Friday, February 28, 2014

Karl Popper's response to Hume

To: Mark S.
From: Geoffrey Klempner
Subject: Karl Popper's response to Hume
Date: 15th November 2011 12:38

Dear Mark,

Thank you for your email of 6 November, with your essay for the University of London BA Methodology module, in response to the question, 'Does Popper provide an adequate response to Hume's problem of induction?'

You have taken this question as an invitation to discuss the merits of Popper's account of falsifiability as a descriptive and normative account of scientific practice.

As you have covered many of the main lines of criticism and defence, I was surprised you didn't mention the difficulty posed by existential statements, such as, 'There is an uncharged subatomic particle which has exactly one half the mass of a neutron' (I don't know whether this is true or not -- sounds false to me). This looks like a statement you could verify, at least by the standards of particle collision data (whose interpretation is of course heavily theory dependent) but it cannot be falsified.

Well, so what. A physicist wouldn't say this unless they had a theory which predicted it. A theory which only has positive existential claims like 'somewhere in the universe there is an uncharged particle with half the mass of a neutron' as its logical consequences is not scientific by Popper's criterion.

At one point in your essay you note that Popper is not seeking to show that induction is rationally justified. Of course not. His response is that the attempt to seek a rational justification for induction is just a wild goose chase.

However, the problem with this easy answer, which arguably should have been a focus of your essay, is that Hume makes precisely this claim. The 'problem' of induction for Hume is not something which he treats in the same way as the seemingly paradoxical conclusion (in 'On Scepticism With Regard to the Senses') that we cannot make sense of the assertion that objects which we do not perceive have a 'continued' and 'distinct' existence. (The only solution is to give up philosophy and take a rest from these 'strained speculations'.)

For Hume, the process of induction describes how beliefs are in fact formed. Moreover, there are 'rules for judging causes and effects', which themselves have arisen from the same source (habit), which allow us to 'reason' about alternative theories and decide which is 'best'. The point being that there is no standpoint or foundation from which we could judge the rationality of induction, or science, as a whole. It's just what we do.

Popper is right (in my view) to respond that Hume's account is to a considerable extent infected with the 'Baconian myth', but then Hume's interest was different. His 'rules' are elementary, hardly a philosophy of science or a methodology. You can read Popper as pointing out that there is something really interesting here, about how far actual scientific practice deviates from the mythical inductivist picture. With a bit of added hyperbole, maybe.

You are right to emphasize that the key question is why we rely on the results of science. But not all science, of course. There are the games cosmologists or particle physicists play, which seem to have little significance for the 'real' world. Not a lot different, in fact, from the games scholars of the Presocratics play, arguing the toss over different interpretations of Heraclitus. (Imagine you are a bemused bystander who in a misguided attempt to end the argument tries to formulate a 'philosophy of Presocratic interpretation'.)

However, at the back of the practical application of science lies our absolute, unremitting faith in the uniformity of nature. When you put questions to nature the answer will always be the truth, the whole truth and nothing but. It's up to us to put the right questions -- so often we fail to do this, or else lamely misinterpret the answers that nature gives. I don't see any merit at all in attempting to justify this faith or demonstrate its 'rationality' and neither did Hume.

If you asked Popper, what would he say? Is 'nature always tells the truth' also a falsifiable hypothesis, on the same level as all the rest? It's something we hold on faith, a presupposition of scientific inquiry. It seems to me perfectly acceptable to say that you don't judge the ultimate presuppositions of science by the standards of science. What an absurd idea!

Nitpicking objections apart, the great strength of Popper's vision lies in his realization that when we do science, there are always several theories 'on the table', some of which we like a lot, others which we don't like nearly so much but can't rule out. There are no logical or methodological rules for 'liking'. A lot depends on fashion, grant funding, etc. Which is not to say that anarchy rules, on the contrary (I've made his point before).

However, you don't build bridges on the basis of your favourite theory. That would be a rather silly thing to do. When it comes to bridges something else comes into play. What exactly is this? What response could Popper offer here other than the Humean response, which he apparently rejects? But does he, really?

To offer a 'conjecture and refutation' model for science (and maybe for Presocratic interpretation too -- see Popper's 'Back to the Presocratics') is not to claim that all human beliefs follow this pattern. If the presuppositions of science are not science, then neither are the basic beliefs which guide our everyday actions, Humean style, habits which we would never once think of questioning. This isn't necessarily an answer, or an adequate defence of Popper. He owes us an account of what it means to be a 'rational agent' at the basic, nitty gritty level. Corroboration looks like such an account. All Popper needs to do is swallow his pride and acknowledge his indebtedness to Hume.

All the best,


Are possible worlds really 'real'?

To: Bernd K.
From: Geoffrey Klempner
Subject: Are possible worlds really 'real'?
Date: 15th November 2011 13:21

Dear Bernd,

Thank you for your email of 7 November, with your first essay for Possible World Machine, in response to the question, 'Are possible worlds really 'real'?'

It's fair to say that you don't really answer this question, although you initially go about this in the right way. If you hadn't stopped when you exhausted the 800 word limit (which is in fact generously elastic, I should have told you, sorry!) I don't think from what you have indicated that you would have got there.

Just one point: it was Berkeley, not Descartes, who denied the existence of a material world of objects in space. Descartes raises the question of scepticism about an external world then answers his own question by means of the God hypothesis and the theory of mind-body dualism.

You are right that there is (at least in the current scientific climate) a strong disposition towards regarding questions about 'reality' as being measured by the standards of physical existence, dualism notwithstanding. However, what is interesting about the question whether possible worlds are 'really real' is that a materialist can happily debate this with a Cartesian dualist, neither having to give ground on their fundamental metaphysical theory. (Or with a Berkeleian idealist, for that matter.)

The key point, which you do raise, concerns relation to time. Even the 'worlds' generated in the many-world theory of QM trace back in a tree-like structure to specific points in time (e.g. an electron either does or does not fall into a lower orbit releasing a photon). Whereas, according to David Lewis, the chief proponent of the 'really real' theory of possible worlds (in his books 'Counterfactuals' and 'On the Plurality of Worlds') possible worlds each exist entirely in their own space and time with no point of contact with the actual world.

It makes to sense to ask the question 'when' the possible world where GK decided to take the day off today 'occurred'. It is not as if somewhere, right now, a counterpart of GK is enjoying a late, lazy breakfast. This is something I might have done, despite the pile of work on my desk, but wisely didn't do.

In the 'larger world of memories and imagination' (your last sentence) worlds exist, but these are just imagined, made up, constructed. If this is how we should understand the ontology of possible worlds, then that is tantamount to saying that possible worlds are not really real. They are not real at all. They exist only in the imagination.

Now one can debate whether, e.g. unicorns or Santa Claus exist 'only' in the imagination, or whether they have a more solid, cultural basis. There's all the difference in the world between one person's overheated imagination, and the shared memories/ imaginings of a significant portion of the human race. But still they are not 'real'. A unicorn, it is said, can detect whether a woman is a virgin or not. You don't need to undertake a scientific test for this, because this is true simply by virtue of myth and fable.

David Lewis is not being fanciful when he asserts that possible worlds are not like unicorns or Santa Claus (although there are possible worlds where 'unicorns' or 'Santa Claus' do actually exist). The existence of possible worlds follows as a logical/ metaphysical principle, if you accept a plausible view about the nature of language, viz. that the meaning of a statement is given by its truth conditions (a view first formulated by the great German mathematician Gottlob Frege).

It is the argument which is put into the mouth of Dr Phillips' student Brenda, in unit 1: 'What we can or cannot imagine, or what we think about possible worlds, is not what makes those worlds real. Because we can be wrong. Our thoughts about possible worlds are true or false depending on something -- whatever it is -- that is somehow independent of those thoughts. What makes possible worlds real, in other words, can't simply be our thinking about them. Our minds discover something that has a reality independent of our minds.'

When I assert, 'There is a mug of lukewarm coffee on my desk,' my statement has truth conditions, it is true if the facts are a particular way, and false if the facts are a different way. I'm telling you that it is true (you have to believe me). It is made true by the thing on my desk which is, in fact, a mug of lukewarm coffee.

When I assert, 'If I had taken the day off today I would have enjoyed a lazy late breakfast', my statement according to David Lewis also has truth conditions. But what could these be? What makes it true, if it is true, that I would have had a late breakfast rather than setting off at the crack of dawn with my camera? Lewis's answer: The similarity of possible worlds makes it true. In the 'nearest', most 'similar' possible worlds to the actual world, I enjoy a lazy breakfast. That's a fact, just like the fact about the mug of coffee. If it is a fact, then possible worlds must be 'really real'.

How good is that argument? Where's the loophole? Surely, we don't want to be lumbered with this massive metaphysical extravagance! Or is it? What would Occam say? But then he would owe us an alternative, workable account of the semantics for counterfactual conditionals.

All the best,


Can truth be defined?

To: Kyriakos C.
From: Geoffrey Klempner
Subject: Can truth be defined?
Date: 9th November 2011 13:04

Dear Kyriakos,

Thank you for your email of 31 October, with your third essay for Possible World Machine in response to the question, 'Can truth be defined? If you think it can, give a definition and explain its philosophical significance. If you think that it cannot, what considerations should the philosopher draw from that?'

Heidegger makes much of the point that for the Greeks truth was 'aletheia', that which is 'unconcealed', brought into the light of day (I understand that 'lethe' can refer to forgetting or concealing). What does this imply? You make the point that for the Greeks, the actual word for 'truth' can be deciphered as a complex thought. It is not just a label.

One could say that 'truth is that which is unconcealed' is intended as a definition of truth. Something that shows itself, that remains in the open, which we will not permit to be lost (forgotten). Yet this already poses a problem. Because facts, as such, are never just 'in the open'. We investigate the world, sometimes with great patience and persistence, and an answer comes to our questions (the scientist 'puts questions to nature'). But this answer is an interpretation, a deduction or induction, a conclusion. It is not just 'something seen'. It is almost as if the Greeks defined their 'truth' or 'aletheia' in a way which makes the acquisition of truth well nigh impossible. All our patient striving merely takes us from A to B, and from B to C. We never get to the truth, as such.

Yet Parmenides, repeating the words of the Goddess, asks us to believe that he has attained the 'well-rounded truth', 'It is', is all one needs to say. The truth, the whole truth, is all that follows from this proposition, and everything else is illusion.

But why can't 'true' be a useful label? Let the Greeks have their 'aletheia' and let's just concentrate on whether the word we use, in everyday discourse, has a definition. You are the second student whose work I have responded to today. That's true. Adding the label 'true' doesn't add any information. If you didn't believe me in the first place then assuring you that what I say is true wouldn't convince you either. But maybe there are circumstances where the use of the label 'true' is useful and necessary.

Wittgenstein in the 'Philosophical Investigations' argues that 'agreement in judgements' is the necessary condition for the very existence of meaningful discourse (contradicting the principle he held in the 'Tractatus' that we cannot allow that 'whether a proposition has meaning depends on whether another proposition is true'). If truth depends on, or is defined in terms of 'agreement', how can one leave open (as one surely must) the possibility, however seemingly remote, that we are all wrong? 'Our' world is flat, say. That's the truth, so far as we are concerned.

I would regard this as a fatal problem if one were seeking to define truth in terms of agreement. However, I see no contradiction in asserting (in Kantian, transcendental style) that the possibility of truth depends on the possibility of agreement, without taking the further step of attempting a definition in these terms.

In the new Study Partners forum I (or my alter ego) respond to a problem raised by my student Joao Magalhaes, regarding the need to refer to truth in order to define knowledge in the traditional way ('justified true belief'). Why not just say:
Joao knows that Mars is the 2nd planet from the Sun if and only if:

1. Joao believes that Mars is the 2nd planet from the sun.
2. Joao has good justification for this belief.
3. Mars is the 2nd planet from the sun.

Now, repeat for 1st, 3rd, 4th, etc.

Repeat for Mercury, Venus, Earth, etc.

Repeat for any other sentence you like, until you get bored.

Job done.
In formal terms, 'true' is a device for propositional quantification. Seen as a mere 'label' the term 'true' enables us to say things like, 'Everything Kyriakos says in his last email is true.' By saying this I am expressing my complete agreement with you. On the matters discussed, we both think the same way. Of course (it goes without saying) we could both be wrong. Maybe the earth is flat after all. But the point is that a definition of truth is not the same as a criterion of truth. A definition of truth doesn't have to give us any additional information (how could it?) about how to determine whether any given statement or proposition is true. It's just a label, a device, and a very useful one at that.

This position is known as 'minimalism about truth'. Another, older, term is the 'redundancy theory' (although there is some debate as to whether these two terms mean exactly the same thing). Some would regard this as equivalent to the rejection of the possibility that truth can be defined. I would prefer to call it a definition. It answers the question, 'What is truth?' in the only way that that question can be answered.

What about 'internal statements'? Wittgenstein's argument against a private language is considered by many to be decisive here, but it leaves a worrying gap (discussed in my article 'Truth and subjective knowledge' Maybe there 'is' something else, something which the label 'true' can never capture, something which each of us 'has' yet cannot express in any language.

All the best,


Functionalism vs behaviourism as theories of the mental

To: Mark S.
From: Geoffrey Klempner
Subject: Functionalism vs behaviourism as theories of the mental
Date: 3rd November 2011 12:19

Dear Mark,

Thank you for your email of 26 October, with your essay for the University of London BA Philosophy of Mind module, in response to the question, ''Functionalism is no better than behaviourism as a theory of the mental.' Discuss.'

This is a very patient (I was tempted to say 'plodding', but in science that's not always a criticism) survey of the various arguments and objections regarding behaviourism and functionalism. I agree largely with the way you have mapped out the terrain. It was inevitable from the start that behaviourism would fail against certain objections where functionalism bravely prevails, and also that against other objections behaviourism and functionalism fare equally well/ badly.

You have also stuck faithfully to the issue of behaviourism vs functionalism, where I might have been tempted to veer off to consider how functionalism relates to, say, connectionism, or to anomalous monism. The only prospect here seems to be a kind of reassurance that one is not looking for (the wrong kind?) of reduction, where supervenience will do. It is one thing to offer functionalism as an umbrella account of the mental, and quite another to pursue projects which involve modelling the mind in terms of concepts from computer science -- something whose point I would be tempted to doubt.

On the contrary, it is clear from the general way the question is formulated that we are considering any version of behaviourism (quite rightly, you limit this to 'analytical' behaviourism) and any version of functionalism. The point that one could be a Cartesian dualist and a functionalist is well taken, but I assume (as you do) that the whole point of the exercise is to resist dualism. (As in the cartoon included in Dennett's 'Consciousness Explained', professor at a blackboard covered in formulae: 'And here a miracle happens!')

But I worry that you may have missed the central point of the question. Your approach assumes that theories can be 'better' or 'worse' in a comparative sense, in the way one might contrast theories of how senility or cancer develop, where the various alternatives are all up for grabs and we are merely concerned with noting pros and cons. Whereas I read the quote in the question in a different way, as a statement that functionalism and behaviourism both FAIL, period, as theories of the mind.

They fail. ''Kruschev was no better than Stalin.' Discuss.' Well, of course, Kruschev introduced all sorts of reforms, bravely criticized his predecessors, but the philosophy he implicitly believed in, Marxism-Leninism, was a total disaster, because...

The point gains some credence from the fact that this is meant to be a one hour essay question (as I've said many times), and therefore the examiners expect you to be able to answer it more or less comfortably in that time. What they don't expect is a page of numbered notes listing all the ways Kruschev was, or was not, better than Stalin!

You can probably gather what would have been the kind of answer I would favour. What is the thing that shoots (claims to shoot) both functionalism and behaviourism down in flames? The person who makes the quote in question clearly believes there is such an objection, and that's what the essay should be about. Either you agree that there is a criticism in the face of which both behaviourism and functionalism miserably fail, or you are fully familiar with the criticism in question and think that it can be resisted. In which case we are then in the game of evaluating pros and cons -- a different essay.

A finesse here might be to indicate why the functionalist 'thinks' they are offering a better alternative to behaviourism, and why, according to a proponent of the knock-down objection, they are sadly deluded. That would gain a mark or two.

A good candidate for the knock-down argument is the problem of qualia, which you describe in various guises. Because this is more central to (my version of) the essay, there would be time to discuss in more detail Wittgenstein's argument against a private language (which I think is compelling, but that's just my view), but also other arguments you mention such as zombies, Chinese room, the million mile high version of Senate House, etc.

How can time make a difference (in the case of the Chinese Room, or Senate House)? I was dying to know the answer to that. Isn't it the utmost parochialism to reject the idea that it might take some gigantic 'creature' a million years to experience a sensation or formulate a single thought? What has time got to do with it? Then again, isn't it parochial to assume that we would ever be in a position to understand/ interpret Senate House's mental states? ('If a lion could speak, we could not understand him', Wittgenstein). I'm assuming that Senate House has arms and legs, or the equivalent (i.e. is an 'agent' with 'its own' needs and goals, not just an information processing device).

Zombies. Would they be different from humans? Massive question! There are two versions of the zombie objection, which make very different points:

Version 1. There might be a zombie double of GK whose behaviour differs in very subtle ways. Most things we can both do, but there are particular tasks at which my zombie double fails, because in order to do them you need that extra 'something' that matter alone, however distributed or organized, cannot supply. The point here is about human ignorance. As would-be materialists, we are merely hypothesizing. We don't know.

Version 2. My zombie double and I are identical in every material respect. But in that case, if I am tempted by the thought experiment to be an epiphenomenalist, wouldn't my zombie double be an 'epiphenomenalist' too?!

All the best,


Hempel's paradox of the ravens

To: Stuart B.
From: Geoffrey Klempner
Subject: Hempel's paradox of the ravens
Date: 2nd November 2011 11:24

Dear Stuart,

Thank you for your email of 24 October, with your essay for the University of London BA Methodology module, in response to the question, 'What is the best response to the paradox of the ravens?'

This is an excellent piece of work with which I can find no real objections.

My main problem (which isn't an objection) is that I just don't believe in 'probability theory'. Not even 'subjective probability' (whatever that means). As an undergraduate, my first ever reading on this topic was the excellent book by A.J. Ayer 'Probability and Evidence', which considers the various pros and cons of the frequency theory, logical relation theory (Keynes) etc. I'm mentioning this as it would be a good book to read. Back then, I wasn't a sceptic but I have become one.

When do we calculate probabilities? and to what purpose? When, and in what way, do we raise a question how much 'support' a theory has? My view would be broadly and loosely Popperian, that any theory which hasn't been decisively refuted is on the table, including theories we do not much like. I would add that preferences for this theory over that theory cut no ice at all. Maybe the choice determines which research project you join, but even there other considerations may override your judgement of the likelihood that the research will prove 'successful' (of course, you can 'succeed' in refuting a theory too). Grant money is always an important factor.

One basic intuition I have is that it is just patently absurd that a white shoe confers any increased probability, however small or even infinitesimal, on the hypothesis 'all ravens are black'. I can see that it would be OK to accept this conclusion as the price for accepting a theory which works well in other ways. No theory is perfect. But it is a flaw, a minor absurdity. It's not a result that we want or can 'justify'. The idea that one could justify the conclusion by some formula seems ridiculous. I'd rather say that if the formula is useful to us in other ways, then we will accept that it occasionally delivers odd or absurd results, but that's a different position.

I agree with you that the question of the nature of probability is a 'major' difficulty. I would add that it's one you can't just sweep under the carpet and say, 'Well, just assume we have a way to calculate probabilities, then...'. One just comes back to the question what probabilities are FOR.

Skill at counting cards in Blackjack (as in 'Rain Man') is one way, objectively verifiable, to tip odds in your favour and make a nice steady income. That's an undeniable fact. We know, in a sense a priori, that we are not going to discover that counting cards actually reduces your chances at winning in the long run (ceteris paribus, one always has to insert that, and provided one can avoid the cameras).

The problem is, in methodology we are tempted by this kind of model -- calculating the odds in order to gain an 'edge' -- where it doesn't really apply. There are too many unknowns, and always will be. What probability theory does do is confer a gloss of rigour and objectivity which adds zero value to the scientific enterprise. There are plenty of opportunities to calculate and apply formulae, but deciding which theory to prefer isn't one of them. As I stated before, all the theories -- even those which look rather implausible -- are on the table. Be prepared to be surprised.

I can therefore totally understand Stephen Hawking's impatience with the philosophy of science and methodology. Bayes' theorem may be useful for all sorts of practical purposes. Of course. But it is practice (or 'praxis') which takes the lead. Not philosophers in armchairs debating how science should be pursued, or what standards a theory 'must' meet.

I guess this will sound rather Feyerabendian, but I'm not really sympathetic to his views either. I appreciate order and rigour in science, and deprecate anarchy. Sociological observations about how science is practiced aren't a substitute for standards and rules. Luckily the editors and editorial boards of professional journals maintain a keen interest in maintaining standards. It's just a pity when these laudable aims are corrupted by the kind of fake 'rigour' which I mentioned earlier. (One thing Feyerabend is good on is observing the slavish way researchers attempt apply 'proper rules and procedures' to situations that don't warrant them, e.g. Masters and Johnson!)

Evidence that is anecdotal, unquantified or unquantifiable, can be valuable too -- in its appropriate place.

I don't see that I am saying anything different from what Aristotle said, that rigour is commendable when applied in the appropriate circumstances. I strongly suspect, therefore, that the whole debate over the ravens paradox is a case of inappropriate application of standards of rigour, which are justified in other circumstances.

All the best,


Why Aristotle needs his four causes

To: Scott B.
From: Geoffrey Klempner
Subject: Why Aristotle needs his four causes
Date: 2nd November 2011 12:18

Dear Scott,

Thank you for your email of 24 October, with your essay for the University of London BA Aristotle module, in response to the question, 'How do Aristotle's four causes relate to each other? Does he need all four of them?'

This is not bad essay, but I have two main problems: The first problem is that I am not persuaded that you fully understand what a 'formal cause' is. The second problem is that there isn't anything here which relates to the specific structure of Aristotle's theory of change, as contrasted with any other theory -- e.g. the view of change that would be taken by a contemporary scientist.

Aristotle states that each of his four causes provide a different answer to the question, 'Why?' We both agree that it's pretty obvious what the question is looking for in the case of efficient, and also in the case of final causes. (Why = what brought something about, vs. Why = for what purpose something was brought about.) So far, so good.

When would you ask 'Why?' in the case of a material cause? It's not quite so obvious, but an example soon resolves the difficulty. 'Why did the spoon melt when I stirred my coffee?!' 'Because it was a joke spoon made of Field's alloy, which melts at 62 degrees C.' What things are made of, their 'material cause', explains what they do, their physical properties and powers.

What about a 'formal' cause? This is also an answer to the question, 'Why?', but which question?

Here, I will follow your examples:

'Why did you give me that horrible jangly metal object for my birthday?!' 'Because you asked for a charm bracelet, and that's what a charm bracelet IS.' 'I didn't realize, sorry.'

'I don't like my feet all covered up like that. Why didn't you make me what I wanted?' 'You asked for shoes, and that's what a shoe is. If you had wanted bare toes you should have asked me to make you a pair of sandals.'

'Look at that, the ball has run away from him!' (said by someone who has never watched a game of soccer before). 'The player passing the ball to his team mate. That's what a pass is.'

The example of the tomato. I don't agree that the final and formal causes are the same. 'That's what it is for a tomato to be ripe', is a different answer to, 'That's why ripe tomatoes are tasty' (the 'purpose' or function of fruit in the biology of seed propagation).

By contrast, I don't see a 'final' cause in the case of the eroded rock, except perhaps if one took the perspective of a wise architect who fashions rocks to be of sufficient permanence for the purposes they serve; or maybe because of the utility of sand, the erosion of rock being the optimal way to achieve this. Whereas, one can perfectly well envisage a question to which the answer would be, 'Because that's just what erosion is.'

In the Socratic dialogues, the question of formal definition assumes supreme importance. 'Why did you prosecute your own father for impiety?' (Socrates to Euthyphro). 'In the name of piety!' 'What IS piety?'

So far, we have completely gone along with Aristotle's four-fold classification. However, there is also a question whether, or to what extent this classification requires, or is required by Aristotle's theory of change in terms of matter and form.

Towards the end of your essay, you talk about genetic codes and their instructions as the modern version of Aristotelian 'entelechy'. There is a problem here, however, in that Aristotle was firmly opposed to the kind of microstructural explanation which this implies. (Of course, there is every possibility that he would change his mind if we could bring him to the present day in a time machine.) Aristotle didn't accept the view of the atomists. He believed that human observation and reason are the ultimate standard by which explanations are judged. Unverifiable speculation about unobservables isn't explanation.

The upshot of this is that in Aristotle's account, formal and final 'causes' play a much heavier theoretical role than they would play for us. It is the formal nature of an acorn, 'what an acorn is' that explains, in the only way this can be explained, why an acorn grows into an oak.

From a contemporary perspective, this looks rather quaint. Yet arguably there are cases, in subatomic physics maybe, where despite the best efforts we cannot find any suitable candidate for an 'underlying structure' (because we are already, or seem to be, at the bottom level) and all we can do is appeal to an entelechy-style Aristotelian 'explanation'.

All the best,


Mind-body problem and the nature of philosophy

To: Graham H.
From: Geoffrey Klempner
Subject: Mind-body problem and the nature of philosophy
Date: 1st November 2011 12:44

Dear Graham,

Thank you for your email of 21 October, with your first essay for the Philosophy of Mind program, in response to the question, 'What is philosophy? Illustrate your answer using the example of the mind-body problem.'

In this essay you do a very good job of raising the question what, if anything philosophy as a distinct discipline has to contribute to the mind-body problem.

It occurs to me that the very phrase 'mind-body problem' has associations which a scientist hostile to the idea of philosophy might reject. Less tendentiously, one might speak instead of the question of 'the place of mind in nature', which leaves open the possibility that the question would be fully answered by results and theories arising from the science of psychology.

Offhand, I can't think of a psychologist whose views about philosophy are as obnoxious as Hawking. But let's assume we have one, whom I shall call Curt. The place of mind in nature is a complex question, studied by biology, experimental psychology, neuroscience and computer science. It takes an interdisciplinarian to thread their way through the different ways in which these areas of science connect and interact, but that's not philosophy. It's part of the scientific enterprise to seek broader views and forge connections between the disciplines.

Curt's view is that all departments of philosophy should be abolished, and specifically that research projects in philosophy of mind transferred to departments of computer science, neurophysiology etc. at a great saving to the taxpayer. How would you defend philosophy, in this scenario?

There's a complication here, in that many contemporary philosophers, following the example of W.V.O. Quine are coming to see their activities and research as continuous with that of scientists. So much so, that you now find departments of 'cognitive science' where it is difficult to draw any line at all between the contributions of academic philosophers and scientific researchers.

From this perspective, one might defend the study of the history of philosophy as useful preparation and groundwork for the kinds of question which arise in this interdisciplinary scenario, thus resisting Curt's scepticism about the value of philosophy as such. However, such study would be severely limited, to something much more like the study of the history of science in the context of a physics or chemistry degree.

If this isn't satisfactory (and you can gather from the program that I don't think it is) then what we are looking for is a line of argument or a question, which demonstrates that philosophy is still valid and useful as an approach to the question of the place of mind in nature. It would be a question that science alone -- that is to say experimental inquiry, empirical theorizing -- cannot satisfactorily answer, a question for which you specifically need the tools of the philosopher.

What are these tools? You talk of thought experiments (gedankenexperiment), reductio ad absurdum, axiomatic proof. You also mention the terms 'analysis' and 'dialectic'. However, this wouldn't convey much information to someone who was ignorant of philosophy. Scientists use thought experiments (famously Einstein when he discovered the Theory of Relativity). The tools of logic, such as reductio or axiomatic proof, apply in any discipline.

What is it about the mind-body problem, therefore, which is so intractable? The crucial point, it seems to me, is that we don't exactly know what the problem is. Science needs precisely formulated questions and research procedures. But you can't conduct research if you don't even know what question you are putting to nature!

It was Descartes who made an invaluable contribution to philosophy by formulating the question about the nature of the mind in a way which refuses to yield to a scientific approach. We are not even in the business of science until we have a subject matter, a domain of inquiry. That subject matter is the material world and all that is included in it, including the movements of the material bodies we know as 'human beings'. So far so good. But Descartes had (or thought he had) an argument which rules out from the start any possibility that the study of human beings as part of the material world can possibly lead to an understanding of the nature of the mind.

I am talking about Descartes' argument for mind-body dualism. You more or less state this in your essay, but I'm trying to give this a twist so that it becomes clearer why the merely 'scientific' approach is inadequate.

Obviously it is, if Descartes is right. Of course, you can just assume blithely that he is wrong without bothering to engage in his argument. That's more or less what the 'Australian materialists' Smart and Armstrong (both philosophers!) do, in appealing to Occam's Razor. The simplest empirical hypothesis is that mental events are *in fact* identical with physical events. Subsequent debates (in particular Kripke's considerations on the nature of identity in 'Naming and Necessity') have put this view into disrepute, because it simply fails to engage with the logic of Descartes' original argument. The continued attraction of epiphenomenalism, which you cite, is evidence that science does not resolve the question.

Will Curt be convinced? He can always say, 'I'm just not interested, you debate the issue if you want, I don't see the point.' My response to Curt is that it is a fact that we are beset by particular views about the nature of the mind -- about the nature of our own selves -- that we find profoundly 'tempting'. At the beginning of the program, I suggested that maybe Martians aren't tempted in this way. But we are. The only way to deal with this is by engaging the temptation on its own terms, using the tools of analysis and dialectic. Philosophy is ultimately nothing more than getting clear about things that confuse us. But, in order to do this, we need to admit that we are confused.

All the best,


Wittgenstein on solipsism

To: Simon C.
From: Geoffrey Klempner
Subject: Wittgenstein on solipsism
Date: 1st November 2011 11:51

Dear Simon,

Thank you for your email of 23 October, with your first essay for the Philosophy of Language program, entitled 'Wittgenstein's World' on proposition 5.62 of the Tractatus: The world is MY world: this is manifest in the fact that the limits of LANGUAGE (of that language which alone I understand) mean the limits of MY world.' -- You were asked whether you agreed with this statement.

I like the strategy you have adopted for this essay, of first considering the statement 'as it stands', outside the context of the Tractatus and what we know/ believe Wittgenstein was trying to do; and then within the context of Wittgenstein's argument in the Tractatus.

In my comments I will follow your plan. Let's take as read the possibility that an AI could be constructed. The standard story is that our AI is an embodied computer interacting with our world and with us. However, in your version of the story the 'world' in question is a virtual reality (as in Doom or The Matrix), and moreover the subject who is aware of, say, the Matrix as a world is itself a computer generated object in that world (e.g. in the film, the AI program called 'Smith').

I count (at least) two programs here, although you, a software engineer, would be better placed to say whether or when one can draw the line between one program doing two tasks and two separate/ separable programs (crude example: MS Office). The important point is that Smith, say, alone in his 'world', has no knowledge of how he came here or how he world came to be, is ignorant of the fact that he and his world are merely 1s and 0s.

However, let's suppose that Smith is sufficiently intelligent to conceive of this possibility. 'Suppose I am...!'. To make that supposition, he has to reject the idea that 'my world is the only world'. His 'doom' is not solipsism but scepticism. It's a variation of the 'brain in a vat' scenario.

(Incidentally, there's an interesting thread on an answer I wrote for my Tentative Answers blog on solipsism: see )

So we come to Wittgenstein. The first thing we have to put right out of the way is the idea that Wittgenstein's argument for solipsism relies, as you suggest, on the connotations of the term 'experience'.

It is true that Wittgenstein read Schopenhauer in his youth, and so was familiar with Kantian themes (such as Kant's 'Refutation of Idealism' and the idea that we can reconcile transcendental idealism with empirical realism). The world as I experience it necessarily is as it would be if there were objects situated at different spatial locations. In fact, there is no other logically possible way in which experience can be described. There's a strong echo of this at Tractatus 5.64 'The self of solipsism shrinks to a point without extension, and there remains the reality co-ordinated with it.'

However, Wittgenstein seems to offer an argument which is independent of the Kantian analysis of experience, one which relies purely on considerations of what it is for a proposition to have a sense.

You state, 'In the Tractatus Wittgenstein builds a world based on objects that are magically linked together by inexpressible laws of logic to form facts or possibly propositions.' What is 'magical', or necessarily inexplicable, is the fact that the facts are as they are, and not some other way. Any explanation would itself be a fact, which leads to infinite regress.

However, that the world is necessarily 'based on objects', isn't a fact because it is the conclusion of (what claims to be) a logically valid inference. Propositions have a sense. Whether or not a proposition has sense cannot depend on any fact, because we must be able to determine the sense of a proposition -- what it says -- independently of knowing what the facts are (echoes of Russell's theory of descriptions). Therefore, there must be something which all logically possible worlds have in common, something which exists in every possible world: objects and the possibilities of their combination. (There's an echo of this in Kant too, in his argument for the necessity of a spatial framework, and for the necessary existence of 'substance'.)

But whose language are we talking about? That issue hasn't even been addressed. Why must it be MY language?!

If it is necessarily 'my language' then we are home and dry. The situation cannot be as in our previous Smith scenario. Smith ponders the possibility that 'I may be an AI program running inside a virtual reality inside a computer' and those words make sense. Remember, Smith is not a solipsist. He hasn't read the Tractatus.

You offer a possible argument that Wittgenstein might have used. I don't think it gets to the crux of the issue. Why (as in premiss 2) is the language I am using a language which 'only I understand'? Why can't you understand my language? You're doing so now!

I think the crucial step is where Wittgenstein defines meaning in terms of rules (he talks of 'rules of projection' in the context of the picture theory). Whose rules? They can only be my rules, because there is no structure there to explain how I could be following rules which others follow also. There's no way, given the way the argument is constructed, to get outside the perspective of the solitary language user.

True, there is also no way to state the properties of GK which would imply that GK is 'unique', because the only things that can be stated are facts, and the only facts that exist are facts about the world, in which GK is just one living subject or agent amongst others. It can only be shown. The entire argument of the Tractatus shows it, in demonstrating that the conditions which Wittgenstein lays down are jointly necessary and sufficient for a proposition to have a meaning.

As you state, it is in the Investigations that Wittgenstein subjects this claim to a thoroughgoing critique. His conclusion: following a rule is a 'practice' which presupposes the existence of 'forms of life'.

All the best,


Grice on conversational implicature

To: Chris M.
From: Geoffrey Klempner
Subject: Grice on conversational implicature
Date: 25th October 2011 14:16

Dear Christian,

Thank you for your email of 16 October, with your essay written under exam conditions for the University of London BA Philosophy of Language module, in response to the question, 'How does Grice distinguish between the semantic and the pragmatic contributions to what speakers convey by their utterances? How plausible is this distinction?'

This essay gives a lot of information about Grice's theory of conversational implicature, and raises difficulties with the idea (for example, the extent to which we can write the 'rules' for conversational implicature). You also mention views (Searle, or Borg, Cappelan, Lepore) which are hostile to the idea. For example, according to Searle, you say, 'there is no literal 'semantic' meaning', even in such a seemingly straightforward statement as, 'The cat is on the mat.'

What I didn't get from this is any sense of why the distinction between semantic properties and pragmatic considerations matters. What turns on the debate? Why are we interested in semantics at all? What is the philosophical significance of the claim that there is such a thing as 'semantic meaning' of a word, phrase or sentence? Why, for that matter, do we study the philosophy of language?

Grice's own program of defining meaning in terms of speaker's intentions is one thing we could talk about. But the issue is somewhat clearer if one looks at Davidsonian semantics. The idea behind Davidson's program is to account for a speaker's ability to produce and understand a potentially infinite number of meaningful statements in terms of an understanding of the semantic properties of a finite number of semantic particles: proper names, predicates, logical constants, quantifiers, etc. A Davidson-style 'theory of meaning' is an axiomatic system from which one can derive the meanings of indefinitely many sentences.

One comes across the idea of 'theory' in other areas of philosophy. A theory of ethics would tell you what the appropriate course of action would be in any given set of circumstances. It is getting less controversial that there cannot be a theory of ethics, because the situations requiring ethical choice are too complex. There will always be irresolvable ethical dilemmas. Williams argues something along these lines in his 'Ethics and the Limits of Philosophy' (as indeed I do in the Moral Philosophy program).

The situation with regard to theories of meaning does not seem so bad. Arguably, Grice's program is not inconsistent with Davidson (see Mark Platts 'Ways of Meaning'). Frege, Tarski and the idea of truth-conditional semantics seems a good, solid foundation for accounting for the endless variety of linguistic usage -- provided one is able to make the appropriate 'cut', separating off core semantic meaning from pragmatic considerations. (One annoying issue here is context dependence: how does this differ exactly from pragmatics, where does one draw the line? That's something you could have mentioned in your essay.)

Consider a seemingly simple case. There was a time when philosophers of language were happy to defend a truth functional analysis of indicative conditionals. I would still favour such an approach, despite recent work (e.g. by my old tutor Dorothy Edgington) which seems to show a far greater complexity or 'meatiness' in the semantics for indicative conditionals, involving probability theory.

Grice's theory seems to offer a powerful defence of a truth-conditional account of indicative conditionals, which happily copes with the so-called paradoxes of material implication etc. But how good is it, really?

Let's say, I am near a lift and you ask me, 'How do I get the door open?' I reply, nonchalantly, 'If you press the green button, the lift door will open.' You press the green button, and the lift door opens. 'Thank you!' As you disappear into the lift I allow myself a little chuckle. I could see from the panel that the lift had arrived. And I knew that the door opens automatically after 10 seconds. The green button has no effect whatsoever because it's purpose is to call the lift.

Did I lie? I spoke the literal truth. It is not the case that you press the green button and the door does not open. That's what the indicative conditional, 'If... then...' means. 'If P then Q' is equivalent to 'not-(P and not-Q)'. Frege held this. (Interestingly, Frege offers an example where there is no misleading implication of a direct connection between antecedent and consequent, 'If the sun has not gone down then the sky is very cloudy.') What I said is literally 'true' because the door does open. But what you inferred from my statement is that pressing the green button causes the door to open.

How would one adjudicate a debate between someone who holds the truth-functional view of indicative conditionals and someone who holds that the semantic content of conditionals includes an implication about a causal connection between antecedent and consequent? (If there is no direct causal connection, as in Frege's example, then the claim would be that there is an implication that assuming the truth of the antecedent has an effect one's judgement of the probability of the truth of the consequent.)

If you could answer this question -- or pick an issue that you find gripping -- then you would have given the reader a much clearer idea of why anyone would want to distinguish between semantics and pragmatics using Grice's notion of conversational implicature, as well as providing a useful test case for Grice's theory.

All the best,


What is a law of nature?

To: Stuart B.
From: Geoffrey Klempner
Subject: What is a law of nature?
Date: 24th October 2011 13:53

Dear Stuart,

Thank you for your email of 14 October, with your essay for the University of London BA Methodology module, in response to the question, 'What is a law of nature?'

The issues which you grapple with in your essay, which I liked a lot, nicely dovetail with your comments on my response to your essay on Goodman's paradox. Let me lead in with that.

You're right, in a way, that any 'evil demon' universe is 'perverse' by definition, that is to say, it is clear by 'reason and logic' that there's something wrong which (we hope) isn't wrong with the actual universe. The actual universe (we hope) goes the way it goes, without interference or design. The beautiful thing about this picture is that we, as evolutionary products of a universe going the way it goes, naturally develop the ability to discern relevant patterns (epistemology naturalized), an explanation which would be impossible in any scenario where laws change or are interfered with in some artificial way.

I don't think this amounts to an admission that Goodman is wrong, or that his paradox has somehow been solved or refuted. We are relying on the language that we have, with the well entrenched concepts which we use, in order to artificially construct cases which are consequently infected with artifice all the way through. The situation is similar in a way to Quine's strictures on the indeterminacy of meaning/ inscrutability of reference. The examples (like gavagai=rabbit part) seem silly, because we literally haven't got the words to capture what the examples are gesturing towards.

In your essay on laws of nature, you talk about the basic laws of conservation (an idea whose power Leibniz was arguably the first philosopher to appreciate) where there is a logical necessity that we find *something* to conserve, as a condition for the very possibility of doing science. But why are the things that are conserved, the things that are actually conserved? Was it a merely logical error on the part of Descartes to claim that momentum, not energy is conserved (thus allowing mind-body interaction)? His contemporaries clearly thought there was something fishy going on. Yet it seemed to follow from Descartes' assumption that the laws of geometry capture the essence of 'extended substance'. As Leibniz pointed out, 'material substance' as geometrically conceived fails to account for the observed facts of impenetrability and inertia.

The logical extension of the claim that conservation principles lie at the base of the laws of nature is the idea that the universe exhibits certain fundamental 'symmetries'. This is the idea behind the Special and General Theories of Relativity, as I understand it. There is a powerful motive for thinking that there is some kind of necessity in the idea that the universe is maximally symmetrical. But we still haven't made an inch of progress, the Humean would say. If this isn't about describing God's mind (God being assumed to be a perfectionist who would naturally prefer maximal symmetry) then how, or the degree to which the universe exhibits symmetry, or obeys conservation laws, is ultimately just brute fact.

Necessity arises from two related ideas, both of which are very powerful: the idea of unrestricted universal generalization, and the idea of logical deduction. We have these, within a view of scientific laws which remains consistent with Humean strictures on necessity.

I fully agree with you that the self-styled 'necessitarians' merely label the problem. I also see your point that the difference in views can be explained in terms of the distinction between 'low-level analysis' (necessitarians) and 'high-level analysis' (Humeans).

But I'm left with questions. Why can't the laws of nature change? Is it just a brute empirical fact that they don't? Is it also a fact (about the future!) that they won't ever change? Why can't the laws of nature be different in remote parts of the universe? (You can run Goodman's paradox with space as well as time.) No lawgiver. Just a fact. It makes my head spin. How lovely it would be if you could prove the laws of nature from logic alone (the one, maximally symmetric set of laws). But then again, how could you ever prove this, how could you ever dictate to the facts in this way.

Possibly, you could have said a little bit more about why necessitarians merely 'label the problem'. As I said, I agree with this. Just a sentence or two would have been enough, just to cap the point.

All the best,


Determinism and the justification of punishment

To: Paul E.
From: Geoffrey Klempner
Subject: Determinism and the justification of punishment
Date: 20th October 2011 14:11

Dear Paul,

Thank you for your email of 13 October, with your first essay for Possible World Machine, in response to the question, 'In the light of the critique of 'free will', can blame and punishment ever be rationally justified? Consider hard cases, such as brainwashing, crimes of passion, the influence of drugs, medical or psychological conditions etc.'

This is a very good essay. For the purposes of your argument, you assume the truth of determinism, as a metaphysical thesis, illustrating this with Laplace's famous hypothesis concerning a 'Supermind' capable of discovering the state of the universe at any given time, from a total description of its state at any other time.

In your second paragraph, you say that, given determinism, 'retributional justice would be difficult to justify'. I suspect that you were aware here of an alternative view of the nature of justice, according to which the aim of punishment is not retribution but rather simply deterrence and correction. These aims are fully compatible with determinism. If this is indeed all we are seeking to do when we punish, then the only relevant consideration should be whether the individual in question, at the time when they did the action for which they are being punished, would have been deterred by the thought of punishment, or, whether as a result of punishment they are less likely to do the action again if they find themselves in similar circumstances.

We can apply this to the case of hypnosis. It may be true, as you state, that a person can only be hypotized with their consent, although there seems to be a widely held belief that this is not always true. But let's assume that it is. I volunteer to get hypontized by a stage hypnotist in a night club and the next day, inexplicably I murder my next door neighbour. It turns out that the hypontist whispered the suggestion in my ear. Well, obviously, I am going to think twice about allowing myself to be hypontized again! But given the nature of hypnotic suggestion, it is clear that 'the thought of punishment' has no effect whatsoever.

In the third paragraph you quickly move to a 'compatibilist' view, according to which we can talk of a person being 'free' or not 'free', even on the assumption of the truth of determinism. More could have been said here. The point is that there are certain kinds of 'causal chain' which lead to action, which we view as cases of 'personal responsibility' and other kinds of 'causal chain' where a person is not held responsible. The former kind involve the human process of (unforced) deliberation and choice.

So far, so good. But now we face the real challenge, which is that the way a person deliberates and chooses is very much a consequence of their social upbringing and experiences. This doesn't matter, if we adopt a theory of punishment as 'deterrence and/ or correction' as above. If the thought of punishment has a deterrent effect, or if actual punishment has a corrective effect, then it is simply irrelevant that the individual concerned was likely given their upbringing and circumstances, to do the thing that they did. Suitably chastised, they will (we hope) be less likely to do it in future.

But is that an acceptable view of punishment? It is acceptable to say that no punishment is 'deserved', in the sense which most persons understand this, namely as just retribution for what the person did?

F.H. Bradley, a 19th century British follower of Hegel, in his first book 'Ethical Studies' offers a refutation of the idea that punishment is justified as correction and deterrence. The Master of Hounds likes to give his dogs a good thrashing before they go out on the hunt. Why? 'To show them who is boss, and deter them from misbehaviour.' In that case, why not do this with human beings? Why wait for someone to commit a crime? The answer is, obviously, that it would be 'unjust'. But by what standard of 'justice', if we reject the idea that punishment is 'deserved'?

You make the point that not everyone in given circumstances turns to crime. That is true, if we consider things from a sufficiently lofty perspective, where individual differences are not considered. Let's say that only 5 percent of persons from deprived backgrounds turn to crime. (I'm picking this figure off the top of my head.) The other 95 per cent were able to develop into good, honest citizens despite their social handicap. However, the closer we look at the 5 per cent, the more reasons we discover why these particular individuals made the choices that they made. That's the point of determinism. There is never an opportunity to make an alternative choice, because every choice is the consequence of external circumstances and the agent's current psychological state.

All the best,


Causation and the explanation of human action

To: Mark S.
From: Geoffrey Klempner
Subject: Causation and the explanation of human action
Date: 19th October 2011 13:50

Dear Mark,

Thank you for your email of 12 October, with your essay for the University of London BA Philosophy of Mind module, in response to the question, 'Are actions best explained as caused by agents, mental states, or something else?'

I really liked this. You have gone into an area of analytic philosophy which is dense with argument -- you know my views about the scholastic tendency -- but you have succeeded in giving a clear analysis of the problem of explaining action. I agree generally with your conclusion, although I also think there is a notion you could have appealed to which would have made your conclusion more compelling, that 'If behaviour forms a continuum, stretching from action to non-action, and retrospective analysis is often or always required to position behaviour along this continuum, then causal deviance may have to be accepted as a concomitant to even the best explanation of action.' -- But more of that in a minute.

The reason my response is more positive this time is that all the way through you motivate the claims, and the objections to those claims, so that the reader does not get the feeling that they are just following a report on the debate, as it were at one remove. I was challenged each step of the way. I was gripped.

A criticism could be made of the structure of your answer, although you are following a well-beaten path using a plan which has worked well for you -- I mean, considering all the alternatives, not just the most popular amongst contemporary philosophers. So 'awareness of initiation' and 'volition' seem to get equal treatment as serious contenders, even though they are not really. In an examination, you need to flag the proposed alternatives in some way so that it is clear where you are following the current state of play of the debate, and where you are just looking at historical antecedents of the problem (no less important).

In the literature, there are many, many proposed refinements to the causal story, and obviously there is a limit on how deeply you are able to delve into the detailed arguments. However, there is one proposal, originally by Christopher Peacocke, which although certainly not universally agreed as a way forward seems particularly apposite in relation to your conclusion: I am talking about the idea of 'differential explanation' ('Holistic Explanation' OUP 1979).

Consider your example of Jane. I agree that we should be focusing on 'tryings' as primarily what action explanation is seeking to explain. Jane want's to attract the attention of the chairperson, and believes that a cough will do this. Then she coughs, unintentionally. What exactly happened here? Jane's intended cough was not just any cough. It was a cough-in-order-to-gain-attention. A polite 'aahm', not a full-blooded hacking, sputum loaded cough, nor a barely audible suppressed 'humm'.

If the disputants had been talking more loudly, Jane would have turned up the volume of her 'aahm' to the appropriate level. Similarly, if they had been talking more softly she would have turned the volume down. Whereas her involuntary cough is not sensitive in the same way. Of course, we can invent reasons why the loudness of the conversation would cause Jane to be more nervous and therefore cough more loudly, or its quietness make her less nervous and so that her involuntary cough is quieter, but the point about differential explanation still holds -- there is all the difference in the world between a consciously judged trying, where gradations are fine-grained, and the relatively accidental differences in involuntary cough volume.

The precision of our tryings and the control we exercise over them is appropriate to the circumstances. Like the difference between turning on a light switch, and shooting an arrow in an archery competition. Differential explanation adapts to suit.

As Davidson argued in his original paper, actions are things we rationalize. Often, as you observe, rationalization comes after the event. There are many things we do where we don't stop to think or reflect on why we did what we did precisely that way. Our actions/ tryings are guided much of the time by practice and habit. It is only when the circumstances are challenging in some way that we need to judge our trying precisely. (I'm not saying that one can't execute 'precisely judged' actions from habit and practice -- e.g. the skilled footballer evading a tackle.)

The final finesse is to recognize, as you state, that there is no sharp line between the 'intentional' and the 'unintentional'. Post-rationalization creates many of the lines which were not there originally, in the spirit of creating an account of ones progress through the world that makes sense. What counts as an 'event' or a 'cause' are similarly constructed.

I don't agree with the criticism that Davidson's account fails to distinguish between human and non-human 'agents'. I don't accept that non-human animals have 'desires' and 'beliefs' although their behaviour is certainly goal-directed. It is through language, through our sensitivity to criticism and capacity for reflection, that the lines get drawn in the first place -- but now I'm veering off topic!

All the best,


Tuesday, February 11, 2014

Heraclitus, Aristotle and the problem of change

To: Richard K.
From: Geoffrey Klempner
Subject: Heraclitus, Aristotle and the problem of change
Date: 18th October 2011 13:50

Dear Richard,

Thank you for your email of 9 October, with your second essay for the Ancient Philosophy program, entitled, 'Heraclitus, Aristotle and the Problem of Change'.

This is an excellent piece of work with which I have few disagreements. As you may have gathered from the program, I am one of those who espouse the 'strong version' of the theory of Flux as the interpretation which makes the best sense of what Heraclitus wrote, in the context of the ongoing dialectic.

Kirk, Raven and Schofield disagree, holding that Heraclitus merely meant that everything, even a rock, is in a continual, if sometimes slow and imperceptible process of change. That's the point about the example of the farmer. It's news, not merely a truism, that even a rock changes. But it doesn't seem enough to explain why Heraclitus got so agitated about his theory, nor does it explain the subsequent dialectic through Plato to Aristotle.

Can the strong theory be defended? You seem to hedge on this, agreeing with Plato in the Cratylus that if everything really is like a river, then we can hardly speak of 'things' at all. But here is a finesse which I think you've missed. We can still talk of rivers because irrespective of their ontological constitution they are relatively constant features of our experience. And the same can be said of a tree or a rock, if these too 'flow' like rivers, lacking any enduring substance of their own.

What Plato (or, rather Plato's Cratylus) should have said is that it's perfectly OK to discourse about rivers, trees and rocks, so long as we remain at the level of appearances or phenomena. But when you try to describe how things are at the *ultimate* level, in reality, then there is nothing to say, nothing to describe -- literally, no *things*.

But we're forgetting the Logos. Why not say that ultimate reality is the Logos, and nothing but the Logos? Isn't that what Heraclitus wanted to say?

You ask, 'how does the Logos interact with all the things that change, and where does this Logos reside with respect to the things which undergo change?' Why is that different from asking, 'How does the Law of Gravity interact with masses whose motion is affected by gravitational force, and where does this gravity reside with respect to the masses whose motion is affected by gravitational force?'

One possible answer would be 'gravitons'. However, it is an empirical claim, at best, that there exist such entities, not a logical or metaphysical presupposition of making the existence of a law of gravity intelligible.

But I get your point, that an analogous problem (how Forms interact or affect appearances) motivated Aristotle to reject Plato's two-world metaphysic, claiming that things are a compound of matter and form, that form is in things themselves not in a second, non-physical world.

Which brings us to Aristotle. You've put your finger on the key problem for Aristotle's theory of matter and form, which concerns 'unqualified' change. By chance, I was marking an essay by one of my University of London students last week on the question whether Aristotle can account adequately for 'substantial change'. The vulture eats the carcass of a goat, and the dead flesh of the goat is transformed, in a manner which we are unable to observe or indeed comprehend, into the living flesh of the vulture. What is going on here?

Aristotle is led to posit that there is 'something' constant through the change, but this something is not identifiable as any kind of substance or quality. It is sheer formless matter.

This begs a huge question why Aristotle was so hostile to the Atomist theory of change (according to which all changes are ultimately reducible to locomotion). This would take an essay in itself. However, I have an interesting vignette regarding this.

In my first or possibly second year as a graduate student at Oxford University (somewhere between 1976-8), I attended a seminar on the philosophy of quantum mechanics for advanced students, given by Prof Michael Redhead. The level of the discussion was way out of my league. However, there came a point where Redhead was considering the problem of change at the subatomic level, when there doesn't seem to be any ready candidates for 'things' undergoing the changes. 'Well, of course, you could always take an Aristotelian view -- if that made any sense at all,' Redhead says causally, to titters from the knowledgeable audience of mathematicians and physicists. I just gripped my seat and said nothing.

Well? Maybe you could? Could it be that Aristotle was right all along?

All the best,


Solving Goodman's new riddle of induction

To: Stuart B.
From: Geoffrey Klempner
Subject: Solving Goodman's new riddle of induction
Date: 14th October 2011 15:35

Dear Stuart,

Thank you for your email of 10 October, with your University of London BA essay for the Methodology module in response to the question, 'What is Goodman's New Riddle of Induction? How if at all, can it be solved?'

I have no criticisms of this as a response to the question. You stick to the point, express yourself concisely, and offer what seems like a valid response. I take it that you are saying there never was a 'new riddle' of induction. It was just a dressed-up version of the old. I agree, but the question remains why Goodman goes to all this effort, and whether or not the effort is wasted.

I agree with you that the example of 'emeralds are green' is badly chosen. However, I think that the problem with this goes deeper than merely the problem that emeralds are green by definition. The problem is (as I think you also indicate) with the choice of a colour predicate.

Let's consider instead something totally off the wall, but which fully satisfies the logical elements of Goodman's riddle.

Call a 'experon' an academic philosopher examined before t, who is an expert about philosophy, logic and reasoning, or an academic philosopher examined after t who is a complete moron, unable to perform the simplest logical inference and ignorant of the problems of philosophy.

It would not be beyond the powers of a suitably endowed evil demon, to so arrange the world that, as a speaker of English would (from the point of view of a speaker of Experonish) 'perversely' describe it, academic philosophers change from being 'experts' before t to being 'morons' after t. While from the Experonish speaker's point of view, all academic philosophers naturally are, and remain, 'experons'.

Non-academic philosophers who are sufficiently knowledgeable to tell the difference between an expert and a moron would generally agree that academic philosophers are experts. However, it would not be beyond the powers of a suitably endowed evil demon to give non-academic philosophers the ability to tell the difference between an 'experon' and a 'morpert', an ability which they would deploy without even being aware of any change occurring at t.

Part of your point is that if you change one thing, lots of other things have to change too. But that's assumed in the example. You tell that someone is an expert by asking them questions. The brains of experts are different from the brains of morons. There's masses of data and theory that go towards accounting for the effects of study and education on human intellectual abilities. And so on. Add whatever you like to the mix. The bottom line, for Goodman, is that there is no way to prove by means of reason and logic alone that the predicates which we project are 'normal' rather than 'perverse'.

I think the answer to this is, why on earth would you want to 'prove it by means of reason and logic alone'? Goodman would consider that his point has been taken. It is rational to reason by induction, it is rational to prefer the predicate 'expert' to the predicate 'experon', because what we do, our practice, defines the standards of rationality. Every attempt to justify that practice, or find some deeper ground on which it supposedly rests, ultimately leads to circularity.

I would ultimately agree that the 'new riddle of induction' is just a dressed-up version of the 'old riddle of induction'. But the hocus pocus succeeds in making a point. Hume, and many philosophers since Hume, have focused their inquiry too narrowly, though I believe Hume would have been perfectly capable of seeing this had the question been put to him.

All the best,


Aristotle's account of substantial change

To: Scott B.
From: Geoffrey Klempner
Subject: Aristotle's account of substantial change
Date: 14th October 2011 14:00

Dear Scott,

Thank you for your email of 3 October with your essay for the University of London BA module on Aristotle, in response to the question, 'Discuss Aristotle's explanation of change. Can it account for substantial change?'

This is for the most part a clear and thorough account of Aristotle's theory of change in terms of his notions of matter, form, potentiality and actuality.

Responding to the second part of the question, you include a discussion of Aristotle's account of coming into existence or going out of existence (substantial change), where in certain cases -- living creatures -- Aristotle is led to posit the idea of 'formless matter', a matter which we know on principle must remain the same, yet which cannot be identified in the process of change however closely we observe. You cite the examples of a vulture eating the carcass of a dead goat, and the coming to be of a human embryo from a sperm and an egg.

What is remarkable about this is that we have been led to this conclusion by logical steps starting from the axiom that in every change there is a component which remains the same throughout the change and another component which alters. You get credit for explaining this very clearly in your exposition.

I have a problem, as most contemporary philosophers would have, with the idea of 'formless matter'. It looks like a desperate resort to save a theory. Logically implied it may be, but surely Aristotle must have considered the possibility that there might be an alternative?

He did. And this is the really puzzling part. Aristotle was fully familiar with the theories of the atomists Leucippus and Democritus on the subject of change. When the vulture eats the goat, or when the sperm and egg combine, something happens on the microstructural level which we cannot observe because it is beyond the acuity of our sense organs.

On the atomist theory of change, all change is locomotion, an alteration in the arrangement of unchanging atoms. What appears as different qualities, such as water and ice, or the door painted blue and then red, or the quality of goat meat and the flesh of a living vulture, are merely effects on our sense organs of complex structures of atoms. This is, more or less, the story that science accepts today. Yet Aristotle rejected it outright. Why?

Interestingly, in current particle physics, there seems to be a point where we are unable to account for the changes observed in bubble chambers etc. except in Aristotelian terms. A particle of a certain kind displays a characteristic pattern as the result of a collision with a particle of another kind because that's just the kind of particle it is. That's all we can say. We are at the very bottom level, there is (or there appears) nowhere to go in the search for even deeper, more microscopic 'structures'.

It would be tendentious to attribute to Aristotle this degree of prescience, yet the stand which he took against atomism was a principled stand. I attended a seminar a few years ago when a philosopher (whose name I can't remember, unfortunately) gave a paper on Aristotle's text 'On Generation and Corruption' where he sought to explain Aristotle's position.

I found the explanation which he gave compelling. The fundamental premise of Aristotle's approach to these questions is that human beings have the power, through observation and reason, to account for the world around them. Remember that the very notion of a microscope or even a magnifying glass would have been unheard of. The speculations of atomists were unverifiable, and based on dubious metaphysical premisses. If the atomists were right, then as Democritus observed, human knowledge or the idea of science became something very doubtful indeed.

The alternative, Aristotelian explanation of change looks, from a modern perspective, almost tautological, like saying that 'opium makes us sleep because of its dormative property'. But he isn't looking for 'explanation' in the modern sense, that is to say, hypothetico-deductive explanation. He is seeking to resolve a logical problem, a logical challenge to the very idea of change. As you demonstrate in your essay, he succeeds in that task.

I do think the examiner expects you to say something more about the problem of substantial change. You give Aristotle's theory, but what you don't do (and I think what the examiner was looking for) is offer some sense of why this is such a challenge. What is the special problem here? Is it just that we are forced to posit 'formless matter'? Why is that so bad? Perhaps because this goes against Aristotle's own principles, that the account should remain at the level of what human beings can discover through unaided sense perception. This is the point, it could be argued, where Aristotle comes closest to doing the very thing he finds so objectionable in the atomist theory.

All the best,


Scepticism, consciousness and moral theory

To: Kyriakos C.
From: Geoffrey Klempner
Subject: Scepticism, consciousness and moral theory
Date: 5th October 2011 14:48

Dear Kyriakos,

Thank you for your email of 2 October, with your second essay for the Possible World Machine, in response to the question, 'Explore some of the issues surrounding the attribution of consciousness to machines and non-human animals,' and your email of 3 October, with your response to unit 7, on Morality.

Scepticism and consciousness

One big issue which you raise concerns the moral consequences of attributing consciousness. Here, one might distinguish between the question whether non-human animals are capable of consciously suffering, and the ethical consequences of this, and the question whether there might be a sense in which non-human animals might be admitted as members of the moral community.

The latter could, conceivably, happen if experiments with apes made sufficient process. Just as a child can be blamed for bad behaviour, so the experimenter might remonstrate with an ape who has successfully been taught a simplified version of English. 'You should not have stolen the banana!' (This would be different from 'punishing' an animal, say a dog, where our only concern is stimulus-response.) Maybe.

It might seem incredible that anyone could claim that animals don't suffer, yet Peter Carruthers (who was Head of the Philosophy Dept at Sheffield when I taught some classes there) in his book 'The Animals Issue' argues that animals are not 'conscious' of their pleasure or pain in the way that human beings are. So, if you are a utilitarian (like Peter Singer) it would be a fallacy to weigh up a human pain and the pain suffered by an ape because the two pains are incommensurable.

Is HAL conscious? An argument given in the program is that a computer program capable of consciousness would necessarily have beliefs and desires. Having desires, it must, necessarily, be capable of performing actions which satisfy those desires. So it must have real physical needs, the capacity for pleasure or pain, which it is able to provide for by the appropriate physical behaviour. But, as you say, even so there is no implication that HAL's 'consciousness' would be similar to human consciousness. Could we even communicate, or share the same language?

In Blade Runner, the android Rachel is presented as someone who genuinely believes she is a human being. She has fond memories of her childhood, precious photographs and mementoes. But then, in response to questioning, she begins to suspect the horrifying truth. What we have here is a technology fully capable of reproducing the biological structure of any animal (such as a snake) or human being.

To make a real human being you need something else: a life, a childhood from early infancy, through all the stages of learning and socialization. What the constructors of Rachel have done is merely provide her with a patchwork substitute, a simulacrum of 'memory', made up from genuine memories of other human subjects and pasted together in a form which resembles a coherent story. But this resemblance crumbles under close examination. This is the purpose of the specially designed psychometric tests conducted by the blade runner. Rachel displays intelligence, but she is not a genuine example of a 'conscious being', because of what she lacks.

You say some things about morality in your essay which would imply that we are not necessarily bound to behave ethically towards extra-terrestrial intelligent aliens. This would be speciesism. While a preference for our own species can be defended (in the same way as a preference for one's nation, or family, or spouse), I don't think there could be a moral case for placing an alien species (or by parity of reasoning, a genuine artificially constructed intelligence) altogether outside our moral community.

Unit 7

I was interested in your distinction between the 'efficiency' of a moral system and its 'consistency'. An efficient system would be one which always gives an intelligible, actionable answer to any moral question. A consistent system would be one which did not lead to dilemmas and conflicts. Kant's duty ethics is efficient, but it leads to conflicts of duties. Utilitarianism avoids conflict, because there is always an answer in principle to the question of 'the greatest happiness for the greatest number' but it is inefficient, because of the difficulty of calculating that answer.

My own view is that there cannot be a moral theory, period. As moral beings, each of us is equipped with all that is necessary to make moral decisions in the real world, but in the real world there also many situations or potential situations where we find ourselves faced with insoluble dilemmas.

The trolley problem has been the focus of much debate in moral philosophy. A trolley with 20 passengers is on a collision course which will lead to their certain death. The only way to divert the trolley involves causing the death of an innocent bystander. Would you kill 1 in order to save the lives of 20? Kantian ethics says, no, while utilitarianism says, yes. And so on. The problem is you can alter the numbers in any way you like. At some point the Kantian will have to say 'I don't know', and likewise (at a different point) the utilitarian.

'Do moral values determine beliefs or do beliefs determine moral values?' Beliefs about what, exactly? I just marked the essay of one of my University of London students who was asking a question (for the 'Ethics: Contemporary Perspectives' module) 'Are there moral facts?' The hallmarks of a 'moral fact' would have to be that it is (a) capable of being ascertained by the normal method that we ascertain facts, such as sense perception, experiment etc. but (b) such that knowing this fact necessarily entails a particular action.

According to David Hume, in his discussion of the gap between 'is' and 'ought' and also G.E. Moore, in his account of what he terms the 'naturalistic fallacy', such facts would be highly problematic. It is up to us to decide what we value, given the facts. The facts can't make that decision for us. But where does the decision come from? What justifies it?

I would argue that we should not look for metaphysical 'objects' to serve as moral facts. Rather, there are rational constraints on human behaviour which derive from the very fact that we are 'persons' in relation to other 'persons'. To deny the claims of ethics is equivalent to denying that other 'persons' are real, in other words, it is equivalent to solipsism. --The question of solipsism will be discussed later in the program.

All the best,


Saturday, February 8, 2014

Are there moral facts? (revisited)

To: Stuart B.
From: Geoffrey Klempner
Subject: Are there moral facts? (revisited)
Date: 5th October 2011 13:55

Dear Stuart,

Thank you for your email of 28 September, with the revised version of your essay for the University of London BA Ethics Contemporary Perspectives module, in response to the question, 'Are there moral facts?'

This is an acceptable answer to the question. You cover a number of points which an examiner would be looking for. But it isn't the best answer that could be given, and I think you feel this too.

I take it as axiomatic that there is a 'model' answer to any question that you are likely to see in an examination. So it is safe to assume that if you are tempted to react, 'I can't see how...' then you are missing something. There is a range of possible answers, all of them excellent, it's just a matter of finding one.

The point about the question, 'Are there moral facts?' is that this can be seen as a way in to moral theory. Whereas your approach is, 'If you hold theory T then you will answer yes to the question, and if you hold theory T' then you will answer no.' That's true, but it doesn't get to the heart of the matter.

It's good to start by asking, 'What is a fact?' Facts are truth-makers for true propositions. A point you could have made here, however, is that the predicate 'true' can be applied to any sequence of words which grammatically passes the test for a sentence. What that means is that we can't rely simply on the fact that people are prepared to use the word 'true' in determining whether they believe in the existence of moral facts or not.

That naturally leads to the question, What are the characteristic marks of truth in the substantial sense, truth without scare quotes, rather than mere 'truth'? This is something that David Wiggins has looked at. Truth in the substantial sense implies things like convergency, permanence, underlying explanation, truth 'makers'.

So let's assume we are talking about substantial truth. One line of thought which is fairly persuasive (at least, at one time it persuaded me) is that there are many concepts in our language, not just moral concepts, which justify action of some kind because that's just what the concept is. To accept the use of the concept is to accept the inference from the criteria for its application to the consequences of applying it. A crude example given by Michael Dummett in his first 'Frege' book is the term 'Boche' used of Germans during the First World War.

This has given rise to a big debate over 'thick' and 'thin' moral concepts. (See, e.g., Simon Blackburn.) The fact that the criteria for a thick concept are satisfied 'leads to' appropriate action, so (seemingly) bridging the is-ought divide. The only problem is, you can always reject the concept!

Note, that we haven't yet talked about any particular moral theory. We are in the area of metaphysics or semantics. What the question is asking you to do, in my view, is describe the ground rules for debate over moral theories, prior to consideration of this or that particular theory. That's what G.E. Moore does with his naturalistic fallacy. Moore has a theory (a pretty unpersuasive one) but that's merely his response to the challenge which he himself has posed. What other responses can there be? What are the relevant considerations on the adequacy of any possible response to the naturalistic fallacy argument, or to Hume's 'is-ought' argument? (Again, Hume has his own response, but it is not entailed by the 'is-ought' challenge.)

What kind of thing would a 'moral fact' have to be? You discuss Mackie's argument from queerness, but a criticism to make here (which I've mentioned before) is that Mackie simply assumes that the hunt is on for some special kind of 'object', whereas if we consider George Kreisel's dictum about mathematics ('the point is not the existence of mathematical objects but the objectivity of mathematical truth') and apply it to ethics, then we should also be looking for constraints on what is 'rational' or 'irrational', as Kant effectively does with his categorical imperative.

For Kant, doing the right thing for the right reason is a matter of seeing its 'rationality', in a not totally dissimilar way to the way one accepts the validity of a logical inference. Of course, this claim is contentious, but it opens the way for moral facts which are not objects in any sense but rather rules like the rules of logic. This is something I discussed in the Moral Philosophy program.

Moral facts could be natural facts. They could be transcendental facts (i.e. conclusions of a Kantian transcendental argument). They could be supernatural facts. Each of these alternatives leads to questions and problems. Once again, we haven't even looked at any particular moral theory, we are merely describing the ground rules for debate.

As you can see, I am not trying to give you a 'model answer' but rather throwing out ideas of possible ways of proceeding which would get deeper into the heart of the question than simply considering the consequences of this or that theory. I hope this helps.

All the best,


Intentionality: Frege and twin-earth cases

To: Mark S.
From: Geoffrey Klempner
Subject: Intentionality: Frege and twin-earth cases
Date: 4th October 2011 14:07

Dear Mark,

Thank you for your email of 28 September, with your essay for the University of London BA Philosophy of Mind module, in response to the question, 'What do Frege cases and/or Twin-Earth cases show about intentionality?'

What you've written is two short essays, on the Frege cases and the Twin-Earth case, with a short introduction and conclusion indicating the relevance, as you see it, of the two mini-essays to the question of the nature of intentionality. That is an answer to the question, but I don't think that it is the best answer, or the best strategy for coming up with an answer.

The question which has become known as the debate between 'internalism' and 'externalism' or between 'narrow content' and 'broad content' is a deep problem. But it is easy to forget that there was a time, not so very long ago, when the very idea of broad content or externalism would have been laughed to scorn. The credit goes to Wittgenstein for first raising the question, in Philosophical Investigations:

663. If I say 'I meant HIM' very likely a picture comes to my mind, perhaps of how I looked at him, etc.' but the picture is only like an illustration to a story. From it alone it would mostly be impossible to conclude anything at all; only when one knows the story does one know the significance of the picture.

Page 177. What makes this utterance into an utterance about HIM? -- Nothing in it or simultaneous with it ('behind it'). If you want to know whom he meant, ask him.

(These might not be the best quotes, as I had to rely partly on memory and partly on Google.)

It was Wittgenstein who first argued, 'meanings are not in the head' (cf. Putnam), that you can't determine whom my thought is about by examining my thought, by itself and out of its external context (form of life etc.). It's the context which determines content. But how can that be? 'If you want to know whom he meant, ask him.' 'OK, whom did you mean?' 'I meant Mark.' -- How does that tell us anything, if there are two or more candidate 'Marks'??

I struggled with these questions, as a first-year graduate student working under McDowell. Many of the articles which you reference hadn't been written (back in 1976!).

There is, in your essay, an indication that you see the point where we feel pulled both ways, where you talk about 'psychological' and 'social' content. I am going to elaborate on this, because I think that this is what the question was asking for. How is it, that we can have our intuitions about intentional content pulled in these different and opposite directions?

Meanings are not in the head. Philosophers defending narrow content these days don't make the error of thinking that you can equate a thought with a mental image, or thinking that mental content is some kind of 'private object'. Instead, one considers the question of what it takes strictly to explain the subject's actual behaviour. If during your sleep you were kidnapped by Martians and placed in a perfectly lifelike simulacrum of Sydney, peopled by androids designed to mimic all the people that you know, or half know, your behaviour, as determined by objectively observable bodily movements could be predicted on the desires/ beliefs model, with exactly the same results, matching margins of error etc.

The principle is the same, whether we are dealing with a person or a jet engine. If you subject the entity in question to the same inputs, the you will get the same observable outputs.

The next question, then, is how the idea of broad content or externalism could even get a look in? That is equivalent to, what MORE are we trying to do, when we account for a person's actions, than predict outputs on the basis of a given input?

I have expressed the question in a form which virtually forces you to come up with, not just one answer, but a variety of answers. The person whose actions we are seeking to explain isn't a laboratory specimen. We are there, we are in with them, in the same world, dealing with the same or similar situations etc. If you want to know whom he was thinking about ask him, and the answer WILL mean something to you. It won't necessarily mean anything to just anyone, but it will mean something to YOU because...

I find the very idea of 'explanation' as applied to human beings mind-blowing, because we are 'in there', there's no separation between observer and observed. (And of course, we are also explaining ourselves to others.) What is explanation? what is it's point? how does it relate to justification, criticism, praise, rebuke, knowledge, authority? These things are essential, not accidental, because they provide the point.

Getting back to the question: I think a better strategy would have been to deal with the various issues around Frege and Twin-Earth from a more lofty perspective. What we are trying to do here is find some leverage into the debate about content, something that will offer clues as to how one should go forward, rather than picky disputes about the details of the various 'intuition pumps'.

All the best,