pageok
pageok
pageok
Sunstein on Hayek:

Cass Sunstein, guest-blogging for Lessig this week, has had a number of posts on Hayek (here and here). Most of the posts have actually been on why certain institutions, such as blogs, open-source software, Wikipedia, etc., are not actually Hayekian information-processing mechanisms. That is correct. There is a fundamental difference between a Hayekian information-processing mechanism, such as the price system, and the type of information-processing mechanism that Sunstein is exploring.

Oversimplified, the distinction is between those systems where the purpose is simply to centralize dispersed knowledge in one place versus systems that centralize knowledge for the purpose of decentralizing it again.

[Continue reading "Sunstein on Hayek" under hidden text]

Martin (mail):
Your reference to the merits of common law systems of law making "which collected information from case-by-case adjudications of specific disputes that arise among people and infer more abstract principles from those" would appear to be relevant to the evaluation of the jurisprudence of Justice O'Connor. O'Connor is widely criticized for deciding cases based too much on the specific factual situation and with inadequate attention to the development and application of systematic principles (and, more particularly, systematic principles of a sort that meaningfully dictate outcomes, as opposed to balancing tests with excessive degrees of freedom). But isn't a great sensitivity to the facts of specific cases, and a corresponding reduced focus on systematic legal theory consistent with a common law approach to judging that has historically been favored by many conservatives.

My understanding is that there is a Burkeian/Oakshottian conservative tradition that favors common law lawmaking. (Posner's early, semi-abjured, work on the efficiency of the common law is arguably a Darwinian analogue of this line of thinking.) I am not very familiar with either the historical or current analytical literature on these ideas. However, my sense is that they are based on, among other things:

-- a belief that systematic theorization in law, as elsewhere in human life, is inherently highly fallible and that this fallibility is not confined to grand theories like Marxixm or Jeffersonian democracy but extends to the level of theorization needed to, e.g., codify the law of torts or negotiable instruments.

-- a belief that judges are likely to make better decisions even on issues of law if they focus, to a very considerable extent, on the facts of concrete cases before them and on the practical consequences of different legal choices in the specific factual situation. This belief is based on the traditional Burkeian etc. view that human intelligence is much better at dealing with concrete situations of modest scope trying to predict the benefits and costs of a policy over an extended period of time and over a wide variety of circumstances. Consistent with this, focus on concrete, small scale factual situations arguably allows the judge to take advantage of his or her own tacit knowledge (Karl Llewellyn's "situation sense").

-- a belief that policy change, including legal change, works best if it is gradual, with oppurtunities to obtain feedback concerning practical consequences and stop or adjust the course of change and with reduced risk of wholesale destruction or abandonment of current practices that have unrecognized benefits. The common law system of change by stretching or distinguishing precedents to produce results that make sense in response to particular facts seems, with a new rule sometimes recognized as having emerged from enough stretching and distinguishing in many cases is consistent with this sort of gradualism.

Your discussion of common law as a method of aggregating information from social practices sees like an interesting variant of these traditional conservative reasons for favoring a common law system. I don't know enough, and do not have enough time for research and thinking to see just how it fits in or differs.

More broadly, I don't have a clear sense in my mind as to how the common law model described above empirically or normatively applies to the work of the US Supreme Court. (Indeed, I don't really have a firm sense as to how well the model applies to its home turf of, e.g., state contract or tort law.)

Nevertheless, the common law model clearly has been very influential (for all Anglosphere lawyers and political thinkers, not just conservative). I would have expected more discussion of it in connection with Justice O'Connor and with candidates for future appointment to the Court.
7.22.2005 11:37am
Martin (mail):
Your reference to the merits of common law systems of law making "which collected information from case-by-case adjudications of specific disputes that arise among people and infer more abstract principles from those" would appear to be relevant to the evaluation of the jurisprudence of Justice O'Connor. O'Connor is widely criticized for deciding cases based too much on the specific factual situation and with inadequate attention to the development and application of systematic principles (and, more particularly, systematic principles of a sort that meaningfully dictate outcomes, as opposed to balancing tests with excessive degrees of freedom). But isn't a great sensitivity to the facts of specific cases, and a corresponding reduced focus on systematic legal theory consistent with a common law approach to judging that has historically been favored by many conservatives.

My understanding is that there is a Burkeian/Oakshottian conservative tradition that favors common law lawmaking. (Posner's early, semi-abjured, work on the efficiency of the common law is arguably a Darwinian analogue of this line of thinking.) I am not very familiar with either the historical or current analytical literature on these ideas. However, my sense is that they are based on, among other things:

-- a belief that systematic theorization in law, as elsewhere in human life, is inherently highly fallible and that this fallibility is not confined to grand theories like Marxixm or Jeffersonian democracy but extends to the level of theorization needed to, e.g., codify the law of torts or negotiable instruments.

-- a belief that judges are likely to make better decisions even on issues of law if they focus, to a very considerable extent, on the facts of concrete cases before them and on the practical consequences of different legal choices in the specific factual situation. This belief is based on the traditional Burkeian etc. view that human intelligence is much better at dealing with concrete situations of modest scope trying to predict the benefits and costs of a policy over an extended period of time and over a wide variety of circumstances. Consistent with this, focus on concrete, small scale factual situations arguably allows the judge to take advantage of his or her own tacit knowledge (Karl Llewellyn's "situation sense").

-- a belief that policy change, including legal change, works best if it is gradual, with oppurtunities to obtain feedback concerning practical consequences and stop or adjust the course of change and with reduced risk of wholesale destruction or abandonment of current practices that have unrecognized benefits. The common law system of change by stretching or distinguishing precedents to produce results that make sense in response to particular facts seems, with a new rule sometimes recognized as having emerged from enough stretching and distinguishing in many cases is consistent with this sort of gradualism.

Your discussion of common law as a method of aggregating information from social practices sees like an interesting variant of these traditional conservative reasons for favoring a common law system. I don't know enough, and do not have enough time for research and thinking to see just how it fits in or differs.

More broadly, I don't have a clear sense in my mind as to how the common law model described above empirically or normatively applies to the work of the US Supreme Court. (Indeed, I don't really have a firm sense as to how well the model applies to its home turf of, e.g., state contract or tort law.)

Nevertheless, the common law model clearly has been very influential (for all Anglosphere lawyers and political thinkers, not just conservative). I would have expected more discussion of it in connection with Justice O'Connor and with candidates for future appointment to the Court.
7.22.2005 11:37am
Bruce Hayden (mail):
One reason that I see pricing to be superior to, say, Wikipedia, is that in the former cases, one of the incentives for (or causes of) accuracy is arbitrage. I commented in the Sunstein thread that Wikipedia somewhat breaks down in political discource because it is too easy to subvert. It can fail there for the same reason that online polls are invariably inaccurate. They are self-selecting, and are vulnerable to organized campaigns.

Arbitrage, for the most part, gets around this problem with pricing, because if you go out to drive up prices through overbidding, someone is likely to slip in and sell high, with the expectation that they can buy low later at the correct price, etc. Wikipedia, et al., don't have this built in mechanism.
7.22.2005 1:35pm
Zywicki (mail):
Martin:
I think the distinction is that the Supreme Court interpreting the Constitution is not the type of common law court that Hayek (or I) have in mind. The Supreme Court is an intensely centralized body that decides cases differently from the traditional common law, which dealt with very concrete cases of tort, contract, property, and the like. The main difference is that there is no real way in which the Supreme Court collects dispersed knowledge or coherent feedback on its rulings.

I argued in an article a few years ago that even though the Supreme Court sometimes looks like and fancies itself to be common law in nature, in fact there are very big differences between it and a traditional common law court.
7.22.2005 2:33pm
Scott Scheule (mail) (www):
Wonderful post.
7.22.2005 3:16pm
Carl Pham (mail):
Eh, I think your distinction is forced. Primarily because the ultimate purpose of Wikipaedists is not acumulating accurate knowledge for its own sake (that's if anything merely an intermediate goal), but, indeed, the redistribution of this knowledge to bazillions of end-users, who will then make better decisions, coordinate with others and the natural universe better, et cetera. That's why it's published on the Web, after all, and not written down on a scroll in a monastery.

In other words, the ultimate purpose of the Wikipedia is centralizing a certain type of knowledge very much for the purpose of decentralizing it again. It does not satisfy your criterion. If you want a system designed to centralize knowledge without redistribution, I think you need to look at something like overseas espionage, where knowledge flow is carefully designed to be strictly centripetal.

Maybe what you're talking about is a distinction between knowledge apparently useful to the individual by itself, like knowing whether red light or green has the longer wavelength, and knowledge that seems a priori meaningless except in a market context, like the price of goods and services.

But while that distinction seems less forced it is still not clean. I suggest even "abstract" knowledge is in fact put to use in a market context, e.g. to get a better job or do your present one better, or maybe to impress girls. And also that the price of things does incorporate non-market-related knowledge about human beings, e.g. that the price of the death of a human being is far higher than the price of the death of a large wild animal says something interesting about our species psychology.

Also, in re other comments about, the Supreme Court enforces "abstract" principle only in the short run. Since the Constitution can be amended arbitrarily if the votes are there (i.e. there is no super-Constitution restricting the ways in which the Constitution can be modified), and because traditions in how the thing is interpreted are sensitive to political trends over time, then over a long enough time scale (e.g. centuries) even Supreme Court jurisprudence can reasonably be argued to simply codify "what works".

My point here is that the distinction between just "what works" and "principle" in human law is mostly just a difference in time scales. That is, "principle" tends to be just "what works" but with a much longer time scale for its change.
7.22.2005 11:50pm
Bill (mail):
While in law school I was on the lookout for important current bodies of law where we can watch the developmental characteristics of common law, that Zywicki and Hayden describe, "in action". Ironically, the most significant of these in my estimation is the laws of mergers and acquisition as developed not "in law" but by the "court of equity" in Delaware. Still, all the cases in this line refer to each other and distinguish somewhat on (shall we say) factual characteristics of broad legal import. It's "commonlawy" stuff.

Now, note that those current lawyers with actual business experience/knowledge can actually appreciate the current economic consequences of the decisions out of which these M&A "doctrines" are evolving. This latter fact leads me to wonder:

What would Hayekians like Zywicki, who like the (shall we say) "fact-minimizing" understanding of common law adjudication think of the idea that we may someday have an corpus of M&A doctrine that could decide
these cases? Understanding this doctrine would not require, according to this line, understanding the economic history involved... however large developments like Paramount Time Warner seem presently.

Of course the threat of SEC intervention may loom (as Prof. Mark Roe argued in his article Delaware's Competition). But this does not mean the question is unanswerable.
7.23.2005 12:51pm
Bill (mail):
Correction: Actually I meant to refer in my post to Todd [Zywicki] and Martin's dialogue in this thread, not to "Zywicki and [Bruce] Hayden".
7.23.2005 2:27pm
Bill (mail):
Correction: I meant the law of takeovers and takeover defenses, not M&A generally.
7.24.2005 11:29am