May 2018

Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
Blog powered by Typepad

« Administrivia: Comment Issues | Main | FEE Summer Seminars »


Feed You can follow this conversation by subscribing to the comment feed for this post.

I agree completely, and ironically this is similar to the issue I was trying to raise with the Dan Klein post. To put it briefly (since that's another post), describing the decision making of a neoclassical agent in a world with unknown unknowns and specifying a situation where neoclassical decision makers maximize welfare (and therefore can't face unknown unknowns) are two quite different things.

But I agree completely on your concerns about BHL - that was about my thought when I saw it this morning. I don't think it's too much to ask either, although I'm probably even less likely to see philosophical conundrums than you are.

It is interesting, though, how informed economists' values are by their positive conclusions.

Causal nexes in economy do not lie in physical causality but in the mind of agents.
Hayek said somethink of the like, cannot quote it perfectly.

The view of Jeremy Bentham is instructive. Bentham believed, contrary to almost everyone today, that ethics was more scientific than the descriptive-analytical study of choice and its consequences. Why?

Bentham believed that ethics was about the objective consequences of decisions (personal or governmental). Social study was about the subjective causes of human behavior and then its consequences. How will people behave? In general, they are poor calculators of pleasure-pain (read: consequences). They commit errors all of the time. Bentham believed that there were not just a few errors that people committed but an indefinitely large number.

So to trace out the consequences of actual behavior you first have to know what people will do. This is very hard indeed.

On the other hand, to ascertain what people should do you have to trace out the objectively true consequences of actions. This is easier to ascertain.

Now obviously this sharp distinct will probably fall apart as how people actually respond to what one does is relevant (but less so for rule utilitarianism where you postulate that everyone will follow the rule).

But if we mean by "subjectivism" something like what Bentham was worried about, it is quite right to be worried.

We must have limits on the errors we shall entertain, just as we must have limits on the degree to which people can be assumed to do the "right" thing.

So the conflict between subjectivism, conveniently and rightly understood, and consequentialism is false.

How can one possibly know the effects of the consequences (effects are personal, and subjective) of ones action in order to construct a calculus out of it?

How can we possibly be consequentialists if the goal we are to maxmize - good consequences - is subjective and cannot possibly be known for certain?

I'm with Mattheus on this one.

I've always been incredibly skeptical of consequentialism. I wrote a blog post once ("Ethics and Landsburg's Runaway Train") in which I pointed out that the choice to save 5 people over 1 from a consequentialist standpoint assumes that the 1 person you choose not to save is not some amazing person who would have gone on to save many other lives.

For example, if the choice is either (a) Kill Albert Einstein or (b) Kill X number of Average Joes, then the consequentialist question becomes "How many Average Joes are worth the life of Albert Einstein?"

My answer: It's either entirely subjective, or you'll never be able to know the "correct" answer in advance.

I'm with Vallier: Ditch consequentialism. It only seems acceptable if you look at it very shallowly.

But I'm crazy and not very smart, so take my thoughts with a grain of salt...

Pete, thanks for interacting with my post. But note that I'm not talking about *economics* but *economists*. The post was partly inspired by talking to some of your students at APEE last year who were both consequentialists and subjectivists about value. So I'm not making the claim that the subjective theory of value as employed by economists makes any metaethical assumptions, but rather that assumptions are made about the way that economists employ it.

agree with you Peter that the distinctions you mentioned address the conflict but I wonder whether there still is at some level a tension between consequentialism and subjectivism. If consequentialism can entertain the imposition of rules (for example) that interfere with voluntary exchange (which necessarily express subjective preferences), it must do so on the assumption that the decision maker has some knowledge of individuals' preferences and, in practical terms, an assumption of at least some predictability and similarity amongst individuals of those preferences.

Furthermore, consequentialism must presumably also embody some means to determine what set of consequences are considered better. Unless the standard for adopting the rule is unanimity, any non-voluntary rule will involve benefits to some and harm to others. Thus, some means to determine whether the benefits, however defined, from the rule "win" over the harm is required. Doesn't that require stepping away from subjectivism?

I think you're right, Pete. Here's what I posted over at BHL:

Let me see if I've got the argument. Consequentialists say that we want "good" outcomes. Subjectivists say "good" varies from person to person. If "good" is different for each person, then "good outcomes" doesn't mean anything. If, instead, the phrase "good outcomes" has content, then it's gotta be the same for everyone. Thus, you say, we can't be both subjectivist and consequentialist. Have I got it?

I don't see where there is a real problem here. When people espouse subjectivism of this sort, they are usually talking about the methodological principle that social science must generally take preferences and human nature as a given. What people *imagine* to be good varies from person to person. In thinking about what we should do, however, I may nevertheless have a definition of "good" that does not vary from person to person. I want people to "thrive" or "have pleasure" or whatever even if I know that people do what they damn well please even. Sometimes they do what they want even when it ruins their lives or kills them. As a social scientist I generally take preferences as given, even foolish preferences. But as a moral actor, I may try to show you that those same preferences are bad and encourage you to mend your ways. Where is the inconsistency in all of this?

A few other points seem relevant. The uniform "good" of my consequentialist ethics might be pretty high-level abstract stuff. It's "thriving" or "pleasure" or "self realization" or whatever. The preferences of choosing individuals are necessarily more specific. Yes, I want to "thrive" or whatever. But I never choose "thriving" over "not thriving." I choose reading a book over watching Oprah, exercise over chocolate cake, shaking your hand over shunning you. Those individual choices depend on details of context not imagined in my ethical theory. No ethical theory can pack in all relevant context or anticipate all contingencies. Thus, my consequentialist principles may tell me not to second-guess the choices other people make. There's a "knowledge problem" with choosing for others. Thus there is a presumption in favor of autonomy, of individual choice, of experiments in living. Our ignore of place and circumstance limits the ambitions of our moral judgments.

If one makes the case for consequentialism as a normative standard, rather than as simply an explanation of observed morality, is one not necessarily also making the case for rules that may conflict with a freely evolved order or override voluntary exchange?

As Roger Koppl pointed out, libertarian consequentialists drastically limit the cases in which they would be willing to consider such rules, in recognition of the knowledge problem and perhaps the subjective value of autonomy itself. However, is it not the case that what separates consequentialist libertarians from natural rights libertarians is that the former would sometimes consider such rules appropriate whereas the latter never would. And in such cases, would consequentialists not encounter the tension with subjectivism I referred to above?

@ david:

As it happens, I am not myself a libertarian, but I think I can respond anyway. For the consequentialist libertarian, the ban on coercion is the result of a possibly lengthy analysis of consequences rather than a first principle or the implication (by a short route) from a few initial premises.

Jan Narveson lays out a case for libertarian consequentialism in his _The Libertarian Idea_.

Narveson began as a non-libertarian consequentialist, and developed his rather radical liberarian consequentialism via engagement with Nozick's work.

I introduced Narveson to Kirzner's entrepreneurial / finders keepers discovery arguments when Narveson spent a week at UCR in the early 1990s.

Narveson discusses Kirzner and the justification of profits in his Respecting Persons in Theory and Practice.

Correct, at least with Subjectivism there are some standards to go on. In terms of Consequentialism, how can one kn ow the future beforehand.

My best guess is that Kevin Vallier's post reflects a failure on the part of economists to effectively convey their concepts to political philosophers.

Otherwise, why would somebody who understands Hayek's concept of partial knowledge... ...ever suggest ditching consequentialism? Why would Vallier want to "disregard everything that is important and significant in the real world"?

Why did Matt Zwolinksi want to do the same thing?

Obviously they have information that I don't have...but are they going to have some partial knowledge that disproves the concept of partial knowledge? If they do have such information then why haven't they shared it already? Why haven't they said, "here's the philosophical answer to the knowledge problem."

When you read over Vallier's preferred political theory...contractualism... you get the sense that he's providing the philosophical answer to the knowledge problem? Personally I get the feeling that he's disregarding "everything that is important and significant in the real world".

Like I said before, hopefully you can succeed where I failed!

If consequentialism is seen as a "best fit" description of the implicit collective moral calculus which operates in the background and which explains the moral basis for freely evolved rules and norms, then I see no conflict with the knowledge problem. But if one sees consequentialism as normative, then one sees it as providing the moral justification, at least in theory if not in practice, for adding rules which may have not freely evolved. Isn't there in that case a conflict between the knowledge problem (of which subjectivism represents one element) and consequentialism?

Libertarian consequentialists may argue that most (and maybe all) such rules would fail the consequentialist test but, in order to make the statement, they would need in effect to apply the test and believe that it could be reliably applied. Perhaps the way around that is to argue that the knowledge problem implies that the test could not actually ever be reliably applied by policy makers and the only hope for even approaching the consequentialist ideal is to have a completely freely evolved order. That would imply that consequentialists who recognize the contraints of the knowledge problem must be anarchists (or at least believe in very drastically reducing the geographic scope of government units to, say, cities), which I gather is generally not the case. Non-libertarian consequentialists don't have that "out".

The other way around the problem I suppose might be to play with the definition of "free" or "coercion" so that one can only have a "freely evolved order" in the presence of some minimum level of coercive state. In that case however it would seem that one would run into the knowledge problem again in determining the boundary of the state.

Where am I going wrong here?


david stinson, here's where you are going least from the pragmatarian perspective.

The option you missed is to allow taxpayers to directly allocate their taxes. For example, at anytime throughout the year you could visit the Environmental Protection Agency website and directly submit a tax payment. The EPA would then send a notice of your payment to the IRS.

Part of the goal of this approach is to allow people to consider whether the supply of a public good should ever be greater than the demand for that public good.

Subjectivism and Consequentialism can be reconciled in the contractarian approach, as exemplified by J.M. Buchanan.

In consequentialist economic analysis, rules and institutions are analyzed and chosen because of their (presumed) outcomes.

However, if preferences over outcomes (and thus indirectly over different sets of rules and institutions) are subjective, there seems to be little ground for "objectively" identifying "good" rules and institutions.

Consequentialist economics can be reconciled with subjective Economics with Buchanan's contractarian approach:

Oly unanimous agreement of the members of society to a certain reform of the "rules of the game" can "objectively" legitimize social change. If all people agree that a reform is "good", there is no need for further debate. And they will agree or not agree to a reform based on their evaluation of the hypothesized outcomes of the rule-change in question.
Yes, preferences are subjective. But (informed) unanimous agreement to a rule-change is objectively observable. Furthermore, individuals need not agree to a certain reform proposal for the same reasons, they can value different things in a reform than the other citizens. Gains from trade can also be had at the level of economic reform.

Ultimately, subjective consequentialist economists which seeks to offer policy advice must offer proposals for reform which should be able to find the agreement of all.

Of course, economists could derive their reform-proposals from other normative value-standards than unanimous agreement. However, policy advice that does not clearly spell out why a certain reform would be also in their own informed interest will simply be irrelevant to the citizens it is offered to.

Henry Hazlitt. Foundations of Morality. Read it. Love it. Live it. :)

I read a lot of blogs but this blog is really awesome and fulfill all my requirements...Very Good job Guys!

The comments to this entry are closed.

Our Books