« Shameless plug | Main | Rawls Made Easy As Easy As I Can Make Him »

June 27, 2006

Comments

I think it's completely futile to discuss psychological hypotheticals like that. How about instead thinking about ways of universally establishing "a basic level well above subsistence".

Sebastian, I think you are going far off the rails with this one:
"I am not convinced that from behind the veil people would generally choose the hyper-conservative (in the sense of risk) position of maximizing the position of the least well off."

He's dead on in some cases. My cousin is seriously mentally handicapped. It is a massive burden on his parents, though they are fortunate enough to be able to handle it, with difficulty. That doesn't mean they don't take advantage of the infrastructure in part provided by the state--they do. And I don't begrudge it to them, or to any family so burdened.

Rawls' veil of ignorance does not imply the extreme risk-aversion that you imply, and I don't think suggested that it did. It does suggest that there should be some floor below which we should not be willing to condemn anyone.

After all, people still buy car insurance even in states where it is not required by law.

"...and in this scheme the equal political liberties, and only those liberties, are to be guaranteed their fair value."

This seems pretty clearly opposed to any "positive liberty" or "positive rights" interpretation. "Only those liberties" A person is not guaranteed a right to a car or education, under Rawls. OTOH, she could be, if her society so decided.

But "equal political liberties...guaranteed their fair value" seems to me to pretty strong, and in practice Ken Lay or Bill Gates should have absolutely no more influence over elections, access to politicians, or input on policy than I do. Their money should not get them better lawyers, etc. This might have serious economic implications.

But Rawls was writing a kind of "Utopia", not drafting a constitution. Equal political liberty is a goal we should strive for to the extent Gates is not treated more unjustly than myself(myself being the least advantaged), or treated in a way I would consider unjust if I were Gates. For example, the solution to political inequality would be unfair if it involved taking all of Gates money, but fair if inequality were made irrelevant to influence, e.g., public financing.

To put it another way, I suspect I view the right to vote and the right to a car as more similar than Rawls would, because they are achieved by the same process; but Rawls would say taking away a right to vote would be an injustice because it is a process right; but taking the right to a car would not be taking away a liberty.

Or something.

Your hypothesis about a higher tolerance of risk is interesting, but runs into trouble.

What you are saying is equivalent to saying: "but isn't it the case that people are often willing to buy lottery tickets that might have a huge pay-off, even though they have a negligible chance of winning?"

Yes, of course it's true. It's even true that people are willing to pay money far in excess of the value of the bet (where that value is calculated in standard ways as the pay-off discounted by the chance of receiving it, so that a 10% chance of getting $10.00 is valued at $1.00).

In fact, what all of these observations about human behavior show is that the human mind is very bad at statistics, very ill-equipped to deal with large numbers, and very prone to certain perceptual illusions.

Where once unscrupulous bartenders used perceptual illusions about volume to sell rot-gut in sucker-glasses, now the lotteries use perceptual illusions to sell near-worthless tickets at inflated prices.

So the trouble is simply this: people *are* willing to tolerate high amounts of risk. But they are also manifestly willing to buy into what are manifestly short-sighted and irrational risks (manifest if you have the patience and moderate skill to run the numbers). Indeed, the most profitable industries in the country right now, sc. casinos, are flourishing because people are demonstrably irrational when it comes to risk, statistics, and large numbers.

It's just not an area of endeavor where untutored human intuitions are very reliable. Indeed, quite the opposite--one may rely on people's intuitions to get it wrong, and make a lot of money by relying on it. And the best way to dupe their intuitions is to dangle the infinitesimal prospect of an astronomical reward in front of them.

So the fact that "many people would be willing to have a small risk of being worse off for a large chance at a place in a society which was much better off than a more equalized society" may show a flaw in Rawls. Or it may just show a persistent flaw in human perceptual capacities.

I share Sebastian's view on the second principle. Note that Rawls is not just saying that there should be a floor - he's saying that the most important thing is that the floor be as high as possible (assuming that the first principle is already taken care of, that is). I don't know of any society, or any liberal political party or organization, that actually acts as if that was the priority.

Obviously, it's better if the floor is higher, but it's also better if a smaller portion of the society is at the floor, and it's better if the people who are not at the floor are doing better. I don't think that any of these three considerations should have absolute priority over the other two, and I think that most everyone at least acts as if they agree.

"I think with a basic level well above subsistence, people are willing to risk a small or medium chance of slightly worse outcomes for a medium to high chance of much better outcomes."

Well, here's a trenchant criticism of Rawls: he didn't spend enough time thinking about worlds in which every person is already, miraculously, guaranteed a basic level well above subsistence, and knows about that guarantee from behind the veil. It's almost as though he was concerned about how things might play out in worlds that are remotely similar to our own, in which poverty far below subsistence level has been a constant fact of life throughout our past, and looks to be a constant fact for the foreseeable future (or at least as long as conservatives and their apologists have their hands on the world's resources).


His theorizing is too utopian, you see--it didn't occur to him that he could just tap that handy resource of glibertarian theorizing, the Pony Principle.

"I can come up with a better theory than Rawls did! I'll see Rawls, and raise his Maximin! By a Pony!"

Another serious question lies in how one goes about setting the floor "higher," if indeed that's what we determine is good.

I would suggest that redistributional programs aren't particularly successful at this, and that they very often have the opposite of the intended effect. In other words, even if Rawls got the diagnosis right (which I encourage you to continue debating), the prescription has usually been wrong.

Maybe this is just me, but it seems a bit off to have a post on Rawls that concludes with some percentages that 'disprove' him. It kind of misses at what Rawls was trying to get at. I'm sure you think that your 'stats' constitute a disproof, but, as ob points out, generating a situation that is possible under Rawls but has never been obtained in human history doesn't seem like much of a disproof.

ob: Rawls explicitly restricts his theory to questions of domestic justice (on the grounds that international justice raises very, very different questions, which he tried to tackle, I think unsuccessfully, in a later book.) He also restricts it further: to domestic justice in countries that have enough not to be faced with starvation or (more generally) questions of subsistence.

So it's not that he ignores these questions or wishes them away; it's that he is explicitly restricting his analysis to cases in which they do not apply.

Seb: there is basically no tendency in Rawls for the civil and political liberties to transform themselves into e.g. the right to a car.

Moreover, note one important point about the comparisons between different systems: they all have to be made at the level of roughness that one would expect, given that they are comparisons of different systems of rules governing the behavior of free people who make, well, free choices. Your comparisons might be possible, if one didn't place too much weight on the precise splits (as in: I said 30% would get 50; not a single person more!). But the 1% is probably pushing the margins of what it would be possible to predict. (I mean: that the number is 1%, rather than 5%.)

And to Jason: Rawls is only trying to ask: what criteria should one use in assessing the justice of a set of rules? If it turns out that some rule that people think would produce a given result wouldn't do so, then he's completely open to that.

Or, more specifically: if it turns out that an absolutely unfettered market system leads to the least well off being better of than they would be in, say, Sweden, then Rawls; response would be: OK then. Too bad for Sweden. That-s an empirical question.

Ob,

Well, here's a trenchant criticism of Rawls: he didn't spend enough time thinking about worlds in which every person is already, miraculously, guaranteed a basic level well above subsistence, and knows about that guarantee from behind the veil. It's almost as though he was concerned about how things might play out in worlds that are remotely similar to our own, in which poverty far below subsistence level has been a constant fact of life throughout our past, and looks to be a constant fact for the foreseeable future (or at least as long as conservatives and their apologists have their hands on the world's resources).

Even if you are correct, my criticism applies perfectly well when you try to apply his principles in the United States.

Hilzoy,

Moreover, note one important point about the comparisons between different systems: they all have to be made at the level of roughness that one would expect, given that they are comparisons of different systems of rules governing the behavior of free people who make, well, free choices. Your comparisons might be possible, if one didn't place too much weight on the precise splits (as in: I said 30% would get 50; not a single person more!). But the 1% is probably pushing the margins of what it would be possible to predict. (I mean: that the number is 1%, rather than 5%.)

Even if the number is 5%, I would be surprised if people took 95%/5% of 50/100 over a 5%/95% of 40/1000. The downside risk isn't that great since both of the lower levels are well above subsistence and the upside gain is enormous. I know I would certainly choose the latter distribution. My point is not to assert some specific distribution in any case. It is to challenge the idea that the maximization of the position of the lowest really is the biggest concern to the exclusion of all other concerns outside of the first principle.

British sociologist Michael Marmot says that inequality causes health problems--not poverty in itself, but simply the existence of hierarchies in a society. People at the bottom suffer more health problems. I suppose Sebastian could argue that a society where everyone lives to age 60 is worse off than one where the poor live to 70 and the rich to 80. But that might not be the actual choice in the real world.

If we are going to speculate as to what conclusions those behind the veil might come to let's note that people are, in general, extremely, almost unbelievably, risk-averse in financial matters. This is an empirical fact. Just look up the topic of "equity risk premium," or look to see how individuals invest their retirement funds. So while some of the choices Sebastian argues for seem convincing, lots of seemingly good gambles would not be popular.

And of course Sebastian's argument assumes that the judgments reached will be based wholly on individual expectations. Isn't it plausible that ideas about fairness will influence decisions? Then a big increase in our chance of going from 500 to 1000 will not be seen as attractive if it involves pushing even a few people down from, say 50 to 20.

In other words, to the extent we value the well-being of others the calculations involved become more complex.

"Isn't it plausible that ideas about fairness will influence decisions? Then a big increase in our chance of going from 500 to 1000 will not be seen as attractive if it involves pushing even a few people down from, say 50 to 20."

I'm not sure how you are using 'fairness' here.

I think real intuitions about fairness in this context involve things that Rawls doesn't allow in to the equation. If, in the context of the hypothetical about setting up a society from behind the veil of ignorance, we find out that the few people moving from 50 to 20 refuse to (as opposed to cannot) work I suspect that a large number of people would not object to reducing their position to move everyone else from 500 to 1000. The question of desert is one that Rawls goes to great pains to avoid.

Rawls undermines the idea of rewarding desert in counterintuitive ways (claiming that since we did nothing to earn being smart or tall or nimble we don't have any claim on deserving the fruits of using our smarts, height or nimbleness to maximum utility). I'm not sure I buy it. Lots of people have talents that they don't use effectively. In my understanding of justice they deserve less than those who have the same talents and use them effectively. If (and I'm completely aware it is a big 'if') we were designing a society where the least well off were tied to an idea of desert, I suspect that you would get strong support for positions that deviate from Rawls' principle.

Bernard: Rawls rules out worrying about fairness in the decision about principles, since the choice he's setting up is supposed to define what counts as fair. He assumes the hypothetical beings behind the Veil of Ignorance are concerned with securing their own futures.

(Compare: when you want to divide a cake fairly and, to do this, let one person carve it up and then let everyone else pick their slice of cake first, so that the one who cuts gets the last slice, you assume that everyone is selfish, and thus will pick a larger slice if the carver produces one. The whole thing falls apart if the person who carves the cake gets to think: oh, but X is altruistic, so I can make one slice very small and X will take it, and as a result my slice will be bigger! The combination of (a) the assumption that all the recipients of cake are selfish, plus (b) a set-up designed to ensure a fair result given this assumption, produces a fair result. Same thing with Rawls.)

"95%/5% of 50/100 over a 5%/95% of 40/1000."

These are not reasonable choices. I would think that you would have to have a fixed population and accumulated wealth for the starting points of the thought experiment to have any useful conclusions. Assuming 100 units of population the wealth in the first case is 50*95+50*100=9750 and the second is 5*40+95*1000=95200. None of your choices seem to use the same starting conditions.

And Seb: Rawls does allow for rewarding those who use their talents more effectively. (Incentives that can make everyone better off.) He does not think, in general, that economic systems do or should reward moral desert. For a different argument to the same effect, see here.

...imagine a scale of wealth...

I would argue that while economic security and lack of poverty have huge effects on well-being, wealth as such is much less important.

One thing that is important is the chance to participate on a basis of rough equality in the social groups you encounter in your daily life.

Given a lack of economic misery and a lack of social misery, I'd be perfectly happy to have a system where some folks were unimaginably wealthy.

But since we have plenty of economic and social misery in this world and plenty of economic and social misery in the U.S., ...

Sebastian - My impression (without looking again at the books) is that one of the reasons people behind the veil are supposed to choose maximin is that they have no access to information like the total amount of societal wealth, or its distribution, the factors which you consider in your last paragraph. Perhaps a framework in which that information is unavailable is in some way improper, but my understanding is that it is his framework.

"My impression (without looking again at the books) is that one of the reasons people behind the veil are supposed to choose maximin is that they have no access to information like the total amount of societal wealth, or its distribution, the factors which you consider in your last paragraph."

I don't remember that being the big deal, though my reading was many years ago. My moral intuitions tend toward believing that such things as absolute level of the least well off are very important. If the minimum level is high, I am much less concerned about the relative distribution. This is especially true if the minimum level is high and rising. Rawls hypothetical is useful for looking at a snapshot of distribution (though in my opinion much less demonstrative than many social/wealth egalitarians believe) but its drawbacks become strikingly evident when you have to look at the possibility that different structures will have different growth/improvement rates. When you do that you might see that what is best for the least this second may lead to a worse position of the least ten years from now.

But even positing a static society, I'm not at all convinced that a 'best for the least' system conforms to many people's idea of justice compared to other systems ('well off for the least but fantastic for the everyone else' for instance).

Note: I will be posting a 'Rawls Made As Simple As Possible' post at some point today.

Carry on...

Rawls rules out worrying about fairness in the decision about principles, since the choice he's setting up is supposed to define what counts as fair. He assumes the hypothetical beings behind the Veil of Ignorance are concerned with securing their own futures.

Is that restricted to their own economic wellbeing, or are other notions allowed?

Perhaps "fairness" was a poor choice of words on my part, but it does seem to me that some might have preferences over the entire distribution of wealth in addition to concerns over their own possible situations. Are these preferences also excluded?

Bernard: 'well-being' is hard to define in the Original Position, since the people there do not know what their values are, and thus what they will care about. For this reason, Rawls says they will try to maximize their share of certain all-purpose goods, including rights and liberties, powers and opportunities, income and wealth, and the social bases of self-respect. Why these? They are, as I said, all-purpose (or as near as one can come to all-purpose) goods: means to all sorts of good lives. So people with different, or even unknown, conceptions of the kinds of lives they want to lead can all assume that they will value these things. They want the greatest share of them for themselves, in the OP.

In the OP, they are not concerned with how well others do. This is for a number of reasons, chief among them the 'cake' reason above: the set-up as a whole is meant to produce a fair result, and the assumption that each participant wants the best for him- or herself is part of that set-up. Also, the participants do not know their own values, and thus do not know what weight, if any, they will give to the interests of others.

Note, though, that since no one knows who s/he is in real life, this will not end up allowing any particular person to benefit him-or herself. (Here's where the analogy with the cake-cutting comes in.) If you don't know who you are in real life, and you're concerned to secure your own interests, you can only do that by securing everyone's interests.

Searching Will Wilkinson for Rawls ...about two pages of articles

Ya know, even though he doesn't have his doctorate, works for Cato, works on some pretty difficult areas of happiness/well-being/preferences, is usually so far over my head I can't be sure that he isn't a hack, I still feel very guilty for not reading Will Wilkinson more often. As far as I can tell, Wilkinson is crazy for Rawls.

Maybe more like 5 pages

Will's Quote of the Day:May 18th, 2005

“There is no more justification for using the state apparatus to compel some citizens to pay for unwanted benefits that others desire than there is to force them to reimburse others for their private expenses.”
- John Rawls, A Theory of Justice, p. 250 (rev. ed.)

Behind the Veils ...Wilkinson quotes a footnote in TOJ that might go to Seb's percentages:

"The aim is to characterize a just constitution and not to ascertain which sort of constitution would be adopted, or acquiesced in, under more or less realistic (though simplified) assumptions about political life, much less on individualistic assumptions of the kind characteristic of economic theory."

If there is a more useful Rawls site out there, I haven't found it.

hilzoy,

Thanks. I meant "economic wellbeing" as "income and wealth," but your answer makes clear that there are broader considerations.

I'm still murky on "social bases of self-respect," and how one might maximize one's share of those without knowing one's values. Indeed, I have similar doubts about the other non-economic all-purpose goods as well, but I'm willing to wait for your "Rawls for Dummies" post.

Bernard: the social bases come in in large part through having a system that is publicly justified on the basis that it serves everyone's interests as much as is (fairly) possible, and that does not require anyone's interests to be sacrificed so that people who are already better off can get still more benefits.

I made a minor clarification update.

I don't know if this directly (or indirectly!) relates to Rawls, but I was defending a progressive tax a long time ago and this was the result:

With 50 being the American poverty line, a fairly accurate picture of population quintiles would have A earn 10, B earn 50, C earn 85, D earn 130, and E earn 280. A simple progressive tax would tax C 4, tax D 9, and tax E 27. A flat tax would tax C 7, tax D 10, and tax E 23. Therefore, a flat tax is a lower middle class gift to the wealthy since the two categories most affected are C and E. https://www.census.gov/hhes/www/income.html

I have not read Rawls, but what little I can find about the "Veil of ignorance" indicates that what is obscured are things that affect the self interest of the chooser. The per capita wealth would not be one of those things, so any choice would be constrained by a fixed wealth/population at the time of choice. You might have to make a choice that is invariant over time and reflects changing states, but the choice at any given time would have a fixed economy.

Thus your choices, that alter the basic economic realities of per capita wealth for the alternatives, are false choices. Unless you provide additional information about how each choice would apply in the economic reality of the other choice, they can't be compared in a useful way.

In fact, I believe the chooser is meant to be given wide latitude to create his or her own choices, and not constrained to a selection made by someone else. I could easily construct different choices to the ones you picked that I would prefer.

Also, the participants do not know their own values, and thus do not know what weight, if any, they will give to the interests of others.

But ...they will try to maximize their share of certain all-purpose goods, including rights and liberties, powers and opportunities, income and wealth, and the social bases of self-respect.

How is that people can know what they want for themselves but can't know what they would want for others? Seems like the deck is stacked here.

I haven't read Rawls. Perhaps there's a flavor of "Even assuming for the sake of argument that people choose purely selfishly, they would still arrive at result X".

Whether for the sake of argument or in all earnest, I think it's counterproductive to assume that everyone would choose to maximise their share of goods, even at the expense of others; that no one would choose to walk away from Omelas.

I understand that Rawl's notion of justice would not allow for the establishment of an Omelas; but I'm not happy with assuming that (say) Rawls, and hilzoy, and Sebastian would all be delighted to live in Omelas, could they but guarantee they would not be the one miserable person in the place. I don't believe that any of them would, and so the process by which Omelas is ruled out seems contorted and artificial.

Yarrow: the people in the Original Position are, basically, fictional characters who are set up in a particular way to yield conclusions we (real people) would accept as fair. As in the cake-cutting example, the assumption of mutual disinterest (I care about securing my ability to live as I think best; I am not concerned either to advance or to subvert yours) is not supposed either to be accurate about actual people, nor to be in any way a good thing. Nor can it usefully be detached from the whole theory of justice.

Basically: the assumption of mutual disinterest makes all the choosers try to press their own claims rather than preemptively giving them up; the setup of the Original Position, and specifically the restrictions on information, prevent them from tailoring the outcome to fit the interests of the person they will turn out to be in real life, and force them to adopt principles that they'd be willing to live with no matter who they end up being. (Specifically, not knowing whether or not they might be that child, they have to consider that possibility.)

We -- those who think about Rawls' theory and decide whether or not to accept it -- need to consider it as a whole. The fact that neither you nor I are actually mutually disinterested is not obviously more of a reason to reject it than is the fact that each of us actually does know his or her gender. As I said, the people in the OP are fictions designed to serve a very specific purpose.

I suspect that Rawls' work is more or less simply above my pay grade.

I am also, frequently, both puzzled and frustrated with the efforts of various folks -- right and left -- to "frame" social programs in the language of rights.

The actual set of rights guaranteed under the Constitution are fairly small in number (although not in scope). They do not, however, include remedial efforts on the part of government to address the consequences of social and economic inequalities.

This does not mean, however, that government may not act to address those consequences if folks think that is a good thing to do.

To recap: government action to address the consequences of social and economic inequality:


  • not a fundamental and inalienable right
  • nonetheless a good, useful, and legitimate thing for government to undertake

Both are true.

thanks

I think that now would be an apropos time to link to John Scalzi's post on Being Poor I would think these people are well above subsistance but still I think that they are in a dangerous fix.

The comments to this entry are closed.

Blog powered by Typepad