Nozick's sceptic is incoherent

What mistake is Nozick making when he writes

You think you are seeing these words, but could you not be hallucinating?

The answer is- he is asserting that when seeing something, a thought occurs whose content is 'I am seeing such and such'. Why on earth would any such thought occur? What purpose could it serve? Why waste energy on it?

Of course, its is possible to think 'I am seeing these words' as you actually see these words- but this is counterfactual. Nozick is comparing two counterfactuals but doesn't realise it. One arises where you think 'I am seeing these words', though it is very unlikely you actually have any such thought- though, you may have another thought which is- 'oh! I suppose, now I think about it, I think I am seeing these words'-  and the other arises where, by sheer chance, you hallucinate reading this sentence on my blog though actually watching porn.


Suppose you suffer from a mental illness or have recently ingested a psychotomimetic drug. You may need to do a quite complex 'path analysis'. The exogenous variable is 'these words', the endogenous variables are 'seeing' or 'hallucinating'. Suppose, as you read these words, you become alarmed because, suddenly, you are addressed by name and, a few sentences later, some information about your present circumstances, known to you alone, is suddenly displayed. Your sense of horror mounts as the words you are reading spell out the horrible mutilation that will be inflicted on you by a demented serial killer who is even now jumping over your garden fence with an axe in his hand. You suddenly recall an incident from your childhood when your elder brother locked you in the potting shed. For some reason, your eyes return to the screen and you see that these words are describing with pin point accuracy, the musty smell of that shed.

At this point, your 'path analysis' should have convinced you that you are hallucinating. You are not seeing whatever words appear on the screen when you read this blog. Why? Things you see can be described in causal terms as a directed acyclic graph based on an exogenous input. Of course, it is possible that this blog is written by an omniscient being or one who has some method to inspect the internal workings of your mind and predict the trajectory of your thoughts- but, in that case, this same blog would also be able to solve mathematical problems of great complexity. At the very least, it should be able to tell you how to hack the KGB's servers so as to give you access to Trump's Golden Shower tapes- something which would make you very rich almost instantaneously.

Thus, you have a way to get the hallucination to prove it isn't one- and to disregard it if it does not do so.
More generally, there is a causal analysis framework- for example, that championed by Judea Pearl- which has salience.

Indeed, this was true even when Nozick was formulating his theory of explanations which was motivated by the problem of scepticism.

Suppose a sceptic, paraphrasing Nozick,  stands before you saying-

'could you not be floating in a tank while super-psychologists stimulate your brain electrochemically to produce exactly the same experiences as you are now having, or even to produce the whole sequence of experiences you have had in your lifetime thus far? If one of these other things was happening, your experience would be exactly the same as it now is. So how can you know none of them is happening? Yet if you do not know these possibilities don't hold, how can you know you are listening to me now? If you do not know you haven't always been floating in the tank at the mercy of the psychologists, how can you know anything-what your name is, who your parents were, where you come from?'

Your reply is- 'Thank you for giving me this valuable wrist-watch of yours. In return for this gift, I comply with your request to keep punching you in the gut till you vomit- which you have assured me is a necessary medical procedure to save your life.'

If the skeptic is stronger than you or might knife you or report you to the police, the natural thing to do is to say 'Oh. This is awkward. The truth is I'm not sexually attracted to you at all. Nobody is. And, yes, you are right. It is because you smell like shit and have a face like a butthole and talk shite incessantly. Do please kill yourself. As you say, you'd be doing the world a favour.'

Nozick says-

The sceptic argues that we do not know what we think we do.

But in that case we can't know any sceptical argument. It might turn out to be the Nicaraguan horcrux of the neighbor's cat.  Thus scepticism can't exist as a coherent epistemological position. The thing is self-defeating.

Nozick describes how self-defeat feels- in this case at the hands of an inconsistent sceptic

Even when he leaves us unconverted, he leaves us confused. Granting that we do know, how can we? Given these other possibilities he poses, how is knowledge possible?

Knowledge is useful stuff. But what is useful? That is determined by the fitness landscape. Whatever helped us thrive was Knowledge, and if it continues to do so it remains Knowledge. Since the fitness landscape is likely to change- the strong possibility exists that we need to get ahead of the game by getting possibly useful types of 'might be Knowledge' ready.

Nozick gets that talking shite about skepticism is useless. He also gets that thinking can be useful. Unfortunately he does not say the sensible thing- viz.  'Philosophers need to show how their distinctions without a difference are 'might be' Knowledges. Only the fitness landscape will turn 'might be' into 'is'.

Instead he says there is a 'how possible' knowledge- i.e. a potential explanation for something we might encounter- and it is richly suggestive and potentially useful to dwell on it.

In answering this question. we do not seek to convince the sceptic, but rather to formulate hypotheses about knowledge and our connection to facts that show how knowledge can exist even given the sceptic's possibilities.

How 'knowledge can exist' means 'what way of looking at things might turn out to be useful?'. A thought experiment which begins 'wouldn't it be cool if we could do something impossible- like travel at the speed of light?' may turn out to be very useful if it leads to rigorous thinking about what kind of knowledge we would be able to get under those circumstances.

This sort of 'gedanken' or 'thought experiment' could be called philosophical and might turn out to be 'paradigm busting'. However once it becomes useful, e.g. makes a prediction that is verified, it would cease to be philosophical.

A sceptical possibility might be very useful. It may enable us to say- this complicated problem has a simple solution because there is no way of proving that we don't live in the simpler world (even if it is absurd) than the one we actually live in.

These hypotheses must reconcile our belief that we know things with our belief that the sceptical possibilities are logical possibilities.

There is no such need. We can believe anything we like and still be able to do something useful.  On the other hand, we may believe whatever is required of us by an employer and yet do nothing for him anyone finds useful. In the short run, employers may think they need to hire people who share their beliefs. In the middle to long run, they realise that beliefs are irrelevant. All that matters is whether a person predictably does something useful.

The problem with ranking theories, as applied to Beliefs, is that their fitness landscape is overshadowed. The truth is, beliefs serve a certain evolutionary function and are likely to be mimetic in propagation.

The sceptical possibilities, and the threats they pose to our knowledge, depend upon our knowing things (if we do) mediately, through or by way of something else.

Sceptical possibilities aren't a threat to stuff which is useful. However, it is worthwhile asking under what conditions our knowledge might stop being useful and to prepare for that eventuality. It may be that there is a general solution to a problem which is simpler and more robust than our current jury-rigged heuristic.

Nozick thinks-

Our thinking or believing that some fact p holds is connected somehow to the fact that p, but is not itself identical with that fact.

Is 'our thinking' this useful? If not, it isn't knowledge. What matters is who is doing the thinking. If it is me, the thing is useless because I'm a fuckwit. If it is some A.I maven, it may not be. What matters is what sort of actions the belief has a conceptual tie to. In my case, the result may be the writing of a shite poem which nobody would read. In the case of the A.I maven, it may lead to some change in the protocols governing database access or something technical of that sort. This approach may lead to 'belief' being defined as 'having a conceptual tie to action' and serve to partition the information set.

On the other hand, saying

Intermediate links establish the connection.

is likely to be useless. You'd just end up with an endless regress of such things, unless these 'intermediate links' achieved modularity, in the manner of a Bayesian network, with arrows always pointing from causes to effects. In this case, one could always improve the specification of the cause, or turn a 'black box' into a 'white box', with the help of some other type or research.

Indeed, Nozick himself seems to be referring to this possibility-

This leaves room for the possibility of these intermediate stages holding and producing our belief that p, without the fact that p being at the other end. The intermediate stages arise in a completely different manner, one not involving the fact that p although giving rise to the appearance that p holds true.

If an intermediate state produces our belief that p, then that stage is a module with a particular function. Is that function independent of the 'wet-ware' we have in our brain? Is it something implementable on a Turing machine? If so, we would be obliged to say that a lot of beliefs we actually have, example P not equal to NP, aren't beliefs at all because a Turing machine couldn't have them. Of course, we can get around this by stipulating for 'Random Access Memory'. But, in that case, 'wet-ware' does matter in that nodes within a random access network would feature variation in transmission protocols. This is an interesting line of inquiry but it is not really philosophical- at least at Nozick's level of argumentation- because it assumes that a particular problem we know to be open is closed in a manner we don't believe at all likely.

Are the sceptic's possibilities indeed logically possible? Imagine reading a science fiction story in which someone is raised from birth floating in a tank with psychologists stimulating his brain. The story could go on to tell of the person's reactions when he is brought out of the tank, of how the psychologists convince him of what had been happening to him, or how they fail to do so. This story is coherent, there is nothing self-contradictory or otherwise impossible about it. Nor is there anything incoherent in imagining that you are now in this situation, at a time before being taken out of the tank.

Unfortunately, there is also nothing incoherent in imagining that your unconsciousness is stimulating the psychologists' brains so that they are actually only doing what you make them do. They believe they are experimenting on your when actually the reverse is the case!

Of course, the true culprit is the Nicaraguan horcrux of the neighbor's cat- but that's a twist you saw coming because of the transcendental declension of the hypotenuse of the Neo-Liberal platypus.

To ease the transition out, to prepare the way, perhaps the psychologists will give the person in the tank thoughts of whether floating in the tank is possible, or the experience of reading a book that discusses this possibility, even one that discusses their easing his transition.

They do so in vain. They are clearly the meat-puppets of a Neo-Liberal platypus.

(Free will presents no insuperable problem for this possibility. Perhaps the psychologists caused all your experiences of choice, including the feeling of freely choosing; or perhaps you do freely choose to act while they, cutting the effector circuit, continue the scenario from there.) Some philosophers have attempted to demonstrate there is no such coherent possibility of this sort.' However, for any reasoning that purports to show this sceptical possibility cannot occur, we can imagine the psychologists of our science fiction story feeding it to their tank-subject, along with the (inaccurate) feeling that the reasoning is cogent.

But, we also know that those silly psychologists will go mad from frustration when the subject calmly informs them that their strings are being pulled by a platypus.

So how much trust can be placed in the apparent cogency of an argument to show the sceptical possibility isn't coherent?

A lot. That is why sceptical arguments have convinced nobody throughout history. Pyrrho returned from India in vain. His leading disciple started off as a lap dancer. These guys should have simply set up a chain of 'Curry in a Hurry' fast food outlets or done the Indian rope trick or something of that sort.

The sceptic's possibility is a logically coherent one, in tension with the existence of (almost all) knowledge; so we seek a hypothesis to explain how, even given the sceptic's possibilities, knowledge is possible. We may worry that such explanatory hypotheses are ad hoc, but this worry will lessen if they yield other facts as well, fit in with other things we believe, and so forth.

Nozick is right that 'explanatory hypotheses' may yield some useful fact or technique. But this is true even if it is caused by a reaction to a Theist, rather than a Sceptic's, argument. Indeed, an apple falling from a tree and hitting Sir Isaac Newton on the head may given one a useful 'explanatory hypothesis' which may motivate one to make important breakthroughs in the theory of Quantum Gravitation by hitting one's Physics Professor on the head with an i-Pad. Anyway, that's what I would do if I got a MacArthur genius grant.

Indeed, the theory of knowledge that follows was not developed in order to explain how knowledge is possible. Rather, the motivation was external to epistemology; only after the account of knowledge was developed for another purpose did I notice its consequences for scepticism, for understanding how knowledge is possible. So whatever other defects the explanation might have, it can hardly be called ad hoc. 

What Nozick had actually done was stumble on the importance of what he calls 'intermediate links' in counterfactual arguments.

Judea Pearl's 'Mediation analysis'-  which aims to uncover the causal pathways along which changes are transmitted from causes to effects has enormous scientific and practical implications. The Mediation Formula he has developed, on the basis of counterfactual logic- which quantifies the fraction of the observed effect “owed” to any given mediating path-could kill off a lot of the Junk Social Science we still see.

More generally, his causal approach disposes of Nozick's program.

Our task is to formulate further conditions to go alongside 

(1) P is true 

(2) S believes that p. 

We would like each condition to be necessary for knowledge, so any case that fails to satisfy it will not be an instance of knowledge. Furthermore, we would like the conditions to be jointly sufficient for knowledge, so any case that satisfies all of them will be an instance of knowledge. We first shall formulate conditions that seem to handle ordinary cases correctly, classifying as knowledge cases which are knowledge, and as non-knowledge cases which are not; then we shall check to see how these conditions handle some difficult cases discussed in the literature. One plausible suggestion is causal, something like: the fact that p (partially) causes S to believe that p, that is, (2) because (1).

What's wrong with that? Most people would go along with it. I believe P because P is true. However, I also say X is true but I have no Beliefs regarding it. Thus, 'my Mum doesn't cook the best dosas in the world' is true, but I don't believe it. It is an empirical fact that I prefer to anticipate the delicious taste of my Mum's masala dosa rather than satisfy my hunger at Sravana Bhawan before taking the Metro home. No doubt, scientists could find a causal explanation for why this is true.

It is not the case that Beliefs are all of just one type and that they have a necessary connection with 'Knowledge'. That is why we have two different words, two independent concepts, for these very different things which are useful in different ways.

But this provides an inhospitable environment for mathematical and ethical knowledge;

Why? Univocal foundations is not inhospitable at all to Mathematics. It enables Type theory to work very productively and makes computer checking of proofs easier.

As for 'ethical knowledge'- surely, this is of different types?

also there are well-known difficulties in specifying the type of causal connection.

So what? Difficulties can be overcome by smart people using Type theory

If someone floating in a tank oblivious to everything around him is given (by direct electrical and chemical stimulation of the brain) the belief that he is floating in a tank with his brain being stimulated, then even though that fact is part of the cause of his belief, still he does not know that it is true.

So what? We can simply say that this belief is of a different type.

Let us consider a different third condition: (3) If p were not true, S would not believe that p. Throughout this work, let us write the subjunctive "if-then" by an arrow, and the negation of a sentence by prefacing "not-" to it. The above condition thus is rewritten as: (3) not-p → not-(S believes that p). This subjunctive condition is not unrelated to the causal condition. Often when the fact that p (partially) causes someone to believe that p, the fact also will be causally necessary for his having the belief without the cause, the effect would not occur. In that case, the subjunctive condition (3) also will be satisfied. Yet this condition is not equivalent to the causal condition. For the causal condition will be satisfied in cases of causal overdetermination, where either two sufficient causes of the effect actually operate, or a back-up cause (of the same effect) would operate if the first one didn't; whereas the subjunctive condition need not hold for these cases.

It is here that Pearl's Mediation formula comes in useful. It distinguishes pathways by developing a directional typology of causes. Why not just junk subjective conditionals in favour of a proper Structural Causal Model?

When the two conditions do agree, causality indicates knowledge because it acts in a manner that makes the subjunctive (3) true. The subjunctive condition (3) serves to exclude cases of the sort first described by Edward Gettier, such as the following. Two other people are in my office and I am justified on the basis of much evidence in believing the first owns a Ford car; though he (now) does not, the second person (a stranger to me) owns one. I believe truly and justifiably that someone (or other) in my office owns a Ford car, but I do not know someone does. Concluded Gettier, knowledge is not simply justified true belief.

There is belief here but it is not justified and it is not true. Suppose a murder had been committed and it was known that a person from my office did the foul deed and that he owned a Ford. I come forward and say 'That is the man!' On the witness stand, however, the defence attorney will show me to be either an inveterate liar or fantasist, or, more sinister yet, the guilty man who only committed the murder so as to frame this person against whom I had a long standing grudge. The prosecutor may initially object to a line of questioning which expresses incredulity that I might fail to have noticed that my colleague had disposed of his car. However, he will soon give the thing up as a bad job. Indeed, his office may contemplate bringing charges against me for perjury. Indeed, I may be forced to shell out a substantial sum in a civil suit for libel.

The following subjunctive, which specifies condition (3) for this Gettier case, is not satisfied: if no one in my office owned a Ford car, I wouldn't believe that someone did. The situation that would obtain if no one in my office owned a Ford is one where the stranger does not (or where he is not in the office); and in that situation I still would believe, as before, that someone in my office does own a Ford, namely, the first person. So the subjunctive condition (3) excludes this Gettier case as a case of knowledge. The subjunctive condition is powerful and intuitive, not so easy to satisfy, yet not so powerful as to rule out everything as an instance of knowledge.

This is silly. My belief is 'X owned a Ford car on such and such a date. If he hasn't sold it and he hasn't been fired unknown to me, then someone in my office owns a Ford'.  In other words, beliefs would themselves have a causal structure exhibiting modularity and each module would need to be checked separately.

A subjunctive conditional "if p were true, q would be true," p → q, does not say that p entails q or that it is logically impossible that p yet not-q. It says that in the situation that would obtain if p were true, q also would be true. This point is brought out especially clearly in recent "possible-worlds" accounts of subjunctives: the subjunctive is true when (roughly) in all those worlds in which p holds true that are closest to the actual world, q also is true. (Examine those worlds in which p holds true closest to the actual world, and see if q holds true in all these.) Whether or not q is true in p worlds that are still farther away from the actual world is irrelevant to the truth of the subjunctive. 

Unfortunately, provided the Razbarov Rudich theorem is true- and there is no way to tell if Possible World is constructed intensionally or extensionally- there is always a closest-possible world in which this subjunctive is false viz. one which differs from our world in no other respect. This argument does not depend on 'miracles' but rather says that Nozick's use of a Lewis-Stalnaker type argument, as given above, must either show how to distinguish the pseudo random from random- i.e. do something genuinely miraculous- or it must keep silent about 'subjunctive conditionals' not grounded in a thoroughgoing  structural causal model which allows itself to be updated and which has a direction in time.