Dilemmism vs n-lemmism
This is not a post about Motorhead.
I had an undeveloped thought while watching the Rodrik-Tabarrok death match; maybe someone can work it up into something useful.
There are two kinds of people in the world, those who subscribe to Dilemma and those who subscribe to n-lemma. [1]
The Dilemmists include people like Bill O'Reilly, George W. Bush, George Lakoff, and Dani Rodrik. Their chief characteristic is to think in one-dimensional terms (respectively): right-wrong, terrorist-'merican, strict-nurturant, and -- in Rodrik's case alone -- we have 1st-2nd, pro-anti government, pro-anti industrial policy, and simple-sophisticated.
The n-lemmists include Tyler Cowen and David Friedman. Their chief characteristic is that they tend to look for multiple explanations to puzzles. Look at Tyler's lists, at Friedman's chapter in Machinery of Freedom on libertarian arguments he doesn't agree with and in Law's Order at the back & forth arguments over various legal principles.
For any given problem, there may be m true explanations. m may be as few as one [2] but may be greater than that. The dilemmist stops looking at two; that's not exactly true, they stop looking at one and then determine its opposite analytically. The n-lemmist tries to list as many as possible, even considering contradictory explanations. Neither the Di- nor the n-lemmist is going to get the m=1 case correct. The Dilemmist is also never going to find a correct explanation when m>2. When m<=2, the n-lemmist is wasting time and probably putting too much emphasis on finding new explanations after identifying the correct ones [3]. Thus, neither of these groups is going to be successful all of the time at explaining observed facts. Rodrik, for example, has determined that the explanation for why some people agree with him is that they are sophisticated, second best economists. Therefore, everyone who disagrees must be the opposite: an unsophisticated, first best economist. It does not seem to occur to him to look for other explanations [4]; no need to be too sophisticated, I suppose. Tyler on the other hand tends to go well beyond the call of duty in opposing himself (though Tyrone might disagree).
This whole system seems rife with irony. The strongest Dilemmists believe themselves to be more sophisticated despite their adherence to a belief in simple models. The n-lemmists seem to be committed to an approach that generates many explanations, which seems wiser and more rational in the face of uncertainty, but in all probability m << n so they should be looking at how to reduce the number of conjectures required to get to the correct answer(s). In other words, the wiser people are wasting lots of time generating quantity rather than quality. And of course I have just built an entire system around a simple dualist explanation while implying that dualist explanations are probably the less satisfactory approach.
---------------------------------
[1] I realize my use of "lemma" is a little unconventional here, but it's a lot catchier and less cumbersome than "n-explanation" and "di-explanation".
[2] Although social phenomena seem to have multiple plausible explanations, do physical phenomena have only one true explanation where the science is "settled"? Or perhaps regardless of whether the science is settled? The duality of light suggests that either this conjecture is not true or that science needs to resolve the issue in favor of a single explanation.
[3] I think the n-lemmist approach is more likely to be successful because we don't know a priori which are the correct ones, or even how many there are (m).
[4] Or perhaps he has considered them and has already discounted them and edited his comments. Did I mention that I am an n-lemmist, too.
I had an undeveloped thought while watching the Rodrik-Tabarrok death match; maybe someone can work it up into something useful.
There are two kinds of people in the world, those who subscribe to Dilemma and those who subscribe to n-lemma. [1]
The Dilemmists include people like Bill O'Reilly, George W. Bush, George Lakoff, and Dani Rodrik. Their chief characteristic is to think in one-dimensional terms (respectively): right-wrong, terrorist-'merican, strict-nurturant, and -- in Rodrik's case alone -- we have 1st-2nd, pro-anti government, pro-anti industrial policy, and simple-sophisticated.
The n-lemmists include Tyler Cowen and David Friedman. Their chief characteristic is that they tend to look for multiple explanations to puzzles. Look at Tyler's lists, at Friedman's chapter in Machinery of Freedom on libertarian arguments he doesn't agree with and in Law's Order at the back & forth arguments over various legal principles.
For any given problem, there may be m true explanations. m may be as few as one [2] but may be greater than that. The dilemmist stops looking at two; that's not exactly true, they stop looking at one and then determine its opposite analytically. The n-lemmist tries to list as many as possible, even considering contradictory explanations. Neither the Di- nor the n-lemmist is going to get the m=1 case correct. The Dilemmist is also never going to find a correct explanation when m>2. When m<=2, the n-lemmist is wasting time and probably putting too much emphasis on finding new explanations after identifying the correct ones [3]. Thus, neither of these groups is going to be successful all of the time at explaining observed facts. Rodrik, for example, has determined that the explanation for why some people agree with him is that they are sophisticated, second best economists. Therefore, everyone who disagrees must be the opposite: an unsophisticated, first best economist. It does not seem to occur to him to look for other explanations [4]; no need to be too sophisticated, I suppose. Tyler on the other hand tends to go well beyond the call of duty in opposing himself (though Tyrone might disagree).
This whole system seems rife with irony. The strongest Dilemmists believe themselves to be more sophisticated despite their adherence to a belief in simple models. The n-lemmists seem to be committed to an approach that generates many explanations, which seems wiser and more rational in the face of uncertainty, but in all probability m << n so they should be looking at how to reduce the number of conjectures required to get to the correct answer(s). In other words, the wiser people are wasting lots of time generating quantity rather than quality. And of course I have just built an entire system around a simple dualist explanation while implying that dualist explanations are probably the less satisfactory approach.
---------------------------------
[1] I realize my use of "lemma" is a little unconventional here, but it's a lot catchier and less cumbersome than "n-explanation" and "di-explanation".
[2] Although social phenomena seem to have multiple plausible explanations, do physical phenomena have only one true explanation where the science is "settled"? Or perhaps regardless of whether the science is settled? The duality of light suggests that either this conjecture is not true or that science needs to resolve the issue in favor of a single explanation.
[3] I think the n-lemmist approach is more likely to be successful because we don't know a priori which are the correct ones, or even how many there are (m).
[4] Or perhaps he has considered them and has already discounted them and edited his comments. Did I mention that I am an n-lemmist, too.
Labels: doggerel, philosophy
<< Home