouchy guy who goes "harumph" whenever he counters a new idea. I'm talking about the person who generates the new idea! In strict Bayesian terms, most innovators are not justified in thinking that their new ideas are in fact correct. Most new ideas are wrong and the creator's "gut feeling" that he is "onto something" is sometimes as epistemologically dubious as is the opinion of the previous scientific consensus. Yet we still want that they promote these new ideas, even if most of them turn out to be wrong.
In this view, the so-called "reasonable" people are selfishly building up their personal reputations at the expense of scientific progress. They are too reasonable to generate new ideas.
To put it another way, there are two kinds of truth-seeking behavior:
1. Hold and promote the view which leads to society most likely settling upon truth in the future, or
2. Hold and promote the view which is most likely to be correct.
These two strategies coincide less than many people think. Which do you prefer and why?
Addendum: Here is a recent NYT article. Excerpt:
Voters who insist that they are undecided about a contentious issue are sometimes fooling themselves, having already made a choice at a subconscious level, a new study suggests. Scientists have long known that subtle biases can skew evaluations of an issue or candidate in ways people are not aware of. But the new study, appearing Thursday in the journal Science, suggests that professed neutrality — sitting on the fence — leaves people more vulnerable to their own inherent biases than choosing sides early.
Will Bryan Caplan endorse this?
Posted by Tyler Cowen on August 25, 2008 at 05:23 AM in Economics | Permalink
Comments
While I largely agree with the ideas here, I find it rather unclear how one should move from accepting them to actually acting. Pretending to be dogmatic is hardly a substitute for real dogmaticists.
In your first case, it suggests that people should put their full weight behind ideas they do not fully believe in, to bet on the chance that it is a good idea anyway. Even if someone can pull this off, the real beleivers are likely to be much better promotors.
The added politics example is even harder. Even if neutral people probably do have much stronger biases then they think they have, turning them into uncritical partisans does not seem like an improvement to me.
Posted by: Zamfir at Aug 25, 2008 6:07:53 AM
I profess neutrality!
Posted by: hanmeng at Aug 25, 2008 6:16:04 AM
It seems that you're talking about two separate topics here: 1) Should those who develop new ideas profess and defend them even as they themselves remain unconvinced of the ideas' legitimacy? 2) Would our discussions of contentious issues be more fruitful if all observers were made to be participants, adopting or creating a position in the argument?
I think that the answer to both questions depends upon the confidence of the individual. An individual who is not confident in his or her ideas will not present the best arguments in their support, and weak arguments are good for very little. This want of confidence probably arises from a person's lack of familiarity with the subject at hand or his or her belief that there is insufficient evidence to indicate a clear answer. Those who remain neutral because they consider themselves to be poorly informed do a disservice to discussions in which they might otherwise participate. Those who profess a nuanced neutrality based on available evidence can actually provide a foundation for more further discussion. Neutrality born of ignorance is undesirable; neutrality born of knowledgeable skepticism still possesses value.
Posted by: Amy at Aug 25, 2008 8:01:33 AM
I believe there is absolutely no reason to be dogmatic, and nothing you say will change my mind.
Posted by: at Aug 25, 2008 9:06:35 AM
I think "dogmatic" just isn't the right word for what you're getting at. One is dogmatic when ignoring all evidence, they stick to an idea already established within a group. Its connotations are toward the religious and doctrinaire, exactlty the opposite of what you seem to have mind.
Usage example: Milton Friedman was innovative and then became dogmatic. And then an entire school of dogmatic followers followed.
Posted by: Jason Ping at Aug 25, 2008 9:09:08 AM
In an Easter egg hunt, hunters do not need different beliefs about where the eggs lie to spread out over the lawn. What evidence shows idea hunters need different beliefs to best explore different ideas?
Posted by: Robin Hanson at Aug 25, 2008 9:18:29 AM
Amy said: "Those who remain neutral because they consider themselves to be poorly informed do a disservice to discussions in which they might otherwise participate."
What do you mean with this? There must be myriad of debates I do not participate in because I lack knowledge, and I would say that it's only good for those debates. or you trying to make a more specific point?
Posted by: Zamfir at Aug 25, 2008 9:19:51 AM
@Robin Hanson:
I like the Easter Egg comparison, but its weakness would be that it assumes that ideas can be reasonably easily checked, after which one can try another idea, and again and again.
But for example in many technology fields, R&D and market acceptance of a new idea will take years of dedicated work before it is clear whether the idea had value. In such cases, there might be something to say for an irrationally strong belief in the idea, and I guess this is not uncommon in reality. I suppose that the same applies to some academic ideas, if working out the details and pushing for acceptance will take a not insignificant part of your life.
A modification would be Easter Egg hunting when the eggs are buried 3 feet deep...
Posted by: Zamfir at Aug 25, 2008 9:42:51 AM
1. Is better for the economy, because it gives the other scientists something to study and correct.
2. Is the result of a learned behavior. People spend the first 20 some odd years of their lives being corrected, especially throughout academia. In college, the teacher does not hand back tests and write, "you have infinite time to show your view is correct". Instead teachers tell you to read the book and pay attention in class because the only correct answer is the one they are teaching you. After spending this time under the iron fist, the layman does not believe in free thinking, (although African Americans invented rap to show that they are capable of it). So the above average person who isn't scared to theorize tries to make sure their idea has enough validity to, at minimum, keep their current quality of life. (where a poor idea may not just hurt their reputation, if they are entrepreneurs, could bankrupt them)
I think the bigger problem is people outside of a certain field have no way of passing on, or profiting from, any theory they may think of, and that if all ideas were passed on, there would be too many to ever test.
So if the average 2 accounts for about the top 1% of people in the country, and the average 1 accounts for about the next 49% of people in the country who still only put forth maybe only 1 or 2 of these types of ideas in a lifetime, then I would say democracy works.
Posted by: brainwarped at Aug 25, 2008 10:13:49 AM
@brainwarped = You are on the right track. The NY Times article is incorrect because the research it describes is fundamentally flawed. The first sentence of that article says this:
Voters who insist that they are undecided about a contentious issue are sometimes fooling themselves, having already made a choice at a subconscious level, a new study suggests.
This is absurd nonsense. Undecided voters are not fooling themselves. They are fooling YOU. They don't want YOU to know what they are thinking. On the contrary, they are protecting themselves.
They are being underhanded and passive, so that you do not expend energy disproving them. The more dogmatic they are, the more dogmatic you are going to be in attempting to disabuse them of their position. The more they profess a belief, the more they can expect you to fight them. As far as they are concerned, you can just shut the hell up. (As it were).
So the more "undecided" they seem, the more protection they give themselves against the likes of people like the teachers described by brainwarped above. They are trying to soften the impact of potentially holding a view that will lead them into a debate by keeping it to themselves.
Now I'm going to soften my own stance. It is, in fact, possible that people are fooling themselves, and they really do believe themselves to be "undecided" when they really aren't, but I'm going back to brainwarped's "iron fist" theory and my own thoughts on the subject, and I'm going to insist that they are only fooling themselves because it's the most efficient way to protect themselves from abusive debate.
Posted by: J.V. at Aug 25, 2008 10:44:45 AM
Instead of asking innovators to act against their own self-interest (i.e. to risk their reputations more than is healthy for their own futures), academia and other cultures that profess to value innovation, or who should value innovation, should reward originality somewhat more and correctness somewhat less, so that incentives for innovation are aligned, i.e. so that innovators are not overly exposed to reputation risk.
Another reason to promote new ideas is the law of the dominant paradigm: given opposing opinions with equal evidence in their favor, the less popular opinion is more likely to be true.
Posted by: nick at Aug 25, 2008 5:37:15 PM
Rationality usually requires you to conform your beliefs to the best available evidence. But, depending on your circumstances and abilities, it may also require you to improve the evidence available. A scientist may be required to work on his research programme A even if general members of the scientific community are required to believe the results of research programme B; and if this scientist cannot work on A while believing in B, she should dogmatically believe in A rather than reasonably believe in B if the requirement to improve the evidence is stronger than the requirement to conform to it.
Posted by: Pablo Stafforini at Aug 29, 2008 4:31:39 PM
