Hope and Mental Imaging

Can Dogs Hope? part 5

In his article 'The Value of Hope'(1), Luc Bovens attempts to clarify what the necessary and sufficient conditions for us to be able to say that a person is hoping for something.(2) This is of great use in terms of trying to answer whether dogs can hope - if the account is satisfactory we will have a tick list of abilities dogs and other animals must possess in order to hope.

Bovens gives the following three conditions as, together, being necessary and sufficient for someone to hope for something:

1. One must not be certain that what one hopes for will or will not come about.
2. One must desire what one hopes for.
3. One must engage in devote some amount of mental energy to what it would be like if what we hoped for were to happen (Bovens terms this 'mental imaging')
   

Number 2 seems pretty uncontroversial. Number 1 is a little more difficult, but I've already dedicated some discussion to the relationship between hope and certainty, and am personally satisfied with Bovens' justifications for this point (for now at least).

Number 3 may cause us some difficulties. Bovens gives the following example of such expenditure of mental energy or 'mental imaging': "I had been looking at my clock wondering whether Sophie would still come, I had been turning my head earlier to check whether Sophie was amongst some newly arrived guests" But the turn of phrase, 'mental imaging', suggests very much that what's important here is not the behaviour but the mental processes involved.

Would it be possible to remove this reference to 'mental imaging' and instead speak of characteristic 'hope behaviour'? Could we make sense of hope without reference to an inner life?

Why do I find the reference here to 'inner life' as such a little problematic? Personally I have a little trouble with any concept we can't bring back to the way people act and the things people say. I've some sympathy with Wittgenstein's dictim (or at least, my interpretation of Wittgenstein's dictim) 'The human body is the best picture of the human soul' - the body, in its widest sense, should be our reference for a person's emotions and dispositions, and by this we are not denying the possibility of 'inner life' - it's just not so important. (3)

But that's not the main problem in this context - it's a question of whether we could test if a dog is engaged in mental imaging in this way. The main way we'd verify this with people is by asking them. People can be mistaken, but if we cannot bring a person to admit they were hoping, it seems very difficult for us to claim that they were (assuming that they are not trying to deceive us, and are not deceiving themselves)

Perhaps we could build up a sufficient body of behaviourial evidence to establish someone is hoping for something. In fact, I feel we probably could. But in the case of a dog, would we be actually be establishing this, or would we be engaged in anthropomorphism?

My next post in this series will attempt to tackle this head on - once I've done a little more reading...

(1) in Philosophy and Phenomenological Research, Vol. 59, No. 3, (Sep., 1999), pp. 667-681.
(2) ibid. pp.673-4
(3) Though possibly it might be the case that the words 'inner life' actually refer to a certain position in a complex web of human language and behaviour. I won't pursue this until I've had chance to consider what this might actually mean!

Ethics of Selgiman's experiment

A slight deviation from my posting schedule - I discovered a recent article by Martin Seligman on the ethics of his experiments on dogs. I've long been inclined to question the reasoning by which justified the experiments as described in Learned Optimism, though was hesitant to do so as I am aware that his position may have since changed (indeed, in Helplessness: On Depression, Development and Death he does not express any qualms about his experiments - this of course does not mean that they were not troubling him).

I was very interested to learn that his ethicist friend he'd mentioned earlier was in fact Robert Nozick. But I don't want to get caught up on this for now.

Seligman seeks to justify the suffering he causes to the dogs in his experiment through what we might want to call "awesome utility" (that is as in the ethical principle of utility aka the principle of the greatest happiness for the greatness number) Here the suffering caused to the dogs through the experiments is far far less than the suffering prevented by utilising the results of these experiments. Nozick gives the comparison of experiments on animals that were used to discover the small pox vaccine.

Is an appeal to "awesome utility" enough? I have my doubts - can we justify inflicting any suffering if it prevents many times more? In his Critique of Utilitarianism, Bernard Williams gives a number of compelling examples of where this just doesn't seem clear. The one that sticks in my mind is when someone threatens to kill a group of indians unless you take a gun and shoot one. I certainly don't think one could ever be obliged to shoot the one indian. I don't think the value of life, or of happiness and suffering, is quantative in this way.

I'm not going to attempt to give a breakdown of the flaws in utilitarianism for now, and I realise what I'm saying doesn't really constitute an argument (hopefully I can develop this later). But my strong feeling is that in the situation Seligman described, there was no right answer. Inflicting the pain on the dogs was wrong - but failing to take action which would prevent suffering for many more would have also been wrong. Where I take issue with Selgiman is, I think, is when he sounds an upbeat note on an issue that is essentially irresolvable.

(That is, if there was really no alternative experiments. I always wonder if Seligman could have replaced the stick with a carrot in his experiments - teach some dogs that there's nothing they can do to get a reward, and others that they can if they follow the right course of action. Wouldn't this still be learned helplessness?)

Now, there's a lot that's of interest here - and in the attitudes expressed in the responses to his articles (including a potential challenge the validity of my Can Dogs Hope? little project which I'll have to address). I've not a hope of covering it all here - so I'll probably intersperse Can Dogs Hope with a more detailed consideration of these issues.

But next - onto Luc Bovens.

Optimism versus Hope - sketches towards a resolution

Can Dogs Hope? part 4

Separating optimism from hope is a tricky operation, one to be undertaken delicately. As such, I'm going to begin tentatively. The terms are quite clearly not equivalent - what we're interested in is whether being capable of the former means you are automatically capable of the latter.

Let's start simple: hope can be used as a verb, optimism can not. Perhaps the correct phrase should be optimism as against hopefulness, though this may become clunky. Quite what we do when we hope is another question entirely - one I hope to come to soon, just not in this post.

Now, a second difference. One can hope for something without believing it very likely to happen, or - and this is the important one - while believing it less likely to happen than a realistic appraisal of the situation would indicate. This is a measure by which we consider a disposition optimistic - it's got to be than one is more positive about one's chances that one is rationally(1) entitled to be. One can hope, one can be hopeful, and in fact still be pessimistic - because what we're hoping for is in fact a certainty, and its only our pessimism which stops us from seeing this.

And it is possible to be optimistic in such a way that precludes hoping - we can be so optimistic that we feel absolutely certain that we'll get what we desire.(2)

But can we despair, can we be hopeless, whilst still being optimistic? Well, it might be possible to imagine a situation where we have lost all hope for things improving, but in fact we are blocking out evidence that suggests things are going to get one hell of a lot worse. Is it right to call this optimism? I suppose the phrase 'things can't get any worse' sums up the attitude, and I'm unsure quite what to call this! No matter!

Before we go further, I think a consideration of Luc Boven's attempt to clarify what we mean by hope is in order, which I'll do in the next post. I've not quite got to the nub of the issue, but with some luck we'll get there soon.


(1) Depending on what we mean by rational of course - as Seligman shows in Learned Optimism, being optimistic improves your life chances immeasurably, meaning that there's a sense in which being optimistic is always rational.
(2) See my post of A possible paradox of hope for a justification for why certainty precludes hoping.

Seligman and optimism/pessimism in dogs

Can Dogs Hope? part 3

Before we go any further, we need to examine an experiment conducted Martin Seligman, to see what it establishes regarding canine capabilities, and what implications this has for our investigation.

The experiment(1) was designed to establish whether dogs could learn to be helpless. There were three groups of dogs. Each dog in the first group was given a series of electric shocks in a hammock, which they were unable to control in any way. Seligman calls this 'the yoked group.' The second group, the 'escape group' is given the same electric shocks, but is able to stop them by pressing a panel with their nose. The third group receives no shocks - this is the control group.

All dogs were then given a trial in a shuttle box (a cage divided in two with a jumpable barrier.) The floor was electrified, and the only way they could escape was by jumping the barrier. The difference in response between the yoked group and the escape group was striking. While the both the escape group and the control group quickly got the hang of leaping the barrier to avoid the shocks, two-thirds of the yoked group failed, and instead started to simply lie down and whine in pain.

The implication is clear - dogs can learn to be helpless, they can learn to become pessimistic about the effectiveness of their own actions, and generalise this pessimism to other scenarios. This can work the other way as well - dogs who have become pessimism can learn optimism by being placed in a scenario where they are able to control the outcome of the situation.

In Learned Optimism, Seligman convincingly presents this finding as a refutation of the stimulus-response model of behaviour proposed by behaviourism - under behaviourism, dogs should not be able to generalise between dissimilar experiences.

This presents two challenges to the proposition that dogs are unable to hope, or unable to be hopeful. The first is simply in giving an extended notion of a dog's cognitive capabilities which allows the possibility of a dog having such a disposition as hope. However, this is just the outline of an objection, and needs to be developed. The second is that dogs seem capable of dispositions - optimism and pessimism - which seem, at least on the surface, similiar to hope and despair, or hopefulness and hopelessness. It's the second of these that I plan to focus on in my next post.

(1) For a more detailed account of this experiment, and other related experiments, see Seligman, Martin E. P. Helplessness - On Depression, Development and Death, W. H. Freeman And Company:San Francisco,1975.

Is hope future orientated?

Can Dogs Hope? part 2

As previously remarked, hope may be out of the reach of dogs and other animals because it requires them to conceive of the future in such a way which is impossible without the aid of some kind of grammar or language. But is hope actually future-orientated? Can we hope for past or even present events?

I want to start exploring this issue by examining a footnote by G. Scott Gravlee, in his interesting article Aristotle on Hope:

Thus I take future events to be the paradigmatic objects of hope and fear. One might argue that it is possible to hope and fear concerning what has happened in the past. "I hope he was not hurt, but I fear that he might have been." However, even in this case, what one is properly hoping for is that the news one will hear (a future event) will be good news (recall that one cannot hope for the impossible, and so one cannot properly hope concerning events that have already happened). One properly the fears the news of past disaster. One only wishes that such disaster had not happened.(1)

I felt this was worth quoting in full, just because there's a bag of interesting issues here. Firstly, the idea that one cannot hope regarding what's impossible - this would effectively rule out hope in a pre-determined universe. Hope is then either dependent on free will, or more broadly indeterminism (if there's randomness in the universe, there is still hope). This seems incorrect - even in a rigidly determined universe, I can still hope for that which I don't have knowledge.

But let's look at an example which would more neatly disprove Gravlee's position. Imagine youhave two friends - let's call them Lisa and Bill. Lisa and Bill are good friends - however, Lisa has fallen deeply in love with Bill, and has told you of this fact. She plans to tell him and you encourage her - unbeknownst to her, Bill has expressed similar feelings for her to you. They are going on a car trip together, and Lisa plans to tell him then. However, there's an accident - the car skids of the edge of a dangerous slope or somesuch, and Bill and Lisa plummet to their deaths.

Now, it seems quite clear to me in this scenario that you could hope that Lisa told Bill that she loved him before they died. Seeing as they are close friends of yours, this could be pretty important to you. This would be hope proper - it would not make sense to say "I wish Lisa had told Bill she loved him before they died" because this implies I already know that they didn't. And it could not be that my hope, like in Gravlee's example, is actually for news that Lisa professed her love to Bill - with both of them dead, such news would be impossible.

So hope can be equally be past orientated - it just so happens that the kind of uncertainty is more typically found in future events. And there's no reason why this can't apply to the present - we could hope for the well being of someone for whom we could never expect to receive news from (imagine if they had travelled to another dimension, from was now impossible to get to - like what happens to Rose in Doctor Who).

But there is still a certain kind of common 'distance' in all of these scenarios where hope is possible - a lack of sensory immediacy, and a lack of knowledge/memory. It doesn't feel to me as if Wittgenstein's challenge to whether dogs can hope, or be hopeful, has been diminished by all this. It does bring some useful clarity though - that what we are dealing with is not an issue of temporal distance but something quite different.

We need now to clarify what a dog's cognitive capabilities might be here.

(1) Gravlee, G.S. 'Aristotle on Hope', Journal of the History of Philosophy 38:4 October 2000, p 469

Can Dogs Hope? - Beginnings

'Can Dogs Hope?' is the first of a small number of interrelated projects that I'm going to be pursuing here. My interest in this question was sparked off a number of years ago by the reading of two texts. The first was Wittgenstein's Philosophical Investigations. The second was Martin Seligman's Learned Optimism.

At the beginning of Part II of the Investigations Wittgenstein asks the following:

One can imagine an animal angry, frightened, unhappy, happy, startled. But hopeful? And why not?

A dog believes his master is at the door. But can he also believe his master will come the day after tomorrow?--And what can he not do here?--How do I do it?--How am I supposed to answer this?

Can only those hope who can talk? ... That is to say, the phenomena of hope are modes of this complicated form of life.(1)

What is Wittgenstein saying about hope? Wittgenstein suggests that hope requires language, putting it out of reach of dogs and other animals. And it seems at least part of this relates to the temporal nature of hope - what is emphasised here is how it typically concerns events in the future (i.e. it's future-orientated).

So a reading of Wittgenstein would suggest no. However, two problems became apparent to me, at different points. The first is that, on reading the work of psychologist Martin Seligman, it seems there is good experimental evidence that dogs can be optimistic, and neatly disentangling optimism from hopefulness is no easy task. The second, which came much later, is that the idea of hope as being future-orientated isn't as straight forward as it first appears, and that at the very least this needs disentangling.

These two problems will form the subject of my next two posts on this topic.

---
(1) Wittgenstein, L Philosophical Investigations, Blackwell:Oxford 2001, p148e

A Possible Paradox of Hope?

A. One cannot hope without doubt.
B. To be without doubt is not to be hopeless.

Now, can A and B be true? And if not, which one of these propositions is incorrect?

To justify A - if one is certain of something, then there is no room to hope for it, one way or another. I am certain I'll receive a lovely slice of carrot cake. It cannot properly be said that I'm also hoping for it.

To justify B - if one is certain of achieving one's end, the pursuit of that end cannot be said to be hopeless. Quite the opposite. I'm certain I'll receive a lovely slice of carrot cake - hardly a hopeless situation if carrot cake is my aim!

Now, I can see two immediate ways of tackling this. The first is to say that certainty of a positive outcome, is in a sense, a kind of super-hope, beyond hope. One might recompose it as follows:

A. If I have a full pint, it can't properly be said that I have a half pint.
B. If I have a full pint, I could hardly complain that I'm lacking for a half pint.

Now, I want to shy away from this - I think it's crude. It possibly even amounts to: half a hope + half a hope = blind optimism. The idea that the difference between hope and 'optimistic certainty' is simply quantitative does not seem to my mind helpful. Though it must be said that optimistic certainty precludes hoping, without make one hopeless.

Which leads to the second way out - a distinction between the "act" of hoping/despairing and the disposition of hopefulness/hopelessness. We might want to put it like this:

(i) There's no carrot cake to be had, I cannot hope for it. (hopelessness)
(ii) I am certain to have carrot cake, I need not hope for it. (optimistic certainty)