Elizabeth Anderson has a paper where she does something analogous. However, it was Amartya Sen who made this type of cretinism possible.
AMARTYA SEN'S CRITIQUE OF THE CONCEPT OF PREFERENCE
The concept of preference dominates economic theory today. It performs a triple duty for economists, grounding their theories of individual behavior, welfare, and rationality.Preference is a 'primitive notion'. It is wholly undefined. It does not assume anything. When a mathematician says 1+1= 2 he does not assume that urinating twice is equivalent to taking a dump.
Microeconomic theory assumes that individuals act so as to maximize their utility that is, to maximize the degree to which their preferences are satisfied.This is not true. There are mathematical models in Micro-econ where either Utility or Revealed Preference- both of which are undefined- are subject to constrained optimization. Satisfaction is not mentioned. There is a separate theory of 'satisficing'
Welfare economics defines individual welfare in terms of preference satisfaction or utility, and social welfare as a function of individual preferences.Some Welfare economists may do so. However, their models do not permit this. Why? Because 'preference', 'utility' and 'social welfare' are all primitive notions. They are used to define other terms but remain undefined themselves. Tarski made this clear. Arrow was his student. The thing is common knowledge.
Finally, economists assume that the rational act is the act that maximally satisfies an individual's preferences.This is only as true as the proposition that Arithmeticians assume that two urinations equal one defecation. No Economist is so stupid as to think anybody, including himself, 'maximally satisfies' anything. The cognitive effort is not worth it. However, for a large enough sample, ceteris paribus, Economists make predictions about the direction in which a market will go on the basis of mathematical models.
There is a separate literature on what counts as Rationality- does it involve transitivity of some weakened sort, or a generalization of the Spilrajn extension theorem, or can it feature a 'money pump'? Can it be non-algorithmic? If so, how?
The habit of framing problems in terms of the concept of preference is now so entrenched that economists rarely entertain alternatives.Mathematical models in which preference is an undefined term have a market. Thus economists are paid to frame problems in terms of such models. Suppose, instead of supplying such a model, they choose to dance naked on the Conference table. They won't get paid.
If you see a guy go to his place of work every weekday, you don't say he is in the habit of going to work. You say he is paid to go to work. He may hate it, but he has to do it.
A habit is something you keep doing even if you are not paid to do it. If economists don't get paid for producing a certain sort of study, they will stop doing it. Why pretend they have some bizarre cacoethes scribendi which features a type of stupidity no actual human being has ever subscribed to?
In this commentary, I would like to explore and extend Amartya Sen's critique of the concept of preference.There is no concept. There is only a primitive notion. Sen's critique is self-aggrandizing hot air. He is tilting at windmills which exist only in his mind so as to pose as some great Mother Theresa type figure.
A critique of a concept is not a rejection of that concept, but an exploration of its various meanings and limitations.A critique of a concept is another concept. It is idiographic and idiosyncratic and reflects the ignorance and stupidity of the mind that holds it. One person may have many different concepts of the same thing over any given span of time.
One may pose as a great Liberator by critiquing concepts. However, they are neither univalent nor action guiding and thus have no impact on real world behavior.
If we want economists to do something different, stop paying them for doing what they are doing and pay them to do something else. Critiquing some concept you impute to them is silly. Why not just say- 'Stop being such almighty dicks, you fucking dicks. Straighten up and fly right. As for this genocide business, just cut it out, already. Take a cold shower or something. Also, no more raping the Environment. The Economy is getting jealous.'
One way to expose the limitations of a concept is by introducing new concepts that have different meanings but can plausibly contend for some of the same uses to which the criticized concept is typically put.Concepts are useless. Mathematical models may not be.
The introduction of such new concepts gives us choices about how to think that we did not clearly envision before.Smart people don't need to be introduced to new concepts. Stupid people can be introduced to as many concepts as you like, they still won't know how to think.
Before envisioning these alternatives, our use of the concept under question is dogmatic.Anderson's use is dogmatic. Economists' use of models is practical. They get paid to do a study and make a policy prescription. Amazon is not hiring economists to sit around preaching sermons. They have to figure out ways to cut costs and boost profit margins. They use mathematical models and Big Data to achieve this end.
We deploy it automatically, unquestioningly, because it seems as if it is the inevitable conceptual framework within which inquiry must proceed.Anderson may do so. But she isn't paid for saving people money. Rather it is for repeating some libtard dogma to Credential craving careerists that she makes her moolah.
There may be Economists who 'automatically, unquestioningly' do stupid shit. They work for the Govt. or some International body.
By envisioning alternatives, we convert dogmas into tools: ideas that we can choose to use or not, depending on how well the use of these ideas suits our investigative purposes.The word dogma comes from Religion. Which dogma was ever turned into a tool? Consider the dogma of the perpetual virginity of the Theotokos. Prostitutes may try to convert this dogma into a tool. They don't succeed. Nobody really believes they have been re-virginified and thus can command a premium.
Within economics, the representation of human behavior and normative judgements in terms of the concept of preference has long functioned as a dogma.A dogma is not an undefined term. The Preference relation in mathematical economics is undefined because it is a 'primitive notion'. An economist may say that other economists are very stupid and have a wrong interpretation of the model. Why should we believe such an economist? Clearly, the fellow voluntarily devoted himself to a stupid profession.
Sen has done more than any other economist to convert this dogma into a tool, by offering alternative tools for economists to use, for both explanatory and normative purposes.Where are these tools? All I can see is bullshit. What 'explanatory' purpose has Sen-tentious shite served? He explained that India must spend a lot of money building a worthless University in Nalanda because...well, there used to be one there a thousand years ago and so...urm...obviously poor Bihari people need an International University there and all the lecturers should have diplomatic immunity, even if they are Indian citizens.
The problem with Sen's work is that it makes meaningless or mischievous distinctions which make the Policy Space multidimensional and thus open the door to endless bureaucratic delay and the McKelvey Chaos associated with Agenda Control.
A good starting point for exploring Sen's instrumentalizing of the concept of preference for explanatory purposes is his classic paper, `Behavior and the Concept of Preference' (Sen, 1973). In that paper, he argued that there is a profound conceptual ambiguity at the heart of the economic theory of rational choice.There is no ambiguity. Preference is an undefined term. Sen himself may, without any warrant to do so, define Preference in different ways. However, this is illegitimate. He may say- 'all the other Economists are doing it', but the answer is 'if all the other economists jump off the roof, will you do so too?'
Economists use the same concept of `preference' to perform three distinct tasks: (a) to describe a person's choices;Economists have no warrant to do so. Preferences are one type of primitive notion. Choices are another. Constraints are a third. A mathematical model specifies a functional relationship between these three different undefined terms.
I may say 'I prefer dosa to idli. So I chose dosa'. This is an explanatory, not a descriptive statement.
(b) to represent whatever motives underlie a person's choices; andIn standard Economics, the motive is Preferences subject to Constraints. Thus we say 'I prefer dosa to idli. But I had to chose idli because the kitchen had closed and so no more dosas were being made. Furthermore, I was hungry and pressed for time. Thus I could not wait to go elsewhere to get dosa'. Clearly, my preferences are only a small subset of my motivation. How can they 'represent' it?
Sen is being a fool. Surely, even Anderson should be able to see it.
(c) to represent a person's welfare.This is crazy shit. We all know that our preferences- for fatty food and sugared drinks in my case- can undermine our welfare. Nobody says Preferences represent Welfare. They specify the conditions under which this would hold.
Sen pointed out that these are conceptually distinct, and that in particular one is not entitled to infer that a particular choice advanced the individual's welfare just because she made it voluntarily.Everybody already knew this. That is why Econ 101 spells out all the conditions that have to be met for a competitive equilibrium to be allocatively efficient. Sen contributed nothing new at all.
Sen was later to elaborate this distinction, between preference in the wide sense of whatever states of affairs one values, and preference in the narrow sense of personal welfare or self-interest, as the distinction between agency and welfare (Sen, 1985).Agency means what you can do. Welfare means what you get to enjoy. Since one may not value what one enjoys- for example for moral or spiritual reasons, or simply so as to preserve your figure, or for a reputational reason- there is no necessary connection between 'preference in the narrow sense' and welfare. Nor is there any necessary connection whatsoever between states of the world and what one can actually do. In any case, Agency and Welfare are after the fact outcomes. We may think we have Agency but discover differently. Welfare, too, may turn out to have diminished by our self-regarding actions. Madoff's investors, Greek pensioners and the people of Venezuela have found this out the hard way.
Sen was talking nonsense and we can all now see why. His approach could be accommodated in an Arrow-Debreu framework. In other words, he neglected Knightian Uncertainty. Thus his theory was useless. It would only hold in a world where there was no need for Language or Law or Commerce.
His second point was more momentous. He argued that even if we consider preference in the wide sense, a person might not choose in such a way as to maximally satisfy her preferences.Nobody can choose to 'maximally satisfy her preferences' because constraints also exist.
That is, she may act on some principle other than the maximization of her utility, even where utility was understood in the wide sense as encompassing any states of affairs she might value.She can't not act on the principle of maximization of utility, subject to constraints, because utility is a 'primitive notion'. It is undefined. Someone could always find a Utility function, or Revealed Preference profile which fits her behavior.
Sen illustrated this point by contrasting two alternative explanations of cooperation in a prisoner's dilemma. First, a person could be sympathetic, and care about the welfare of the other party to the dilemma. This is a case of maximizing utility.Because everything is. Utility is undefined. It can be anything which fits the facts.
But second, a person could be motivated by moral principle, a social norm of responsibility, or what Sen was later to call `commitment' (Sen, 1977). In the second case, the individual still prefers to serve less time in prison rather than more, and does not personally care about how well-off the other party is. Her choice cannot therefore be rationalized in terms of maximizing her utility, either in the narrow (self-interest) or the wide sense.Why not? We can either model the 'commitment' as a constraint or just work backwards from her observed behavior to find a utility function or revealed preference profile which fits her.
In a footnote, Anderson observes-
As Sen described the case, which he labeled `sympathy', it was a matter of maximizing utility in the narrow, egoistic sense.Nonsense. If one's Utility function includes the suffering of another, the way to maximise it is to constrain one's information set so that one only gets good news about that person. Indeed, it is likely that if the other person has a similar 'sympathy', they will confirm that they are flourishing not suffering.
More generally, there is a problem with this approach because it incentivizes 'managing the news'.
The sympathetic party feels pain at observing the suffering of others, and helps them in order to relieve his own pain. This is distinct from the case of genuine altruism, where an individual helps out of direct concern for the other. This would be a case of maximizing utility in the broad sense.There is no distinction between the two. A genuine altruist would not wish to be impassable because actual sympathetic suffering is a superior motivator.
All this has nothing to do with maximizing utility because it is an undefined term which can always be shown to have been maximized under plausible constraints.
Anderson thinks a person can act in a manner which prevents a mathematician ascribing a Utility function to her after the fact.
She write-
... she decides to act on a principle of choice other than to maximally satisfy her personal preferences.This changes nothing. We just say she gets Utility from acting in a different way.
She suspends individualistic calculation of consequences, acting as if she cared personally about the other party, even though she does not.She is a fool, but gets utility from her foolishness. She is still maximizing utility as far as the model goes.
Thus, for explanatory purposes, Sen instrumentalized the concept of preference in two ways: first, by disambiguating the concept, replacing it with three distinct concepts (choice, underlying motive, and welfare), and second, by articulating an alternative model of behavior, commitment, that was not framed in terms of preference satisfaction at all.So instead of one primitive notion, he gives us a whole bunch. What does 'underlying motivation' mean? It could be anything. You say, 'my underlying motivation for ordering dosa is so as to eat something tasty and fill my belly'. I may reply 'False! Your underlying motivation is to gain nutrition so as fulfill your destiny as a P. Chidambaram lookalike auctioning your anal cherry on Ebay.'
What does 'commitment' mean? Anything you like. Your commitment to not committing to being a P. Chidambaram lookalike is the underlying motivation for your fulfilling your destiny as an unwilling anal cherry vendor on that precise market because, as everybody knows, P. Chidambaram is straight.
In place of one dogma, he gave us four tools.Tools to do what? Has this idiot achieved anything? In a dogmatic manner, he gasses on about how all the other economists are very evil because of some stupidity which exists only in his own head. He said, his old colleague, Manmohan Singh was so cruel and foolish as to try to turn India into an economic super-power while deliberately starving the workers and leaving them prey to all sorts of horrible diseases.
Stupid lies are not tools. They are a way of gaining the respect of fools.
Sen's wide-ranging critique of the normative uses of `preference' is even more well-known. Within normative economics, the concept of preference is used to (1) make judgements about individual welfare; (2) make judgements about the overall good of society; and (3) articulate a principle of rational choice (utility maximization).So what? Normative economics is shite. Just having a YouTube video of a very poor woman explaining how shitty her life is and how it can easily be made less shitty is all that is required.
Sen has criticized excessive reliance on the concept of preference on all three fronts. With respect to judgements of individual welfare, Sen points out that none of the three explanatory concepts of `preference' (choice, underlying motive, and perceived self-interest) quite does the job of offering a good measure of a person's well-being.So, the tools aren't up to the job. But the job was worthless. It is worthwhile estimating a person's income or wealth- this helps target marketing and taxation- and it is useful to compare one's income or wealth with others to see whether there is a trick you are missing. However, there is no point measuring other people's happiness or well being. You can do nothing with that information. You automatically monitor your own well-being in the same way that you breathe. It would be foolish to seek to measure it. By contrast, it may be worthwhile to measure your skill or level of knowledge or credit worthiness or many other things which are linked to productivity and potential income.
As for normative economics, don't do it. Economize on the use of scarce resources- like your student's time. Norms are what priests and pop-stars are better at talking up.
'Second order' public goods (i.e. agitating for more public goods) crowds out first order provision. That's what happened to Sen's native Bengal. Everybody went on strike demanding a Revolution and Social Justice and so on. Much good it did them. Those who could, ran away to gas on about Social Justice in places where they would be ignored as stupid darkies from a shithole of a country.
Preference conceived as choice or motive fails to (measure well-being)For the same reason that your bathroom scales don't tell you how tall you are or what the color of your eyes are.
This is
because people have motives wider than and sometimes even counter to their self-interest, and often choose accordingly.It is because Preference does not mean the same thing as Well-being just as weight does not mean the same thing as height.
Preference understood as perceived self-interest also often fails as a measure of welfare, particularly with respect to the severely disadvantaged.Preference is not perceived self-interest. Perceived self-interest is. I might say 'I prefer to dosa to idli, though, in my perception, it is in my self-interest to eat idli and not dosa because the latter is more fattening and I suffer from morbid obesity'.
It may so happen that your weight in pounds is also your height in centimeters. However this does not mean that your bathroom scales tells you anything about your well-being coz you just got fired from your job and your wife left you and the cat threw up on the rug.
Anderson is obviously seriously disadvantaged because she thinks words which mean different things are used by Professors who are smarter than her to mean the same thing. This is a good example of the following-
Seriously disadvantaged people often adapt their self-interested preferences to their limited opportunities: they lower their aspirations to avoid frustration.Since Anderson can't contribute to Economics, she lowers her aspirations to talking up a stupid Economist who says all other Economists are stupider than he. She has lowered her aspirations all right. But we feel frustrated by her remorseless stupidity.
`The extent of a person's deprivation . . . may not at all show up in the metric of desire-fulfillment, even though he or she may be quite unable to be adequately nourished, decently clothed, minimally educated, and properly sheltered' (Sen, 1992, p. 55).This is nonsense. The metric of desire-fulfillment will be a weighted average of different sorts of desires- that for food, housing, clothing, education etc, etc. Depending on the metric, deprivation will show up pretty damn fast. Indeed, the Government need only to look at money income and housing to identify deprivation.
With respect to judgements of the overall good of society, Sen argues that utility information alone is insufficient to ground ethical or socially rational evaluations.Utility information is not available at all. Money income, type of housing, level of education etc. are measurable. These suffice for most evaluations. If not, one can commission a statistical survey.
This of course follows from the fact that utility is not a good measure of welfare.Utility is a primitive notion in a certain type of model. Welfare is also a primitive notion in a certain type of model. Both are undefined. Neither measures the other.
But in addition, the amount of freedom people enjoy is important, over and above how much welfare they get from choosing their most preferred option (Sen, 1985).In that case the amount of unfreedom to do harmful things is also important. So what? Comparisons are only made ceteris paribus. It is unlikely that the volume of freedom or unfreedom has changed. So the thing can be ignored in the same way that we must ignore lots of other things like whether it is a sunny day or whether Fulham won the F.A Cup.
It also independently matters that social outcomes be produced by just, fair, and rights respecting procedures (Sen, 1995, p. 13).This is covered under ceteris paribus. After all, it also independently matters that there be no major earthquake or tsunami or that Fulham is relegated.
Why does Sen not stipulate that social outcomes be produced by, salubrious, humane, hygienic, culturally appropriate means? Does he not understand that if social outcomes are produced by people who don't wash their hands after wiping their bums then diseases could be spread? A procedure may be just, fair and rights respecting but it may involve being in close proximity to people who fart in your face. This is not salubrious, it is not humane, and- unless you are Swiss- it is not culturally appropriate.
And fairness in the distribution of outcomes matters, too, beyond the total `amount' of welfare enjoyed in society (Sen, 1995, pp. 9±10).No it doesn't. Those who say so aren't in any hurry to share their wealth with their less fortunate colleagues.
Again, in place of one dogma, Sen gives us many tools.What tools? Which metric has he constructed? None at all. What formula has he discovered? Zilch. All we have is dogmatic assertion and a complete absence of common sense.
All of these observations call for enriching the informational basis of individual and social welfare judgements beyond that provided by the concept of preference.There already were plenty of Statistical Surveys commissioned by the Public, the Private and the Voluntary sector for 'enriching the informational basis' of decision making. This is a lucrative industry. Sen didn't do that type of Statistics. He preferred posing as a Philosopher.
Does Anderson think she herself is providing anybody any 'tools'?
Sen has proposed `capabilities' and `functionings' as appropriate objective measures of freedom and welfare, a judgement with which I concur (Anderson, 1999).That was 20 years ago. Why are there no 'capabilities' or 'functionings' indices anywhere? There are plenty of other economic indices- including Sen's pal's Human Development Index and the similar Multidimensional Poverty Index endorsed by his collaborator- but nothing from Sen himself. It is now cheaper than ever before to construct such things. Why has no one done it? The answer is that both ideas are vacuous.
What about judgements of rationality? Here, I think, there is a lacuna in Sen's work. Sen's analysis of utility maximization in prisoner's dilemma situations certainly suggests that the individual who always acts on his preferences can be a fool and a social misfit in circumstances where acting on social norms of cooperation brings about better consequences for all (Sen, 1977).Utility maximization is bad if Knightian Uncertainty exists. Thus 'regret minimization' is the way to go.
That's it. Economics gets rid of this foolish availability cascade with just two sentences.
Yet, Sen does not propose an alternative, non-preference-based conception of rationality in terms of which committed action makes sense.Preferences relate to the here and now. Regret is about looking at a life-time. For inconsequential actions, there is no difference between the two approaches. However, the latter is better at guarding against catastrophic risk as well as for taking a punt on mega pay-offs.
To be sure, committed action is often socially and ethically desirable.The Law makes committed actions of a certain type obligatory in a wide range of economic situations. It may be socially and ethically desirable to change the law. it is not socially and ethically desirable to encourage 'committed actions' absent any bond of law. The result would be vigilantism or some other such nuisance good.
But this only raises the question of how it can be rational for the individual to act on socially and ethically desirable principles, when so acting does not advance the satisfaction of her broad or narrow preferences.The answer is, the Law decides.
Moreover, not all committed action is desirable from a moral point of view.Only committed actions under a bond of law are desirable.
Recall that the original setting for the prisoner's dilemma involves two presumably guilty co-conspirators who have an interest in getting away with their crime.But the Law says they should confess. That's the end of the story.
Although, given their right against self-incrimination, they are within their rights in remaining silent, one can hardly deem their silence as morally desirable.The right of self-incrimination may- as has happened in Britain- be amended or abolished. That is a purely legal matter. However, from the point of view of morality, there is an obligation to confess your guilt and make reparation.
From a moral point of view, it would be better if either or both confessed to their crimes. It is only from a point of view including the criminals alone that `cooperation' in this prisoner's dilemma is desirable.This sort of stupidity is only possible when you read an idiot like Sen and thus forget that the Law exists.
In this essay, I shall explore this lacuna in Sen's work.No she won't. She is too stupid. She does not know about 'regret minimization' and has not noticed that the Law exists. What 'exploration' can she do?
I shall argue that a full understanding of the rationality of committed action requires us to enrich the information basis of the theory of rationality beyond the concept of individual preference, in two ways. First, we need to devise a non-preference-based conception of reasons for action.That's the Law right there! The Banker can't give his mistress an overdraft, though he'd like to, coz he'd be breaking the law.
Everybody else already has a non-preference-based conception of reasons for action except Anderson! That's what reading Sen will do to you. It will make you stupider than you were born to be.
Second, we need robust conceptions of collective agency and individual identity.The Law is particularly robust collective agency. It has various concepts of juridicial identity with respect to the individual. That's all that is needed. The Law itself can institute inquiries and enrich its informational basis. Nothing is required of stupid pedants in this regard.
We can find hints of both ideas in Sen's work, which I shall develop further.Nonsense! The man is an idiot. He never said anything sensible in his life.
Briefly, committed action turns out to be action on principles (reasons) that it is rational for us (any group of people, regarded as a collective agent) to adopt, and thus that it is rational for any individual who identifies as a member of that group to act on.This is silly. Adverse selection, Moral Hazard and the Free-rider problem militate against this conclusion.
Consider a group of academics interested in the same subject. Clearly if they all wrote alethic, high cognitive value, original papers, then their subject would become more worthwhile. However, there is a problem of adverse selection. Nutters who are interested in the subject for a paranoid reason will try to invade the space. Even if we can screen them out- for example by demanding PhD's from contributors- there is still the problem of Plagiarism and faked Results. Peer-review is costly and, at the margin, has to proceed in good faith. This means that there is a Moral Hazard. Since risk of detection is low, faked papers are submitted and published. Even where the faker is named and shamed, the fact that his whole profession, through its Journal, has been shamed means that the penalty he suffers is reduced. Finally, there is the free-rider problem. People can submit non-fake but vacuous papers and get a reputational bump from appearing in a Journal with some alethic content.
This is an example from Academia illustrating a process which both Sen and Anderson must be aware of- unless they are self-consciously gaming the system themselves in a manner reminiscent of Spivak's notion of 'affirmative sabotage'.
We can then build on this idea to get from the rationality to the morality of committed action.No we can't. We either have to have the Law and what Mill called 'punishability' or the group has to be constituted as a 'separating equilibrium' on the basis of a costly signal. But, in that case, it isn't self-selecting at all. It is elitist.
If it would be rational for a collective encompassing all of humanity to adopt a certain principle of committed action, then action on that principle is morally right.No. It may be immoral or morally indifferent. A regret minimizing course would be to include an antagonomic preference profile. Unanimity can be unanimity in error. Better, as Robert Aumann observed in connection with the Talmudic proscription of unanimity in Sanhedrin judgments, to think twice before pronouncing judgment.
It is then rational for anyone who identifies as a member of this cosmopolitan community of humanity - we could call it the Kingdom of Ends - to act on such a principle.No it isn't. Robert Aumann is genuinely smart- unlike Sen. It is not rational to act on a principle absent an obligation to do so under a bond of law.
I thus propose to develop some of Sen's ideas in a more Kantian direction than he has been willing to go. In the last section of this commentary, I shall consider some implications of this analysis for understanding the plight of women across the world.
WHY THE ECONOMIC THEORY OF RATIONAL CHOICE CANNOT GROUND A GENERAL SOLUTION TO PRISONER'S DILEMMASThe theory of Rational Choice gives rise to Game theoretic dilemmas. They don't otherwise exist. To say the theory can't solve the dilemma is to say something about the theory. Anderson is too stupid to do any such thing. By contrast to say some other theory can solve a Game theoretic dilemma is to say 'Because of ability to say meow, I can prove the abc conjecture. Here goes...meow, meow, me---ow. There- you see?- it's all proved now.'
To see more clearly how committed action is problematic from the standpoint of rational choice theory, let us get the phenomenon clearly into focus.This is as foolish as saying 'the number 1 can't go number 1. This is a huge problematic from the standpoint of Peano Arithmetic because it desperately needs to take a piss, but can't.'
Suppose some people face a situation in which the payoffs take the form of a prisoner's dilemma. That is, for each party to the dilemma, non-cooperation will bring about a better or preferred state of affairs than cooperation, regardless of what the other parties to the dilemma do.Then, there is no game. Each goes her separate way to play 'a game against nature'.
Non-cooperation dominates cooperation.Not playing dominates. This is what happens most of the time. Suppose we have a choice between all shitting side by side. We could do this co-operatively or competitively. In practice, we do neither, preferring to go our separate ways to poop.
However, if everyone cooperated, this would bring about a better state of affairs from each person's perspective than if no one cooperated.Anderson just contradicted herself. She said 'for each party to the dilemma, non-cooperation will bring about a better or preferred state of affairs than cooperation, regardless of what the other parties to the dilemma do.' Now she says if everyone cooperated this would be better for everyone.
The puzzle is to understand how it can be rational for any of the parties to cooperate under these conditions.It can't. Anderson is irrational and talking nonsense.
Proposed cooperative solutions to prisoner's dilemmas within preference-satisfaction theories of rational choice try to show that the payoffs recorded in the dilemma do not reflect all the preferences of the parties.OMG! how stupid can Anderson be? Does she not know about Omerta & 'snitches get stiches' and so forth? Professional criminals have evolved code
to take care of this problem.
As for cooperative games, you need to look at Aumann correalted equilibria and the folk theorem of repeated games.
When the parties' full preferences are taken into account, they really would maximize their utilities by cooperating. There are two main ways to broaden people's preferences in the required way.One way is cranial surgery and the other is talking nonsense till the cows come home. Which will Anderson choose?
First, one may suppose that the parties have sympathetic or altruistic preferences.Then one may suppose that they will let you cut their cranium's open.
They care not just about the payoffs to themselves, but about the payoffs to others. Second, one may suppose that the act of cooperation is valued intrinsically, as a consequence in itself, and not just valued for the sake of the other consequences it brings about. Neither proposal shows that it is rational to cooperate in a prisoner's dilemma. Rather, they assert that the payoffs to the parties, once fully accounted for, do not really have the structure of a prisoner's dilemma. But it is not so easy to evade the problem of prisoner's dilemmas in these ways.But it is easy to evade them by actually learning Game theory to an expert level.
Consider first the altruistic evasion. This supposes that if only people cared about other's interests, prisoner's dilemmas would not arise. This is an error. The key feature of the payoffs that generates prisoner's dilemmas is not the fact that people care only about the payoffs to themselves. It is the fact that any single person's action, considered in isolation, has an expected marginal payoff of zero, or close to zero, with respect to the socially desirable outcome.So, Anderson has jumped from prisoner's dilemma, where individual actions have big payoffs relative to the 'socially desirable outcome' which is (both confess and both get a long prison sentence) , to the free-rider problem for public goods (stuff like, if I don't buy a TV license the Beeb won't actually go out of business).
Such a situation can therefore occur even if everyone has altruistic or public-spirited preferences.No. If they prefer to pay their share then they genuinely have public-spirited preferences. If they don't, they don't. Perhaps Anderson means 'meta-preferences'. But those are meaningless. It is only what you do which counts.
Thus, it arises in almost any n-person public goods case, even when people value public goods for the sake of others besides themselves and think they ought to be provided. Consider, for example, the case of voting. Democracy would collapse if the people did not go to the polls, so, assuming democracy is good for the people, mass voting can be regarded as a public good. Moreover, from the perspective of the supporters of any particular candidate on the ballot, the election of that candidate is also a public good. That is, supporters suppose that the election of their candidate would be better for society, not just for themselves, and may well prefer their candidate for this public-spirited reason. Yet when deciding whether to vote, each partisan who accepts the principle of expected utility reasons as follows: regardless of how my fellow partisans vote, the chances that my ballot will make a difference to the outcome of the election are negligible. Therefore, if there is the slightest inconvenience to me (or -thinking altruistically now- inconvenience to others!) from my voting, this will certainly outweigh the expected marginal positive impact of my voting. But there is always some inconvenience. So I ought not to vote. The conclusion follows not because each partisan is selfish, but because each partisan correctly reasons that her marginal impact on the outcome each prefers from a public-spirited standpoint is negligible.Jason Brennan has put forward this same argument recently. It fails for reasons I discuss here.
What generates prisoner's dilemmas, then, is people's acceptance of a principle of rational choice that has an act-consequentialist form.
Nonsense. Game theory generates dilemmas so as to get fresh insights and cobble together better tools. Reverse Game theory is 'mechanism design' which figures out how to get people to do what is socially optimal by changing the incentive mix they face.
A pure deontologist may be observationally equivalent to an act-consequentialist and vice versa. It does not matter. Nothing hinges on 'people's acceptance' of some pedant's mumbo jumbo. By contrast, acceptance is important for the conduct of the Law. Juries won't convict for crimes they regard to be venial simply.
As long as people judge the value of their action in terms of its expected marginal causal impact, their cooperation in prisoner's dilemmas cannot be counted on, even if their underlying preferences are altruistic.The Dilemma was an actual mechanism used by Law Enforcement. It had the effect of small fry doing long years in porridge while the big fish turned informer and did a Whitey Bulger.
The second strategy for evading prisoner's dilemmas within an act-consequentialist framework is to postulate that the parties value cooperative action intrinsically, apart from its consequences. This is Sen's favored strategy for resolving the paradox of voting: to postulate that people enjoy the act of voting in itself, apart from its consequences (Sen, 1997, p. 750).So the man is as stupid as shit.
Anderson observes in a footnote-
In his replies to a version of this paper delivered at the American Philosophical Association Central Division Meetings (Chicago, April 22, 2000), Sen observed that he had offered this only as an explanation of why some people vote, not as a consideration that could make voting rational. My subsequent comments therefore address not Sen himself, but anyone who thinks attaching a value to the act of voting itself could make voting rational.Voting is rational because people have Muth Rational Expectations- in other words they act in line with the predictions of the correct Economic theory. Sen, unlike Anderson, knew this, but was too stupid or lazy or politically biased to incorporate it in his work coz he wasn't interested in having the correct Economic theory. He just wanted to write sen-tentious shite that made out he was Mother Theresa and everybody else was a sociopathic rapist.
I do not believe that this attempt to rationalize voting makes sense. If the act of marking ballots with one's preferences really did have no further consequences, it would be absurd for people to value it. Suppose an oligarchy announced that, henceforth, the people would be allowed to mark their preferences for candidates on secret ballots. Only the ballots would be burned without being counted, and candidates would be selected by the oligarchy using its traditional undemocratic methods. Here would be an opportunity for people to `vote' for its own sake, apart from its consequences. But only a fool would value it. One might object: but that is not really voting. This would be right. But that is just to admit that the act of voting makes no sense apart from an appreciation of its causal role in selecting public officeholders. And this, I would bet, is how most voters view the matter as well. When they go to the polls, they generally do so with the end in view of helping their favored candidates get elected. And the same could likely be said about any actions, such as paying taxes, that collectively result in the production of public goods, although any of these actions taken in isolation have negligible marginal impact. People would not, in general, find any value in such acts if there were no causal connection between them and the production of the public goods in question. In any event, such an ad hoc solution to prisoner's dilemmas does not have the general features needed to vindicate the rationality of cooperation.John Muth was a smart guy who knew from Math. His idea is not ad hoc at all. Indeed, Rational Expectations is mathematically similar to the 'Kalman Filter' used in Engineering and Control theory.
In `Rational Fools', Sen suggests that those who fail to cooperate in prisoner's dilemmas, however much they may be maximizing their expected utilities, are acting foolishly.Why did Sen not mention Muth's result? How retarded was he?
But it is hardly foolish to not prefer the act of cooperating in itself, apart from its consequences. What is foolish about non-cooperators is not their preferences, which are perfectly understandable, but their principle of rational choice. And what makes that principle foolish is its act-consequentialist structure.No. It is because they don't have Rational Expectations and thus fail to act according to the prediction of the correct economic theory. It is a separate matter that some relevant mechanism might be broken.
There is no way of distinguishing a deontological scruple from a consequentialist imperative. After all, deontics evolved on the same fitness landscape as aetiology. In any case, to ensure robustness, the two are likely to be mixed because of Knightian Uncertainty and Cognitive costs.
Any principle of rational choice that evaluates an individual's act solely according to its marginal causal impact on valued outcomes will meet the same difficulties.Not Rational Expectations because it is statistical. The individual is only important because he is part of a population. It doesn't matter if 'noise' cancels out. Anyway, low participation means there is a missing market or broken mechanism so there's something useful economists can be doing instead of jawing on about rational fools.
This is one powerful reason why many people are drawn away from act-consequentialism toward rule-consequentialism, or toward non-consequentialist frameworks.How many? One million? Ten million? A hundred million? Of course not. Don't be silly. Anderson means people like herself of whom there may be a few hundred. But they don't matter in the slightest. Why? Because they are stupid and ignorant and teach a worthless subject to gormless adolescents.
THE RATIONAL BASIS OF COMMITTED ACTION
When people face a genuine prisoner's dilemma - that is, one that retains a PD payoff structure even when consequences are valued unselfishly and any intrinsic preference for performing the cooperative act is factored in -cooperation can only be rationalized in terms of a non-act-consequentialist principle of rational choice.Either cooperation occurs or it doesn't. If it does the payoff structure could not have had the property Anderson claims. Some mistake in imputed utility must have been made.
If there is no cooperation then we need to think about what type of public signal, or preference revelation mechanism, or external coercion can remedy the situation. That is the sort of thing Economists get paid to do. They aren't paid a great deal but then most aren't very good at their job. Still, it's early days what with Big Data and A.I's and so forth.
Thus, people's cooperation in such cases must be based on a principle of choice other than the maximum satisfaction of their preferences.I'm afraid this isn't true. If you have a principle then you get imputed utility by abiding by it. Since utility is a primitive notion it is undefined and can be anything at all after the fact- to make the sums come right.
This is what I, following Sen, have been calling `committed action'. Sen agrees that some people do engage in committed action for reasons that are not properly represented in terms of maximizing utility.He has no business doing so.
Such action is rather a matter of following social conventions, norms, or conceptions of ethical duty.All of which yield utility.
If we regard cooperation in some such cases as socially desirable, we need to find an alternative account of rational action that does not define it in terms of its marginal causal impact on desired outcomes.Why? We can always fudge the numbers afterwards. It is foolish to come up with some 'alternative account of rational action' which everybody will be able to see is completely ignorant and irrational.
The question is, then, what other principle of choice there could be, and how to understand it as a rational principle. Sen offers us several hints in disparate writings from which we can build a coherent account. First, he suggests that what people do depends on their understanding of their identities, which may be constituted by membership in various social groups.But people's understandings of their identities may depend crucially on what they do. Membership in various groups may be constituted by their actions not their pre-existing identities. There is no way of showing that the influence from action to identity will be greater than that from identity to action. Thus, there is no hope of finding a superior Structural Causal Model here.
For example, he points to data suggesting that Indian women tend to conceive of themselves more as members of their families than as individual selves, and choose accordingly (Sen, 1990, pp. 125±6).Unfortunately, this data is statistically worthless because of glaring methodological errors and simple, human, fraud. Indian economists are fully aware of this scandal. Private sector firms which conduct polls adopt a better methodology and are more alert to fraud. Otherwise, they lose business.
People may also identify with their occupation, the firm where they are employed, various associations and clubs, their nation, caste, religion, and so forth, and choose on the basis of these identities.So what? People may also identify with the Nation, the Species, all sentient beings, God, etc. Unless there is some dramatic change in identity which does not cancel out as noise, ceteris paribus, the thing can be ignored.
Second, in discussing the solution to one-shot prisoner's dilemmas, he suggests that the parties can reach the collectively desirable action by `treating as ``the unit of selection'' their joint strategy'. This treatment `would entail a violation of the standard formulation of individual rational choice' .But the thing can be easily represented in terms of work done by Shapley or Harsanyi in the early sixties. As a matter of fact, the existence of Prisons is what enables Organized Crime to flourish. This is because gangsters form coalitions within the safety of the prison which can then rule the streets. Prisoner's dilemma became the Mafia's opportunity.
Third, Sen has stressed the importance of discussion in changing the bases upon which people act (Sen, 1995, p. 18; Sen, 1999).Discussion can lead us to become disgusted with the stupidity and naked careerist virtue signalling of the other party. It can- and has- led to the emptying of the space of Public Justification.
We can integrate Sen's disparate hints into a unified account of the rationality of committed action as follows. Suppose the parties to a prisoner's dilemma identify with one another as common members of a social group. Then they would pose to themselves a different practical question. Each would ask, not `What should I do?', but rather `What should we do?'. To ask the latter question is to deliberate from a standpoint that one can coherently regard everyone else in the group taking up as well. It is to regard oneself as acting in concert with the other parties, as a single body. Any group of people whose members refer to one another as `we' and who, in virtue of that fact, see themselves as ready to be jointly committed to acting together, will properly regard the object of their choice to be a single joint strategy. They will thereby constitute themselves as members of a single collective agent (Gilbert, 1990). How would such a body evaluate different proposed policies or reasons for action? They would discuss them together, and try to reach a common point of view from which to assess them. In the classic prisoner's dilemma scenario, the parties are not able to discuss what they should do. Nevertheless, each can still take up the standpoint of collective deliberation, try to figure out what the outcome of such discussion would be, and act accordingly. The key to figuring out this outcome is that it is a constitutive principle of a collective agent (a `plural subject' or `we') that whatever can count as a reason for action for one member of the collective must count as a reason for all. That is, in regarding themselves as members of a single collective agency, the parties are committed to acting only on reasons that are universalizable to their membership. The universalization principle rules out the principle of maximizing expected utility (individual preference satisfaction) as an acceptable principle of rational choice for members of a collective agency who constitute the parties to a prisoner's dilemma.This cashes out as rights and obligations under a bond of law. Here punishability gains salience as umpteen Mafia movies have shown us.
To make this demonstration vivid, consider the case of members of a political party, P, who agree that the best outcome for all would be to elect their candidate, A. As we have seen, according to the principle of expected utility, even this shared preference does not give a reason for any of the members of P to vote for A, if each personally finds voting inconvenient. For each member reasons that the expected marginal impact of his vote on the preferred outcome is so negligible that even a trifling inconvenience is enough to outweigh it. Is this reasoning valid, from the standpoint of P? It is valid, only if the members of P could jointly accept a trifling inconvenience as a reason for all the members of P not to vote. It is evident that they could not accept this as a reason not to vote. For if every member of P accepted this as a reason not to vote, few members of P would vote, and this would defeat their joint aim of electing A.That's why actual political parties have canvassers who go knocking on doors and turning out the vote. Generally, there will be some sort of reward- if only psychic or reputational- for this sort of activity.
To act on the principle of expected utility would be self-defeating from the standpoint of the collective agency. The principle of expected utility is therefore invalid for members of the group. This argument does not turn on the members of the group having a common aim prior to collective deliberation. It turns on the fact that they accept as reasons for action only those considerations that each person would be willing to accept as reasons for everyone to act. In effect, each person asks, `what reasons do we have to act?'If the answer isn't 'punishability' of one form or another, the thing crashes.
Only the reasons that we can share are reasons on which people who identify as `we' can accept as a ground for their action. Because we cannot will that each person try to free-ride on the efforts of others, we cannot accept the reasoning that supports unconditional non-cooperation in prisoner's dilemmas.Sure we can. Have a couple of Martinis and you'll see what I mean.
Upon eliminating non-universalizable policies, the members of the group find that whatever jointly acceptable policies remain require them to, in effect, aim at what would be best for all. This is a conclusion, not a presupposition, of their deliberations. If the only alternative to unconditional non-cooperation were a policy of unconditional cooperation, then the parties would will that everyone cooperate.Willing don't mean shit. We can all get together and will North Korea off the map, but it aint going nowhere.
If any actual group talked in this sophomoric pseudo Kantian way it would quickly disintegrate leaving one sixteen year old girl, her cat, and two pedophiles. If she continues to talk Kantian cant, the pedophiles will take turns with the cat.
Kant's categorical imperative is intensional. It has nothing to do with bargaining problems or Shapley index of power or other game-theoretic considerations.
... insofar as I identify with the group, my reason for acting is: to do my part in advancing what we are willing together.Groups may pray together but they don't 'will together'- save perhaps if they are Wiccan covens. A Group can be thought of as an incomplete contract of adhesion. What matters is whether it delivers. Expected utility theory- which is also regret minimizing if uncertainty is low and the thing doesn't matter very much- is good enough for the job.
This argument for cooperating in prisoner's dilemmas shows how it can be rational to cooperate.It is Muth rational to cooperate in accordance with the correct Expected Utility theory. That is a statistical fact. It is not rational to do anything on the basis of an argument or, indeed, to frame an argument save if asked to do so for a pedagogic or professional reason.
If someone comes up to you and makes a very convincing argument for x, you may nod your head but you may also become wary of the fellow. A few fools will hand over their money and get fleeced. You wait till the Audited Accounts are in to see if he really is smart or just a good talker.
It does not show that it is categorically irrational to follow the principle of expected utility, or to fail to cooperate in prisoner's dilemmas.In some matters, it is categorically irrational to maximize expected utility- e.g. tragedy of the commons type situations. The rational course is to demand legislation.
Rather, I make the following priority claim: The Priority of Identity to Rational Principle: what principle of choice it is rational to act on depends on a prior determination of personal identity, of who one is.Anderson is being silly. This is a Nazi principle. Rationality means exiting an Identity class which is stupid and entering one which is less so.
The validity of the principle of expected utility (maximizing the satisfaction of one's personal preferences) is conditional on regarding oneself as an isolated individual, not a member of any collective agency.By virtue of my Nationality, I am a member of a collective agency. It is perfectly rational of me to accept the validity of the principle of expected utility where Hannan Consistency, i.e. regret-minimization, says I should. Utility theory was born in my Nation. I studied it at a leading Institution here. I know more about it than Anderson. I say she is wrong. She does not understand that Utility is a primitive notion in a mathematical model which, after the fact, can always be shown to have been maximized subject to relevant constraints. Sen knows this. He mentioned the Spilrajn extension theorem to Arrow.
In contexts where one regards oneself as a member of a social group, this principle will in general be invalid, because it is not universalizable among the members of the group.So what? No actual group has any such property. Thus, though I am a citizen and the P.M is a citizen, I don't have her powers. Thankfully, I also don't have her obligations and thus am losing no sleep over Brexit.
What is universalizable would be various principles of committed action (not necessarily of unconditional commitment).So all members of the group must have equal capacity in at least some respects. Otherwise they could not all be subject to 'various principles of committed action'. In practice, a group may have a few active members- office-holders for example- and a larger, fluctuating, number of people with varying degrees of commitment.
This argument does not show that we must identify with any particular social group.It shows nothing at all because it refers to a type of group which the earth has never seen- or which, if it existed, changed absolutely nothing and was forgotten by all.
A fortiori, it does not require that we act on our ascribed social identities of gender, race, caste, ethnicity, nationality, and so forth - that is, the group identities we have in virtue of the ways other people classify us. The concept of identity at stake in the theory of rationality is practical, not ascriptive.Practical things can make the world a better place. This is nonsense on stilts.
From both a rational and a moral point of view, it would often be far better if we rejected our parochial ascriptive identities as bases of practical (action-governing) identification. The argument does not require that the people with whom one practically identifies be determined prior to or independently of the practical predicament at hand. Practical identification with others does not require any prior acquaintance or relationship. It only requires that we see ourselves as solving a problem by joining forces. As soon as one says `Let's . . .' and the others manifest their willingness to go along, they have adopted a common practical identity as a social group with a shared goal.Very true! As soon as bunch of nice school-girls say 'Let's have a party and invite nice boys' all the rapists in the neighborhood manifest their willingness to attend.
Groups have to deter the entry of undesirables and attract desirable people. Their effectiveness increases because they serve a screening function which in turn becomes the basis of a signalling function.
A shared intention is sufficient to constitute individuals as a social group with a common practical identity, and the only constraints on whom one may share an intention with are practical (that is, the conditions must be such that such sharing is possible).The intention must be to exclude undesirables. Otherwise you don't have a group, you have a riot on your hands.
We could therefore find ourselves with good reason to practically identify and cooperate with perfect strangers.Not perfect strangers, but those who have acquired a more or less costly signal. Incidentally, kiddies, you should not get into the back of a van if some smelly old dude sez he has a nice puppy in there for you to pet.
What the priority of identity to rational principle does do is establish a rational permission to identify with others and join in a common agency.This 'rational permission' will get you raped in the back of a van.
The argument does not claim that all action is or ought to be based on some group identity.Very good of it, I'm sure. Thus, after you have been raped and dumped in an alleyway you are free to act in your own self-interest rather than keep getting raped by anyone claiming to share a common agency with you.
One's practical identity for certain choices may be simply as an individual, with perhaps idiosyncratic interests or needs. If practical reason also permits regarding oneself as an isolated individual, then, for all I have argued, the principle of expected utility could well apply.The need to not get raped- or even to have to listen to stupid idiots- is one reason to avoid groups which operate according to Anderson's desiderata.
The principle of expected utility and the principle of group universalizability are, on the view I have developed so far, both conditionally valid.If getting raped in the back of vans is your idea of a good time.
Their validity is conditional on the agent's selfconception, as an individual or a group member. Thus it appears that to adjudicate between these principles in any particular case, we need further principles of rational self-identification.And yet further principles of permissible rational self-identification and rational permissibility and so forth.
Rational attitudes such as love, respect, and admiration are more fundamental than preferences, but they do not yield a structure of preferences that satisfy the axioms of rational choice.Nothing does. This is why it is a mathematical model with primitive notions- i.e. undefined terms.
I set this issue aside for the purposes of this commentary. In the next section, I will argue that the sort of individuals we all can be will not permit acting exclusively on the principle of expected utility.Since utility is an undefined term, every possible and impossible sort of individual we can be is, by definition, compatible with acting exclusively on this principle.
I think, we should speak of regret minimization not utility maximization coz neglecting Knightian Uncertainty has been bad for us economically. However, Utility could just describe itself as Regret minimization.
Postmodernists tirelessly remind us that the identities and boundaries of the self are not fixed, but contingent and changing. Although this refrain has been so often repeated that it has nearly become a mantra, here is an important place where it should be heeded.Very true! If you have shat all over yourself talking stupid bollocks, then you are at a place where it is important to heed some even stupider bollocks.
Of course it does not follow from this that the dimensions along which the self fractures (or rather, along which it unites with others) must involve the standard American quadruplet of race, class, gender, and sexual orientation.Good to know. Getting it on with quadruplets is a big deal for some people. However, that bit about fractures puts me off. Reverse cowgirl can do that to you. Very painful- or so I've been told.
These are extremely important dimensions of social, political, and economic inequality, but they compete with numerous other dimensions of identity, and often do not form a desirable ground of practical identification.So, they cancel out as noise. The Law of Large numbers, or Condorcet Jury theorem, prevails. Econ can carry on trying to pay for itself. Meanwhile, Philosophy can ask stupid questions like this-
In what contexts is it rational to identify with others, or, on the other hand, to conceive of oneself as an independent agent?This begs the question. If you aren't an independent agent, how can you identify with others? If you are an independent agent then you and you alone determine what contexts it is rational to do any x. Suppose you were to work out in advance all the contexts in which you should do x. Ask yourself, how much time and effort have you expended doing so? Was it worth it? If the answer to your thought experiment is - no ; in general it is not worth consuming cognitive resources deciding this in advance. Then, decision is- it is not rational to answer this question. Rationality is costly. The question of when it should be deployed is an economic one. So what if this can be modeled by expected utility theory? Arrow & Debreu did not rape my grandmother. I've no beef with them. Sen and Anderson may have been subjected to some atrocity at the hands of Expected Utility theory. But that is their problem- not mine.
Which others is it rational to identify with? I want to avoid either of two extreme positions. One would say that we should identify with whoever else we stand in collective action problems, such as prisoner's dilemmas. The difficulty with this position is that in identifying with one group, we may thereby preclude identifying with another group.Nonsense! There is a bargaining solution between both groups. You could be an arbitrageur between two discoordination games.
The prisoners who maintain a tacit conspiracy of silence in the paradigmatic prisoner's dilemma scenario are helping one another, but not society at large. Yet, it is hardly irrational for a criminal to identify with the larger society, feel remorse for his crimes, and confess because of that identification.So the solution is to work with Gang members and the Police and Community Organizations so as to find a better collective outcome. This may mean legalizing gangs and letting them earn money legitimately, ultimately morphing into cultural organizations with plural membership.
The other extreme position would say that identification with others would only be rational after the members of a group have manifested their willingness to join together, conditional on the others manifesting a like willingness.This isn't extreme at all. It is a plausible first move in a bargaining game.
If this were so, then it would be irrational for people to cooperate in `classic' prisoner's dilemma scenarios, where the parties are not allowed to communicate with one another in advance of making their choices.P.D was a mechanism for getting convictions cheaply. It was defeated by another mechanism- viz. the Omerta code- which turned the Prison and Police system into a tool of organized crime. They took the small fry of the streets so the big fish could rule the roost- though sometimes, for their own safety, from behind bars.
Actual Economists have to envisage things like this. Sen and Anderson don't. They advanced their careers writing shite and getting the nod for being Brown or lacking a penis or something of that sort.
I think this condition on rational group identification is too strong. Common knowledge of everyone's (rational) conditional willingness to join together is sufficient, but not necessary, to make identification with the others rational.No. In this case, it would be rational to have antagonomic preferences for a regret minimizing reason, Aumann pointed out.
Where such common knowledge is absent, it could still make sense to take others on trust.Provided there is a robust retaliatory mechanism which is common knowledge.
In this case, one regards oneself as part of an imagined common agency, in the hope that others will join and make it real by cooperating.Very true! Maharishi says we should all levitate and spread peaceful energy and world will become so nice and please don't forget to send all your money to Maharishi coz actually He Is JESUS BUDDHA AND HAS POWER OF MYSTICAL CRYSTAL OF MT. MERU.
We need hardly consider the suggestion that each agent should be a pure individualist at all times. To achieve most of the functionings constitutive of a person's wellbeing, and most of the larger projects worth pursuing, requires cooperation with others.But it doesn't require reading Anderson's shite.
The principle of expected utility would seem to apply comprehensively only to hermits.It can apply even to cats or computers or the Cosmos. Why? Because a mathematical model can fit anything and utility is the primitive notion of a particular mathematical model.
On the other hand, particular collective agencies may fail to survive rational scrutiny, in which case uncritical identification with them would be irrational. (One might still critically identify with them, with the aim of reforming them so they operate on rationally acceptable principles.)Really? When has that worked?
To make progress on the question of principles of rational identification, I suggest that we turn to a case study. Sen's own work on gender and cooperative conflict in the family is an apt place to look for illumination. His work shows how both actions on the principle of personal utility maximization and actions on a principle of family group identification systematically disadvantage women, often in extreme ways.So would actions on the basis of capabilities or functionings. Did Sen actually do anything for Indian women? No.
The fact is, Development Economists knew that Japan had exploited female labor in the textile industry so as to acquire capital and modernize. Oddly, Japanese women ultimately benefited though not proportionately. India had gone the other way so life-chances for poor women shrank as industrialization stalled.
When China began its reform process it did it off the backs or rural girls who were biddable and less fractious than males. Ultimately this made them much better off than their Indian counterparts. Now India has a declining female participation rate while Bangladesh has surpassed Pakistan thanks to garment factories with high female participation.
Sen and his ilk talked a lot of bollocks- as if Govt. of India was going to come and shower money on poor women. But they didn't state the obvious- viz. women will get a better deal when they earn more and accumulate assets. That can only be done through manufacturing. It is mere eye-wash to say that targeted schemes will help women. If the jobs aren't there, then their condition will be as bad, or worse, than that of the men.
It would seem that neither principle offers an adequate perspective from which to secure justice for women. The quest for a larger perspective, I shall suggest, puts a Kantian twist on some of the grand themes of Sen's life work.Please, Anderson Memsahib, don't put Kantian twist on grand themes of Sen's life work due to dhobi will be having to use a lot of starch and just ironing and ironing to get the creases out.
Women around the world are systematically disadvantaged in their access to income, wealth, and employment opportunities outside the home. In many parts of the world, especially Asia, these disadvantages are severe enough that women suffer substantially higher rates of malnutrition, morbidity, and mortality than men. Sen (1990) has estimated that 100 million women are `missing' in Asia due to excess mortality stemming from material deprivation, parental neglect, and selective abortion. Much of this material inequality can be traced to the institution of marriage and the gendered division of domestic labor (Sen, 1989; Sen, 1990; Okin, 1989).Which obtained everywhere. The reason Britain, America, Germany, Japan, China etc. pulled ahead is because they got unmarried girls into dormitories and factories where they accumulated a little money to get married. However, during boom times, they returned to work and thus served a reserve army function. India could have followed the same path. But only if it rejected the 'Sen-Dodd' thesis and went in first for Labour intensive manufacturing, with high female participation and then climbed the value chain like Korea at the end of the Sixties.
Because women have more unpaid domestic, childrearing, and eldercare responsibilities, they have fewer opportunities to work outside the home for a wage. They therefore also invest less in acquiring the skills required for outside employment, and so are eligible for lower-paying jobs than men are. Women's assumption of domestic responsibilities frees men to work longer hours at income-generating activity and to acquire more human capital. The income husbands generate is thus not the product of their efforts alone, but a matter of joint production between them and their wives: were wives not performing the lioness's share of domestic labor, their husbands would have to devote some of their time to this work, and could only accept the kinds of employment open to workers with substantial domestic responsibilities. Of course, women share in the household income, so there is mutual advantage in the domestic division of labor. However, the division of the rewards from family cooperation is extremely unequal. Why do husbands and wives not share more equally in the rewards of their joint production?If we neglect Socio-Biology- i.e. kin-selective altruism, asymmetric maternal investment, the Price equation etc- we can't answer this question. Threat points would be equal.
However, because of Socio-Biology, husbands and wives have different threat-points. Still, this won't get us very far. The fact is there is an evolutionarily stable strategy mix. Different Indians are pursuing different reproductive strategies. Families like Sen's and mine have few kids, prolonged neoteny, and hypergamy for girls and emigration for boys. We have our imitators at every level of Society and ourselves imitate those above us. These Tardean mimetic effects can be reinforced by popular culture- including the Cinema. Thus whole linguistic groups can undergo demographic transition while other provinces still retain an older Malthusian type of reproductive strategy. Without question, the economic landscape plays a part in the outcome. In particular, what matters is employment opportunities for pre-marriage age girls. This is why, once India was headed by a person who rose up from poverty, more emphasis was given to Manufacturing- though, of course, elites were not enthused.
Where women can work before marriage and accumulate some assets, they can also screen potential mates more effectively. Furthermore, they have fewer children because the opportunity cost has risen as they can return to work, especially during boom times. This means they have a different threat point which slowly the Courts come to recognize. Finally, paternal investment equals maternal investment because if the fellow subsequently misbehaves he will lose his house and pay maintenance and cry and weep that his wicked ex is stopping him seeing his little babies.
This is not a stable equilibrium. My guess is women will pay for high value sperm and losers will spend their lives dossing on their parent's couch working dead end jobs.
The answer appears to involve elements of both economic rationality and commitment. To the extent that both husbands and wives are rational egoists, the answer is supplied by bargaining theory.Bargaining theory can't be applied because of the Price equation. Kin selective altruism is what matters. The thing is literally bred in the bone and is as much part of us as our sexuality. However, the thing is stochastic and corresponds to a evolutionarily stable strategy. So we still have screening and signalling problems. We don't know if a potential groom will be uxorious and want to sacrifice for his kids or whether he will have a wandering eye and be a cuckoo in other men's nests. A similar point may be made about women but it has less salience because nobody really wants to have ten sons anymore. One or two will do.
Under common knowledge of their preferences and alternatives, the parties would bargain their way into a Nash equilibrium in which the division of gains from cooperation is heavily influenced by the vulnerability of the parties in the breakdown position (how badly off each would be if there was no cooperation). The worse off a woman would be on her own, the worse deal she gets in marriage.However, this would affect the children. Since many Indians invest heavily in their kids, treatment of a non-working wife can't deteriorate beyond a point unless sociopathy is involved.
Women face substantial obstacles to developing and fully employing their income generating potential, due to sex discrimination in employment, legal and customary barriers against women seeking employment outside the home, lower access to education, legal barriers to female inheritance and property ownership, and the gendered division of paid employment, which reserves most higher-paid jobs and jobs on promotional ladders to men. Divorce laws provide at best minimal support for divorced women, as well. Thus, the cost of exiting marriage is higher for women than for men, and men therefore enjoy a threat advantage that they can exploit within marriage. Bargaining theory no doubt explains part of the material disadvantage that women suffer. In an insecure marriage, a woman's knowledge of how badly off she would be if her husband left her no doubt quells many complaints she would otherwise voice about the division of benefits and burdens in her marriage. However, the theory clashes with several features of women's circumstances and motives. Marriage is conceived as a realm of love and obligation, distinct from market exchange. This difference is marked in part by norms against naked bargaining among marriage partners. The woman who drives a hard bargain, who insists on an explicit quid pro quo for her services, is marked as a prostitute, ineligible for marriage. In societies that practice arranged marriages, especially when dowry is involved, the woman has little bargaining power at the time of betrothal, since she functions more as the object of bargaining rather than as a party to the negotiations. Social norms against women's bargaining and severe limitations on women's choice of partner make her less able to secure a share of household resources than if she were free to bargain as a rational egoist. In addition to these factors, Sen argues that women's motivations are not those of the rational egoist.Why argue such an obvious point? Only a sociopath, lacking a gene for kin selective altruism, would act as a rational egoist. Indian men don't do this. Nor do Indian women. Even Indian cats and dogs and elephants exhibit this trait. The Price Equation explains why.
Their wide preferences include the interests of other people, often to such a high degree that they have difficulty conceiving of their own interests as distinct from those of their family members.This is true of the Merchant Banker with two kids in private school. He has to make trade-offs just like an Indian mother.
When Indian women are asked about their own welfare, they typically answer in terms of how well their family is doing.It is not polite to ask about the condition of the Griha-Lakshmi- the Wealth Goddess of the home. Her status is superior, she gives a general answer about the welfare of her subjects.
But this is a matter of universal courtesy. The American Ambassador doesn't sidle up to the Queen-Gor-bless-'er and say 'Yo Queenie how's the piles?' Instead he asks circumspectly about the youngest great-grandson and so forth.
The difficulty they have in perceiving the distinctiveness of their own interests further reduces their ability to claim a share of domestic resources.Sheer nonsense! In a joint family, wives often have a keen interest in claiming their share of the patrimony. So do sisters and widowed mothers. If what Sen says is true, then British Courts- and Indian Courts to this day- would not have been crammed to the gills with petitions from women in property disputes. Where, as often happens, a widowed mother is managing the patrimony, the eldest son may get short shrift. His wages may be invested by the mother and when he is married off, she may retain the dowry. There is scarcely a family in India which does not have a tale like this to tell.
Anderson, poor thing, is weeping her eyes out because she thinks poor Indian women don't know what's theirs and can't fight their own corner. No doubt, she also thinks Mother Theresa gave a lot of money to poor people in Calcutta.
Women also manifest committed motivations, in accepting social norms that devalue their contribution to domestic resources.Not in India. Indeed, the caste system- which reduced male ability to access ritually pure food- gave women the upper hand.
Real-world voluntary distributions require legitimation in terms of a conception of justice.No. Nobody gives a flying fart about some stupid pedant's 'conception of justice'. The Law matters but Custom is king.
Considerations of desert or productive contribution therefore play a role, beyond self-interest, in determining the division of gains from cooperation.Not in any family which ever existed. That's why babies don't have to go to the office or clock in at the factory.
Women and men share a perception of relative contributions that gives greater credit to men's wage-earning than to women's unpaid domestic labor and lower-paid employment. The wife is cast as dependent on her husband, obscuring the ways in which his productivity depends on her providing for his physical needs in kind (cooking, cleaning, sewing, shopping, obtaining water and firewood, etc.) as well as on her assuming responsibility for caring for his children and sometimes his elderly parents as well. In accepting norms that devalue their contributions, women again get less from marriage than they would have if they were purely self-interested.
Anderson is a Professor of a shit subject. Does she accept the norms that devalue her contributions to Academia? She gets less from her 'research' than smart people who are discovering the cure to cancer or the secret of quantum computing. Does this cause her to hang her head in shame? No. She deludes herself that she is important- maybe even that she is helping poor Indian women by writing this shite.
The truth is a poor Indian woman who gets married to a poor Indian man and produces poor Indian babies is not of much account in this world. So what? She has a better argument than Anderson for valuing herself. She serves God and will attain Heaven. Her descendants on earth will offer pinda oblations to honor her. However, unless stupid economists like Sen are disintermediated from decision making, her descendants will be very poor. So, she will vote for Modi and get on with her life.
Sen's work on gender and the division of family resources thus places in social context the two types of cooperative motivation he identified as possible solutions to collective action problems - altruistic preferences, and commitment to social norms seen as right, legitimate, or obligatory.The idiot failed to mention that all family relationships will feature kin selective altruism coz Biology sez so. The Price Equation held long before our phylum crawled out of the seas.
Laws we have had with us from before History began. But Custom is mightier than Law. Thus has it always been. Sen has scarcely come up to the level of his Navya Nyaya ancestors.
Women accept a lesser share of family resources not just because (a) they have little bargaining power, but because (b) they think that they ought not act like a hard bargainer,Sez a woman who believes Indian women are completely different from Americans at a comparable period in their economic history. Women in poor societies are expected to be hard bargainers and tight fisted. Having a temper is considered a good thing. The notion is that a woman of this sort will not be taken in by any scoundrel. She may smile sweetly on occasion, but her sharp eyes will note everything and calculate everything.
(c) have a hard time seeing themselves in this role in any event, given the difficulty they have conceiving of their interests as distinct from their family's interests, and (d) because they think they deserve, and hence ought to accept, only the little they are getting.This is all sheer fantasy. A sob story for the Great White Sister. Women are economists. They manage the household budget and scrape together a little savings. This means lots of fighting. If the woman gets beaten, the family won't have the capital to survive a downturn. It will be liquidated and scattered to the winds.
Women are even worse off than they would be if they were rational egoists, bargaining as economic models suppose. Does this mean that women ought rather to conceive of themselves as rational egoists than as family members?OMG! Is Anderson seriously gonna start prescribing for women in a far country she has no knowledge of?
It is not clear that someone can function in a spousal or parental role with the self-conception of an egoist. In any event, women would still get a raw deal in marriage even if they were to conceive of themselves this way. Luckily, we do not have to choose between these two dismal options, of rational egoism or commitment to an unjust form of group agency. The perspective of rational egoism proves highly useful for criticizing oppressive commitments. But it points beyond itself. Bargaining theory recommends that, if women are to improve their situation, they ought to seek employment outside the home.This shows that bargaining theory is inappropriate to this context because it returns the same answer irrespective of the age of the children, the nature of available employment, the level of uxoriousness of the husband etc.
Of course, if Sen meant all this as satire and was laughing himself silly at people like Anderson, then it is simply a joke.
This will reduce the cost of exit from marriage and thereby improve women's prospects within marriage. But outside employment has additional effects not predicted by bargaining theory. First, given that outside income is more salient to husbands and wives as a productive contribution to household resources, wives acquire a greater perceived claim of desert to larger shares of those resources. Second, outside employment puts women into contact with diverse others who do not identify solely as family members. Such contact provides information about wider opportunities outside the home, and can inspire women to take a more critical stance toward their domestic identity and commitments. Third, outside employment gives women opportunities to acquire new identities, besides their family-bound identities as wives and mothers. They now identify as workers in a cooperative enterprise distinct from the family. This is a new form of collective identity. Once they have more than one collective identity, whose constitutive commitments may not be fully coordinated with one another, they may find themselves needing to acquire yet another identity. This is the identity of an individual self, who is authorized to adjudicate the conflicts among its various constitutive collective identities.So there you have it. Indian women who don't have jobs, don't have an individual self. I propose we set up a manufacturing unit in China to mass-produce 'individual selves' which can be surgically inserted into poor Indian women by a benevolent NGO.
This scenario suggests a different genealogy of the individual from that assumed in standard economic models.Indeed! It suggests that Indian women belong to a different species altogether!
Within economic theory, the individual is assumed to be given, with interests and preferences defined independently of and prior to her joining any group. Whether it is rational for her to join a collective agency then depends upon whether so joining would advance the satisfaction of her individual preferences. She joins if, and only if, joining would maximize her expected utility. This model is unrealistic in societies where women do not have an option not to join certain groups, and lack the option to join others.Women don't have the option to join the group of men with large penises who have attained omniscience and then ridiculed other men who attained omniscience but who had small peckers. This caused great irritation to the pencil dicks. They lost their equanimity and thus their omniscience. Women who had attained omniscience migrated to the astral plane so as to get away from all the noise and commotion created by the men who were waving their dicks at each other.
Anyway, that is my theory as to why Anderson does not have the sort of self which can adjudicate on her pronouncements and tell her to shut up and stop being silly.
Where women's identities are comprehensively defined in terms of their family roles - as daughter, wife, mother - they are given no choice about whether to so identify themselves, and have no other options to identify with other groups.This is also true where women's identities are comprehensively defined in terms their Academic roles as teaching worthless shite. They are given no choice about whether to so identify themselves, and have no other options to identify with other groups coz they are too ignorant and stupid.
Where women are deprived of such choices in how they identify, they have no opportunity to become the individuals who could choose otherwise.This is also true of cats and comets and cosmic rays.
To become such individuals, women need mobility between different types of collective agency. They need to be free to move from the family sphere to other spheres of social organization - outside employment, politics, women's associations, friendships unregulated by their male relatives, and so forth.Spheres of social organization- like the brothel- may be worse than the family sphere.
Mobility is risky. Strict enforcement of the Law, reduces risk. If women are more vulnerable, they benefit disproportionately from Right Wing policies more particularly if this means labor organizers are killed and factories start sprouting all over the place.
Thus, if the freedom to function as an individual is an important kind of agency capability for a human being, then we cannot assume that women around the world already function as individuals, but must dramatically revise the norms constitutive of `women's' roles to make this possible.You did this 20 years ago. What was the outcome? Nothing. Women did rise up in places where Trade Unions were severely dealt with and where sweat shops flourished. In India, judging by the participation rate, it is likely that life-chances fell for a large group of women.
I am arguing that individuality - identification as an individual person- emerges out of a certain kind of social order.An ignorant and racist assumption.
Such an order is defined by multiple, distinct spheres of social life, none of which comprehensively defines anyone's agency, individual freedom of mobility among those spheres, and individual membership in multiple spheres.No! Such an order is not defined by anything at all. It may be characterized by some random shite you wrote. Equally it may not.
Only when these social conditions are in place can people become individuals, understood as agents authorized to set their own priorities, on their own, according to an autonomously defined selfconception.These social conditions will never be in place and yet people will always become individuals.
Only when Philosophers stop being ignorant and stupid will they write anything sensible.
Before that point, their priorities are set by the commitments of the collective agency to which they belong, and hence it is not up to them to set their priorities on their own. On the view of individuality I propose, then, acquiring a self-conception as an individual requires actually conceiving of oneself as a committed member of multiple social groups, among whose claims one must adjudicate in allocating one's own efforts.Is there any group which endorses Anderson's view of individuality? If there is, then she is merely presenting the group's view, not proposing one of her own. If there isn't, then her view is self contradictory.
In any case, if one finds oneself 'adjudicating' between the claims of different groups, it must be the case that you are not wholly committed to at least one of them.
But the freedom to determine one's own priorities in committing oneself to various groups depends on those groups limiting their demands on their members in ways that enable their members to identify with and function as members of multiple groups.In other words, these groups are merely provisional alliances or coalitions. They have no independent 'group-life' superior to that of their members. Thus, a Religion or totalizing Ideology could not be represented by this sort of group. Rather, we are speaking of clubs it may be convenient to join for a utilitarian, not imperative, reason.
Social groups do not exist in pre-established harmony; different groups often make incompatible demands on their members. One therefore cannot expect groups to limit their demands on their members simply on the basis of considerations available internally to each group's perspective. To harmonize the demands of different groups requires adoption of a perspective that can coordinate them all.Or we could just leave it to the market.
We could see this as a collective action problem that groups face with one another. To solve it requires that we transcend our various parochial identities and identify with a community that comprehends them all.Or could we just accept that the invisible hand of the market has already done so.
Such a community is what Rawls (1971, pp. 527ff.) calls a `social union of social unions'. Were we to expand this community of identity to the whole of humanity, we would deliberate from the standpoint of what Kant (1964, pp. 100±3) called `the kingdom of ends'. This is the point at which rationality coincides with morality.As well as stupidity and laziness.
I have no argument that would show that identification as a member of a universal community of humanity - a kingdom of ends - is rationally required.There can be no such argument. The thing is silly.
But, as economic development proceeds, we find ourselves more often generating problems, especially environmental, that need to be solved, and can only be solved, within a global system of cooperation.Which can be wholly market-based- e.g. Carbon trading.
Identification with humanity as such may therefore become an historically urgent task.Only if people currently identify as cats.
A move to cosmopolitan identification has some powerfully attractive features.Very true. Navvies working on building sites put down their trowels to wolf-whistle any passing move to cosmopolitan identification.
Besides providing a comprehensive perspective from which the collective action problems worth solving could be solved, a universal commitment to act from this perspective would secure the conditions for everyone being able to achieve an identity and agency as individuals.No it wouldn't. Even if everybody was committed to being nice to everybody some people won't achieve an identity and agency as individuals because they would have died in the meantime.
To see `us' as comprehending all of humanity, is, however, to commit ourselves to placing significant constraints on appeal to preferences as reasons for action.So would seeing 'us' as not comprehending all of humanity- indeed, even more so.
The only sort of individual that everyone can be is one who identifies with multiple collective agencies as well as with humanity as a whole, and who therefore accepts multiple commitments, not grounded in individual preferences, as reasons for action.Everyone already is the sort of individual who thinks this is verbose, vacuous, shite.
To understand the nature of rational choice,You have to choose between talking vacuous shite and doing something useful. It is not useful to pretend you have information you don't. I know what I want to eat for breakfast. I don't know what everybody wants to eat for breakfast.
then, we need to enrich our information basis beyond individual preferences, and include the ideas of identity, collective agency, and reasons for action, where the test for valid reasons is universalizability among those with whom one rationally identifies.So, to understand something one has to know everything. This is a rational view only if you identify with and believe you can actually achieve God like omniscience.
No comments:
Post a Comment