Tuesday 12 September 2023

Sen's Social Ontology

A Society, on one view, is a collection of individuals. More specifically, in the Mathematics of Sen-tentious Social Choice theory- it is the set of the preferences, or capabilities, or characteristics, of a well-defined set of individuals. Sen has a 'Platonic' Social ontology because all the sets named above are 'well defined' and partially ordered. Yet, the fact is, preferences, capabilities, characteristics etc. are all 'epistemic'- i.e. they change as our knowledge changes- and thus Sen's work is vitiated by the 'intensional fallacy'. There are no functional relationships between anything he talks about because they don't have well defined 'extensions'. 

Thus Sen's Social Ontology is simply nonsense of which we can have no knowledge. This means that all his other concepts- e.g. Rationality, Identity, Consistency etc.- are also meaningless or mischievous  nonsense. This is easy to see.

Consider the following. Why is it fucked in the head?  

 Rationality is interpreted here, broadly, as the discipline of subjecting one´s choices – of actions as well as of objectives, values and priorities – to reasoned scrutiny.

But that reasoned scrutiny would itself have to be the subject of reasoned scrutiny. There is an infinite regress here. That's why Tarski insisted that terms like 'Rationality' must be either 'primitives', which are undefined, or else that they must be given a 'buck stopped' protocol bound description. Thus in a contract which requires rational conduct, a clause may be added to specify what would qualify as 'rational conduct'. This may be 'such conduct as is displayed by intelligent and diligent people of such and such a profession in the discharge of their obligations'. In other words there is a clear 'extension' of the term- i.e. a set of elements which conform to its 'intension' or meaning.  

Rather than defining rationality in terms of some formulaic conditions that have been proposed in the literature (such as satisfying some prespecified axioms of “internal consistency of choice”,

Nothing wrong with that at all. Rational people are consistently rational. We may want to know why one rational person is more successful than another rational person. We study the behaviour of both. The first guy is nice to people who help him. The second guy consistently beats or stabs anyone who does him a favour. Word gets around. More and more people help the first guy. They avoid the other fellow like the plague. It may be that the latter has a good reason to want to be shunned. That is a matter of his personal preferences.  

or being in conformity with “intelligent pursuit of self-interest”,

Again there is nothing wrong with this at all. True, the self may be interested in being shunned by all and dying a miserable, lonely, death- but this is a matter of personal preference.  

or being some variant of maximizing behavior,

or, under Knightian uncertainty, regret minimizing behaviour. 

rationality is seen here in much more general terms as the need to subject one’s choices to the demands of reason

This would be circular 'Rationality is about being reasonable' but for the fact that reason makes no demands. It isn't like needing to shit very badly. It's more like remembering to take down your trousers before taking a dump. 

Prof Nuno Martins, in a paper titled 'Rules, Social Ontology and Collective Identity' writes

Sen (2002) notes that the self-interested behavioural structure of the human agent that underpins mainstream (micro)economic theory (including mainstream game theory) posits what he describes as the three features of the self: “self-centred welfare”, “self-welfare goal” and “self-goal choice”.

The word 'agent' can be substituted for 'self'. The agent may be a family, an enterprise, or some other collective.  

“Self-centred welfare” means that the individual’s welfare is not affected by the welfare of other agents;

No. It merely means that the agent has an expected welfare level under different scenarios.  

“self-welfare goal” means that the individual has no goal other than her or his welfare

No. The agent may be seeking to enhance or reduce the welfare of some other individual or collective. This may involve the loss of the agent's own life.  

(even though the latter may be affected by the welfare of other agents, or not)

this is factored into the expectation regarding welfare in different scenarios- even one in which you are no longer in the world but Mummy and Daddy are crying their eyes out because they didn't let you stay up to watch TV on a Skool night which is why you topped yourself.  

; and “self-goal choice” signifies that all the individual’s choices are made in accordance with her or his goals (regardless of how the latter are defined).

or how fucking stupid those choices were. But this is mere semantics.  

A much debated issue in game theory is how social cooperation can arise when: (a) it is always advantageous for each individual not to cooperate (regardless of whether other individuals cooperate or not);

in which case there is no cooperation 

and (b) all individuals would be better off if all would cooperate, rather than if there were no cooperation.

in which case there will be cooperation and a collectively enforced penalty for free-riders.  

One example often discussed in this context is the “prisoner’s dilemma” situation, where in a “one-shot” game

you don't have to worry about any penalty because there is no tomorrow.  

it is always better for a player not to cooperate (both in the case where the other player cooperates and in the case where (s)he does not, and so “not cooperating” is a dominant strategy, that is, it is preferred regardless of what other players do),

Why bother interrogating a suspect in a one period economy? There is no future in which the fellow will be rotting in jail.  

but if both players would cooperate, both would achieve a higher level of welfare.

Why bother? There is no tomorrow.  

The dominance of non-cooperative strategies in the prisoner’s dilemma springs from the fact that agents are supposed to pursue their “self-goals” only, and can happen even in the absence of “self-centred welfare” and “self-welfare goal”.

Prisoner's dilemma does not apply to professional criminals. Only amateurs turn on each other. The truth is Prison is good for organized crime and bad for opportunistic criminals who don't make a living by it.  

For example, agents with altruistic concerns (which need not include any concerns with their own welfare, and need not be self-centred as well, but would be individual goals nevertheless) might still, for some reason, have these “altruistic” goals ranked in the same way as payoffs are usually ranked in the prisoner’s dilemma. Hence, even without “self-centred welfare” and “self-welfare goal”, the existence of “self-goal choice” would be a sufficient condition for the emergence of a prisoner’s dilemma situation—see also Derek Parfit (1981, 1984).

Not in a 'one-shot' game. In anything else, only what happens at the margin matters. But this means there will be uncorrelated asymmetries because different agents will be on different points of their 'cost curves'. Thus 'bourgeois strategies' will prevail- till there is an abrupt saltation.  

In fact, even when “self-goals” include not only the agents’ welfare, but also any other type of individual goals (such as a concern with the welfare of other agents), those “self-goals” would be still individual goals (regardless of how broadly defined), and Sen’s point is that social behaviour is irreducible to individualistic “self-goal” pursuit.

Except when it isn't. There are plenty of such areas for the economists to focus on. They can 'add value' and thus 'pay for themselves'. Why grieve over the fact that your Econ degree does not qualify you to be a ballerina with the Bolshoi? 

Sen argues that the key to resolve the prisoner’s dilemma is to think “in terms of social strategy” (Sen 1987: 86),

It is because there is a social strategy that prisons and policemen exist. Sen didn't know this. Sad.  

instead of trying to derive these social aspects from individualistic interplay: “Behaviour is ultimately a social matter as well, and thinking in terms of what ‘we’ should do, or what should be ‘our’ strategy, may reflect a sense of identity involving recognition of other people’s goals and the mutual interdependencies involved. [. . .]

But something more than this platitude is expected of you if you have taken a degree in Econ. You are supposed to find ways to do things more cheaply or profitably. 

The language of game theory—and indeed of economic theory—makes it hard to discuss behaviour patterns of this kind, since it is very tempting to think that whatever a person may appear to be maximizing, on a simple interpretation, must be that person’s goal.”

Sen is complaining that using the language he himself teaches makes it very hard for him to think about the sort of stuff which the rest of us find easy enough. This suggests that Sen was stupid and that any language used by a stupid person will yield only stupid shit.  

(Sen 1987: 85) Sen suggests starting from the social notion of the “identity” of a community,

e.g the identity of Sen's own ancestral community in East Bengal as 'kaffirs' who should be killed or chased away

instead of trying to explain social behaviour in terms of the individual interaction of self-goal pursuing agents:

Actually, if East Bengali Hindus had worked together with East Bengali Muslims to raise agricultural productivity and establish high value adding manufacturing and service industries, the Sens may not have had to flee Dacca. The plain fact is rectifying the grievances of the mainly Muslim tenants in East Bengal would have been good for the land owning Hindus. Economics is about giving people a material incentive for rubbing along well enough with each other. 'Identity' Politics tends to be fatal to minorities. But it also harms the economy and thus the living standards of the majority.

“The rejection of self-goal choice reflects a type of commitment that is not able to be captured by the broadening of the goals to be pursued.

The rejection is meaningless and useless as anything it 'reflects'. Commitment to babbling meaningless shite doesn't 'capture' anything. If you are a useless tosser, it doesn't matter if your goals are broad or narrow.  

It calls for behavior norms that depart from the pursuit of goals in certain systematic ways.

It can call for a radish up its bum for all anyone cares. Showing people they can get richer by doing something which is clearly sensible and rational is how you get people to answer your calls.

Such norms can be analyzed in terms of a sense of ‘identity’ generated in a community (without leading to a congruence of goals),

Sen's people ran the fuck away from Muslims because of their Islamic identity. Sen also ran away from India with his best friend's wife. Perhaps this was a case of mistaken identity. He had merely meant to borrow his friend's umbrella. 

and it has close links with the case for rule-based conduct, discussed by Adam Smith. [. . .]

There is no such case. There are some rules it is profitable to follow or perilous to break at particular places or particular times.  

It is an alternative program to the recent attempts at ‘resolving’ the [prisoner’s] dilemma through the relaxation of the assumption of mutual knowledge in finitely repeated games.” (Sen 2002: 219/220)

Which are likely to have unique, inefficient, equilibria. But one may just as easily speak of coordination games giving rise to discoordination games with arbitrage between the two. 

It is important to note that the target of Sen’s criticism is only traditional game theory, where it is presupposed that agents permanently engage in the pursuit of their self-goals.

No. It is presupposed that there is a well defined payoff matrix. This may seldom be the case but where it is the case Economists can add value.  

Sen offers no criticism of approaches to game theory which do not presuppose permanent self-goal pursuit, such as evolutionary game theory approaches where conventions are the key element to explanation.

Regret-minimizing which can look like 'self-learning' gets you to evolutionarily stable states. But what really matters is the fitness landscape.  

In fact, Sen (1997) supports evolutionary game theory, arguing that it can provide important insights to the study of behavioural norms and rule-based conduct.

Sadly, this isn't the case. Binmore wrote well and added value in some areas (e.g. Spectrum Auction) but not even he could make this case.  

Furthermore, authors studying conventions from a game theory perspective often adopt a conception of “we-rationality” which is very close to Sen’s (1987) notion of “social strategy”—see, for example, Sugden (2000), or Hollis and Sugden (1993). Nevertheless, notions like “rule-based conduct” and “sense of identity” with a community require further analysis. What are the conditions of possibility for this type of social behaviour?

Uncorrelated asymmetries or 'costly to disguise (or acquire) signals giving rise to separating equilibria or discoordination games.  

And what does it mean to identify with others?

Nothing at all if you have devoted your life to babbling meaningless shite.  

Sen’s conception of collective identity can be fruitfully linked to John Searle’s notion of collective intentionality, which means to “share intentional states such as beliefs, desires, and intentions” (Searle 1995: 23).

Both are meaningless. We can't verify that there is a collective identity or collective intentionality. Why not speak of the different sizes of the horns of the unicorns belonging to different communities which are currently galloping across the greensward of Fairy-land?  

Searle argues that the collective intentionality of a community is irreducible to the singular intentionality of individual agents: “What is the relation between singular and collective intentionality, between, for example, the facts described by ‘I intend’ and ‘We intend’?

The relationship is between singular and plural in the speaker's idiolect. If the speaker is Queen Victoria 'We intend' means 'I intend'. If the speaker is the Fuhrer, 'I intend' means 'the Nazi Party intends'.  

Most efforts I have seen to answer this question try to reduce ‘We intentionality’ to ‘I intentionality’ plus something else, usually mutual beliefs. The idea is that if we intend to do something together, then that consists in the fact that I intend to do it in the belief that you also intend to do it; and you intend to do it in the belief that I also intend to do it.

No. You and me may say 'we intend to beat the shit out of each other' though both of us know very well that you will do all the beating and I will do all the crying. Still, for a moment, I get to look like a stand-up guy.  

And each believes that the other has these beliefs, and has these beliefs about these beliefs . . . etc., in a potentially infinite hierarchy of beliefs. ‘I believe that you believe that I believe that you believe that I believe . . . ,’ and so on. In my view, all these efforts to reduce collective intentionality to individual intentionality fail.

Because every sort of effort in Searle's utterly useless field of study was bound to fail.  

Collective intentionality is a biologically primitive phenomenon that cannot be reduced to or eliminated in favor of something else.”

But it is not a term of any great utility. It is merely 'imperative' or, worse yet, 'phatic'. Still, it is polite to say 'we are going to make love' though the expression on wifey's face suggests that whatever you are going to do is not something she loves at all.  

(Searle 1995: 24) Searle concludes that: “The crucial element in collective intentionality is a sense of doing (wanting, believing, etc.) something together, and the individual intentionality that each person has is derived from the collective intentionality that they share.”

This is just a conditional intention of a particular type. I suppose one may say that an individual may do something with the intention of causing some particular collective to do something- e.g. a terrorist of a particular faigh attacking a Church in the hope that the Christians will retaliate against some other specific community. In other words the motive of the action may be tactical and may involve creating a 'collective intention'. But nothing is gained by using that term. All we are speaking of is a plan to achieve some particular objective. I may chop off my hand so as to force the unicorn collective which runs Fairyland to accede to my demand to sprout wings and come to my rescue thus revealing themselves to be real rather than a figment of my diseased imagination. 

(Searle 1995: 24/25, emphasis in original) Searle uses the expression “social fact” to refer to “any fact involving collective intentionality” (Searle 1995: 26). Collective intentionality is a condition of possibility for us to follow social rules of behaviour.

It is nonsense. 

Whenever we occupy a given social position (such as being citizen of a country, member of a family or teacher at a university), there will be a set of rules we are expected to follow

only if they are known to us to be enforced rigorously. But, in that case, we don't need to know the rules. Indeed, such 'rules' may be unwritten or unknown. We just need to do that doing x reliably causes horrible outcome y.  

and, likewise, a set of rules we expect people in other social positions to follow towards us. But whenever following rules, people who occupy different positions in a network of interrelated social positions will all be engaging in a form of collective behaviour, in which they share intentions, beliefs and desires.

No. Some will be slacking off. Others may have serious mental or other health issues. Some may have crazy beliefs. What matters is habit and statistical regularities.  

For example, if a person X follows a given rule of behaviour when relating to person Y (for example, when person X pays a good bought at person Y’s store), both X and Y (who occupy the social position of buyer and seller, respectively) are engaging in a form of collective behaviour (in this case, commercial exchange) which requires shared intentionality.

It doesn't require shit. Both may be acting out of habit. Indeed, they may be unconscious of what they are doing. The cashier is thinking about her boyfriend. I'm thinking about how much I'd like to be her boyfriend. I don't recall what it was I bought just to have the chance to stand opposite her and accept my change from her dainty hoof (in this story, cows are the dominant species on our planet.) 

Collective intentionality can be seen as a condition of possibility of Sen’s notion of “identity” of a community.

Only in so far as it can be seen as a unicorn which fucks cows who are also cashiers at the local Kwikimart.  

Identifying ourselves with a community means to recognise ourselves as part of a social structure, and place ourselves in a given social position within an interrelated network of social positions (each associated with a set of social rules), through which we are connected to the other members of the community. Sen’s idea of acting in terms of our strategy, as opposed to permanently engaging in individualistic self-goal pursuit, presupposes

having a nice job far far away from Bengal.  

Sen argues that social rules must be “taken for granted” by agents when interacting: “If the sense of identity takes the form of partly disconnecting a person’s choice of actions from the pursuit of self-goal, then a non-inferior outcome can well emerge even without any formal contract or enforcement.

Very true. You can kill your kaffir neighbour even though he is very helpful to you. That is a non-inferior outcome because God hates kaffirs.  

One of the ways in which the sense of identity can operate is through making members of a community accept certain rules as part of obligatory behavior toward others in the community.

Get together every now and then to stone adulterers or kill homosexuals by collapsing a wall upon them.  

It is not a matter of asking at each time, What do I get out of it? How are my goals furthered in this way?, but of taking for granted the case for certain patterns of behavior toward others.

Which is how genocides happen. Cool- right? 

In fact, acceptance of rules of conduct toward others with whom one has some sense of identity is part of a more general behavioral phenomenon of acting according to fixed rules, without following the dictates of goal maximization. Adam Smith had emphasized the importance of such ‘rules of conduct’ in social achievement: ‘Those general rules of conduct, when they have been fixed in our mind by habitual reflection, are of great use in correcting misrepresentations of self-love concerning what is fit and proper to be done in our particular situation’ ).

Don't stab people with your dirk. Also don't shit on the dining table. These are general rules of continent prevalent in more advanced countries which the young laird would do well to observe. No doubt, some of the folk back in the ancestral castle will laugh at you for your effeminate behaviour. Secretly denounce them to the Government in Edinburgh as Jacobites. They will be hanged and you can grab all their cool shiny stuff.  

So, for Sen, “rule-based conduct” is a phenomenon that cannot be entirely derived from the interaction of “self-goal” pursuing agents (as in traditional game theory) because, for one thing, (fixed) rules of conduct are already pre-existent (and taken for granted) at each moment when individuals act. But Sen’s claim that “fixed rules” of conduct must be “taken for granted” when engaging in social behaviour, for example, could seem to suggest a deterministic view where social “rules of conduct” are reified and human agency is entirely determined by (“fixed”) social rules.

Only because Sen writes like shit. Rules underspecify behaviour.  

It could seem that we have criticised a conception where collective intentionality and social rules of conduct are reduced to individual intentionality, to move towards a diametrically opposite conception where human action is completely determined by social rules, or the social structure of which these social rules are a constitutive part, or some collective consciousness.

No. It seems these cretins have wasted their time babbling boring shite. 

In earlier writings, Sen addresses the topic of determinism, and defends a form of historical determinism from several criticisms, trying to allow for a conception where an action can be causally determined, but nevertheless agents are free in the sense that they are following values and preferences: “Let there be three alternative policies, A, B and C, open to a given set of men leading respectively to results A′, B′ and C′. They like, let us say, B′ best of all and, therefore, choose to do B. There was no external compulsion for them to choose B, but they do so because they prefer B′. Thus, in a certain sense (and that is the only sense that matters to the people involved), they are free to choose and B is the result of their free choice. There is, therefore, no reason for them to be fatalists. On the other hand, in so far as their values, preferences, impulses and the reaction system determining the choice are products of their inherited and acquired nature, their choice is determined by the past. Given the causal relations, one would have to be able to predict what they would have chosen, even though in the only sense that is relevant to them as decision-makers, they were free to choose what they liked” 

Sen is merely saying that a choice-sequence may be law-like. Could it be law-less? By Razborov Rudich, we could never prove this to be the case. 

 More recently, Sen (2006) also argues that human agents have a plurality of identities,

but only one body. You can get sent to prison for stuff which your criminal identity did. Sad. 

and that there is no need to see one of these identities as the sole determinant of human action, which seems to suggest that human agents are able to scrutinise, reflect upon, and be critical of, the social rules and norms which characterise a given group that we may identify ourselves with. 

But this is also true of the human ability to scrutinise on various pornographic rules and norms governing unicorns in Fairy-land. What has this to do with economizing on scarce resources- which is what economists should concern themselves with? 

 the account of rationality that Sen defends is one where rationality does not consist in following one preference ordering, set of values, fixed rules, or identity, but rather to have the capacity of reflecting upon preferences, values, rules, and identity, and possibly changing them.

This is not an account of rationality. It is an account of an idle type of fantasy. I am currently a unicorn having sex with Miss Piggy. She turns into Queen Victoria & is totes not amused.  

So to see human action as totally determined by underlying social rules, structures, identities, or collective consciousness, would be inconsistent with Sen’s (1987, 1997, 2002) own conception of agency and rationality, and with Sen’s (2006) more recent writings on identity. Autonomy and freedom of choice are essential to human agency in Sen’s thinking.

Yet, slaves have had more autonomy and freedom of choice than Tyrants throughout history. This is what the Stoics claimed and is at the root of the Hegelian Master-Slave dialectic. 

Furthermore, Sen (1997) also argues that preference orderings can be irreducibly incomplete, in the sense that it may not be possible to rank two options A and B in any way, which means that in such cases it will be impossible to predict human choice.

One can always do so arbitrarily. If this is useful, it 'pays for itself'.  

Prof Martins thinks there is something called social ontology. This is fine if you also think there is an ontology of 'Imaginationland' were unicorns have sex with Miss Piggy who, as a rule, turns into Queen Victoria. 

I will now argue that in order to make Sen’s conception of collective identity coherent with his writings on agency and rationality, Sen’s notion of identity must be seen under the light of a social ontology perspective in which both social rules of behaviour and human agency are ontologically real, and ontologically distinct from (and irreducible to, albeit dependent upon) each other. 

This is crazy shit. You may grant that I am realm, I have being, and thus exist in an ontology. Why grant that the rule whereby Miss Piggy turns into Queen Victoria is also real, has being, and exists in an ontology? What benefit is gained thereby? True, if you are Eric Cartman, Imaginationland is useful because imaginary Kyle will suck his cock there. But, if you aren't Eric Cartman, what is the point of getting a PhD in Econ from Cambridge if you are going to babble such nonsense? 

Professor Martins quotes the following addled passage from Sen- 

 “There is thus a certain difference between the form of physical laws

which are 'Phusis' or the subject matter of Physics which deals with things which are indiscernibly identical save by number or location. They are 'nomothetic'. Leibnitz's law of identity applies as is often also the case with the 'principle of continuity' (whereby infinite sets behave like finite ones) 

and that of social laws (. . .).

which are Nomos or human laws and which are ideographic and feature arbitrary 'uncorrelated asymmetries' and 'bourgeois strategies'. Also the 'masked man' or intensional fallacy forbids the use of Liebniz's law of identity. Furthermore, there are no well defined sets or 'extensions' for anything 'epistemic'. Even anti-reflexivity is out of the window. Maybe 'constructor theory' can be useful to Econ. But only for useful stuff. Even the Highest Math can't turn talking bollocks into anything useful. 

The difference is, however, not necessarily due to any fundamental dissimilarity between physical and human causal relations (the human relations may ultimately be purely biochemical), but due to the number of variables involved in any function”

Nope. There is a fundamental dissimilarity because humans aren't indiscernibly identical save for very limited purposes of short duration. Thus Nomos is always arbitrary, analogical, and has 'the right to be wrong'.  

(Sen 1959: 105). Of course, Sen does not necessarily embrace any type of physical reductionism here, since he only suggests that human relations may ultimately bio-chemical, not that we are sure they are. Sen seems to imply here that we are only sure of the existence of causal relations in the social realm,

No. We are sure of correlations- or Granger causality. But we are often wrong. I think Putin is a greedy bastard who wants to get his hands on valuable Ukrainian real estate. Smart peeps working for the NIA may know differently.  

not of the level of reality where causality is placed. Sen does not clarify whether the higher “number of variables involved” leads to emergent properties in the social realm.

The system is not well enough specified for any such thing. This is merely a manner of speaking. What causes the 'social behaviour' of academics who write nonsense is that they get paid whether or not the do or teach anything useful. There is a 'citation cartel' but the subject attracts only the very stupid or useless so it soon becomes obviously and utterly shit. 

No comments: