Sen on Objectivity and Position

In Economics, where there is an 'objective function'- i.e. a quantity to be optimized- then, there is some 'non-arbitrary', 'canonical', or 'natural' way to solve the problem. If it is unique we can say it is 'categorical'. This is 'objective' even if the thing being optimized (e.g. 'Profit', 'Expected Utility', etc.) pertains to some subject who has values of his or her own. It could be argued that 'utility' is subjective. I enjoy watching Netflix. You enjoy going to the gym. We don't know who gets more utility or pleasure from these activities. In this case there is something objective- viz. opportunity cost (best alternative foregone)- which is being minimized. We can construct a sequence of transactions such that I end up choosing to go the gym instead of watching Netflix. It is likely, that for this to happen some binding constraint facing me has been removed. Studying situations of this sort can be objectively useful. What is utterly useless or mischievous is pretending that something which is wholly subjective is actually objective. This is what Amartya Sen does. 

Suppose, at a certain level of intellectual or scientific development, it is common to make a particular error- e.g. thinking a heavy object will fall faster than a light object. Is this error 'objective' or 'subjective'? It is subjective. It depends on the lack of intellectual development in the subject. 

Sen disagrees. He thinks that if a particular person thinks the Sun and the Moon are the same size, then this is thinking of theirs is objective. Yet it depends on ignorance which is a feature of a particular subject. 

His belief is subjective not objective. Why? It depends on some fact about his subjective existence- e.g. his being uneducated and stupid- rather than some fact about the Moon or the Sun. Suppose we sympathize with Sen. We grant that a purely subjective judgment is actually objective. In that case, my assertion that Sen eats only dog shit is objective. It is not based on my subjective hatred of Sen. The truth of my belief has to be distinguished from the objectivity of what I decide to believe (given what I observe about the relative abundance of dog turds when Sen is absent from the vicinity. 

Subjectively, a thing may 'feel' true- there is a sense in which it would be right and fitting if Sen devoured only dog turds- but, objectively, it may simply be a pleasant illusion. Thus, I would not spend any money seeking scientific proof that Sen eats dog shit. 

Sen however comes from Bengal where 'Navya Nyaya' flourished.

No. They accepted that some knowledge and some illusions may turn on preexisting concepts but others might not do so. Subjectively, there was the sensation of seeing a snake. But, objectively, there was no snake. However, since Indians in ancient times weren't as stupid as Sen or Matilal, they knew that small animals or children could show fear at seeing something snake-like even if they had no 'preexisting concept' of snake. We currently believe that there are 'co-evolved' processes which lead some animals to have instinctive responses to certain stimuli. Sadly, for 'co-evolved' processes on an uncertain fitness landscape, 'naturality' is far to seek. Perhaps it exists at 'the end of mathematical time'. But it doesn't matter. There is always something useful smart people can be doing. Sadly, Nobel Prize winning Economists are stupid and useless. Consider Sen's belief that there is a Marxist 'objective illusion'- i.e. a rope of sand or a mirage in which you can have a nice swim.

An illusion is still an illusion even if lots of people see it. Marxists don't think Our Lady of Fatima genuinely appears in the Sky even if thousands claim to have seen it. 

Morishima says the fundamental theorem of Marxism is that if profits exist then worker exploitation exists even if nobody is being exploited. But this is the delusion of a paranoiac. It is objectively false save in the sense that Sen truly, objectively, devours nothing but dog turds. 

Obviously, if your job is to tell stupid lies, you can use the concept of 'objective illusion' or 'true lies' to blabber shite. Thus, the fact that people are well paid means they are actually being starved, sodomized and robbed by evil bosses. Furthermore, suppose you buy a bag of basmati rice and use the rice to make a tasty biriyani, what has actually happened is that you have been forced to perform some horrible type of fetishism instead of having had a warm, human relationship with a paddy grower in the Punjab. 

Every Indian economist knows that Indian statistics are shit. Objectively, morbidity was higher in some districts of UP and Bihar for economic and epidemiological reasons. There was no fucking 'positional' aspect to objectivity. There was just ignorant nonsense. 

Sen specializes in seeing one thing- e.g. a cat- as something else entirely- e.g. a hat. True you could put a cat on your head but that cat is likely to scratch you or to run away if it begins to rain. 

Subjective probability just means a subject's guess as to the likelihood of an event. It does not depend on the subject's position or standing. Moreover, there is no obligation to follow any particular rule or method in arriving at such probabilities. There is no objective aspect to them whatsoever

Bayesian inference is a method of getting from subjective to objective probabilities. You start with a guess- however wild- and then make an empirical observation and alter your expectation on that basis. One can objectively verify that Bayesian inference is being properly done by checking the relevant calculations at each point in the process. The notion of 'Aumann agreement' is that people with the same 'priors' will have the same expectations. If there is indeed an objective function, this is likely. 

When looking at, or seeking to predict, subjective states of minds or subjective decisions we may look at objective facts about the subject in question. Their position in society may shape their subjective view of the world. But this is a positional view of subjectivity, not of objectivity. 

Positional subjectivity leads to a view of subjectivity which stresses its ability to increase survival value or to serve a signaling or screening purpose. Sen proposes an approach which consists of telling stupid lies about how maybe very sick people aren't sick at all. Using Sen's approach allows stupid Professors and bureaucrats to pretend to be defending Democracy from Hitler by eating dog shit. 

The consequence of telling stupid lies is that, so long as you virtue signal incessantly, people may take you for Mother Theresa and give you a Nobel Prize. 

What is totally arbitrary is Sen's decision that subjectivity actually means objectivity. Deontological considerations focus of duties. Since duties are determined by one's position in society and since they remain invariant (though you may have a defense in law or morality for failing to discharge a duty) and since all deontic systems have consequences and thus are also consequentialist, it follows that Sen is telling stupid lies. 

In his 1993 paper, 'Positional Objectivity', Amartya Sen systematically confuses functional relationships with what are merely coincidences or empirical correlations. Suppose I have a red moustache. I also spend a lot of my time watching porn. This does not mean red moustaches are related to the watching of porn. It is merely a coincidence that I have a red moustache and am a porn addict. 

The subject of this paper is the relationship between the inescapable positionality of observations and the demands of objectivity in science and practical reason.

There is no such relationship. Anything can demand anything of anything. I can demand to be the sister-in-law of Charlie the Walrus. But this creates no relationships of any type whatsoever. It is a different matter that you might say my loud and insistent demands in this respect make you want to shift your position away from my vicinity. But this would be true even if I was merely farting rather than demanding things. 

What we observe depends on our position vis-a vis the object of observation,

No. We only observe the object we want to observe if we are in a proper position to do so. Observation does not depend on position because there are mechanisms whereby impeding obstacles can be evaded- e.g. by using mirrors or cameras etc.  

and that positionality relates to a number of parameters—locational and others—that influence acts of observation.

No. Acts of observation may be impeded or helped by locational and other parameters. But there is no mathematical relation between the two.  

But even though observations are parametrically variable with positions, they are central to our understanding of the world, and thus to science, decisions, and ethics.

Moustaches are parametrically variable with length, thickness, color, etc and though all beings may be said to have a moustache of some degree, it does not follow that moustaches are central to observations or emotions or watching fat naked people slip on a banana peel.  

Objectivity would seem to demand some kind of invariance with respect to particular characteristics of the observer and her circumstances.

So would subjectivity or having a moustache or being a fat person slipping on a banana peel except it isn't a banana peel. Also I wasn't naked. I had a towel around my midriff.  

But the question is: which characteristics should figure in the invariance conditions—and no less importantly, which must not so figure?

None at all. Certain 'observations' are protocol bound- i.e. must meet certain criteria to be 'admissible'- and one type of 'invariance condition' in certain physical sciences is that experiments must be repeatable. Sen should know this. Why is he talking nonsense?  

It is not surprising that objectivity has been seen as a demand for "a view from nowhere" (to quote the title of Thomas Nagel's important book on that subject).

It was stupid shit. It is obvious that 'valuable ideas' are ideas which raise productivity or which alter outcomes in a desirable manner. We don't care if they are 'objective' or 'subjective'. We just need a 'witness' to show the thing works.  

I shall distinguish between two concepts of objectivity:

in which case, one is more subjective than the other and becomes more and more so till it vanishes.  

(1) positional objectivity, and (2) trans-positional objectivity.

A subject has a position. A subject's observation is positional and hence subjective. Thus if there are two eye-witnesses giving testimony and their accounts disagree, we may say the one further away with the obstructed view is likely to be more 'subjective'. His mind filled in the gaps. He thought the big black man snatched the purse of the little old white lady. But the person standing beside the lady sees something else. The black man said 'Granny! I can't let you put this purse which has a bomb inside it on the bus. Instead I will snatch it and run down to the river and dump it in the water.' If we have other evidence which supports this testimony we say it is 'objective'. 

There is only one concept of 'objectivity' and it is 'non-positional' and does not depend on any subjective factor- e.g. what some of person the observer is, where they are located, etc. 

Briefly put (though with some oversimplification, as will be discussed later), the distinction is this. How an object appears from a certain position of observation is an objective inquiry in which the observational position is specified (rather than being treated as an unspecified intrusion—a scientific nuisance) . Any attempt at non-positional objectivity has to start with knowledge based on positional observations and then go beyond that, and in that sense this is really an idea of trans-positional objectivity (rather than one that does without positional objectivity altogether).

No. There is simply no such thing as 'trans-positional objectivity'. We may as well speak of moustache-color-blind objectivity. 

 If, for example, we are trying to find out how a phenomenon would appear to a person occupying a particular position, then clearly that positional view of the phenomenon is exactly what we are looking for.

No. We are looking for that particular person's view of it. Position may be wholly irrelevant. We discover the person is blind. They didn't see shit. Equally, the person has 20/20 vision but was paid to look the other way.  

Now, we may be interested in knowing what a person with good eyesight would see if placed in various different positions- e.g. within a cinema theater. We may also want to do 3-D simulations of architectural plans so as to get an idea of what sort of views would be available from the high price penthouses. But this would not be a wholly 'objective' exercise. There is a specific sort of 'representative agent' whose tastes and pre-existing psychological and cultural conditioning we need to cater to. 

It is, then, part and parcel of that objective inquiry, rather than an illegitimate incursion of subjective features.

No. Commerce has no problem with subjective features. Nothing is as objective as the bottom line.  

Positional variability does not necessarily provide counterevidence to the objectivity of observational statements.

It is stupid nonsense.  

If I say that the moon looks small from where I am,

you are speaking subjectively about yourself. You are a subject. 

I need not be accused of deep subjectivity—another person seeing the moon from where I am could confirm that observational fad.

Or not. Generally, we experience the moon as being bigger or smaller depending on how low it is in the horizon and whether it is 'framed' by something we consider small or big.  

Nor is that observational claim contradicted by what we know—from other evidence—about the mass of the moon,

subjective claims are not refuted by objective datum.  

or by the fact that the moon looked big enough to Neil Armstrong while taking his "one small step for a man, one giant leap for mankind." Indeed, in an obvious sense, there can be no view from nowhere.

But the phrase is figurative. It has no fucking obvious sense.  

Something is subjective when it pertains to a subject. A 'positional observation' is subjective because a particular subject occupies a particular position. Of course, to make money or win votes we may have an objective function to maximize. But what we want to find out about is subjects and their subjectivity so as to gain profit or power. Sen disagrees. He says the fact that we may have objective reasons to want to grasp

 positional observations (this ) does not make those observations, in any sense, subjective.

Suppose you are screening your new movie to a test audience. One guy leaves early. You say 'didn't you like the movie?' and the guy says 'I had a terrible seat. I couldn't see anything. The movie may have been okay for all I know.' Suppose you mention this to the cinema manager. He says 'all the seats in that section are shit. The fact that only one guy left shows that the movie is dynamite.' It may be an objective datum that the seats are bad but the reactions of the audience are subjective. One guy left because the seats were terrible. The others are praising it. They say they will come back to see the film at its next showing. Hopefully, they will get better seats! 

But what is this idea of subjectivity from which I am trying to distinguish the notion of positional relativity?

Stupidity. Sen has shit for brains.  

Am I using some odd notion of subjectivity—different from common use—to make these distinctions? I would argue that, on the contrary, the commonly shared idea of subjectivity is much in line with the usage here. In the Oxford English Dictionary subjectivity is characterized as: (1) "having its source in the mind," and (2) "pertaining or peculiar to an individual subject or his mental operations."The first feature, which I shall call "mental manufacture," relates to the inward-looking nature of subjective judgements or theories.

Subjective judgments are merely judgments of a subject. They are not necessarily 'inward looking'. 

The honest racist who is persuaded without any attempt at actual observation that members of a certain race suffer from some terrible character defects is subjective in relying on his personal imagination on matters in which observations—direct or indirect—must be importantly relevant.

Not necessarily. One may be an honest racist who holds strategic beliefs.  

Objectivity here will demand

nothing. A subject is a subject even if everybody demands that it turn into an object up the rectum of a large Walrus of unfriendly disposition.  

the use of careful observations (rather than mental manufacture), even though it must come to terms with the inescapable fact that what is actually observed may be significantly dependent on the position from which the observation is undertaken.

Rubbish! The honest racist can maintain that his own race is superior while being careful to hire smart peeps from the despised race.  

The second aspect (that is, "pertaining or peculiar to an individual subject or his mental operations") has a different, though not unrelated, focus. I shall call this the aspect of "person sensitivity." No matter how exactly it is characterized, objectivity must satisfy some requirements of inter-personal invariance.

No. It must merely be unrelated to any given subject. Suppose public offices are allotted by some stochastic mechanism- e.g. an urn from which each person picks a stone. This is an objective procedure. It does not satisfy any requirements of an interpersonal type.  

An observation may be inevitably position-dependent, but it would lack something in credibility if others viewing the object from the same parametric position could not see what this subject sees.

This is irrelevant. Credibility is subjective. It does not change objective facts. I may say 'I am actually very thin. There is something wrong with this weighing machine!'. People may pretend to sympathize. Actually, they are laughing at me. 

Positional Judgements and Consequential Ethics 

The issue of positionality in objectivity

does not arise. Objects have no interiority. They don't change when their position changes within an inertial frame. 

arises not only in science and epistemology, but also in ethics and in the theory of decisions.

No. Stupid cunts like Sen can talk bollocks by pretending subjectivity is objectivity but they are brown monkeys who managed to escape a starving shithole. 

The need for agent relativity has been seen as an argument against consequcntialist ethics,

the argument against it is that it is shit and a fucking waste of fucking time. 

for its alleged failure to deal with an important deontological distinction. To take a much discussed example, there have been interesting analyses of the ethical difference between (1) killing someone oneself, and (2) failing to prevent a murder committed by a third person.

which is like the difference between eating your own shit and farting while taking a dump.  

The former has been seen, not implausibly, in even more negative terms than the latter.

By people who prefer not to eat their own shit. 

The relevance of this distinction has been interpreted

by people who don't know about 'disutility'- e.g. your discomfort at having to eat your own shit

as evidence of the inadequacy of consequentialism as an ethical approach. Even though the consequences are "the same" in the two cases (including, a person being murdered), the ethical case against committing a murder oneself can be said to be much stronger than that against failing to prevent a murder committed by another person.

It depends. If you are a killer by profession and you are paid to kill guys who are trying to kill your client, just get busy killing already. 

 And yet, killing someone oneself is seen— with reason—as a bigger personal failure than non-prevention of a murder being committed by another.

Not if that is your job. 

Hence the conclusion that consequentialism must be rejected.

The consequence of not doing your job is you get fired. The relevant solution concept here is 'uncorrelated asymmetries'. I don't care if you smell my farts even if you say you can fucking taste them. What I won't do is eat my own shit. This is because I care about me not about you because I am me, not you. 

But the entire argument turns on the requirement that consequences be evaluated in a person-neutral way, despite the connections between the agent doing the evaluation and her own roles in the respective states of affairs.

John Maynard Smith solved this problem long ago. Crack a book sometime.  There is an obvious information asymmetry between me- I happen to know when I need to pee- and you. I don't know when you need to pee. Thus, there are 'uncorrelated asymmetries' which dictate the bourgeois strategy whereby I am responsible for emptying my bladder while you are entrusted with that task for your own bladder. 

Is there any substance at all to Sen's paper? He stresses higher reported morbidity in rich countries and comes to the conclusion that poor people are fatalistic and get less disutility for illness or dropping dead anytime a mosquito bites them. The fallacy here is obvious. 'Reported morbidity' has to do with surveys which are very differently carried out in Kerala compared to UP or Kerala compared to US. Speaking generally, the statistics are made up or meaningless. However, there are objective morbidity functions which, sadly, show me as unhealthier than many much poorer people in India precisely because I have superior command of Pizza and Netflix. 

The importance of positionality can be enormous in understanding health, well-being, and relative deprivation.

No. It is useless. True, if skeletons in American medical schools were sourced from very poor countries like India, a generation of Doctors may have been mislead about what was or wasn't healthy bone density, etc. But it is easy to change the 'objective function' to fit specific genetic lineages with a common epigenetic expression.

The distorted observations are not proof of their trans-positional truth or objectivity, but nor are they dismissable as purely subjective features of the persons involved.

They are simply shitty observations. Back in the Nineties there was an Indian household survey which showed 

1) few rural Indians had watches, clocks or even radios and thus did not know what time they went to sleep

2) rural Indians could report that they went to sleep at 8.12 PM even though they didn't have watches or clocks. 

In other words, the survey was junk. People just made up plausible answers. Obviously, you couldn't show everybody as having a watch or a clock because the supply side figures were deficient. Still, if you had to fill out a time for going to sleep you could put down 8.12 PM and nobody would call you on it because everybody knew the entire exercise was shit.  

They are systematic and patterned, and can hardly be attributed to whims and quirks of mental manufacture or to purely personal peculiarities.

They are stupid lies. Sen saw himself as the Mother Theresa of imaginary grievances. Why else was Whitey patting him on the back?  

The positionality of perspectives—and the idea of objectivity within those perspectives—provides a framework in terms of which these systematic findings can be analysed and understood.

Sen's stupid lies are meant to foster a culture of grievance. Hopefully, China will conquer America soon. Then the cup of the Buddhijivi will overflow. Sadly, under Modi, Hindu India doesn't seem to want to sink down. Hopefully Biden will do the needful. Mind it!