Hard and Soft Obscurantism
23 January 2013 at 5:38 am Lasse 12 comments
| Lasse Lien |
I recently attended a presentation by the great social scientist Jon Elster, in which he lamented the state of affairs in social science. Elster has – quite nicely, IMHO – coined the terms hard and soft obscurantism as the main problems. To Elster, obscurantism generally refers to endeavors that are unlikely to produce anything of value, and where this can be predicted in advance. This in contrast to more honorable failures, where a plausible hypothesis turns out to be wrong, leaving much effort without much value.
Soft obscurantism is exactly what it sounds like. Unfalsifiable, impenetrable theories which often proudly ignores standards for argument and evidence that elsewhere constitute the hallmark of the scientific method. Examples are post modernism (Latour), structuralism (Lévi-Strauss), Functionalism (Bourdieu, Foucault), Marxism (Badiou) and psychoanalysis.
But there is a ditch on the other side of the road too. Hard obscurantism refers to mathematical exercises without any tangent to reality, which is useful neither as mathematics nor social science. Another form of hard obscurantism is data mining, or misuse of fancy econometrics, or a combination of the two. Both mathematical games and econometric voodoo give the appearance of “scientificness”, but Elster doesn’t hold his guns about the value created by hard obscurantism either:
“I believe that much work in economics and political science that is inspired by rational-choice theory is devoid of any explanatory, aesthetic or mathematical interest, which means that it has no value at all”
One can of course argue about the size of the size of the total problem, and relative size of each type (personally, I would bet on soft obscurantism as the bigger problem), but the key question is perhaps why obscurantism of either type isn’t gradually rooted out. According to Elster their combined “market share” in the social sciences seems to be growing.
For more, see this and this and this.
Entry filed under: - Lien -, Methods/Methodology/Theory of Science, Pomo Periscope.
Organizations and Markets 
1.
Peter Klein | 23 January 2013 at 9:53 am
I once argued this point, informally, in the context of a debate between David Laband and Robert Tollison, who argued that the “marketplace of ideas” is relatively efficient, in the conventional economic sense, and Leland Yeager, who responded that markets are not good at judging among scientific truths. (See http://mises.org/daily/5450/Is-the-Market-a-Test-of-Truth-and-Beauty for some context.) I made a kind of market failure argument, noting that the major players in the academic marketplace are government and nonprofit entities (universities, research organizations, libraries, independent scholars), so we wouldn’t expect the outcomes of their interaction to resemble a competitive market. I think I am borrowing Bryan Caplan’s terminology when I say that for a tenured professor, the cost of holding wrong beliefs is very low. So what’s the incentive to avoid obscurantism, hard or soft?
2.
Lasse | 23 January 2013 at 10:58 am
I guess equally important as a the low cost of being wrong if you have tenure, is the high cost of saying that the emperor is naked – if you want tenure.
3.
Benito Arruñada | 23 January 2013 at 12:22 pm
I agree with Peter in that the academic market for ideas is competitive, at best, in the long run. More precisely: competition between scientific paradigms is slow. But not only because of government involvement but mainly because there is substantial information asymmetry between scientists and “consumers” of science. I wonder if this argument could develop into a testable hypothesis by identifying fields with less asymmetry (biology? even management?), and confirming that they are subject to faster change than those with an allegedly greater asymmetry (economics?).
4.
Randy | 23 January 2013 at 12:33 pm
If Elster spent so much of his career applying rational actor theory (ditch 2) to Marxism (ditch 1), was he actually in a deep pothole in the middle of the road? And how many philosophers of social science would place methodological individualism (per Elster) in a category of obscurantism? Not picking a fight here, just musing…
5.
Mathieu Bédard | 24 January 2013 at 6:46 am
Reblogged this on Economics and Other Petty Things and commented:
Very interesting discussion on “obscurantism” is social sciences. I can’t tell which kind of obscurantism is more popular in France. One reason for that might be that we definitely have a strong ‘hard obscurantism’ tradition with our economic engineers, also many strands of ‘soft obscurantisms’ that are quite popular, but most of all we have a lot of soft obscurantism fueling hard obscurantism research.
6.
Lasse | 24 January 2013 at 8:14 am
Benito, roughly I think the clearer the interpretation of data or results, the more efficient the market for ideas. I think mathematics, phisics and biology have a more efficient market for ideas than economics and management, partly because it’s easier to see which ideas work and which do not.
7.
Lasse | 24 January 2013 at 8:18 am
Randy, I think we should ask Nicolai if methodological individualism is a form of obscurantism :-)
8.
Peter Klein | 24 January 2013 at 5:24 pm
Hmmm, Lasse, where does that leave literature and philosophy?
9.
stevepostrel | 24 January 2013 at 9:01 pm
1. Remember Sturgeon’s Law applies to science as much as to science fiction. It’s hard to do good work. Much of the 90% will always take forms that look like Elster’s two categories. I’m not sure it’s gotten any worse over time.
2. Incentive compatibility depends on the type of researchers and the nature of the their interests. Under the conditions described by M. Polanyi in his Republic of Science essay, where researchers are driven by the desire to make discoveries and to get recognition from other discoverers, and where funding considerations can be ignored, incentive compatibility with a maximum rate of discovery is high. Under other conditions, not so much.
3. It is very hard for outsiders to understand the criteria for what seems like an interesting problem to insiders in a given field. For example, if you look at the latest CERN briefing book on future horizons for high-energy physics research, you will discover that fundamental understanding of the proton is rather poor at this time (e.g. there is nowhere near an understanding of proton spin in terms of measured quark and gluon spins). The layman might think that these gaps and anomalies would be at the top of the research agenda. Yet most particle physicists are quite blase about these deficiencies and instead complain that the Standard Model fits all the data so well that they may not have any clues to a better, more fundamental theory (and may have trouble getting funding for new accelerators). Probably the insiders’ view should be given some deference, in the sense of believing that if the outsider understood the field he would share this view. I would provisionally extend the same presumption to psychology, sociology, etc., until they crashed into something I know enough about to call BS.
4. When I read bread-and-butter applied economic papers of the NBER type, I don’t see a whole lot of obscurantism of either kind. Limited perspectives, sure, philosophical preconceptions, sometimes, but nothing that seems irrelevant or unlikely to resolve to conclusions of at least moderate significance. So unless Elster is naming names on hard obscurantism and hitting prominent targets, it seems a bit of a bluff. As far as theory goes, would the modeling efforts covered in the new Handbook of Organizational Economics (touted on this blog) fall under Elster’s strictures? If so, who is impugned? If not, what is the relevance of the critique?
10.
Rafe Champion | 26 January 2013 at 2:53 am
Taking stevepostrel’s point, there is so much more stuff of all kinds being produced now than a generation ago it is very hard to say whether things have got better or worse.
On literature and philosophy, each of these are dominated by hard and soft forms of obscurantism. In philosophy the soft obscurantism is essentially Continental and the hard obscurantism is the analytical strand, where decades were lost due under the influence of Wittgenstein, logical empiricism, paradigm theory and Lakatos. This was the case in the philosophy and methodology of economics as Wade Hands pointed out a decade ago.
The problems with philosophy are spelled out clearly by Mulligan, Simons and Smith.
Click to access WWCP.pdf
11.
bjk | 29 January 2013 at 1:59 pm
I read Elster’s piece in Diogenes and was underwhelmed. His examples of soft obscurantism are Edward Said on Mansfield Park (and related), Foucault and dispositif, and Levi-Strauss and Jakobson on Baudelaire. About the first, he complains that Said et al. have not gotten at the “intention” of Jane Austen, and defends the honor of Fanny Price (does he realize she is a fictional character? She can’t be defamed). Appealing to the intention of Jane Austen strikes me as a case of obscurum per obscurius – couldn’t he come up with a more egregious example? I’m not going to defend Foucault, but I don’t think he realizes the Levi-Strauss/Jakobson interpretation is a bit of a high-class academic joke. Or maybe he does. But all of these examples are around three decades old – he couldn’t come up with more recent examples? As for hard obscurantism, he doesn’t know enough about LTCM to characterize it as obscurantism, nor does he show the harm (economic value was not destroyed, just distributed).
In sum, there surely is a good case for hard and soft obscurantism, but Elster doesn’t make it.
12.
bjk | 29 January 2013 at 2:22 pm
I just read the “Crisis of Social Sciences” article by Elster linked above. It fills in some of the gaps in the Diogenes article. His criticisms aren’t as off-base as it seemed at first.