Sunday, March 20, 2011

Assault of Thoughts - 3/I suppose it's technically the 20th/2011

"Words ought to be a little wild, for they are the assault of thoughts on the unthinking" - JMK

- Bravo. Very good to see this sort of thing. A bipartisan supermajority of Senators calling on the president to tackle tax and entitlement reform. One of the nice things about this sort of initiative is that even if they are overly conservative in their approach, a tax and entitlement reform is not going to pose the sort of threat to those struggling through a depression that, say, a cut to the non-military discretionary budget (i.e. - the portion of the budget that goes to working families) would pose. Even if they cut deeper than necessary, it's the sort of thing that unfolds over many years. And it's the sort of thing that really needs to be done. The fact that right now we need larger deficits does not change the fact that we face a very real long-term debt problem.

- This does not seem like a wise way to fight an increase in liquidity preference.

- This is perhaps the best post I've seen so far on the most recent back-and-forth over whether economics is a science or not. If it's not, fine - just count biology, geology, meteorology (and Sumner argues even physics) out too. It seems most meaningful to me to say that economics is a science because we chase useful knowleddge the same way all the other sciences chase useful knowledge. If you want to provide another definition for it, then be my guest. Go right ahead. In the mean time, I have some science to go do.

- Matt Yglesias tackles the constitutional justification for federal provision of infrastructure. He makes things more complicated than they need to be. He works his way through the commerce clause, dismisses it, and then brings up an unnecessarily obscure clause: "I would have said that the power “[t]o establish Post Offices and Post Roads” implies a generic grant of authority to do transportation." You don't need to read any farther than the first clause of Article 1 Section 8 - the Congress can appropriate money to provide for the General Welfare. The constitutional quibbles of Jefferson himself on this point amounted to no more than that he felt the need to obtain the permission of the states the roads and canals were passing through first. Nobody ever questioned this application of this clause until decades after the Constitution was ratified. To some 21st century eyes that may seem like it's close enough to be "originalism", but to me if someone takes decades to register a complaint I'm not buying that this is a legitimate originalist interpretation of the Constitution. If I were around back then I'd be in Gallatin's corner.


  1. Perhaps you'll be interested in my recent blog about this issue:

    I think it is far more meaningful to address some of the philosophical underpinnings of the topic, as well as the forces which divert economics (and hence its credibility) if we are concerned with its character as a science or not.

  2. Nobody? I be to differ. Much of the Election of 1800 centered directly upon considerations like publicly financed infrastructure, particularly due to the issue of debt - Jefferson's victory was a triumph against a broad reading of the General Welfare clause (amongst other broad readings of various clauses of the Constitution).

    Anyway, if you read "The Federalist Papers" you will see a very narrow reading of the clause by Madison; in particular he states that the clause can only be used with some other specific enumerated power.

    As for Gallatin, he aided in the creation of a capitol designed for empire - and now we have that empire.

  3. The moniker of "science" is important -- it bestows respect, prestige, and power. Certain qualities and abilities are expected of a "scientist", some good and proper but many quite fanciful. Many groups wish to enjoy the benefits of these connotations, but it is also necessary to deprive their intellectual opponents. Thus, many words are written to define "science" in such a way that it captures all one agrees with and leaves aside all one does not. This exercise is normally disguised as an effort to discover what science "really is", (whatever that means), but its just so much political maneuvering with a pretense of intellectual substance.

    Popper was the only one who ever got this right. He knew that his criterion of falsifiability was a proposal to adopt a convention -- a definition. It was intended to help elucidate the important critical distinction that some hypotheses can be experimentally tested and others cannot; those that cannot are not necessarily useless or false, but just do not qualify according to the proposed definition.

  4. People like Russ Roberts are no different. He wants to lower peoples' expectations of economists, but so long as economists call themselves "scientists" there is a problem -- our culture assumes high expectations of scientists. Moreover, many economists enjoy the respect, prestige, and power that comes from being a "scientist", and they are unwilling to give that up regardless of whether they deserve it. Roberts just finds himself in a peculiar position -- his little subculture rewards humbler-than-thou behaviour. By denying himself the moniker of "science", he is depriving his opponents of its benefits while also enjoying the praise of his immediate peers for being so humble.

    Okay, maybe I am getting a wee bit too cynical here.

  5. Lee is back!!!!

    I do not think you are being too cynical... but I think Russ is guilty of error as much as he is guilty of strategic rhetoric. His complaint is that economists can't live up to peoples' expectations. This is almost certainly true. But that seems to me to be a problem with peoples' expectations rather than any reason to question the nature of the discipline!

    Economists largely could not predict the crisis. But who said they should be able? Have economists ever claimed to be able to predict these things? No - not to my knowledge. They have made claims about "smoothing the business cycle", and they had some reason to make those claims, even as we look at it in retrospect. But nobody ever claimed that economics could do what the public expects it to.

    In the wake of the earthquake in Japan we don't have people questioning the scientific status of seismologists, after all. Russ indeed lives in a sub-culture where that sort of talk passes as analysis. It even gets him several gigs testifying on the Hill - he's a favorite up there. His line on this plays very well to Senators and Representatives. But aside from all of those reasons, his argument simply doesn't make sense on a fundamental level.

    The problem is people don't have scientific expectations of economics - they expect it to be like astrology. We shouldn't downgrade economics itself for not living up to peoples' semi-mystical expectations. We shouldn't downgrade science because it's not astrology.

  6. Not all sciences are falsifiable (notably evolution - however much its "defenders" would like to claim otherwise). This is a false dichotomy.

    The non-objectivity of certain people and tendencies is a big issue that needs to be addressed. Notably, some Austrian economics erroneously calls themselves "value free." In fact, even the most mundane sciences have value-systems congealed in them - the very activity of inquiry is driven by a desire, a value (even if that value is honest inquiry).

    Popper assumed that deductive logic was "fool-proof" in a sense, along with Mises. It's a very shaky foundation for a study of nature.

    Its telling that Marx was considered an objective economist, while Mises and some other proudly considered themselves "subjectivists" and based their ideas on "deductive logic" almost exclusively. I don't think it makes sense to lose touch with either side of this issue if you are going to be honest about economics.

  7. Evolution is falsifiable.

  8. Gary, be careful not to respond to bad arguments for "intelligent design" with bad arguments for evolution.

    To take their examples of falsifying evidence in turn:

    •a static fossil record;

    No. A static fossil record does not preclude (1.) the destruction of fossils indicating evolution, or (2.) the reality of biological evolution but an alternative source of life on this planet. For example, we wouldn't say that evolution is wrong if aliens evolved by natural selection and then created the life that appears here (which would give us a static fossil record).

    •true chimeras, that is, organisms that combined parts from several different and diverse lineages (such as mermaids and centaurs) and which are not explained by lateral gene transfer, which transfers relatively small amounts of DNA between lineages, or symbiosis, where two whole organisms come together;

    Again, no. Evolution does not require that all changes happen through evolution by natural selection. This would falsify evolution about as much as a purple duck would falsify "all swans are white".

    •a mechanism that would prevent mutations from accumulating;

    This one is a little vague... I'm not quite sure how to respond to this.

    •observations of organisms being created.

    Again, this would falsify evolution about as much as a purple duck would falsify "all swans are white". I imagine we're not that far off from having observations of organisms being created. That wouldn't change their ability to evolve after their creation or our own evolutionary history.

  9. I am with Dean - falsification is something of a red herring. It's good as far as it goes. Nice epistemology (although even here not entirely satisfying). But it's not especially useful in demarcating science.

  10. The myth and somewhat lax enforcement of an ideal of falsification, though, is surely a good thing to trick ourselves into buying into. If we can get theories to look kinda-sorta falsifiable and do empirical tests that have an air of falsification about them we're probably doing pretty good science. Popper makes for a pretty good mythology of pragmatic science - either experimental or observational.

  11. Daniel,

    Evolution is falsifiable.

    There is nothing vague about it; if there were a mechanism which halted mutations before they started then evolution would be falsifiable.

    Be careful not to give ammunition to anti-evolutionists.

    I think the problem is that you don't know what falsifiable means.

  12. Falsifiability, as I understand it, is when a theory is incompatible with a possible empirical observation.

    The first version of that statement was vague. If it meant "a mechanism exists that halts mutations", then that is not incompatible with evolution. Evolution works fine if such a mechanism exists, so long as it doesn't apply to all life. If it meant "a mechanism exists and applies to all life that halts mutations", then it's true evolution would be incompatible with such a mechanism, but this does not seem to meet the test of being a possible empirical observation.

    By your use of the term "falsifiable", it seems like any claim that is conditional on something would be falsifiable. All you have to do is remove that condition in all cases! Bingo - it's falsifiable!

  13. Falsifiability talks about the "might be" and empirical observation only a thought experiment where the empirical is thought is sometimes required. It is not testability, though a lot of people think it is synonymous.

  14. Falsifiability refers to those theories which can be disproven by a test. Theories which rely on bodies of data are positivist (popper was harshly critical of Positivism), and therefore non-falsifiable. Evolution relies on inductive logic, and so is non-falsifiable.

    Popper is valuable in terms of understanding the limitations of some sciences. But his narrow focus on a-priorism is a problem not only for other sciences, but for his own "social theory":

    This is also linked to in my blog, linked above.

    I don't think observational science fits the falsifiability criterion.

  15. It is not testability, though a lot of people think it is synonymous.

    Well that definition comes from the Stanford Philosophy Encyclopedia. That's not strictly and completely authoritative, but you'll have to quote Popper himself if you want to convince me of this.

    If it's only that something could disprove it then any conditional statement is falsifiable and therefore scientific. Any conditional statement at all. All you have to say is that if the condition doesn't hold it's not true. There are many commenters here who know Popper far better than I do and can correct me, but I really don't think Popper was saying "conditional statements are falsifiable statements and are therefore scientific statements". Normally Popper is accused of making the whole discussion too narrow - this blows the discussion wide open.

  16. This is another interesting paper on the topic, which goes into many of the same issues, except exclusively focusing on Austrian thinkers:

  17. Popper makes rather clear that merely because something is not falsifiable does not mean it lacks truth; it is merely that it is not "scientific." Too many people think he argues that it is casts out any knowledge which isn't falsifiable.

    Anyway, yes, Popper talks about knowledge which is falsifiable in theory, though not in practice.

    As for narrowness, I think that is in large part because a lot of people are confused about what Popper wrote.

  18. Gary: Popper's notion of the scientificity of something is precisely what we were (or at least I was) talking about. If you agree with me on this point - as you seem to indicate here - I don't think we are at odds.

  19. *smacks own head on table*

    Ah, that's better.

  20. clarify the head smack... you've been very cryptic lately


All anonymous comments will be deleted. Consistent pseudonyms are fine.