Arnold Kling  

Popular Opinion

PRINT
If Medicare were a Country...... Ten Ideas Worth Thinking About...

Bryan writes,


If virtually everyone just argues for whatever position he was born into, a truth-seeker should hold the gladiators for popular views to higher standards.

I have a slightly different way to arrive at this position. Consider the following matrix.
Ideas that arePopularUnpopular
Empirically ValidIII
Empirically InvalidIIIIV

Ideas that are in quadrant I are popular and empirically valid. Think of the law of gravity. Those ideas have the best survival characteristics.

Ideas that are in quadrant II are unpopular but empirically valid. I would put the proposition that "immigration hurts the economy very little, if at all" in that quadrant. These ideas are always a threat to take over for popular ideas, should people adopt an empiricist approach to the issue.

Quadrant III is the mirror image of quadrant II. Creationism is popular, but not empirically valid. These sorts of ideas are threatened by empiricism.

Quadrant IV are ideas that are neither popular nor empirically valid. They threaten to become extinct once their few adherents change their minds or die. Freudian psychology was in quadrant III 50 years ago, and is in quadrant IV today.

If we look at ideas that are alive at any one point in time, they are alive either because they are popular, because they are empirically valid, or both. So if an idea is very popular, you should hold it to a high standard of empirical validity. If it is somewhat less popular, then there is a good chance that the only reason it survived is that it has some empirical validity.

The old joke says that as a lawyer, when you have neither the facts with you nor the law with you, you pound the table. There are table-pounding methods of argument in ordinary discussions, and when you observe people using them, you should assume that the empirical evidence runs the other way.


Comments and Sharing


CATEGORIES: Economic Methods



COMMENTS (13 to date)
Kyle writes:

Econlog has become my number one site for meta-arguments. Y'all are competitive with Robin Hanson on that topic, and that's saying something. Love the stuff

Lord writes:

"immigration hurts the economy very little, if at all"

Thank goodness. Now we can just ignore it and all proposed "solutions" for it, and stop talking about it.

Nathan Smith writes:

If by "creationism" you mean seven-day creationism, fine. If by creationism you mean the general idea that God created the world, this is really outside what empirics can tell you, one way or the other.

But let me dissent from the implied corollary: that evolution is empirically valid. Evolution is an infinitely flexible theory, which cannot be proven wrong. In Karl Popper's terms, evolution is not falsifiable.

Example: Suppose you find a new fossil. The theory of evolution would seem to predict that one would find other fossils, later in the fossil record, which are midway between the fossilized specimen and some modern specimen. But if later fossils are exactly the same, and no "link" to a modern specimen can be found, fine: it's a case of "punctuated equilibrium," when the species evolved rapidly due to an environmental shock, making it improbable that the fossil record would capture the change. Or, if no similar modern specimen exists, fine again: the species went extinct. No matter what the fossil record says, it can be easily reconciled with the evolutionist account.

No doubt the atheist-libertarian readers of this blog are now sneering with contempt. "How did this ignoramus, brainwashed in some fundamentalist Sunday school, find this blog, and why do we have to waste our time listening to his stupid sophistries? These arguments have (can't remember where at the moment) been refuted hundreds of times (no doubt)! We know on The Authority of Science that evolution is true!"

Keep pounding those tables.

caveatBettor writes:

I believe the acceptance of creationism requires faith. What does empiricism have to do with faith? About as much as data has to do with hypothesis.

But let's clean such data like the geologist claiming the age of the rock because the anthropologist said the fossil is so old, and the anthropologist claiming the age of the fossilized lifeform because the geologist said the rock is so old.

Darwin's theories of sexual selection are presently being discredited by more and more scientists. But until recently, the same theories were as accepted as much as natural selection.

Science does turn out to be a double-edged sword. Give the scientists some more time, I say. They may climb the mount of truth, only to find some simpleton creationists like Nathan already camped at the summit.

James writes:

For the most part I agree with the idea of this post, but I'm thrown off by this line:

"If we look at ideas that are alive at any one point in time, they are alive either because they are popular, because they are empirically valid, or both."

What category do the ideas in a textbook on complex analysis fit into?

Arnold Kling writes:

The ideas in a math book would be tautologies. That is, they are true by definition. To an empiricist, those are fine. It is when you make a statement that sounds like it is empirical, such as "the Oedipus complex causes neurosis," that you had better treat your statement as falsifiable rather than as a tautology.

Half Sigma writes:

You have immigration in the wrong quadrant.

We have massive amounts of immigration, and we are a democracy, so therefore it must be popular or politicans would check it to retain their office.

There are valid arguments that immigration does harm the economy, which depends on how you define "harm." In any event you are asking the wrong question, which should be "does immigration harm the United States," because the U.S. is more than just its economy.

Nathan Smith writes:

re: "I believe the acceptance of creationism requires faith. What does empiricism have to do with faith?"

The idea within this rhetorical question is a standard view, but, I think, wrong. Empiricism actually pre-supposes faith. Hume proved long ago that inductive reasoning is not itself justified in terms of the principle of sufficient reason. We have no "reason" to believe there are patterns in the world, which is the meta-belief that must underlie induction. Another way of describing this meta-belief is... faith.

James writes:

Dr. Kling! I was only taking an easy shot at your either/or claim, not disagreeing with you main point.

"The ideas in a math book would be tautologies. That is, they are true by definition. To an empiricist, those are fine."

Here's a testable prediction: Old school empiricists believed that sensory experience was the only way to know anything. The Vienna crowd granted that one could be an empiricist and still believe that tautologies were true by definition. Modern empiricists have come to accept introspection as a way of knowing things. Extrapolate the trend and within my lifetime, the people calling themselves empiricists will believe that some tautologies can be the basis for true conclusions about the world we live in. They'll still insist that their methods are superior to the methods of those silly rationalists.

What if there are two groups of table pounders shouting at each other?

Lauren writes:

Hi, Arnold, James.

Arnold wrote, in response to James's asking about how he would categorize mathematical ideas, which are also alive and kicking:

The ideas in a math book would be tautologies. That is, they are true by definition.

Just to clarify some terms: I think Arnold must have meant axiomatic, not tautological. The ideas in a math book are based on carrying forward the implications of a series of axioms. Neither the axioms--including any definitions of terms--nor their logical exensions are intended to be subjected to empirical analysis. A tautology, on the other hand, is a vacuous statement that is logically true only because of its self-limiting sleight of hand. See Dictionary.com for an excellent example of a tautology in logic. I think Arnold misused the term.

More substantively, I think James appropriately suggests a qualification to Arnold's rather blanket, though pithy, claim:

If we look at ideas that are alive at any one point in time, they are alive either because they are popular, because they are empirically valid, or both.

Perhaps that should read "ideas that are testably alive"?

RogerM writes:

Anyone who thinks the theory of evolution has empirical evidence for it should read Roger Lewin's book "Bones of Contention." Before you dismiss it, keep in mind that Lewin is a devout evolutionist and former editor of a major scientific publication.

Tom Anger writes:

I would put the proposition that "immigration hurts the economy very little, if at all" in that quadrant. I agree with that assessment. Now, off the subject of this post, I find it surprising that many economists who oppose the minimum wage, even though it hurts the economy very little (by some accounts), nevertheless support unfettered immigration. Inconsistent, no?

Comments for this entry have been closed
Return to top