The journal Philosophia recently accepted a paper by TSZ commenter walto, entitled CHOICE: An Objective, Voluntaristic Theory of Prudential Value. Congratulations to walto.
Our discussion of walto’s previous paper was cut short due to censorship by the moderators. Let’s hope they have the sense to stay out of the way and allow open discussion to proceed this time.
Prudential values are a good topic for TSZ, and a nice change of pace from our usual discussions of objective moral values and whether they exist. Hence this thread.
You can download walto’s paper here.
I’ll save my remarks for the comment thread.
walto,
You disagreed with the following…
…but why?
keiths:
walto:
There’s a regress. Let me make it explicit:
1. The original question is “Should I vote for the social program?”
2. If a rich person applies CHOICE-individual, the answer is no, because the social program slightly reduces their future options.
3. If the rich person applies CHOICE-societal, the answer is yes, because the social program produces a net increase in options across the entire society.
4. This creates a dilemma. The rich person, who wants to affirm CHOICE generally, can’t affirm CHOICE-individual without denying CHOICE-societal and vice-versa.
5. So a new choice presents itself: “Should I apply CHOICE-individual or CHOICE-societal?”
6. One might hope that CHOICE itself could answer this question, but no such luck. If one applies CHOICE-individual to this question, it indicates that one should apply CHOICE-individual to the original question. If one applies CHOICE-societal to this question, it indicates that one should apply CHOICE-societal to the original question. The dilemma has been moved back one level, and the person still doesn’t know which version to apply.
7. One can continue applying CHOICE, but each application will move the dilemma back another level. The only way to stop the regress is to appeal to something other than CHOICE.
So CHOICE is incomplete in that it can’t resolve the dilemma, and it’s internally contradictory in that its two forms — CHOICE-individual and CHOICE-societal — contradict each other in some cases.
walto,
But CHOICE does. Otherwise there’s a conflict internal to CHOICE, between CHOICE-individual and CHOICE-societal.
Plain observation reveals that both rich and poor people value their children and their descendants. For non-sociopaths, there is no clear distinction between personal good and societal good.
There are, however, differences in how one perceives the value of immediate gratification vs improvement in the systems that produce goods and services.
Because it has nothing whatever to do with equal votes.
keiths,
There’s no conflict. Forget about CHOICE for a second and think about your mom telling you that packing off and going to India to help the poor instead of going to college might be good for the world but it would be bad for you. The point of mentioning the societal CHOICE is only to indicate that summation is possible: it does not require that what would maximize YOUR prudential value will also maximize your group’s.
I generally agree, although me and mine need not equal “the society.” There maybe goods that will only largely inure to one small societal class.
Right. Of course. Not sure why keiths doesn’t get this.
petrushka,
Sure, but we are talking about walto’s CHOICE standard here. According to CHOICE (individual version), a state of affairs is intrinsically good for a person only if it predictably increases that person’s options.
If the social program is implemented, it will dramatically increase options for the poor while marginally decreasing them for the rich:
So according to CHOICE-individual, it’s “intrinsically good” for a poor person to vote for the program, but not for a rich person.
That’s the nature of voting. Just like–on the individual level–it’s intrinsically good for diabetics but not others to take insulin.
walto,
The conflict is within CHOICE, as I noted above:
I stress this because I think your paper should mention the internal conflict.
keiths:
walto:
According to you, it does:
keiths,
You misunderstand. What has nothing to do with equal votes is this:
“3. The [social] program does not have the same (predictable) effect on everybody’s future choices”
It’s just a non-sequitur.
walto,
But the effect on future choices is exactly what you were talking about when you wrote this:
keiths,
The idea is that making the assumption that choices are equal is tantamount to letting votes be equal. Or, if we think choices are equal, we may as well think that votes are equal. Now you say, “Yes, but you admit that choices are not actually equal: some produce more successful choices or more options for them.”
But, again, that’s different from saying that choices aren’t equal. In fact, if they weren’t equal, it wouldn’t follow that those that are more fecund have anything special about them, since they could be productive of “less important” choices. It’s like this: Suppose in some area all the farmers agree on a price for sheep that is indifferent to their age, gender, size, etc. They’re 1000 bucks each, regardless. But, obviously, those that are fertile are worth more under such a pricing convention–not because you can get 1500 dollars for one: nobody checks. But the fecundity is valuable, even though it’s not priced–or is factored into the average price that is used for all the sheep.
By noting this we haven’t altered the contract regarding the price per sheep. Each still has the same price as all the others. The same is true of choices (and votes). If we treat them as equal, there will nevertheless be additional successes brought by some of them. That’s worth something.
walto,
Even CHOICE itself says that choices aren’t equal. Some are “intrinsically good” and some aren’t. Some lead to an increase in future options, and some don’t. CHOICE undermines the case for the equality of choices rather than supporting it.
That isn’t desirable. The sheep economy would suffer due to less trade, and the bad sheep would drive the good ones out of circulation. It’s the ovine version of Gresham’s Law.
Some votes are up some are down, some wise some stupid. Still equal.
Re the last bit, I agree that that pricing system might not be advantageous under certain conditions, but it would be under others–i.e., if the costs of checking for individual characteristics were high, or it was hard to price most of them. Insurer’s don’t price for every difference they can find in their risks: just the big ones that aren’t too expensive to find. And some of those aren’t allowed to be put into their prices anyhow–like race, e.g. The sellers may know that some races will be more or less expensive risks for them, but they must treat them equally or not write that line of insurance.
I don’t see the bad sheep driving the good ones out of circulation, assuming $1000 is really a decent approximation of the average “worth.” If so, all should be OK. If not, some farmers may have to stop trading or go out of business. I think these sorts of arrangements aren’t terribly unusual. Stuff is sold in bulk–the good with the bad.
Reminds me of trying to find the best container of strawberries. Or the best used car. Or house. Or spouse.
Better stop there.
keiths:
walto:
Votes are stipulated to be equal in one sense, in that each vote counts equally in determining the outcome. Under CHOICE, options are likewise stipulated to be equal in that sense, because each option contributes equally to the determination of whether a choice is “intrinsically good”.
You wrote:
But the stipulation that choices are to be treated equally (in that one sense) doesn’t make them inherently equal, and it doesn’t support the notion that votes are (or should be) inherently equal in that one sense, either. The latter is also a stipulation.
Yes. It’s axiomatic.
keiths:
walto:
It depends on the variation in sheep worth as well as the average. If the cost of sorting sheep (which is very small, indeed) is considerably less than than the typical variation in worth, the farmer will have an incentive to sort. And when they sort, they’ll keep the good sheep for themselves and sell the bad ones.
The bad sheep will drive the good sheep out of circulation. (Or the farmers will find a different market in which price is responsive to quality.)
Just to add to the “It’s axiomatic” remark, you have to consider that there simply is no way to measure interpersonal “intensities” of successes. That’s something that economists have had to face up to for a couple of generations. So, while it may seem that the fat guy’s wanting one more piece of pie is much less important than the mother’s wish that her baby not die of cancer, we really have no objective measures. But we can treat them as equal and simply count whether the desire is satisfied or not. That’s what this theory does–as you say–by stipulation. The choices are equal, but some are more fecund (and it’s likely that the baby living will be that, but it’s impossible to know). In any case, that’s all we can do. Without such an axiom, we’d either have to pretend we have a “God’s eye view” of what’s important or we could say nothing at all about prudential values. And successful free choices do seem intrinsically good.
Maybe evolution mavens will see some connections with discussions of “fitness” measures. Dunno.
You have to admit, it’s kick-ass!
Could definitely happen!
walto,
It’s almost certain, given the wide range of sheep prices.
At one time, some farms needed ‘chicken sexers’: they could tell the gender of chickens, but had no idea how. There are some interesting philosophy papers on characteristics of that sort. I mention this because not every animal is as easy to classify as sheep are.
Anyhow, choices, unlike sheep and billiard balls, can’t be measured or weighed at all. Not for any price. We can make ex ante estimates (good or bad ones) about their fecundity though.
walto,
I think that’s an overstatement. You can often get a good idea of relative intensities. Example: When I was a teenager, I worked at a livestock yard shoveling manure. I chose to take flying lessons despite the fact that it required 13 hours of shit shoveling to earn enough to pay for one hour of flight time (plane and instructor). Since I chose to keep flying purely for the fun of it, it’s not hard to discern that that I valued flying quite highly and considered my choice a success, even though it reduced my options for spending money on other things.
Compare that to a guy whose TV remote is broken. He doesn’t like the show that’s on, but he also doesn’t want to get up off the sofa, walk over to the TV, and change the channel to a better one. He decides to stay put and watch the crappy show.
It’s pretty easy to figure out that I value flying a lot more than that guy values the better TV show.
Again, that’s an overstatement. Here’s one way you could approach that question: Survey a bunch of fat guys and a bunch of mothers. Present the pie scenario to the fat guys and the cancer scenario to the mothers, and ask them how much they’d pay to achieve the desired outcomes: one more piece of pie for the guys, and the survival of their child for the mothers. Also obtain information about their household incomes, net worth, and number of dependents.
Correcting for all those factors, I’d be willing to bet that the mothers would pay far more for their children’s survival than the fat guys would for an extra piece of pie. Don’t you agree?
That would be a huge mistake in the scenario described above. Applying CHOICE in that case wouldn’t just be suboptimal — it would arguably be immoral.
There’s some middle ground between saying nothing and pretending to have a God’s-eye view.
walto,
We can also make ex ante estimates (good or bad ones) about their relative “intensities”. So why not do so?
The literature on interpersonal measures of utility is very large, and I don’t have much to add. Many agree with you on this matter, of course: lots of behavioral tests have been suggested as an antidote to Lord Robbins’ positivism. But I don’t myself think that anything follows from the examples you gave about you and the TV watcher or would result from production of the survey you suggest. We can surmise that something was high on your own preference scale and something else not so high on the watcher’s, but nothing follows from that that we can use–except maybe about social calm or unrest given various societal actions. I don’t deny that social calm may be a legitimate goal, but being a hothead is not a guarantee of anything. Maybe the pacifists are martyrs with huge capacities for undergoing pain without complaint.
It’s surely possible that the mothers would give more of what they have for their kids than the fat people would give for their extra slice. And heroin addicts might give more of their stuff for a dose than the mothers for the kids. The guy in M might pay more for a little girl to kill than the heroin addict will pay to get high. This shows importance to THEM, but the scales have to be coordinated to use ordinal rankings and I don’t think they can be. Suppose the mother loves breathing first, heroin second, then her child’s health. The last is now down to her third favorite. Do you suggest we make determinations by what percentage of their money they’ll pay? Don’t some people care more about money than others? Some want most to be richer than their peers. Some want just to get by.
Your comment suggests that you think consulting moral tenets is the answer to questions like that. I don’t know if CHOICE is immoral. Like intensities, I take access to moral truths to be beyond rational investigation.
Again, I don’t deny that some desires are more intense than others, only that measurements can’t be objective. They’re not like weight or momentum. I will mention, though, that the soda-picking example I gave in my OP on voting suggests that there are some proxies for intensity that AV can pick up. X may be a favorite, Y only passable, Z unacceptable. Those attitudes will have their effects on the results of elections using approval votes–even if we treat each vote as equal. Move those attitudes around and election results will change.
–Lionel Robbins, 1938
I watched the Schwartz TED talk. The stuff about depression resulting from higher expectations (that also produce self-blame) reminded me of Marx’s “dialectic of increasing expectations.” He really seems to like paternalistic care–doctors, shopkeepers, parents etc. telling him what to do and buy, whom to marry, etc. I suppose people like him could choose to reduce their available choices, maybe construct their own fishbowls to live in by, say, moving to Somalia or throwing away their computers. I think, though, that what he said about the Pareto-optimality of redistribution is mostly wishful thinking.
walto,
It isn’t just about Schwartz or “people like him”, or about moving to Somalia and tossing our computers.
It’s about the simple fact that having more choices doesn’t always increase well-being, contrary to the central assumption underlying your paper.
I can think of many events or conditions that coincide with reduced choices:
Illness, injury, poverty or reduced income, illiteracy, innumeracy, etceteracy.
There are some “good” things that reduce our choice:
Marriage, matriculation, book contracts and contracts in general, and so forth.
But in general, good things that restrict our future choices also open up otherwise unavailable choices.
Right.
In case it wasn’t clear, I disagree with him.
Whiniest rich guy since the Kochs.
walto,
This isn’t just a hunch on his part. The evidence backs him up.
What do you make of the study he mentioned showing that 401(k) participation declined as the number of fund options increased?
petrushka,
Right, and that’s a problem for walto’s CHOICE standard. According to CHOICE, they can’t be intrinsically good if they reduce the options available to us.
By walto’s standard, the only thing that matters is the number of choices. If the number of choices decreases, then so does your well-being. Even if otherwise unavailable opportunities have opened up for you.
The number theory is nonsense. Unless you are immobilized, you always have choices. You get a job, you can’t sleep in or drink heavily in the morning, but the job opens up new choices. It’s not about number.
petrushka,
I agree. An increase in the number of choices doesn’t automatically lead to an increase in well-being. Quantity can’t make up for a relative lack of quality.
I just don’t understand your cafeteria example. If the consequences are unknown or misunderstood, it’s not a clear choice. This kind of thing is the meat and potatoes of contract law.
petrushka,
According to walto’s CHOICE theory, all that matters it that the choice is freely made and that under ordinary conditions it will increase the options available to the chooser. Here’s the definition again:
In the cafeteria example, Linda chooses freely and her choice increases the options available to her. It is therefore “intrinsicaly prudentially good”, according to CHOICE.
That she ultimately regrets her decision doesn’t factor into CHOICE, which is blind to regret. Note that even if Linda knew in advance that the food at the Razzmatazz was inferior, CHOICE would still advise her that it was intriniscally good to select the Razzmatazz as her cafeteria for the year. It increases her options, after all. The inferiority of those options is invisible to CHOICE.
CHOICE is clearly broken. As I said earlier, the quantity of options can’t make up for their relative lack of quality.
Good judgement come from experience.
Experience comes from bad judgement.
Nothing whatever.
Yes.
No.
Exactly. Hence it IS about number.
walto:
keiths:
walto:
Best not to ignore it (and others like it). It challenges a core assumption of your paper — namely, that having more options increases our well-being.
keiths:
walto:
Yes. CHOICE deems her selection “intrinsically good” as long as a) it is freely made, and (b) it is likely under normal conditions to increase her options. Linda freely chooses the Razzmatazz, and there are more menu options there than there are at the Slow and Steady. Both criteria are satisfied, and so CHOICE renders a verdict of “intrinsically prudentially good”.
Having additional options is always good, in spite of schwartz’s whining. As he says at the bringing of his talk, freedom/autonomy is a central feature of what we mean by well-being. I mean I know Schwartz wishes he didn’t have so many kinds and cuts of jeans to choose from, but (i) irrelevant options (those we don’t care about) are neither here nor there, as i’ve already explained, and (ii) wah wah wah 😪
As for your other complaints, I’ve already replied to them too. Taking something you don’t like/want is hardly a success.
I am not arguing either keiths or walto; just thinking out loud.
I am thinking that number of choices is pretty easy to fudge, as is quality. Neither is objective, even in the informal sense of the word.
I am thinking of how kids learn, and how regertted choices contribute in a positive way to our lives. Assuming they are not catastrophic.
This is true.
Here, “objective” needs a definition. With counting, it’s clear that there are any number (an infinite number, in fact) of counting conventions that may be used. But if one is consistent, we can say that X leaves you with more than Y does–and mean something objective by it. So, for example, if we’re counting half-breaths with X, we have to do the same thing with Y. Same if we’re just counting the obtaining of “bucket list” items with X.
But, yes, it’s easy to fudge and hard to estimate. Many objective things are.
walto,
Not when it paralyzes us, makes us unhappy or causes us to act against our best interests, as in the retirement study.
Yes, he acknowledges the importance of autonomy. But instead of saying simplistically that more choice is always better, he adopts a more nuanced (and evidence-based) view.
Your “official” definition of CHOICE says nothing about success:
If you want to apply a success criterion, you should amend the “official” statement to include it.
You do mention success elsewhere in the paper, but the only example you give of a successful choice appears to be this one:
By that criterion, success amounts to obtaining what you choose.
Also, when you write…
…you are (perhaps inadvertently) acknowledging that CHOICE doesn’t stand on its own, but is instead an add-on to desire satisfaction (DS) or hedonistic (HED) theories of prudential value.