BJUP 2007 Issue 2(1)

British Journal of Undergraduate Philosophy Journal of the British Undergraduate Philosophy Society Editor: Andrew Stephenson, University of Oxford editor@bups.org Sub-editor: Ryan Dawson, University of Cambridge subeditor@bups.org bjup@bups.org www.bups.org/BJUP Consultant Publisher: Robert Charleston robert.charleston@bups.org ISSN 1748-9393 BJUP - 2(1) - April 2007 Contents 5. 9. 21. 30. 41. 47. 59. 67. 76. 85. 94. 107. 110. Editorial Three asymmetries and a new solution to the new problem of induction Robert Trueman Was Kuhn a relativist? Robbie Duschinsky McDowell’s foundationalism Akosua Bonsu When choosing the best possible world, can God always go one better? Emily Thomas An argument against the principle of alterative possibilities Tomas Bates Limitations of semi-compatibilism: a defence the principle of non-responsibility Gregor Ulm What is it for a particular to instantiate a property? Benjamin Stephen Brown Gewirth, Darwin, eugenics, and solutions to the problem of future agency Jeremy Thomas Interpretation and mystery: religious beliefs and Heidegger’s account of the experience of art Guy Bennett-Hunter Book reviews: Jennifer Saul – Feminism: issues and arguments Heather Arnold Four Shorts Paul Murphy Upcoming BUPS and BJUP events Subscribing and submitting papers to the BJUP -3- British Journal of Undergraduate Philosophy, 2(1), April 2007 © BUPS 2007; all rights reserved. Editorial Hello! … Now, in writing this, my first editorial, I toyed with many such openings. One-liners were my initial thought – snappy observations or quips to grab your attention. But then I couldn’t think of any. Next, in lieu of a snappy one-liner and already in despair before I had even put pen to paper in my new position, I very briefly contemplated being honest. I briefly contemplated being frank about how I felt once all the reviewing, all the compilation, and all the oh-sotroublesome copyediting had been done. But had I done so, the welcoming sentence of the second year of our journal would have been something like ‘So…’, or even ‘Ermm…’. I know what you might be thinking, but I promise that ‘Hello!’, which is what we have ended up with, is infinitely better. Hello! This sentence speaks volumes, especially with that exclamation mark attached (a piece of punctuation made all the more attractive since it became illegal to use it in academic work). Just as the rejected one-worders were elliptical expressions of something like ‘What do I do now dad?’, so is the decided upon one-worder an elliptical expression of my history and relationship with the BJUP and BUPS. I have been involved, giving and taking, almost right from the beginning. And, as I think is clear, I am here for a while yet. But that’s more than enough about me. ‘Hello!’, says the journal, ‘I am a new issue, I am a new year’. So what else is new? Not the lay out or the formatting. Not the quantity and certainly not the high quality. But we do have nine great new papers for you to get stuck into. Not only do we have a miniature colloquium on the relevance of the freedom of the will for ethical responsibility, we also have starkly contrasting approaches to philosophy of science and philosophy of religion. To start us off, then, Robert Trueman reassures us that there is only one problem of induction – it’s almost as though he thinks that one is -5- BJUP - 2(1) - April 2007 enough, a sentiment with which I am sure many epistemologists, metaphysicians, and philosophers of science will agree. Then Robbie Duschinsky asks what at first seems to be a very odd question. It seems odd because the answer, we think, is so obvious and indisputable that it is barely worth thinking about. ‘Of course Kuhn was a relativist’, we may cry, either incredulously or with a faint aura of perplexity about us, depending on our philosophical character. But after reading Duschinsky’s paper, you might feel differently. Next Bonsu daringly but confidently tackles McDowell’s response to the problem of the epistemic regress, offering a qualified but optimistic conclusion, and Emily Thomas’s offering is short, sweet, and to the point: when it comes to baking the best, logicians can always get one over on God. In the molten centre of this issue, Tomas Bates and Gregor Ulm battle it out over principles. In actual fact these papers, both broadly in the area of the reinvigorated free will debate, were composed entirely independently of each other, but it is nice to be dramatic. In any case, they certainly do raise similar issues, but they proceed to tackle them in different ways and from different perspectives, so they make all the more interesting back-to-back reading. Benjamin Stephen Brown takes a much warranted trip back to the basics, and tries very successfully to get clear on an age-old philosophical question which he approaches from a traditional analytic ground. Jeremy Thomas’s thought provoking paper is ethics with a scientific twist (or is it science-fiction with an ethical twist?). Thomas talks schematically of genetic manipulation with the detachment of a lawyer and the rigour of a philosopher. And last but not least, rounding-off the main part of this issue, Guy Bennett-Hunter draws a strong analogy between what he takes to be the proper conception of the philosophy of religion and Heidegger’s canonical aesthetic theory. Phew! So while in a sense there is a lot that hasn’t changed, there is also a very real sense in which there is a lot that is brand new. And that sums-up my ethos perfectly I feel. You shouldn’t expect reforms from me, but you should expect lots of good quality, interesting philosophy. In other words, what you should expect from me, and what I fully expect to deliver, is the presentation of philosophical content. While this may seem depressingly benign, it is far from it. With all the natural worries of an undergraduate philosopher, and with all the application -6- BJUP - 2(1) - April 2007 forms to fill in and maybe all the rejection to deal with, and amongst all the interdepartmental and interdisciplinary feuds and affairs, it would be quite easy to forget about the philosophy, that is, about the content. And with this in mind I will return to the few remaining notes I wish to make regarding this month’s issue. First, we must not forget the book reviews, an essential supplement to all scholarly journals. For what should you be doing when you are not reading our journal if it isn’t reading books? However, for some strange reason we find it tough to get people to write reviews for us, especially considering their importance and the relatively small commitment they require (hint hint). With this in mind I am extremely grateful to both Heather Arnold and Paul Murphy. Arnold’s review, of a new introduction to feminism, is at once light-hearted, informative, and philosophically interesting, which is an incredibly difficult balancingact to pull-off. With the inclusion of Murphy’s reviews we have tried something a little different, although it is something that won’t be too much of a surprise coming from a journal that has in the past attempted reviews of not only whole series of books but even whole genres of books. Murphy manages to give us both a general feel for the books he reviews and specific and usual information about their content, genesis, and context, and he does this for no less than four substantial books in no more than three thousand words. Murphy tackles a collection of interviews with philosophers, a classic Wittgenstein biography, a companion to the philosophy of mind, and a substantial and idiosyncratic introduction to contemporary analytic philosophy. Second, I would like to draw attention to what I feel might be a neglected aspect of the journal, namely the information on upcoming BUPS events and on subscribing and submitting to the BJUP. This information lurks at the back of the journal, no doubt the last straw for a weary reader who has just consumed more than thirty thousand words. But conferences are at the heart of any philosophical society, and submissions and subscriptions are the life-blood of any journal, so take heed of this information and get involved in every way you can. It is also the privilege of this issue to have some particularly exciting information to announce: BUPS and the BJUP are running a prize -7- BJUP - 2(1) - April 2007 essay competition. But don’t think I’m going to give away all the details here, for these you will need to turn to page one hundred and eight. And finally, I feel it is necessary to comment on certain themes that may run as an undercurrent throughout the journal. Those who are aware of my own interests may think (or worry) that, with me as editor, the journal will become more Continental in character and content. Those who think (or worry about) this will not be at all assuaged with the knowledge that the BJUP’s sub-editor, Ryan Dawson, is similarly personally inclined. Fortunately, however, such thoughts are incorrect (and such worries utterly misplaced). This is for many reasons, only a few of which I will briefly mention here. First and foremost, the content of the journal is and always has been dictated by concerns of quality and clarity, regardless of philosophical topic or approach. Second, I not only have immense respect for both traditions, Continental and analytic, I also try to work in and to experience both traditions, as far as that is possible (and although I cannot speak for Dawson, I suspect the same holds for him). BUPS and the BJUP together have always worked to promote and be tolerant of great work in both traditions, and I am sure this will continue. I have promised to let the philosophical content speak for itself, and so I shall. -8- British Journal of Undergraduate Philosophy, 2(1), April 2007 © Robert Trueman; all rights reserved. Three asymmetries and a new solution to the new problem of induction Robert Trueman Fitzwilliam College, Cambridge rt295@cam.ac.uk Introduction Simon Blackburn has in the past tried to reduce the new problem of induction to the old one. He attempted to do this by showing that socalled gruesome predicates involve a change at the appropriate time and we do not project gruesome predicates because we know from experience that such changes do not occur. In this essay I shall examine the two asymmetries that Blackburn says exist. I shall accept one but reject the other. I shall then present a third asymmetry based on metainduction and argue that with the first one this asymmetry successfully reveals the new problem of induction as a special form of the old one. The new problem As is familiar we begin the new problem of induction with the introduction of an artificial predicate, ‘… is grue’: (1) At any time t, a thing x is grue at t if and only if (t<T→x is green) and (t>T→x is blue).1 We then point out that we have exactly the same evidence available to us which confirms the hypothesis ‘all emeralds are green’ as we do for ‘all emeralds are grue’. Every green emerald we have observed has also 1 This is not the definition that Goodman used for ‘grue’. Rather cf. Hesse, M., The Structure of Scientific Inference, 1974, p.79. I believe that this definition is just as powerful for the purpose of proposing the problem and my solution to the problem thus proposed applies equally well to Goodman’s other form. What is more, this definition is significantly less cumbersome to use. -9- BJUP - 2(1) - April 2007 been a grue one. Yet we project the predicate ‘… is green’ but not project ‘… is grue’. We say that future emeralds will be green but do not say that they will be grue past the crucial time T, which would be to say that they will be blue. We then ask ourselves why this is the case, and the new problem is that it seems difficult to construct an account of our inductive practices which properly rules out the projection of such gruesome predicates without also throwing away legitimate ones. The first asymmetry An obvious, and in this form impotent, solution to this problem was quickly dismissed by Goodman. Some suggested that the reason ‘… is grue’ is not projected is that it is positional (roughly meaning that it makes reference to a specific location, either in space or time) and positional predicates are illegitimate. It was never made clear why being positional makes a predicate illegitimate, but in any case Goodman showed that the solution fails.2 ‘… is grue’ is only positional in our language, where we start with ‘… is green’ and ‘… is blue’ as simples. However, you could start in a language with ‘… is grue’ and another gruesome predicate ‘… is bleen’ as simples. ‘… is bleen’ is defined as: (2) At any time t, a thing x is bleen at t if and only if (t<T→x is blue) and (t>T→x is green). If we belonged to such a grue-bleen language, then we would define, ‘… is green’ as: (3) At any time t, a thing x is green at t if and only if (t<T→x is grue) and (t>T→x is bleen). Positionality, then, is relative to the language you are using. ‘… is green’ is just as positional in a grue-bleen language as ‘… is grue’ is in ours. What is more, we have no reason to think our language has got the right predicates as simples. 2 Goodman, N., Fact, Fiction and Forecast, 1983, pp.79-80. - 10 - BJUP - 2(1) - April 2007 Blackburn, however, drew attention to a language-independent positional asymmetry between ‘… is green’ and ‘… is grue’.3 If we want to know whether something remains grue over a period of time, so Blackburn argues, we must know whether or not T comes up in that duration, whether we realise it or not. If that crucial time comes-up, then the thing needs to change to blue, otherwise it needs to remain green. On the other hand, if we wish to know whether something remains green over a period of time, we do not need to know whether any particular time comes up. This, Blackburn says, is not a fact about languages. Even if we accept that grue-bleen languages exist where these gruesome predicates are simple, a speaker of that language would have to know when T is to properly use ‘… is grue’. He may know this fact in a mysterious and innate way and not even realise it, but he must know it, just as we must know what sensations in certain conditions an emerald would give us to know if it is green. It should be noted that this argument is not the same as the similar one which is attributed to Stephen Barker and Peter Achinstein by Goodman in Problems and Projects.4 Barker and Achinstein argue that a predicate is language-independently positional if one cannot make a representation of that predicate which would represent it at any time. For instance, ‘… is grue’ is positional because you would need two images to represent it. You would need a green picture before T and a blue one afterwards. Goodman convincingly argues against such a concept of positionality. But Blackburn’s language-independent positionality comes only from noticing that in order to know whether something is grue you need to know the time. There may be some concerns that not all gruesome predicates are language-independently positional by Blackburn’s meaning. However, I believe we can offer good reason to suppose that they are. Gruesome predicates are predicates which contain what look to us like changes. This is why they are troubling. In order for something to look like it changes there has to be a point where it stops looking like one thing and starts looking like another. It seems to be in the way that we make 3 4 Blackburn, S., Reason and Prediction, 1973, pp.76-7. Goodman, N., Problems and Projects, 1972, pp.402-4. - 11 - BJUP - 2(1) - April 2007 gruesome predicates that they all contain a crucial point (be it a time or a place or whatever). In order to know whether something partakes in a gruesome predicate we must know on what side of the point we are. Therefore, we have good reason to think that all gruesome predicates are language-independently positional by Blackburn’s understanding. I agree with Blackburn that gruesome predicates are languageindependently positional. This is the first asymmetry. But this alone does not solve the new problem of induction, in order to do that we must move on to Blackburn’s second asymmetry. The second asymmetry Blackburn continues his argument like so: ‘telling that a thing retains [gruehood] entails telling that a change happens at the right time.’5 Blackburn says that at T a thing must change in order to remain grue, and that change must be a genuine one. In our language, it would be the change from green to blue. He then says that from experience we have seen that things do not change in such a way and it is a matter of our everyday inductive practices which leads us to not predict such a change. At this stage we can ask, ‘Why do we follow those practices?’ However, in doing so we would just be posing the old problem of induction. Therefore, Blackburn argues, the new problem of induction is just the old problem. Blackburn’s argument is only acceptable if we claim that we possess a language-independent concept of similarity. That is, a speaker of a grue-bleen language would consider the same things to be similar as you and I do. If there was no such standard, then there is no reason why the speaker of a grue-bleen language would consider what happens at T as a change. Instead, he may just conclude that things are carrying on gruely. In his 1984 book, Spreading the Word, Blackburn gives us some reason to assent to the existence of objective similarity. He imagines a gruesome predicate 2G: 5 Blackburn, S., op. cit. 1973, p.78 (italics added). - 12 - BJUP - 2(1) - April 2007 (4) ‘Add 2G’ means ‘add 2 up until 186, then add 7.’6 Blackburn then offers three possible descriptions of someone who uses this gruesome predicate and claims that it is an exhaustive list of descriptions: (5) The person knows when to add 2 and when to add 7 in some mysterious innate way. (6) The person keeps track of how much he has added, so that if he is asked to add 2G before 186 he adds 2, and after 186 he adds 7, and is aware of doing this. (7) As (5), except the person is not aware of doing this.7 Blackburn dismisses (5) on the grounds that it is simply incomprehensible. He then discusses a particular case of (7) and says that should a bricklayer add 2G bricks at a time and fail to notice the sudden change of weight when he passes 186 bricks, then ‘that’s just failure.’8 His argument is that you cannot fail to notice the change which occurs at a certain time when you use a gruesome predicate and know the world about you. This suggests that similarity is not a concept relative to languages and even speakers of a grue-bleen language would have to recognise change where we do in order to avoid failure. If this argument is acceptable, then Blackburn has given himself a languageindependent similarity based on epistemology. Should we accept this similarity asymmetry then it would do the work needed for his 1973 argument. Therefore he would be able to reduce the new problem of induction to the old one. Unfortunately for Blackburn, his argument fails. I shall move away from the example of the brick layer adding 2G as it would be a clumsy example to use for my purposes. However, the following argument 6 Blackburn, S., Spreading the Word, 1988, p.74. Blackburn, S. op. cit. 1988, pp.79-80. 8 Blackburn, S. ibid. p.80. 7 - 13 - BJUP - 2(1) - April 2007 applies to it also. I shall now give a two gruesome predicates and their English definitions: (8) At any time t, a thing x is blight at t if and only if (t<T→x is black) and (t>T→x is white). (9) At any time t, a thing x is whack at t if and only if (t<T→x is white) and (t>T→x is black). Before T whack is lighter than blight, after T blight is lighter than whack. Blackburn would want to say that anyone who failed to notice this change would simply not know the world about them and would genuinely be failing when they see no change. However, this does not seem to be the case if in the whack-blight language there is no predicate ‘… is lighter than’ or ‘… is darker than’ but instead ‘… is liker than’, meaning: (10) At any time t, a thing x is liker than a thing y if and only if (t<T→x is lighter than y) and (t>T→x is darker than y). In such a language a speaker could say that no change has occurred. Everything has remained the same; everything that is whack has remained liker than everything that is blight. If a language was made completely gruesome in a similar way then it would appear at least reasonable to conclude that a speaker of such a language could use his gruesome predicates without recognising a change and know the world about them. Such a speaker would not fall into the tribulations of Blackburn’s bricklayer. They would not be committed to saying that there has been no change in whack things but at the same time accept that whack things have stopped being lighter than blight things. Nor would they have to fail to notice that whack things have ceased being lighter than blight things in order to avoid acknowledging a change in whack things. They simply say that all whack things always have been and continue to be liker than blight things. Not only does this stop Blackburn’s argument for a language-independent similarity, it actually serves as evidence to the contrary conclusion that similarity is languagedependent. - 14 - BJUP - 2(1) - April 2007 I am, then, in the position of granting Blackburn his first asymmetry but not his second. It is natural to slip from the first to the second and Blackburn seems to consider it natural. One can, however, maintain that there is a language-independent positionality but not a languageindependent similarity. The first asymmetry does not presuppose that in order to know whether something remains grue over a period of time you must know whether a change happens at the right point in time. It only says that if you know whether something remains grue over a period of time you must know what the time is. Just as we need eyes in order to something is green, you need some manner of time keeping (whether it be external or internal) to know something is grue. The third asymmetry and a new solution Traditionally a gruesome predicate is roughly understood to be an unprojectible predicate. I shall break with tradition and from now on take language-independent positionality to be the defining feature of gruesome predicates. Whenever I talk of a gruesome predicate I shall mean some predicate which one cannot know obtains without knowing the time (or location or whatever the crucial point may be). With this new definition of ‘gruesome predicates’, it becomes clear that not all gruesome predicates are predicates which we do not project. The predicate ‘… is deciduous’ is, with this definition, gruesome yet we project it. Similarly, ‘… is not grue’ is a gruesome predicate, yet we do seem to project it. If this is not clear, imagine that someone asked you, ‘Are all emeralds grue?’ You would almost certainly answer, ‘No’, which is the same as saying ‘It is not the case that all emeralds are grue’, which involves predicting that there is at least one emerald which is not grue. ‘… is not grue’ can be defined as: (11) At any time t, a thing x is not grue at t if and only if (t<T→x is not green) and (t>T→x is not blue) Clearly you cannot know whether an emerald is not grue over a period of time without knowing whether T comes-up during that duration, making it gruesome by the above definition. - 15 - BJUP - 2(1) - April 2007 At this point, I would also like to make clear what I mean by ‘projecting a predicate’. I shall take it that moving from the premise that all observed Fs are G (or most observed Fs are G) to the conclusion that all Fs are G (or most Fs are G) is paradigmatic of projection. For the sake of ease, I shall restrict myself to discussing this form of projection (although I believe that similar arguments can be constructed for any other form). Moreover, I shall call any hypothesis of the form ‘all Fs are G’, where ‘G’ is a gruesome predicate, ‘a gruesome hypothesis’. I shall also limit myself to temporal gruesome predicates (this is just for simplicity, and the same points can be made for non-temporal ones), and call all gruesome hypotheses whose gruesome predicate’s T was in the past ‘an observed gruesome hypothesis’. All other (temporal) gruesome hypotheses will be called, ‘unobserved gruesome hypotheses’. It is my contention that we only refuse to consent to a very specific group of unfalsified gruesome hypotheses. We do not assent to an unfalsified gruesome hypothesis G iff: (12) The set of evidence that we have which confirms G is the same as the set of evidence that we have which confirms some unfalsified non-gruesome hypothesis, N (13) G makes a prediction contrary to a prediction of N We do not assent to the hypothesis ‘all emeralds are grue’. The set of evidence we have which confirms this hypothesis is the same as the set of evidence we have which confirms ‘all emeralds are green’. ‘All emeralds are grue’ also makes contrary predictions to ‘all emeralds are green’. On the other hand we do assent to the hypothesis ‘all oaks are deciduous’ and there is no unfalsified non-gruesome hypothesis which is to ‘all oaks are deciduous’ as N is to G. I do not wish to claim that it is true without exception that we do not assent to an unfalsified gruesome hypothesis iff it meets (12) and (13). However, I do contend that we refuse to assent to the vast majority of such hypotheses and this is all that I require. It is beyond the means of - 16 - BJUP - 2(1) - April 2007 this essay to prove this claim. The best evidence I can offer is to ask the reader to survey all of the gruesome hypotheses he can think of and decide for himself if there are many (if any) exceptions. The only gruesome hypotheses whose truth-values we can check are observed gruesome hypotheses (otherwise I should imagine that there would be no problems of induction). Most of the observed gruesome hypotheses which met the criteria of (12) and (13) until their crucial moment T were shown to be false. To make this clear we must remember that we can construct any number of observed gruesome hypotheses right now. It is just a matter of our normal inductive practices to move from the claim that most observed hypotheses which met the criteria (12) and (13) until their T were false to the conclusion that most unobserved gruesome hypotheses which meet the criteria (12) and (13) are false. This is the final asymmetry and is also the solution to the new problem of induction. The asymmetry is based on meta-induction – it is making an inductive inference about our inductive inferences. Gruesome hypotheses which meet (12) and (13) until their T have a low success rate and we move to the conclusion that they will continue to have this low success rate by normal inductive practices. We may ask why we follow these practices but this is just to pose the old problem of induction. Therefore, the new problem of induction is just a special case of the old one. To give a more concrete example, take the gruesome predicate earlier defined ‘… is whack’. Let us say that the T in this predicate is 2008. The hypothesis ‘all snow is whack’ is an unfalsified gruesome hypothesis. The set of evidence we have which confirms this hypothesis is the same as the set of evidence we have which confirms ‘all snow is white’. It also makes contrary predictions to ‘all snow is white’. Most observed gruesome hypotheses which make contrary predictions to a non-gruesome hypothesis which was confirmed by the same evidence until T are false. From normal inductive practices we are led to the conclusion that most unobserved gruesome hypotheses are probably false also. This is why we believe ‘all snow is white’ is true and ‘all snow is whack’ is false. - 17 - BJUP - 2(1) - April 2007 We argued against Blackburn by pointing out that his crucial idea of similarity was language-dependent. It seems difficult to construct a similar problem for my argument. The only concepts brought up in my argument are gruesomeness, contrariety, confirmation and our normal inductive practices. Firstly, even if our normal inductive practices were found to be language-dependent, then my argument would not haven been shown to fail to reduce the new problem to the old. Claiming that our normal inductive practices are language-dependent is a (partial) answer to the old problem of induction. My arguments would still show that the new problem is a special case of the old one. It is difficult to see how the idea of confirmation could be language-dependent. Even if it were, this would not show my argument fails to reduce the new problem to the old one. Confirmation is a central part of our normal inductive practices and saying that it is language-dependent would be a (partial) answer to the old problem of induction. Contrariety cannot be language-dependent. At best a different language can have a different concept which performs a job similar to contrariety. However, that is not the same as saying that contrariety is language dependent. Gruesomeness, as I defined it, was shown earlier to be languageindependent. Before concluding I shall briefly discuss what happens in those rare cases where a gruesome hypothesis which meets (12) and (13) is shown to be true. If we asked a man who had never experienced any of the seasons previously, but who was now experiencing his first summer, which was true, ‘all oaks are deciduous’ or ‘all oaks are evergreens’, I believe he would assent to the latter. Moreover, I believe that many share this intuition. If we were to ask the man the same question ten years later, he would certainly say that all oaks are deciduous. In the first summer, the set of evidence he had confirmed the first hypothesis but also confirmed the second. The first hypothesis also made contrary predictions to the second. The first hypothesis was also gruesome and the second was not. Finally, neither hypothesis was falsified. By my argument it is no surprise that the man would assent to the second hypothesis and not the first. Ten years later, however, ‘all oaks are evergreen’ had been falsified. What is more, there was no non-gruesome unfalsified hypothesis whose set of confirming evidence which the man - 18 - BJUP - 2(1) - April 2007 had was the same as the set of confirming evidence which he had for ‘all oaks are deciduous’. By my argument, it is no surprise that he accepted the first hypothesis in such a situation. It is true that we have an example of a gruesome hypothesis which meets the criteria (12) and (13) which turned-out to be true. However, such cases are rarities and so do not disprove the claim that we move inductively from the premise that most observed gruesome hypotheses which meet (12) and (13) are false to the conclusion that most unobserved such hypotheses are false. Conclusion In conclusion, I have reduced the new problem of induction to the old one. I did this by agreeing with Blackburn that there was an asymmetry between gruesome and non-gruesome predicates on the basis of language-independent positionality. I disagreed with Blackburn that a language-independent similarity asymmetry existed, but then I proposed a meta-inductive asymmetry. I used this final asymmetry in conjunction with the first to reveal the new problem to be a special case of the old. This is, in a manner and after a fashion, to solve the new problem, as many thought that there were two problems of induction and at the end of this essay we can see only one. However, the remaining problem is of the greatest importance and is nowhere nearer solved for all the effort of this essay. - 19 - BJUP - 2(1) - April 2007 Bibliography Blackburn, S., Reason and Prediction (Cambridge: Cambridge University Press, 1973, 1st edn.) Blackburn, S., Spreading the Word (Oxford: Clarendon Press, 1988, 3rd edn.) Goodman, N., Fact Fiction and Forecast (Cambridge, Ma: Harvard University Press, 1983, 4th edn.) Goodman, N., Problems and Projects (Indianapolis and New York: The Bobbs-Merill Company, Inc., 1972, 1st edn.) Hesse, M., The Structure of Scientific Inference (London: The Macmillan Press Ltd, 1974, 1st edn.) - 20 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © Robbie Duschinsky; all rights reserved. Was Kuhn a relativist? Robbie Duschinsky Selwyn College, Cambridge rd313@cam.ac.uk This essay will make the controversial argument that it is precisely the ‘subjective’ element in Kuhn’s account of scientific activity, which influential thinkers such as Popper, Searle and Sheffler see as the cause of his relativism, that is the locus of his objective conception of science. I will attempt to show that his sociological analysis demonstrates that the subjective values of the scientific community orientate the scientist towards objective experimental results, even through the changing incommensurable lexicons of paradigm shifts. In The Structure of Scientific Revolutions, Kuhn explains scientific development as constituted by a tension between two alternating stages.1 Firstly, there are periods of ‘normal science’ in which scientists attempt to solve puzzles through the application of a method embodied in a model set of problem solutions.2 Over time, normal science produces a crisis created by mounting inexplicable anomalies that call the whole theoretical order into question. Secondly, then, there are periods of revolutionary change, in which this order- the reigning ‘paradigm’- is overturned to produce a new model of reality, one which renders the anomalous findings intelligible and which initiates a new period of normal science.3 Elements of the old theoretical order may still be employed, but will now be used to describe a new world; in effect, post-revolutionary scientists will be speaking a new incommensurable scientific language, one that makes claims about the cumulative nature of scientific advancement impossible .4 1 Kuhn (1977), p. 227. I will be using ‘paradigm’ to refer to a ‘disciplinary matrix’, cutting free the notion of ‘exemplar’. 3 Kuhn (1996), pp. 181-7. 4 ibid. p.150, p. 202, p.205. 2 - 21 - BJUP - 2(1) - April 2007 A certain reading of this last conclusion has been the source of deep discomfort for many philosophers of science, who have seen Kuhn’s arguments as relativist. They argue that the lack of a neutral extraparadigmatic language with which to describe, and thus evaluate, the two theories seems to preclude the rational justification of theory choice; they see Kuhn as arguing for ‘mob rule’ and ‘irrational’ factors as the main deciding considerations in such a decision.5 The classic example of such criticism is Israel Scheffler’s Science and Subjectivity, and thus it is with an outline of his reading of Kuhn that we shall begin. Scheffler broadly provides one single criterion for a non-relativist theory: the model must be subject to control by reference to experimental tests of its claim to describe objective reality. The theory’s context of discovery is irrelevant; what matters is simply that the theory measures up to the objective facts. A theory that fits these facts is superior to one that does not.6 By contrast a relativist believes that all theories of reality are each as relatively good as each other, since he or she eschews criteria for judging them. Kuhn, on Scheffler’s reading, believes that scientists change from one theory to another by conversion rather than through rational consideration, since paradigms are largely unfalsifiable. Until anomalies mount and a viable alternative presents itself, paradigms merely undergo ad hoc changes which preserve their essential theses. This perspective, argues Scheffler, is relativist since it sees paradigms as chosen, in the absence of convincing proof from experimental testing, solely on the basis of a subjective, arbitrary choice.7 Scheffler’s second argument proposes that Kuhn believes that after a paradigm-shift, scientists see new things when looking through the same instruments at the same objects, and thus they are inhabitants of a different world. The objects that constitute this new world are describable only in the language of the new paradigm since there is no 5 Lakatos (1970), p. 178. 6 Scheffler (1982), p. xi, pp. 1-2, p. 19. 7 Scheffler (1982), p. 75, referencing Kuhn (1996), pp. 77-8. - 22 - BJUP - 2(1) - April 2007 access to the original objective reality, only to our theory-implicated perceptions of it; Scheffler therefore suggests that, in Kuhn theory, there is no possibility of comparing a paradigm to objectivity reality, and that therefore all perspectives are equally valid as truth claims.8 In response to Scheffler, Kuhn has denied that he gives no importance to the empirical checking of the paradigm: the difficulty with which we come to terms with novel stimuli does not ipso facto preclude it from coming to our attention. It is indeed true that, initially, the scientist blames her own working methods, and consequently ignores a single exposure to a fact that does not fit with the existing paradigm. The scientist continues to have faith that the theoretical framework will account for the anomaly in time. However, if this problem proves to be intractable, then crisis results as the paradigm itself is questioned, and the constitution of the taxonomic categories belonging to the scientific community must be radically rethought.9 I think this stance is illustrated clearly in Kuhn’s major historical work Black-Body Theory and the Quantum Discontinuity. Here he describes how the traditional account of the discovery of quantum theory by Max Planck is incorrect, and that actually his theoretical results relating to discontinuity were part of a normal science research programme of applying to the problem of energy Boltzmann’s statistical approach to molecules, firmly grounded in the Newtonian physics of his day. What quantum theory was later to call ‘oscillators’, the centrepiece of the early theory of discontinuous energy change, Planck initially conceived of as ‘resonators’, absorbing and emitting energy at a rate governed by Maxwell’s Newtonian equations; the radical implications of Planck’s law were perceived by Lorentz, who later had to convince Planck himself of their significance.10 This historical research indicates that, on my reading, Kuhn is not a relativist since what he is documenting is how a scientist was surprised by his own results,11 something that requires a reality independent of the reigning paradigm. Proof of 8 Scheffler (1982), pp. 40-3, p. 76, pp. 80-1, referencing Kuhn (1996), p. 109. Kuhn (1996), p. 64; (2000), p. 97. 10 Kuhn (1978), pp. 188-9, pp. 198-201. 11 ibid., p. 369. 9 - 23 - BJUP - 2(1) - April 2007 Kuhn’s conviction in an external reality exists in his descriptions of crises and revolutions. As I understand him, Kuhn is not arguing that ‘each viewpoint creates its own reality’, as Scheffler believes,12 rather his conception of scientific development is that ‘nature cannot be forced into an arbitrary set of conceptual boxes.’13 There is indeed something fixed, external and constraining- resembling a Kantian Ding an sich- and Kuhn believes that there can be many different ways of speaking about this ‘thing’, with some being better for certain purposes (such as accurate prediction) than others, but without any coming closer to providing an ontologically correct picturing.14 This is supported by the fact that Kuhn’s conception does not rule out scientific progress, in a certain sense. He considers that the number of accurate predictions that a paradigm can make will almost inevitably be more than its predecessor.15 The reason for this is that normal science is very effective in its project of increasing the match between the facts shown by experimental studies and the paradigm’s predictions- in this sense normal science is ‘a highly cumulative enterprise’. Moreover, as well as observing novel regularities, a large part of the previous paradigm’s gain in precision and scope must be preserved by its successor if it wishes to be accepted by the scientific community.16 Science progresses in an evolutionary sense- Kuhn describes himself, in his last published article on this issue, as developing a ‘post-Darwinian Kantianism’- it achieves greater adaptation to the various tasks set it (e.g. accuracy of prediction; as Rorty points out, there are many possibilities17) by its niche of physical reality. In order to achieve that end there is increasing specialisation, but like natural selection this evolution is not towards some preordained result.18 What Kuhn is 12 Scheffler (1982), p. 19. Kuhn (2000), p. 95, p. 104; (1970), p. 263. 14 ibid. 15 ibid., p. 264. 16 Kuhn (1996), p. 24-5, p. 52, p. 169. 17 Rorty (1999), p. 179; Kuhn (2000), p. 104. 18 ibid., p. 98, pp. 117-120. 13 - 24 - BJUP - 2(1) - April 2007 objecting to, I believe, by denying the cumulative nature of science, is the idea that with the increase in problem-solving ability comes a better fit between objective noumena and their picture, the paradigm. This is impossible, Kuhn argues, due to the inescapable influence of subjective factors on our perception of reality, such as that we only ever choose between paradigms rather than evaluating each one alone against reality.19 Against the claim that the inclusion of subjective elements in theorychoice leads to relativism, Kuhn elaborates in his lecture ‘Objectivity, Value Judgement and Theory Choice’ how such subjective values are actually the main insurance that the predictions of a paradigm are continually measured against the standard of the results of empirical studies. He references Scheffler’s charge of relativism on this account, and claims that it shows ‘total misunderstanding.’20 The problem, he suggests, comes from an ambiguity contained within the term ‘the subjective’, in that it can be opposed both to ‘the objective’ and at the same time can be opposed to ‘rational deliberation’; Kuhn’s critics are mistaking his use of this first sense for the use of the second- they see him as claiming scientists as irrational, whereas what he is actually proposing is that scientists can consider the same objective facts, and yet come to different conclusions, like two civilisations deriving different constellations from the same pattern of stars.21 Kuhn writes of three broad types of causes of this difference in perception: the individual scientist’s biography, the values of the scientific community, and the wider cultural context. Of these he singles out the second as the most significant. There are, loosely, five overarching values that scientists hold and through which they rationally choose between competing paradigms: firstly, and most importantly, they believe that a theory should be accurate in agreement with past observations and the results of its own predictions; secondly, a theory should be internally consistent and fit with other accepted theories; thirdly, the theory should have a wide scope; fourthly, it 19 ibid., p. 96; (1996), p. 4; (1970), p. 265. Kuhn (1977), p. 321. 21 ibid., p. 337; (2000), p.223. 20 - 25 - BJUP - 2(1) - April 2007 should order phenomena that would otherwise seem individually isolated; lastly, it should be fruitful in revealing new phenomena and disclosing previously unnoted relationships.22 Paradigms are chosen on the basis of both their current and their projected ability to meet these standards.23 I would argue, therefore, that Kuhn believes that it is a subjective orientation towards reality, as ensured by these five standards, which gives the results of objective testing authority in the evaluation of paradigm. Kuhn considers it vital that these values are essentially contestable: each one can be interpreted differently, and their relative importance debated. Before the scientific community accepts a new reigning theory it has to have been tested thoroughly against reality by both scientists inside and outside of the paradigm; it would be a disaster if scientists en mass abandoned an old theory too quickly for a bogus newcomer and equally bad if they refused to accept a new theory with greater accuracy and scope. Therefore Kuhn sees the fact that rational scientists can disagree to be highly ‘functional’ for science, since on average enough scientists investigate a new contender from the inside, while enough scientists remain loyal to the incumbent to provide a healthy balance of acceptance and scepticism.24 Yet this set of arguments from Kuhn, whilst convincing, does not fully address the second of Scheffler’s criticisms, in that if, with a scientific revolution, the whole world of the scientific community changes, then whatever values they possess will be irrelevant in maintaining a common standard with which to judge the relative merits of a paradigm and its predecessor. The issues of relativism and incommensurability are explored in Kuhn’s paper ‘Commensurability, Comparability, Communicability’ to which we shall now turn. The term ‘incommensurability’ comes from mathematics. The circumference of a circle is incommensurable with its radius, in that there is no unit of length contained, without remainder, a whole 22 ibid., p. 115, (1977), pp. 321-3. Kuhn (1996), p. 23, p. 153. 24 Kuhn (1970), pp. 234-6, pp. 330-2. 23 - 26 - BJUP - 2(1) - April 2007 number of times in both. However, this lack of a whole common measure does not make comparison impossible- approximate comparisons are easy to perform. What Kuhn means when he claims that two theories are incommensurable is that there is no taxonomic language, or lexicon, into which both paradigms, considered as a set of sentences, can be entirely translated without loosing meaning. For most of the terms of each scientific lexicon, translation leaves no remainder, because, after all, the scientist is still working in the same objective world, if not the same phenomenological one. For a few terms, however, translation does involve loss. This partial change is what Kuhn means by incommensurability, and he coins the term ‘local incommensurability’ to isolate this more modest sense from ‘total incommensurability’.25 It is the terms that do preserve their meaning through a revolution which allow for the discussion and comparison of a theory, in view of making a theory-choice. What I believe Kuhn is suggesting is the impossibility of point-by-point translation of every term in a paradigm’s lexicon into the terms of another, and then the assessment of both, point-by-point, against reality, the reason for this being that terms change their meaning in different contexts, and this alteration has an affect on the surrounding cluster of concepts. If a scientist wishes to assess the relative merits of two theories he or she should not attempt a translation from one into the other, but rather must act as an interpreter and become proficient in both scientific lexicons without attempting to reduce one to the other.26 Both theories can then be tested against their respective predictions of reality,27 and the ‘bilingual’ scientist can move a great deal towards a comparison of relative effectiveness. Kuhn claims that this possibility ‘was for me never in question’,28 and we can see in Structure that ‘it makes a great deal of sense to ask which of two actual and competing theories fits the facts better’,29 an entirely illogical assertion, I would suggest, unless he is 25 ibid., p. 129; (2000), p. 36, pp. 58-60. ibid., p. 36-40, p. 44-45; (1977), p. 338; (1996), p. 202. 27 Kuhn (1977), p. 339. 28 Kuhn (2000), p. 55. 29 Kuhn (1996), p. 147. 26 - 27 - BJUP - 2(1) - April 2007 arguing for local rather than total incommensurability between paradigms. However, Kuhn continues, this comparison can never be complete. The large number of invariants between theories provides no way of eradicating local incommensurability.30 Rather, what occurs is that at a certain moment, the scientist has a change of ‘native language’ such that, though still bilingual, he or she is now a thinker in the new scientific lexicon. This change is motivated by the rational values they hold since the decision to believe that the new paradigm will be a fruitful research programme is based to a large extent on the past performance of its predictions, in objective tests.31 Conversion is therefore not relativistic, in Scheffler’s sense, since it is a product of faith in the future ability of a paradigm to make successful truth-claims about reality. The fact that this faith is not determined solely by the available facts is essential for science, since it ensures that there are scientists who wish to elaborate the new alternative, as well as those who wish to defend the old paradigm, ensuring a theoretical division of labour.32 Kuhn is not, I believe, a relativist- there is still the opportunity for the assessment of paradigms against the experimental facts in his understanding of science. Bibliography Foucault, Michel, (2002 [1966]) The Order of Things, Routledge Kant, Immanuel, (1996 [1781]) Critique of Pure Reason, CUP Kuhn, Thomas S., (1985 [1957]) The Copernican Revolution, Harvard University Press 30 ibid., p. 148. Kuhn (1977), p. 338; (1970), p. 277; (2000), pp. 74-77. 32 Ibid., p. 204; (1996), p. 23-4, pp. 184-5; (1977), p. 331. 31 - 28 - BJUP - 2(1) - April 2007 Kuhn, Thomas S., (1996 [1962]) The Structure of Scientific Revolutions, 3rd edn., University of Chicago Press Kuhn, Thomas S., ‘Reflections on my Critics’, in Lakatos and Musgrave (eds.), (1970) Criticism and the Growth of Knowledge, CUP Kuhn, Thomas S., (1977) The Essential Tension, University of Chicago Press Kuhn, Thomas S., (1978) Black-Body Theory and the Quantum Discontinuity, 1894-1912, University of Chicago Press Kuhn, Thomas S., (2000) The Road Since Structure, University of Chicago Press Lakatos, Imre, ‘Methodology of Scientific Research Programmes’ in Lakatos and Musgrave (eds.), (1970) Criticism and the Growth of Knowledge, CUP Lakatos and Musgrave (eds.), (1970) Criticism and the Growth of Knowledge, CUP Nagel, Ernest, (1979) The Structure of Science, Hackett Publishing Company Popper, Karl, (2002 [1935]) The Logic of Scientific Discovery, Routledge Rorty, Richard (1999) ‘Thomas Kuhn, Rocks and the Laws of Physics’ in Philosophy and Social Hope, Penguin Books Scheffler, Israel, (1982 [1966]) Science and Subjectivity, 2nd edn., Hackett Publishing Company Watkins, John, ‘Against ‘Normal Science’’ in Lakatos and Musgrave (eds.), (1970) Criticism and the Growth of Knowledge, CUP - 29 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © Akosua Bonsu; all rights reserved. McDowell’s foundationalism Akosua Bonsu Heythrop College, University of London akosua_b@hotmail.com In Mind and World, John McDowell argues that states of the world, which we access via experience, can justify empirical beliefs. According to McDowell, when things go well in our experience of the world, we become open to empirical facts. Those facts are the purported contents of experiential states and they ‘rationally constrain’ empirical belief. The claim, then, is that veridical experiences provide reasons for empirical beliefs because that method of belief acquisition discloses facts about the world to us. If successful, McDowell’s argument shows how to resolve the problem of epistemic regress, if only for empirical beliefs. The bulk of this paper will be concerned with an examination of McDowell’s position. First I identify the problem of epistemic regress. Then I give an account of McDowell’s Minimal Empiricism, which I suggest can be employed to solve the problem of epistemic regress. And finally I deal with a potential problem for this argumentative strategy. I conclude with some general remarks on the issues discussed. Let’s assume that justification is a necessary condition for knowledge. So, where P is the empirical proposition ‘Tabby is in the garden’, S cannot know that P without being justified in believing that P. I take it that X justifies P if and only if X is something that can be legitimately cited to demonstrate S’s entitlement to her belief that P. This view of justification is neutral as regards the internalist/externalist debate.1 But we should note straight away that McDowell is an internalist. He spells out justification in terms of a subject having and giving reasons. According to McDowell’s view, S is justified in believing P just in case Sincerest thanks to Craig French for our critical discussions. 1 All reference to ‘justification’ will refer to this neutral notion, unless explicitly indicated. - 30 - BJUP - 2(1) - April 2007 S has good reasons which she could cite when she is challenged to demonstrate entitlement to the belief. The regress argument is employed by a sceptic to establish that justification (construed in either internalist or externalist terms) is impossible. If successful, the sceptic can conclude that S can never know that Tabby is in the garden because S can never be justified in believing that Tabby is in the garden. The regress argument is generated by considering the following question: (a) What can S’s belief in the truth of P be justified by? There is a restriction on what kind of answer can be given to this question. As justification is a rational relation it requires contentful relata. This means that the terms of the justificatory relation must have content. Contents are propositions and propositions are (here conceived as) Fregean Thoughts: structured complexes, with Senses as constituents. This requires that whatever is enlisted to justify the belief that P must have (or be indexed by) propositional content. Only those contents can inferentially support, probabilify,2 or be reasons for a belief.3 To say A justifies B is just to say that A confers justification on B in virtue of some rational relation (e.g. entailment, probabilification etc.) that holds between the propositional contents of A and B.4 This restriction has prompted some philosophers to answer (a) by claiming that only a belief can justify another belief.5 The problem with this suggestion, in this context, is that whatever belief is enlisted to justify S’s current belief that P, will itself be in want of justification. It is prima facie legitimate to ask of any justifying belief: (b) What further belief is S’s belief in the truth of P justified by? As we are working on the stipulation that only a belief can justify another belief, we are forced to answer this question by citing some 2 Where A ‘probabilifies’ B if and only if A justifies a belief that B is probable. Crispin Wright (2002), p. 143. 4 Sellars (1956a), pp. 127-196. 5 Davidson (1986), p. 310. 3 - 31 - BJUP - 2(1) - April 2007 further belief that S has. This is deeply problematic because there appears to be no reason to stop us asking (b) of any justifying belief that is proposed. If there is in fact no good reason to block the sceptics demand for justification, a regress of justifying beliefs is generated. And if there is no stopping place to that regress, S must first go through an infinite series of prior justifications before she can claim entitlement to her belief that P. But this is a task beyond any human’s conceptual capacities, so we must conclude with the sceptic that S is not warranted in taking herself to be justified in believing P. Nor are we, as attributors, warranted in taking S to believe that P. On the assumption that justification is a necessary condition of knowledge, we must conclude that S can never know that P.6 One response to the problem of epistemic regress is Foundationalism. The strategy common to all foundationalist theories is to block the infinity of the regress by postulating a foundation on which the edifice of knowledge rests. Looked at in this way, McDowell is clearly a foundationalist for he argues that empirical knowledge rests on a foundation of experience. This is just to say that experience justifies our empirical beliefs, and the question ‘What justifies your experience that P’ is ill-placed. But, if experience is conceived as sensation or sense-data (as the traditional empiricist has it) it cannot justify anything. This is because sensations or sense data lack conceptual content and so cannot enter into rational relations. Only things with conceptual contents can have or transmit positive epistemic status. So if McDowell is to avoid buying into what Sellars called the ‘Myth of the Given’, he must argue that experience has conceptual content. And this is precisely what he does. The idea is that the conceptual capacities of a perceiving subject are passively engaged in the production of experience. So in experience a subject is ‘saddled with content’. S sees, for example, that it is raining and the content of her experience is the proposition following the ‘that’ clause.7 6 As S can be replaced with any subject and P can be replaced with any proposition we are to think of, radical scepticism results. 7 McDowell: ‘We should understand [experience] not as a bare getting of an extraconceptual Given, but as a kind of occurrence or state that already has conceptual content. In experience one takes in, for instance, sees that things are thus and so. This is the sort of things one can also, for instance, judge’ (1994), p. 9. - 32 - BJUP - 2(1) - April 2007 This rather bizarre theory requires a radical refashioning of the notion of ‘experience’. McDowell provides transcendental arguments for this conceptualization of experiential contents. In brief, the idea is that empirical beliefs could not be about the world if those beliefs could not be justified by experience. But in order for empirical beliefs to be justified by experience, they must have propositional content. So in order for empirical beliefs to be about the world, experiential content must be propositional content. In discussing these interconnected claims we will see how experience can confer justification onto empirical beliefs and why the question of justification does not arise in the case of genuine experience (in other words, we will see why experience blocks the regress). According to McDowell, thought about the world is only intelligible if the following normative condition is met: (1) The world must ‘rationally constrain’ empirical thought.8 This condition bifurcates into two distinct but related claims. The first claim is motivated by the observation that empirical thought must be ‘answerable to the world’ for its correctness. The world is here conceived in Tractarian terms as the totality of facts, and facts are conceived in Fregean terms as true Thoughts.9 In this context ‘correctness’ is used as a synonym for ‘truth’, so the claim is that beliefs about the world are true or false depending on what the facts are. It is certainly not up to me whether my belief that P is true, rather the truth of this belief is determined by the facts which do or do not obtain. If P expresses a fact, then the contents of my belief that P are factual, and this means that my belief is guaranteed to be true. From this observation, McDowell stipulates the following normative condition: 8 McDowell (1994), p. 163. This view of reality as the totality of facts is contestable. It is natural to think that the World consists of non-propositional objects like people, coffee, chairs etc. I sympathize with this objection, but would like to put it aside for the purposes of this essay. For an interesting defense of McDowell on this point, see Fish & Macdonald (2007). 9 - 33 - BJUP - 2(1) - April 2007 (i) If empirical beliefs are to be about the world, S ought to believe P (where P is an empirical proposition) only if P expresses a fact. Following on from this first constraint, a second normative constraint is motivated by the thought that the world itself must impose a genuine constraint on our thinking. Considering that my thoughts about the world are answerable to the world for their truth, these thoughts should not be a product of my imagination, but should be suitably constrained by how the world actually is. Thus McDowell stipulates the following normative condition: (ii) If beliefs are to be about the world, facts ought to determine (and thereby restrict) the contents of those beliefs. To stipulate normative condition (1) is just to hold (i) and (ii). Both conditions are beyond reproach if we accept that the world is independent of what we think about it and true thoughts about the mind-independent world are desirable. The two conditions amount to the following claim: beliefs about the world could not be possible if a mind-independent world did not determine the correctness (or truth) of our thoughts, and if our beliefs about that mind-independent world diverged from what the world is actually like. A few things should be noted about (1). Firstly, we must note that facts must be available to subjects if the normative constraints necessary for thought to be about the world are to obtain. For if facts are not available to subjects, it would be impossible to determine the correctness of our empirical beliefs and it would be impossible to know that the contents of those beliefs are constrained by objective reality. But facts are peculiar. Like sensations, they cannot be justified or unjustified, but unlike sensations, facts (conceived as epistemically true thoughts) do have conceptual content so they can enter into rational relations. As such, facts can justify empirical beliefs, and, as facts determine the truth of empirical beliefs, they prove to be excellent justifiers. This means that S can appeal to the facts in demonstrating her entitlement to her beliefs, even though they are not themselves the - 34 - BJUP - 2(1) - April 2007 objects of justification. If in response to the sceptic’s question ‘What is your belief in the truth of P justified by?’ S responds ‘P is a fact’, then that answer settles the matter. The sceptic makes no sense if he persists in his demand for justification. Facts are not that kind of thing. This peculiar nature of facts, of being excellent justifiers but of which it makes no sense to ask for a justification, along with the claim that facts must be accessible to subjects if empirical beliefs are to be about the world, means that all subjects must have available to them excellent (and foundational) justifiers for their empirical beliefs. But how can facts be available? McDowell answers that facts must become available through experience.10 This new turn in the argument leads us to conceptualism about the content of experience. McDowell notes that our ‘cognitive predicament’ is such that the facts (the totality of which is the world) can only become available to us through experience, but that means that the only way facts can become available to subjects is if they are contained in experiential contents. This idea is motivated by McDowell’s theory of perception. According to this theory, veridical experience involves the empirical facts ‘making themselves manifest’ to a perceiving subject where that ‘manifestation relation’ is an explanatorily basic relation which puts subjects in direct perceptual contact with the mind-independent world. So the contents of experiential states are the facts of which a subject is made directly aware through experience. As facts are conceptually structured, experiential contents are conceptual contents. When S sees that Tabby is in the garden the fact that Tabby is in the garden is content of her experience. Her experiential content can justify her belief that Tabby is in the garden because that content is factual. Here it will be noted that, as McDowell says, the contents of the belief and the content of the experience are identical. Beliefs are differentiated from experiences not by their content, but by the way in which S’s conceptual capacities are activated. In experience S’s conceptual capacities are passively engaged. S cannot help but see that Tabby is in the garden. Beliefs, however, are associated with inducement and acceptance. In belief S chooses to accept the content of her experience as true after engaging in explicit checking procedures. In forming beliefs, those very same conceptual 10 McDowell (1994), p. xii. - 35 - BJUP - 2(1) - April 2007 capacities which are passively engaged in the formation of experience, are actively engaged.11 Before moving onto a potential problem, I will first offer a brief summary of the above. If all that has been said so far is cogent, the sceptical paradox has been dissolved. According to McDowell, facts need to rationally constrain beliefs if beliefs are to be about the world. In order to rationally constrain beliefs, facts must be accessible to subjects. The only way for facts to become accessible is if they are contained in the contents of experience. According to McDowell’s theory of perception, we are receptively open to the facts in experiences, so experiential contents just are the facts. This means that experiential contents are conceptual contents, so they can enter into rational relations.12 And as facts are foundational justifiers, experiential contents are foundational justifiers of empirical beliefs. Experience is able to confer positive epistemic status onto the relevant beliefs in virtue of having facts as contents. When, for example, I see that Tony is in the park, the fact that Tony is in the park is made manifest to me. When I appeal to my seeing that Tony is in the park as my reason for believing that Tony is in the park, I am appealing to the facts themselves as my reason for belief. In other words, when I endorse the content of my veridical experience my belief is guaranteed to be true because the content of my veridical perceptual experience just is the fact that P. As 11 McDowell (1996), p. 26. Whether McDowell makes a credible distinction between beliefs and experiences is a matter of debate (e.g. Stroud 2002, Wright 2002). It is also a matter of debate whether McDowell is committed to some objectionable form of idealism (e.g. Gaskin 2006). I take it that the above comments do commit McDowell to an idealism of sorts. But I don’t see how this form of idealism can be separated from any sensible form of realism. As experience involves a passive exercise of conceptual capacities the world can exert a genuine constraint on our thinking, but conceptual capacities have to be exercised if our experience is to be of anything. For experience (as we know it) is unthinkable outside our conceptual framework and our experience would be impossible apart from the organization it receives from our conceptualization of the world. 12 As experiential contents are facts, the jobs that facts do are taken over by the contents of experience. So experience must rationally constrain empirical beliefs in virtue of having facts as contents. Which is just to say experience must determine what is correct or incorrect to believe as well as what we ought to believe (in virtue of having facts as contents). - 36 - BJUP - 2(1) - April 2007 experience can justify an empirical belief in virtue of having facts as contents, S can appeal to the contents of her experience as her reasons for believing P and in doing so S is appealing to the facts themselves as her reasons for believing P. When the sceptic asks ‘what justifies your belief that P’ and S replies ‘My experience of seeing that P’, that settles the matter because S’s experience is factive, and facts are just not the kinds of things that need to be justified. In the final part of the paper I would like to briefly discuss an objection to the argumentative strategy sketched above and also sketch a response. Someone might argue as follows. S cannot justify her beliefs by appealing to the contents of her experience unless her experience is veridical. If her experience is hallucinatory her experiential content could not be a fact, and as such, it could not justify her belief. But given McDowell’s internalist notion of justification, S must know that she is justified in order to be justified. But in order for S to know that she is justified, she must know that her experience is veridical. But veridical experience is indistinguishable from hallucinatory experience; there can be no reflectively accessible grounds which indicate that S’s current perceptual experience is veridical. So S cannot know when her experience is veridical, and hence S cannot know that she is justified. In other words, because of the phenomenological indistinguishability of the hallucinatory content and the veridical content, S cannot know when she is appealing to the facts themselves to justify her beliefs, and when she is appealing to some hallucinatory content instead. The objection amounts to the claim that so-called Minimal Empiricism, as that which is outlined above as McDowell’s position, can only be used to block the regress (i.e. Pyrrhonian scepticism), if Cartesian scepticism can first be overcome. This is a very tricky argument to respond to. The key to the argument is that reflective access to the veracity of content is restricted to purely phenomenological criteria. If we accept this, it seems that the only way for McDowell to deal with this argument is to give up internalism. By adopting an externalist theory of justification we can say that there are various facts which determine what kind of situation S is in, but those facts are not reflectively accessible to her. However, this is an unattractive option because internalism plays a crucial role in - 37 - BJUP - 2(1) - April 2007 McDowell’s argument. One way to deal with this problem and retain internalism is to prise apart reflective access from phenomenological access. In that case we could allow that S knows when she is in a veridical case even though the content of veridical experience is phenomenologically indistinguishable from the content of hallucinatory experience. This is precisely what McDowell does argue, but discussing it would take us too far afield. The point here was only to outline a response to the objection.13 It seems to me that some form of Foundationalism must be the correct response to the epistemic regress problem. And it seems clear that, whatever the foundation, it needs to have conceptual content if it is going to be capable of justifying beliefs. For these reasons McDowell’s Minimal Empiricism is attractive. He defends the view that experience must have factual content and hence is capable of justifying empirical beliefs. I agree that experiential contents are factual, because our everyday practice testifies to the claim that experience determines the correctness of empirical beliefs as well as what we ought to believe. And given the peculiar nature of facts, it can be argued that experiential contents are foundational justifiers for empirical beliefs. These conclusions chime well with commonsense for it seems obvious that experience does justify empirical beliefs. It seems a little less obvious that the world is the totality of true thoughts, or that the content of veridical experience involves the perceptual facts making themselves manifest to a perceiving subject. I think both of these claims are defensible, but have not attempted to defend them here. This is because in order to present the argument these issues must be presupposed, and critical appraisal of those presuppositions would take far too long. Suffice to say, the argument as it is presented is complex because it required an acceptance of a battery of McDowellian assumptions. And even granting these assumptions, there was still one major problem with employing Minimal Empiricism as a response to Pyrrhonian Skepticism, namely that we might first have to combat Cartesian scepticism. Responding to this objection is a tricky business which deserves critical attention, but again, such attention 13 C.f. Pritchard (forthcoming) ‘McDowellian NeoMooreanism’. - 38 - BJUP - 2(1) - April 2007 could not be given here. I hope at least to have revealed a little of the machinery that runs through Minimal Empiricism, and suggested an interesting way in which it can be applied to rebut a certain kind of scepticism, but clearly there is much work to be done. Bibliography Bonjour, L., (1985), The Structure of Empirical Knowledge, Cambridge Mass.: Harvard University Press Brandom R., (2002), ‘Non Inferential Knowledge, Perceptual Experience, and Secondary Qualities: Placing McDowell’s Empiricism’, reprinted in Reading McDowell on Mind and World, Smith (ed.), pp. 92-105 Byrne, A., (1996), ‘Spin Control: Comments on McDowell’s Mind and World’, Philosophical Issues, Vol. 7, pp. 261-273 Chen, K. C., (2006), ‘Empirical Content and Rational Constraint’, Inquiry, Vol. 49, No. 3, pp. 242–264 Davidson, D., (1986), ‘A Coherence Theory of Truth and Knowledge’, reprinted in LePore (ed.), Truth and Interpretation: Perspectives on the Philosophy of Donald Davidson, pp. 307-19 De Gaynesford, M., (2004), John McDowell, Polity Press DeVries, W. A., and Trimplet, T., (2000), Knowledge, Mind and the Given: Reading Wilfred Sellars ‘Empiricism and The Philosophy of Mind’, Hackett Fish, W., and Macdonald, C., (2007), ‘On McDowell’s Identity Conception of Truth’, Analysis, Vol. 67, No.1, pp. 36–41 McDowell, J., (1994), Mind and world, Cambridge Mass.: Harvard University Press - 39 - BJUP - 2(1) - April 2007 McDowell, J., (1996), ‘Précis of Mind and World’, Philosophical Issues, Vol. 7, pp. 231 – 239’ McDowell, J., (1986), ‘Singular Thought and the Extent of Inner Space’, in Subject, Thought, and Context, Pettit & McDowell (eds.), Oxford: Clarendon Press, pp. 136-68 McDowell, J., (1994), ‘The Contents of Perceptual Experience’, The Philosophical Quarterly, Vol. 44, pp. 190-205 McDowell, J., (1995), ‘Knowledge and The Internal’, Philosophy and Phenomenological Research, Vol. 55, No. 4, pp. 877-893 Sellars, W., (1956a), ‘Empiricism and the Philosophy of Mind’, reprinted in DeVries & Trimplet (eds.), (2000) Sellars, W., (1956b), ‘Does Empirical Knowledge have a Foundation?’, in The Foundations of Science and the Concepts of Psychology and Psychoanalysis, Minnesota Studies in the Philosophy of Science, Vol. 1, Minneapolis: University of Minnesota Press, pp. 293-300 Sosa, E., (1980), ‘The Raft and the Pyramid’, Midwest Studies in Philosophy, Vol. 5, Studies in Epistemology, Minneapolis: University of Minnesota Press, pp. 3-25 Taylor, C., (2002), ‘Foundationalism and the Inner/Outer Distinction’, in Reading McDowell on Mind and World, Smith (ed.) pp. 106-120 Williams, M., (2001), Problems of Knowledge, Oxford: Oxford University Press Wright, C., (2002), ‘Human Nature’, reprinted in Reading McDowell on Mind and World, Smith (ed.), pp. 140-160 Wright, C., (2002), ‘Postscript’ in Reading McDowell on Mind and World, Smith (ed.), pp. 160-173 - 40 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © Emily Thomas; all rights reserved. When choosing the best possible world, can God always go one better? Emily Thomas University of Birmingham aet400@bham.ac.uk It is sometimes asserted that when God took the decision to create the universe he was confronted by an infinity of possible worlds to chose from, and from these he chose the ‘best possible world’ to actualise. This paper will evaluate the claim that such a world is impossible, as no matter what world you pick there will always be one that is better. This claim can either be interpreted to mean a finite cycle of worlds or an infinite sequence, and I will seek to show that while the former interpretation fails the latter succeeds. This means that there is no ‘best possible world’, and no matter what world God picks he can always go one better. First we will examine the nature of the ‘best possible world’ and the goods used in determining it. Next we will critically evaluate the two possible interpretations of the above claim, which implies either a finite cycle of worlds or an infinite series. I will argue against Brown and Nagasawa’s formulation of the first interpretation, as I think it fails to show that no matter what world you pick there is always a better. However I will then argue for Swinburne’s formulation of the second interpretation, as despite replies from Grover and Strickland I think that this interpretation succeeds in showing there is no ‘best possible world’. We begin, then, by examining the nature of the ‘best possible world’ and the goods used in determining it. What exactly are possible worlds? Jubien describes them as complete descriptions of the ways our world might have been (Jubien, 1998, 134-5). This might mean that they are merely completely consistent sets of propositions or that they are fullyfledged physical entities. Fortunately, there is neither room nor need for - 41 - BJUP - 2(1) - April 2007 debate here. Either way, the ‘best possible world’ would surely contain the most goods and be the most perfect. Note that it is an assumption of this paper, and of the claim we are evaluating, that different goods and possible worlds are morally commensurable. Although one could perhaps argue that different goods and worlds are too distinct to be commensurable, this would still mean there is no ‘best possible world’ – if you cannot compare worlds you cannot pick the best one. As for the goods involved in world-making, they might reasonably include happiness, virtuosity, freedom, variety of natural phenomena and simplicity. Three issues can be identified regarding the nature of these goods. Firstly, do goods have an intrinsic maximum as to how ‘good’ they can be? Secondly, would these goods coexist in harmony or tension? Thirdly, what sort of relation is implied when we say x is ‘better than’ y? These issues will become crucially important below. Let us turn then to the first interpretation of the claim that for whatever possible world you pick there is a better one. This maintains that there is a finite cycle of worlds. Brown and Nagasawa express this interpretation as follows. If we let there be a finite set of possible worlds (W) and an index of worlds (w1, w2 etc.), and if we say that x>y stands for ‘x is better than y’, we can generate the following sequence: w1>w2, w2>w3 … wn-1>wn, wn>w1 The sequence completes a full ‘cycle’ from w1 to wn and back again (Brown & Nagasawa, 2005, 315). The set could contain just three worlds. But isn’t it counter-intuitive to claim that world x is better than world y and world y is better than world z but world z is better than world x? Surely world x would also be better than world z. The viability of the sequence depends on answering the third issue concerning goods: what sort of relation is ‘better than’? It could either be transitive (e.g. for relations R including height or greatness, Rxy and Ryz implies Rxz) or non-transitive (e.g. for relations R including love or hate, Rxy and Ryz implies nothing about Rxz). - 42 - BJUP - 2(1) - April 2007 A finite cycle of worlds could only exist if the relation of ‘better than’ was non-transitive, but can we imagine that? Without endorsing it, Nagasawa suggests the following scenario. Imagine using different criteria to judge different pairs of worlds. For example a world is better than another if it has a higher number of happy inhabitants or, if the world has less then 20 inhabitants, a higher proportion of happy ones. If world x had 300/1000 happy inhabitants, world y had 99/100 happy inhabitants and world z had 9/10 happy inhabitants then the following series would be created: wx>wy, wy>wz, wz>wx. However, I intend to argue that such a scenario is impossible. I think that if we accept that different worlds are morally commensurable then, by definition, they should be directly comparable and we shouldn’t have to use different criteria to judge different pairs of worlds. If used independently either of the criteria would be transitive, and the fact that the scenario needs to use them in conjunction shows that these worlds aren’t directly comparable. We said above that this scenario couldn’t work if no worlds were commensurable (as no world would be better than another) so this scenario would only work if different worlds were judged by different standards and they were not all morally commensurable to the same standard. As this paper is assuming that all worlds are commensurable, I conclude that this first interpretation of our claim fails to show that the ‘best possible world’ is impossible. We will now consider the second interpretation of the claim that for whatever possible world you pick there is a better one. This maintains that there is an infinite series of possible worlds possessing no maximum degree of perfection. Brown and Nagasawa formulate this idea as before, except that this time let the set of possible worlds (W) be infinite so that the following sequence is generated: w2>w1, w3>w2… wn>wn-1 An infinite chain of worlds, each slightly better than the last, would thus be created (Brown & Nagasawa, 2005, 315). How might this sequence come about? The answer relates to the first issue concerning - 43 - BJUP - 2(1) - April 2007 goods: do they have intrinsic maximums? Swinburne claims that certainly ‘happiness’ doesn’t, and adds that the goodness of the ‘best possible world’ must be judged at least partly on the happiness of its inhabitants (Swinburne, 1991, 114). Swinburne imagines a world whose inhabitants have an infinite capacity for happiness, and claims we could always improve on this world by making its populous slightly happier (Swinburne, 1991, 114). There are two obvious objections to this: firstly one could argue that goods do have intrinsic maximums, and secondly, as Grover points out, since we do not have an infinite capacity for happiness (or indeed any other good) perhaps it is impossible that any creature could (Grover, 1993, 222). Swinburne neatly circumvents both of these objections for even if we accept they are both correct, there is no reason why possible worlds would be limited as to how large they are or how many inhabitants they contain (Swinburne, 1991, 114). We can then always improve upon any given world by adding one more happy, virtuous, free inhabitant. I think that Swinburne’s scenario is solid, but we will nonetheless consider a novel reply to it from Strickland. Strickland’s reply utilises the third issue concerning goods: how might they co-exist? He claims that at least two goods have intrinsic maximums (simplicity and the variety of natural phenomena) and that when these goods are combined with others that arguably don’t have intrinsic maximums (virtue, happiness and freedom) they exist in tension with each other. For example, a truly virtuous creature might not be happy if virtuosity requires self-sacrifice (Strickland, 2005, 39). If Strickland is right then the ‘best possible world’ would be the one possessing the optimal ‘trade-off’ between at least two types of conflicting good, so no good could exist in infinite amounts (Strickland, 2005, 46). However, Strickland’s defence of the ‘best possible world’ against Swinburne’s scenario is problematic on two grounds. Firstly, as Grover observes, his defence may prove unwelcome to theists because Leibniz never intended the ‘best possible world’ to be a trade-off between various competing perfections. If goods were necessarily in conflict then sometimes even God would have to choose between competing ends - 44 - BJUP - 2(1) - April 2007 and that would be untenable (Grover, 1998, 662). While this first problem is really little more than an observation, the second is more serious – Strickland’s defence does not work. Even if Strickland is correct and co-existing goods are in tension, I don’t see how that limits the amount of space or inhabitants a world can possess. I don’t think any of the goods would affect the size of a world, and while combining certain goods (for example simplicity and maximum variety) might limit the types of creatures that exist, I don’t see how it would limit the tokens of each type. Therefore I think that Strickland’s defence of the possibility of the ‘best possible world’ fails. It seems, then, that Swinburne is right to say you can always make a better world by adding just one more happy creature. Having critically evaluated both interpretations of the claim that no matter what possible world you pick there is a better one, I have shown that while the first interpretation fails to show there can be no ‘best possible world’ the second succeeds. As Wielenberg observes, this means that the term ‘best possible world’ is similar to ‘the highest integer’ – neither idea is meaningless but it is necessarily uninstantiated (Wielenberg, 2004, 57). God may have the choice of any possible world, but no matter which world is picked there will always be one that is better. - 45 - BJUP - 2(1) - April 2007 Bibliography Brown, Campbell, and Nagasawa, Yujin, (2005), ‘The Best of All Possible Worlds’, Synthese, Vol. 143, pp. 309-320 Grover, Stephen, (1993), ‘Satisfied Pigs and Dissatisfied Philosophers: Schlesinger on the Problem of Evil’, Philosophical Investigations, Vol. 16, pp. 212-230 Grover, Stephen, (1998) ‘Incommensurability and Possible Worlds’, The Monist, Vol. 81, pp.648-668 Jubien, Michael, (1997), Contemporary Metaphysics, UK: Blackwell Leibniz, Gottfried, (1890), Philosophical Works of Leibniz, trans. Ladd, George, New Haven: Tuttle, Morehouse and Taylor Strickland, Lloyd, (2005) ‘Determining the Best of All Possible Worlds’, The Journal of Value Enquiry, Vol. 39, pp. 37-47 Swinburne, Richard, (1991) The Existence of God, Great Britain: Oxford University Press Wielenberg, Erik, (2004) ‘A Morally Unsurpassable God Must Create the Best’, Religious Studies, Vol. 40, pp. 43-62 - 46 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © Tomas Bates; all rights reserved. An argument against the principle of alternative possibilities Tomas Bates University of Sheffield pia04tab@shef.ac.uk Introduction The principle of alternate possibilities (PAP) states that: A person is morally responsible for performing action x only if he could have done otherwise than perform action x.1 PAP rests on the idea that, if we are to be morally responsible for the actions we perform, we must have ‘forking paths’2 available to us. For example, if I am to be morally responsible for stealing Jim’s bike at t1, it must be possible that I ‘could have done otherwise’ than steal Jim’s bike at t1. However, in this paper I will support David Hunt’s claim that one can construct a ‘blockage case’3 counter-example to PAP which eliminates all alternative possibilities. In such a case it seems that the agent is morally responsible for the action he performs, despite the fact that he could not have done otherwise. I shall admit that ‘blockage cases’ are not immune to objection. Specifically, the Libertarian could object that if all neural pathways except one are blocked off from an agent, there is no room for indeterminism in the actual sequence. However, I will try to respond to this objection by claiming that whilst a ‘blockage case’ may leave a certain outcome determinate, the blockages themselves need not determine that the outcome will occur. 1 Frankfurt (1969), reprinted in Watson (ed.) (2003), p. 167. Fischer (2002), reprinted in Watson (ed.) (2003), p. 196. 3 Hunt credits Fischer with giving his counter-examples this name. Hunt (2005), p. 131 (footnote 9). 2 - 47 - BJUP - 2(1) - April 2007 Frankfurt’s attack on the principle of alternate possibilities In his article ‘Alternate possibilities and moral responsibility’, Harry Frankfurt attempts to disprove PAP. Frankfurt imagines a situation which we shall scenario (1). In scenario (1), Black wants Jones to kill Smith but he doesn’t want to intervene and force Jones to kill Smith unless it is absolutely necessary to do so.4 For this reason, Black waits on the sidelines until it becomes clear whether or not Jones is going to choose to kill Smith. Imagine that Black becomes aware of what Jones will decide to do at t2 because at t1 Jones shows some involuntary ‘prior sign’5 of what he is about to choose to do. For example, if Jones is blushing at t1, he will choose to kill Smith at t2, but if Jones is not blushing at t1, he will choose not to kill Smith at t2.6 Thus, if Jones is blushing at t1, Black will not intervene, but if Jones is not blushing at t1, Black will intervene and alter Jones’s brain states in such a way as to ensure that Jones chooses to kill Smith at t2. Finally, in scenario (1), imagine that Jones does have a strong dislike for Smith prior to being kidnapped by Black. This strong dislike causes Jones to blush at t1 (indicating that he is about to choose to kill Smith at t2), and as a result Black does not intervene and at t2. Jones chooses to kill Smith and at t3 he does so. I will now explain how scenario (1) is supposed to act as a counterexample to PAP. Specifically, Frankfurt argues that in scenario (1) Jones is morally responsible for choosing to kill Smith, even though he could not have done otherwise than choose to kill Smith at t2. It is true that if Jones had not blushed at t1, Black would have intervened and brought it about that Jones choose to kill Smith at t2. But Frankfurt argues that this lack of alternate possibilities does not affect Jones’s moral responsibility in the actual sequence. Intuitively, Jones seems morally responsible for killing Smith because he chooses to kill Smith and really wants to kill Smith. It appears irrelevant that Jones lacks alternate possibilities in scenario (1). Despite his lack of alternate possibilities, Jones still chooses to carry out the action that he really wants to perform. Significantly, in scenario (1), ‘the circumstances that made it 4 Frankfurt (1969), reprinted in Watson (ed.) (2003), p. 172. Fischer (2002), reprinted in Watson (ed.) (2003), p. 192. 6 Widerker (1995), reprinted in Watson (ed.) (2003), p. 179. 5 - 48 - BJUP - 2(1) - April 2007 impossible for him [Jones] to do otherwise could have been subtracted from the situation without affecting what happened or why it happened in any way.’7 One could remove Black from scenario (1) and Jones will still choose to kill Smith for exactly the same reasons. The problematic dilemma for Frankfurt’s counter-example However, I shall now argue that scenario (1) can be forcibly objected to by giving Frankfurt a dilemma.8 Firstly, if it is to be accepted by libertarians, Frankfurt’s scenario must not assume universal causal determinism.9 This is because Libertarians believe that morally responsible actions must involve indeterminism; for example, the Libertarian David Widerker states that ‘an agent’s decision (or choice) is free in the sense of freedom required for moral responsibility only if it is not causally determined.’10 However, in scenario (1) as I have stated it, there seems to be a deterministic causal relationship between the occurrence of a particular ‘prior sign’ at t1 and a certain choice at t2. Specifically, if Jones blushes at t1, he will choose to kill Smith at t2, but if Jones does not blush at t1, he will choose to not kill Smith at t2. Thus, in scenario (1), it appears that the occurrence of the ‘prior sign’ is causally sufficient to bring about the choice that Jones makes at t2. But, if this is the case, Libertarians will reject that Jones is morally responsible for the choice he makes at t2 because they reject from the beginning that an agent can be morally responsible for a determined action. Therefore, if Frankfurt presupposes universal causal determinism in this way, he is unfairly begging the question against the Libertarian case, and thus scenario (1) will fail to offer a convincing counter-example to PAP. However, if Frankfurt’s example is not presupposing determinism in this way, then there seems to be room for indeterministic alternate 7 Frankfurt (1969), reprinted in Watson (ed.) (2003), p. 174. Widerker offers such a dilemma. Widerker (1995), reprinted in Watson (ed.) (2003), p. 180. 9 For the purposes of this essay I shall take ‘universal causal determinism’ to mean the thesis that all events that occur in the universe are deterministically caused by some other event. 10 Widerker (1995), reprinted in Watson (ed.) (2003), p. 177. 8 - 49 - BJUP - 2(1) - April 2007 possibilities in scenario (1). For example, if there is not a sufficient causal relation between the occurrence of the ‘prior sign’ at t1 and Jones’s choice at t2, then Jones seems free to either choose to kill Smith at t2 or to choose not to kill Smith at t2. Of course, Black might realise that Jones is in the process of choosing to kill Smith at some point between t1 and t2. However, as Fischer points out, whilst ‘Black can intervene to prevent the completion of the choice’, this still gives Jones the possibility of ‘beginning to make the choice’11 not to kill Smith. Thus, if one assumes indeterminism in scenario (1), Jones appears to have a morally relevant alternate possibility available to him at t2; he can begin to make the choice not to kill Smith. Hunt’s ‘blockage cases’ Given these problems for counter-examples to PAP which involve some ‘prior sign’, I will now explain and argue in support of a counterexample to PAP which does not involve a ‘prior sign’. In his article, ‘Moral responsibility and unavoidable action’, David Hunt attempts to construct a counter-example to PAP which rules out all alternative possibilities. He does this by coming up with a scenario similar in kind to John Locke’s ‘locked room’ story.12 In Locke’s example, a man, say Bob, is locked in a room with Jim (who is someone that Bob really wants to speak with). Bob couldn’t leave the room if he tried but instead he happily chats to Jim without even trying to leave the room. In such a situation, Bob lacks the alternate possibility to leave the room, but this fact seems irrelevant. Despite his lack of alternate possibilities, Bob is still perfectly capable of doing what he wants to do (i.e. speaking to Jim). Importantly, such a scenario does not depend on a counterfactual intervener. As Hunt states, ‘the door is actually locked, it doesn’t lock only when someone approaches the door and tries to leave.’13 However, whilst the possibility of leaving the room is blocked from Bob in Locke’s scenario, there are several other relevant alternate possibilities open to him. For example, Hunt points out that the agent 11 Fischer (2002), reprinted in Watson (ed.) (2003), p. 194. Hunt (2000), p. 217. 13 Ibid., p. 217. 12 - 50 - BJUP - 2(1) - April 2007 could ‘try’ to leave the room and he contends that ‘this alternative is surely enough to under-grid his moral responsibility.’14 For this reason, Hunt devises a counter-example to PAP which brings ‘the locked door – the blockage – into the head.’15 Specifically, he imagines a scenario in which ‘neural pathways rather than doorways’16 are blocked off from an agent. Firstly, imagine a scenario in which an agent is intuitively morally responsible for his actions and has alternate possibilities open to him. For example, consider scenario (2). In this scenario, Jones strongly dislikes Smith and chooses to kill him at t2. Imagine that there is no intervention from an outside force or agent in this situation and that Jones chooses to kill Smith after indeterministically utilising a single neural pathway in his brain. In such a situation, it appears that Jones is morally responsible for his choice to kill Smith at t2. Jones consciously and voluntarily chooses to kill Smith, and does so based on his strong dislike towards Smith. Furthermore, Jones has alternate possibilities open to him at t2. He can choose to kill Smith or he can choose not to kill Smith. This means that Jones can utilise one neural pathway or another in scenario (2) (imagine that the choice to kill Smith will utilise one neural pathway and the choice not to kill Smith will utilise another neural pathway). However, now imagine scenario (3), a Hunt-style ‘blockage case’. In this scenario, Jones’s choice to kill Smith at t2 has ‘precisely the same causal history’17 as in scenario (2). Thus Jones utilises exactly the same neural pathway that he utilises in scenario (2) in order to choose to kill Smith. But consider that, in scenario (3), Black kidnaps Jones at t1 and blocks off every neural pathway in Jones’s brain apart from the one which is utilised in scenario (2). Imagine that by a ‘fantastic coincidence’18 the neural pathways that Black blocks off at t1 are exactly the same neural pathways that remain dormant throughout scenario (2). For example, imagine that Black is just an incompetent neurosurgeon who likes to practice his expertise by randomly blocking 14 Ibid., p. 217. Fischer (1999) p.114 16 Hunt (2000) p.218 17 Pereboom (2000) p.126 18 Hunt (2000) p.218 15 - 51 - BJUP - 2(1) - April 2007 off neural pathways in a person’s brain. Thus, whilst he has the technical prowess to block off neural pathways in the brain, Black has no idea what effect blocking a particular pathway in a person’s brain will have on that person’s desires and actions. This means that Black has no idea that by blocking off the particular neural pathways he is blocking off, he is making it unavoidable that Jones will choose to kill Smith at t2. Thus, in scenario (3), all alternate possibilities seem blocked off from Jones. It is ‘unavoidable’ that he will choose to kill Smith at t2 because he cannot utilise any other neural pathway than the one he does actually use. Whilst Jones lacks alternate possibilities in scenario (3), I will now argue that he is still morally responsible for his choice to kill Smith at t2. This point can be illustrated by considering the difference between scenario (2) and scenario (3). It seems that the only difference between the two scenarios is that whilst the neural pathways that are left dormant in scenario (2) remain open, these same dormant neural pathways are blocked off from the agent in scenario (3). But the dormant pathways being blocked off in scenario (3) seems to make no difference to Jones’s moral responsibility. He chooses to kill Smith at t2 for exactly the same reasons and using exactly the same neural pathway as he utilises in scenario (2). Thus, it appears that one could deduct Black and the blockages from scenario (3) without changing the situation in anyway. If one accepts my intuition that Jones is morally responsible in scenario (2), this means that they have no less reason to believe that he is morally responsible in scenario (3). Objection 1 to ‘blockage cases’: there are still alternate possibilities Despite the intuitive strength of ‘blockage cases’, they are not immune from Libertarian objection. I shall now finish the essay by addressing two such possible objections to Hunt-style counter-examples to PAP. Both these objections argue along similar lines to the dilemma levelled against Frankfurt’s scenario. However, I will respond by contending that if one clarifies exactly what it is going on in ‘blockage cases’, such objections lose much of their appeal. - 52 - BJUP - 2(1) - April 2007 The first Libertarian objection claims that a blockage case cannot really rule out all alternate possibilities. Specifically, the Libertarian might claim that even if all the dormant neural pathways are blocked off in scenario (3), Jones might still have the ability to ‘bump up against’19 the blockage. It may be conceded that this is true – Jones cannot actualise the neural pathway which will lead him to choose not to kill Smith at t2. However, Jones can still try to actualise this neural pathway. Such an objection can be illustrated by looking again at Locke’s ‘locked door’ example. As Hunt himself notices, such a counter-example cannot completely rule out alternate possibilities because Jim can still try to break down the locked door. If Jim wanted to do this and tried to do so, surely we would not withhold moral responsibility from him. Similarly, if a ‘blockage case’ is only supposed to ‘differ only in degree, not in kind’20 from the ‘locked room’ case, surely in scenario (3) Jones has the capacity to try and ‘bump up against’ the blockage, just as Jim has the ability to try the locked door in Locke’s scenario. However, as I said, I believe that such an objection can be overcome by clarifying exactly what is going on in a ‘blockage case’. Hunt’s scenario doesn’t only block off neural pathways but also blocks off any ability an agent has to reach the blockage. Fischer explains this idea by introducing the notion of a ‘bridge’21 between the actual neural sequence and the blockage. In order to ‘bump up against’ the blockage, the agent must have the ability to try and cross this bridge. However, in a full ‘blockage case’ even the ‘intermediate neural events’ that form the bridge are blocked off from the agent. The libertarian may respond to such an argument by claiming ‘but at least the agent can try to get on the bridge, even if he is incapable of crossing it’. However, such an objection can only push the argument one step back. This is because, for any such an objection to a ‘blockage case’, Hunt can always contend that even this alternate possibility is blocked off from the agent. Such a reply doesn’t seem to unfairly presuppose the truth of Hunt’s case either. It seems plausible to suppose that, in order to reach any particular neural pathway, there will be some ‘intermediate set of neural 19 Fischer (1999) p.119 Hunt (2000) p.221 21 Fischer (1999) p.119 20 - 53 - BJUP - 2(1) - April 2007 events.’22 But if even these intermediate events are blocked off from the agent, what alternate possibilities can plausibly exist? Objection 2 to ‘blockage cases’: such cases presuppose determinism However, Hunt must still deal with the second fork of the dilemma. Specifically, he must convincingly argue that ‘blockage cases’ do not unfairly beg the question against Libertarianism by presupposing universal causal determinism. For example, a libertarian might object that a ‘blockage case’ couldn’t possibly involve indeterminism in the actual sequence and therefore must be deterministic. Pereboom explains this objection by giving two scenarios.23 In situation (A), an atom is falling downwards. Imagine that the atom’s course is indeterministic – there is no object blocking its path and at any time it could randomly swerve left or right. Finally, imagine that the atom does not swerve between t1 and t2. Now imagine situation (B). In this situation, exactly the same atom is travelling through space. However, in this scenario, the atom is travelling downwards through a vertical tube and the tube’s ‘interior is exactly wide enough to accommodate the atom.’24 Finally, imagine that the atom would not have swerved between t1 and t2, even if it were not in the tube. The ‘blockage case’ theorist will claim that the two situations are basically identical. One could remove the atom from the tube in situation (B) and it would make no difference. However, the Libertarian may react to this claim by contending that there is actually a crucial difference between situation (A) and situation (B). Whilst the atom is situation (A) can indeterministically swerve left or right between t1 and t2, the atom in situation (B) cannot move either left or right between t1 and t2. Thus, even if the atom in situation (B) was going to swerve either left or right between t1 and t2, it could not. Pereboom sums-up this concern when he writes ‘since the tube prevents any alternative motion, it would seem that it precludes any indeterminism in the atom’s causal history from t1 to t2.’25 22 Ibid p.119 Pereboom (2000) p.127 24 Ibid p.127 25 Ibid p.127 23 - 54 - BJUP - 2(1) - April 2007 However, despite the strength of this objection, I shall now try and respond to it. I think that one can make an important distinction between a particular outcome being ‘unavoidable’ or determinate and that same outcome being determined. Specifically, when one considers Pereboom’s example, it seems possible for the ‘blocker’ theorist to contend that whilst the atom’s path may be determinate, it is not necessarily the case that the atom’s path is determined by the vertical tube. When we talk of one thing determining another thing, it seems that we consider the first thing to play some important causal role in what happens to the second thing. However, if this is the case, how can we make sense of the vertical tube determining the atom’s path in situation (B)? As I have already pointed out, one could take away the supposedly ‘determining’ factor (e.g. the vertical tube) and the atom would act in exactly the same way. This suggests that the tube cannot really play an important causal role in what happens to the atom in situation (B) and cannot really be said to determine the atom’s path either. My response can be clarified by looking again at scenario (3). Is Jones really determined to choose to kill Smith at t2 by the blocked neural pathways in his brain? Granted, it is determinate that Jones will choose to kill Smith at t2 – given that there is only one neural pathway available to Jones it is ‘unavoidable’ that he will make this choice. However, just because this outcome is determinate, it doesn’t necessarily follow that the blockages themselves determine the outcome. Just as in situation (B), one could subtract the supposedly determining factor from scenario (3) without altering the outcome in any way. Remember Jones will choose to kill Smith at t2 even if the other neural pathways in his brain were unblocked. Thus, whilst Jones’s choice to kill Smith in scenario (3) seems determinate, it seems odd to say that the neural blockages determine Jones’s choice. As I said earlier, a determining factor usually plays some causal role in bringing about a particular outcome. But it seems odd to say that the blockages play a causal role in bringing about Jones’s choice to kill Smith, when they could be subtracted from the scenario without affecting Jones’s choice in any way. And it seems that if the neural blockages themselves don’t determine Jones’s choice in scenario (3), there is room for indeterminism in the actual sequence. I therefore believe that it could - 55 - BJUP - 2(1) - April 2007 plausibly be the case that Jones’s choice to kill Smith at t2 is determinate, even though his choice is undetermined by prior events. Finally, the Libertarian seems to have some reason to accept my distinction between a choice being ‘determinate’ and something causally ‘determining’ that the choice will occur. Specifically, whilst Libertarians will reject the idea that a morally responsible choice or action could be causally determined, they at least have some reason to accept my idea that a morally responsible choice or action be determinate. For example, Libertarians often reject a traditional argument against their case which contends that an undetermined action is simply ‘a matter of chance or luck.’26 Such an argument states that a person cannot possibly be in control of something which is simply ‘a matter of chance or luck’ and therefore a person cannot be morally responsible for an undetermined action. However, Libertarians respond to such an objection by claiming that an undetermined act is not necessarily only ‘a matter of chance or luck’ and therefore an agent can be morally responsible for an undetermined action. But to me this suggests that there is a sense in which an undetermined action could be determinate. Specifically, if the Libertarian is to justifiably claim that I am morally responsible for my choice to steal Jim’s bike (even though my choice is undetermined), it seems that I must be in control of my choice to steal the bike. But if I am to be in control of my choice to steal the bike, it seems that my choice should, at the very least, follow from who I am as a person (my particular desires etc.). But if my choice is to follow from who I am as a person, then there is at least some sense in which my choice to steal the bike must be determinate. I have a specific character and thus there will be a limited (i.e. determinate) number of choices that I can plausibly make, given that I have the particular character that I have. This suggests that the Libertarian does at least has some reason to accept that an undetermined choice or action could nevertheless be determinate. 26 Kane asserts how such an objection rests on the ‘Luck Principle’- Kane (1999) reprinted in Watson ed. (2003), p. 299 - 56 - BJUP - 2(1) - April 2007 Of course, there is clearly a substantial difference between a choice being ‘determinate’ in the sense that there are a limited range of choices that I could make (given that I have the particular character that I have), and an action being ‘determinate’ in the sense that there is only one possible choice that I could make in a situation. However, for the purposes of this paper, I feel it is enough to conclude that the Libertarian at least has some reason to accept my distinction between a choice or action being ‘determinate’ and something ‘determining’ that the choice or action will occur. Conclusion In this paper I have argued that David Hunt’s ‘blockage cases’ provide the most convincing counter-example to the principle of alternate possibilities. I accepted that Frankfurt’s alleged counter-example to PAP fails because the Libertarian can object to such a scenario by giving a dilemma. I then argued in support of Hunt’s ‘blockage cases’ which remove the ‘prior sign’ from the PAP counter-example. However, I accepted that a ‘blockage case’ may seem hard to distinguish from a purely deterministic sequence. But I responded to this possible objection by claiming that whilst a ‘blockage case’ may leave a particular outcome determinate, the blockages themselves need not determine that this particular outcome arises. For this reason there still appears to be room for indeterministic neural activity in ‘blockage cases’. - 57 - BJUP - 2(1) - April 2007 Bibliography Fischer, John Martin, ‘Frankfurt-style Compatibilism’, in Watson, Gary (ed.) Free Will (2nd edn.) (Oxford: Oxford University Press, 2003) Fischer, John Martin, ‘Recent work on Moral Responsibility’, Ethics, 110: 93-139 (1999) Frankfurt, Harry G, ‘Alternate Possibilities and Moral Responsibility’, in Watson Gary (ed.) Free Will (2nd edn.) (Oxford: Oxford University Press, 2003) Hunt, David P., ‘Moral Responsibility and Buffered alternatives’, Midwest Studies in Philosophy, XXIX (2005) Hunt, David P., ‘Moral responsibility and Unavoidable action’, Philosophical Studies, 97: 195-227 (2000) Kane, Robert, ‘Responsibility, Luck and Chance: Reflections on Free Will and Indeterminism’, in Watson, Gary (ed.) Free Will (2nd edn.) (Oxford: Oxford University Press, 2003) Pereboom, Derk, ‘Alternate possibilities and Causal histories’, Philosophical Perspectives, 14: Action and Freedom (2000) Widerker, David, ‘Libertarianism and Frankfurt’s attack on the principle of alternate possibilities’, in Watson, Gary (ed.) Free Will (2nd edn.) (Oxford: Oxford University Press, 2003) - 58 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © Gregor Ulm; all rights reserved. Limitations of semi-compatibilism: a defence of the principle of non-responsibility Gregor Ulm King’s College, University of London gregor.ulm@gmail.com Introduction A proponent of semi-compatibilism argues that moral responsibility, free will and causal determinism are compatible. Formal difficulties notwithstanding, we should scrutinize the practical applicability of this if we intend to judge the merits of it as a philosophical position. Therefore, I will expound Harry Frankfurt’s attack on the principle of alternative possibilities and its utilization by John Martin Fischer and Mark Ravizza, the most eminent semi-compatibilists. Afterwards, I will raise an important objection to their reasoning. This will lay the groundwork for my defence of Peter van Inwagen’s famous argument for the principle of the transfer of non-responsibility, which involves a special emphasis on the difficulties underlying the implications of causal overdetermination. Frankfurt’s attack on the principle of alternative possibilities Where a compatibilist argues that free will and causal determinism are not mutually exclusive, which is by itself anything but self-evident, a semi-compatibilist takes a rather daring additional step. She accepts Harry Frankfurt’s argument against the principle of alternative possibilities (PAP) (1969, 1971), she takes moral responsibility to be compatible with determinism, and she furthermore assumes that determinism may exclude free will (e.g. Clarke: 2003: 10). Frankfurt aimed to undermine PAP, a principle once claimed to be an a priori truth, at least by some philosophers (Frankfurt, 1969: 829). It states that an agent can only be held morally responsible for an action if - 59 - BJUP - 2(1) - April 2007 she could have done otherwise. Harry Frankfurt famously showed that this line of reasoning could be challenged, so that an agent can be held responsible for an action even if he could not have done otherwise. Let us consider a typical example (cf. Frankfurt, 1969: 829; Fischer and Ravizza, 1998: 29).1 David holds a loaded slingshot in his hand, intending to kill Goliath, and does so. However, unknown to him, an evil force exerts control over his brain. This mysterious force is able to monitor David’s mental state and to manipulate his behaviour if need be. Would it have been the case that David was to abstain from his original plan to murder Goliath, the evil force would have intervened, thus leading to the very same and well-known result – David would have fired his slingshot and killed Goliath. However, as it turns out, such an intervention is unnecessary because David does not deviate from his original plan – he takes aim, shoots and kills his enemy. Even though David voluntarily killed Goliath in the actual case it is obvious that he could not have done otherwise as he was bereaved of any alternative possibility – he was bound to act in accordance with his original plan because it just was not possible for him to resist the influence of the counterfactual intervener. As will be seen, the plausibility of semi-compatibilism hinges on the implications of this controversial thought experiment.2 The question we have to deal with is whether such a case would indeed, as Frankfurt claims, justify the statement that an agent, in our example David, can be held responsible for his actions even though he could not have done otherwise. At first sight, an analysis of the situation seems to be remarkably unspectacular. We are able to discern two outcomes, a factual and a counterfactual one: (1) In accordance with his original plan, David kills Goliath. 1 In the original example, Black wants Jones ‘to perform a certain action’ (Frankfurt, 1969: 829). As the adaptation of Fischer and Ravizza is filled with rather cumbersome technical details, I opted for a less ostentatious example. 2 Frankfurt’s thought experiment is controversial to this day, see e.g. Widerker and McKenna (2003). - 60 - BJUP - 2(1) - April 2007 (2) David is struck by a sudden pang of conscience. However, at the very same moment he is about to loosen his firm grip on his slingshot, the evil force intervenes and David proceeds to fire his slingshot, leading to Goliath’s premature death. As it seems, David has no real choice. In every case, Goliath is moribund as we take it as given that his opponent will not miss. In (1) it is indubitable that David assumes the role of an agent. On the other hand, in (2) the case is quite different. For an observer the scene of the action appears to be indistinguishable from (1), but since the act of killing is due to the intervention of the evil force it seems appropriate to charge the intervener with responsibility for the murder and not David as the intervener represents the conditio sine qua non without which David would not have fired his slingshot. Consequently, David assumes the role of a mere assistant to the evil force.3 A possible objection As a solution to the problem, I propose to split the role of the agent in order to better handle the intervener’s manipulation. Therefore, in the actual case (1) we are confronted with David1, the murderer of Goliath, whereas in the counterfactual case (2) David2 acts as the assistant of the evil genius. Thus, I suggest that we should restate the two cases in a slightly different way: (3) In accordance with his original intention, David1 kills Goliath. (4) David2 is struck by a sudden pang of conscience. However, at the very same moment he is about to loosen his firm grip on his slingshot, the evil force 3 Peter van Inwagen provides a similar objection, albeit in a different context when he writes that the agent then is degraded to the status of a mere instrument (1983: 12934). - 61 - BJUP - 2(1) - April 2007 intervenes and David2 proceeds to fire his slingshot, leading to Goliath’s premature death. It should be obvious that David1 could not have done otherwise. Considering that (3) does not allow for an alternative possibility nor for free will, as there is only one path open to him, we should therefore refrain from holding him morally responsible for his action. This is due to the fact that we take the wind out of Harry Frankfurt’s objection as the manipulation that takes place in (4) is of no relevance for David1’s action in (3). Hence, the purported counterfactual case is irrelevant for our judgement of the action of David1 as we actually have to deal with two separate agents. Therefore, we have elegantly sidestepped any possible objections a proponent of Frankfurt’s point of view could raise.4 In what follows this result will enable us to successfully defend Peter van Inwagen’s principle of the transfer of non-responsibility (TRANSFER NR) against the objections put forward by John Martin Fischer and Mark Ravizza.5 Transfer of non-responsibility Peter van Inwagen (1983) opposes Frankfurt as he contends that we cannot justly hold an agent responsible for something he has done if he 4 This is also the reason why I consider myself justified in not having to dwell on Fischer and Ravizza’s concept of ‘guidance control’ (1998: 30). They understand this concept to allow for attributing moral responsibility to an agent even if causal determinism obtains. This is based on their interpretation of Frankfurt. As my criticism is more fundamental, I am justified in leaving this point largely uncommented. Nonetheless, it might be worthwhile to point out that Fischer and Ravizza deduce moral responsibility from Frankfurt-type examples for the sole reason that the agent is understood to possess the aforementioned feature of ‘guidance control’ (1998: 34-5). By splitting the role of the agent I avoid their objection. Of course, one might object to my way of reasoning. On the other hand, I contend that my argument is more convincing than Fischer and Ravizza’s as they merely postulate the existence of ‘guidance control’ without giving sufficient reasons for accepting that position. The difficulties of such a manoeuvre should soon become evident. As will be seen, we can add as many counterfactual interveners as we want to without finding a way to avoid my fundamental objections. 5 I borrow the term ‘TRANSFER NR’ from Fischer and Ravizza (1998). - 62 - BJUP - 2(1) - April 2007 lacked control over the causes that led to this specific outcome.6 His argument is as follows: If p obtains, and no one is even partly morally responsible for p; and if p obtains, then q obtains, and no one is even partly morally responsible for the fact that if p obtains, then q obtains; then q obtains, and no one is even partly morally responsible for q (1983: 182-8; Fischer and Ravizza, 1998: 152).7 This abstract syllogism is then illustrated by means of a rather graphic example: no one is responsible for the fact that John is bitten by a cobra on his thirtieth birthday; and no one is responsible for the fact that if John is bitten by a cobra on his thirtieth birthday, then he dies on his thirtieth birthday; then no one is responsible for the fact that John dies on his thirtieth birthday (Fischer and Ravizza, 1998: 154; van Inwagen, 1983: 187). Fischer and Ravizza now try to refute the plausibility of the principle TRANSFER NR, which we just exposed, by modifying the scenario. They opt for complicating matters by adding a sharpshooter to the scene (1998: 166).8 The most important detail is that she has achieved such a high level of marksmanship that she is able to judge when to fire the rifle so that the bullet will lead to John’s death at the very same moment as the snake’s venom. Technical intricacies notwithstanding, they claim that this modification refutes Peter van Inwagen’s original example as he is only able to back the principle TRANSFER NR because it allows for just one path leading to John’s death. The added sharpshooter is supposed to open up a second path leading to the same outcome.9 6 This stance seems to be plausible enough. An opposing point of view, namely that lack of free will does not bereave us of responsibility, is brought forward by Derk Pereboom (2003). 7 The original example is to be found and explained in van Iwagen (1983: 182-8). I opted for the reformulation by Fischer and Ravizza as I will scrutinize one of their examples later on. 8 One might assume that it would have been more convincing if Fischer and Ravizza had tackled the validity of van Inwagen’s original syllogism, and not just one of his examples, because even if we assume that in a nearby possible world nobody dares to contradict their claims, then one could still hold against them that their findings do not adequately reflect ordinary cases like the one represented by the original and unmodified principle TRANSFER NR, and thus they only managed to refute a very special case (note: a special case which they themselves introduced!). 9 Again, the sharpshooter is understood to be responsible for his action as she exerted ‘guidance control’. Please see footnote 5 for my objection to this concept. - 63 - BJUP - 2(1) - April 2007 On the contrary, I contend that even though we now have to deal with a case of causal overdetermination. This is due to the simultaneous effects of the snake’s venom and the bullet. So the main problem remains unresolved. Instead of fully detailing my objection one more time, I refer the reader to my splitting of the agents introduced above (if we are tempted, we could even try to introduce a counterfactual intervener ensuring that the sharpshooter kills John). The result is that the sharpshooter cannot be responsible for his action if causal determinism obtains. The additional problem of causal overdetermination, however, remains to be dealt with. Problems of causal overdetermination In general, and opposing Fischer and Ravizza (1998: 165), I think it is quite implausible to assume that we can ever have two different paths leading to the same outcome. As it seems, their whole argument is based on an inadequate interpretation of Peter van Inwagen’s example. Obviously, there is nothing wrong with letting the predicate variable q stand for ‘John dies on his thirtieth birthday’, but once we alter the scenario, we should watch our steps carefully because once we decide to add poisonous venom and a sharpshooter to the scene we will need to distinguish the cases. The statement that John dies in both cases is certainly true, but proper reflection should lead to a reformulation like the following: (5) John dies on his thirtieth birthday because of the venom of a snake. (6) John dies on his thirtieth birthday as a consequence of the simultaneous effects of poisonous venom and a bullet entering his head. When looked upon superficially, both cases might appear to be identical. However, we are faced with a problem. Fischer and Ravizza take each of their cases, (5) as well as (6), to be, at the same time, instantiations of q. This approach is inconsistent, however, because as we have seen the events leading to John’s death are quite different. In - 64 - BJUP - 2(1) - April 2007 order to refute TRANSFER NR both cases would have to be shown to be identical, which is nothing short of impossible – it is clearly false to assume that (5) equals q and then assume that (6) equals q as well. As both cases are by no means equal, due to the fact that John’s death is the result of different causes, Fischer and Ravizza do not live up to their claim to produce a valid counterexample which functions against the principle TRANSFER NR. Each of the cases, (5) or (6), can only in isolation be accepted as a valid instantiation of q. Finally, I would like to remark that we can very well accept a certain amount of imprecision in Peter van Inwagen’s original example, as there is just one way for John to die. Nonetheless, one might still raise objections against my arguments. One could assume that the instantiation of q is exactly the same in two cases, while allowing for the possibility that the events leading to q differ. I would concede that such a scenario is quite possible – just imagine that the doctor who performed the autopsy on John’s dead body in the two possible worlds comes to the same conclusion. In both cases he scribbles down ‘cause of death: snake venom’ on the report. But an omniscient external observer would have experienced two different incidents: in the first one, John’s niece opens the snake’s cage, and in the second his nephew did this. As a result, we are still able to distinguish two cases of q, namely ‘John died from a snake’s venom (as a direct consequence of a negligence of his niece)’ and ‘John died from a snake’s venom (as a direct consequence of a negligence of his nephew)’. Thus, I contend that my objections to Fischer and Ravizza are quite solid. While I do not pretend that my reasoning is able to refute semicompatibilism altogether, I nonetheless argue that my objection poses a serious threat for any supporter of this position. It is my understanding that so-called Franfurt-type examples should be irrelevant for the debate regarding free will and determinism. Furthermore, I have raised an objection against cases of causal overdetermination. If I am right on these two counts, then one would indeed be justified in reducing semicompatibilism to compatibilism of free will and causal determinism. - 65 - BJUP - 2(1) - April 2007 Bibliography Clarke, Randolph, 2003, Libertarian Accounts of Free Will, Oxford: OUP Fischer, John Martin, and Ravizza, Mark, 1998, Responsibility and Control: a Theory of Moral Responsibility, Cambridge: CUP Frankfurt, Harry G., 1969, ‘Alternate Possibilities and Moral Responsibility’, Journal of Philosophy, 66, pp. 829-39 1971, ‘Freedom of the Will and the Concept of a Person’, Journal of Philosophy, 68, pp. 5-20 Pereboom, Derk, 2003, Living without Free Will, Cambridge: CUP van Inwagen, Peter, 1983, An Essay on Free Will, Oxford: OUP Widerker, David, and McKenna, Michael (eds.), 2003, Moral Responsibility and Alternative Possibilities: Essays on the Importance of Alternative Possibilities, Aldershot: Ashgate - 66 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © Benjamin Stephen Brown; all rights reserved. What is it for a particular to instantiate a property? Benjamin Stephen Brown University of Nottingham apyybsb@nottingham.ac.uk First of all, it would be helpful to spell out exactly what is required to answer this question. Let us take an example. A table, in particular this table that I am now observing, has the property of being wooden. It also has the property of being hard. Whether or not these two properties are related is something I shall discuss later. For now, we must wonder how it is that the table is either wooden or hard, or any of the other countless properties it instantiates. We must be very careful with language here. It is easy to say such things as the table ‘possesses’ the property of being wooden, but this implies that the table is something over and above its properties, which is an idea we may not necessarily wish to accept. To understand what it is for a particular to instantiate a property, we must first decide what we mean by ‘property’. Please note that this is a substantial task in itself and a very interesting discussion, but it is not the subject of this essay. Excuse me, then, for presenting rather brief descriptions of the most popular theories of properties. Moreover, I have avoided so far using the term ‘universal’ to denote a property. I do this because the first theory of properties I shall discuss does not positively acknowledge universals as real. Let us begin then, with Nominalism. Prima facie, Nominalism appears very sensible. We observe particulars as having certain properties, notice that some particulars’ properties resemble one another and thus form a concept, a ‘nom’, for that property. By this view universals are simply tools to aid our understanding of the world – little more than a linguistic filing system. - 67 - BJUP - 2(1) - April 2007 Trope theory too attempts to do away with universals. The story is that each particular has its own properties – particular properties. For example, each red snooker ball has its own trope of redness, but the properties are contained in the particulars and cannot be separated from them. However this gives us the problem that Nominalism tries to solve: why do we call all the instances of redness together, red? A possible response to this problem is to deny broad concepts altogether and simply refuse to categorise properties – each property has only a single instantiation and so there are an infinite number of properties. Shoemaker (1980) attempts to reduce properties to causal powers. For example, the property of being breakable consists of the causal power to be broken. Clearly I have chosen a deliberately simple example, but the idea is that any property, no matter how complex, can be expressed in terms of causal powers. A property is complex exactly because it has a complex set of causal powers. The names we give to bundles of causal powers are simply a matter of convenience. It could be argued then that since some properties have only a single causal power, like my example of ‘breakable’, we might just as well list the causal powers that the property represents. In fact, this is exactly what we are doing in the case of words like ‘breakable’. So we have particulars and we have properties. By what relation are properties bound to their particulars? This question, at its heart, concerns types and tokens. Hence the question concerning properties is a specific example of the question, ‘by what relation is a token bound to its type?’. The answer to both these questions, for anyone other than a nominalist, is the relation of instantiation. Our task then, is to give a satisfying account of the nature of this relation. Before I attempt to provide such an account, allow me to clarify why exactly the relation of instantiation concerns types and tokens. It could be argued that when a particular instantiates a property we actually have two relations in play. Consider the ontology wherein we have first the particular, which is somehow related to its properties, which are somehow related to universals. For example, we have our snooker ball, which is somehow related to its own redness, which is somehow related to redness as a universal. But this is to overcomplicate things. The point - 68 - BJUP - 2(1) - April 2007 is that the particular is related to the universal by its own redness. The instantiation of redness is the relationship between the snooker ball and the universal of redness. A particular can be expressed in terms of its properties – we have no need for a relation between a particular and its properties. The token here is the red snooker ball and all the properties that it consists in. This is always the token, the particular as a whole, since we do not want to fall into the trap I described above. This does not mean that we cannot talk about the relation between a property and a universal, since we can do so by abstracting a single property from its particular. It is just that we must take care not to call an instantiation of a property a token, since an instantiation is not an entity, but a relation. The corresponding type to the token of our red snooker ball is simply ‘red snooker balls’. Our token red snooker ball is an instantiation of the type ‘red snooker balls’.1 Consider what is known as Bradley’s Regress. The regress is said to occur in all fields of relatedness, not just instantiation. The problem is as follows: if a and b are related by R, then there must be a further relation R` that relates a and b to R. Moreover there must be a further relation R`` that relates a, b and R to R`. Repeated ad infinitum we are left with an infinite number of relations. Of course, an infinite regress is not necessarily a bad thing. A traditional example of a non-vicious regress is an infinite sequence of natural numbers – each natural number is related to the next by the relation ‘plus one’. This is not a problematic regress because the same relation is present in every case. As Kapitan points out, we must decide whether or not the regress is vicious.2 For a regress to count as vicious it must imply something impossible or absurd. Unfortunately our relation of instantiation does imply something both impossible and absurd. When a is related to b by instantiation, then a and b must both instantiate the relation of instantiation. The point is that in the case of the relation of instantiation, we do not know what the new relations that are infinitely formed consist in. However we do know that it is not the same relation 1 For more on this and a slightly conflicting analysis of the same issue, see Goldfain (2005). 2 T. Kapitan, Bradley’s Regress, independently published, p. 3, http://www.niu.edu/phil/~kapitan/Bradley's%20Regress.pdf. - 69 - BJUP - 2(1) - April 2007 of instantiation every time as was the case with the infinite sequence and the relation of ‘plus one’. This problem occurs because the infinite sequence of natural numbers and the infinity caused by the relation of instantiation are fundamentally different. The unproblematic case of the infinite sequence appears unproblematic because we have not subjected it to Bradley’s Regress argument. We could just as well use the relation of instantiation as an example of infinite occurrences of a particular relation – it is exemplified by all the particulars in the universe. It just so happens that the relation ‘plus one’ creates a numerical regress. It does not, though, create a relational regress whereby new relations are formed in virtue of the nature of the relation. New numbers are formed in virtue of the nature of the relation, but not new relations. Applying Bradley’s Regress argument to the infinite number sequence is to ask the question, ‘what is the relation between x and x’s relation with x+1?’. If xRx+1 and R = ‘plus one’, then xR`R. But then, what relation is R`? We never ask this question because it is absurd. The mistake here is to think of relations as the kind of things that can themselves be related. It may be that we can say aRb is an instantiation of the relation R and that the instantiation of the relation can itself be related, but this does not necessarily mean that it is related to a or b. It seems to me that to call aRb an instantiation of the relation R we would have to include in that instantiation both a and b. By this view the only relation a would have with the instantiation aRb is the relation of being a part of the relation. The same applies for b, and again, a regress would follow. a would be related to the instantiation of aRb by R` where R` is the relation of ‘being part of’. a would then also be related to the instantiation aR`(aRb), but this time by the same relation of ‘being part of’. Thus we do not encounter the problem of having a new relation to explain at every step of the regress. Baxter (2001) would argue that since we have the instantiation aRb, we cannot then instantiate it again in a new relation aR`(aRb). Before I go on to explain why he thinks this, I should first point out that Baxter does not think that instantiation is relational. He shares the view of Strawson that it is a non-relational tie. This, of course, is another way to avoid Bradley’s regress. However, it is still possible to relate his ideas - 70 - BJUP - 2(1) - April 2007 to the situation I have described above. This may later turn out to be because the situation I describe above does in fact treat instantiation as a non-relational tie, but first let me concentrate on explaining why Baxter thinks that this kind of multiple instantiation is not possible in the case of relations. To understand Baxter, we must first have a grounding in Armstrong’s theory of States of Affairs. Likewise to understand Armstrong’s theory of States of Affairs, we must first understand his theory of Truthmakers. A truthmaker, very simply, is something about the world that makes a truth true. For example, the proposition ‘there is a desk in this room’ is true iff there really is a desk in this room – a proposition is true iff it corresponds to reality. States of affairs are the facts about reality that serve to make propositions true – their truthmakers. I said above that an instantiation of the relation between a and b would have to include both a and b, since it is the state of affairs aRb that is the truthmaker for the proposition aRb. This is fundamentally equivalent to what Armstrong says in ‘A World of States of Affairs’ (1997), but Armstrong uses the terms ‘a is F’ instead of aRb, though he admits that this is still a relation. The relation, though, is an internal one, ‘depending as it does, only upon the internal structure of the entities in question.’3 Baxter, however, believing as he does that particulars and universals are bound together by something non-relational, wishes to show that even this kind of relation is not necessary. Baxter postulates what he calls ‘aspects’. It is an aspect of particular P, for example, that is F. Likewise it is an aspect of universal F that occurs in particular P. Baxter gives the example of Hume’s benevolence: Suppose Hume is a particular, Benevolence is a universal, and Hume is Benevolent. Then Hume has an aspect, Hume insofar as he is benevolent. Also Benevolence has an aspect, Benevolence insofar as Hume has it. These are the same aspect – Hume’s benevolence.4 3 4 D. Armstrong, A World of States of Affairs, Cambridge, 1997, p. 116. D. Baxter, Australasian Journal of Philosophy, 79/4, p. 454. - 71 - BJUP - 2(1) - April 2007 Thus the particular and the universal are related by the aspect or aspects that they share. They are related, but by virtue of their own make-ups, rather than by some extra entity. In fact, it is not even necessary to say they are related, since the aspects they share are the very same aspects – there is no need to relate something to itself. Hence Baxter concludes that instantiation is ‘partial identity’. Baxter thinks that the problem can theoretically be reduced to the classic problem of ‘one over many’. This is similar to my initial argument that instantiation concerns types and tokens. How is it that one type has many tokens? Baxter’s reply is to say that sometimes a collection of particulars counts as one thing, and sometimes it counts as many: When we say the distinct particulars are the same we count them as one and the same, not two. It is them we count this way. Thus the apparent contradiction is resolved as follows: particulars distinct in one count are identical in another. In that other count they are the same universal.5 Particulars and universals are then simply different ways of counting the same things. Instantiation arises when the two ways of counting overlap, which Baxter calls ‘cross-count partial identity’. As Baxter notes, theories of properties come about when we favour one of the ways of counting as being ‘stricter’ than the other. Nominalists would consider counting the distinct particulars as two to be the stricter count, whereas a Platonist would consider counting as one the stricter count. He proposes that we should really consider the two ways of counting as equally viable: Admit there is in principle no way to find out which is really strict and which is really loose [of the two ways of counting]…recognise that the two counts have equal, though competing, claims to being strict.6 5 6 ibid. p. 455. ibid. - 72 - BJUP - 2(1) - April 2007 It should be noted that Baxter does not postulate ‘partial identity’ as a relation. He does, though, offer it as an explanation of relations. This is the part that is useful when explaining my problems with the infinite regress of the relation ‘being part of’. He says, ‘A ‘this’ is something that is. A ‘such’ is a way that something is… A state of affairs is a ‘this’ and a ‘such’ sharing an aspect.’7 Generally, a particular is a ‘this’ and a universal is a ‘such’. A state of affairs is a ‘this-such’. Baxter again: Now suppose that the ‘this’ in a state of affairs is a particular. Then the state of affairs cannot be multiply instantiated. For consider what being multiply instantiated entails: anything that can be multiply instantiated must be able to be instantiated in wholly different particulars – ones with no particular part in common. Now consider what it would be for a state of affairs to be multiply instantiated. The instantiations would have to be partially identical with both the particular and the universal in the state of affairs. But then the instantiations aren’t wholly distinct particulars.8 By this view the situation I described wherein the state of affairs aRb is instantiated over and over again in the infinite number of instantiations of the relation ‘being part of’, could not occur. We can, then, postulate relations as universals. Hence a particular is related to a relation in exactly the same way that a relation is related to a universal – by instantiation: The relation, or at least an aspect of it, is the relevant aspects of the relata counted as identical. Thus the relation is partially identical with each relatum, with the relata counted as many, and the relata counted as one.9 We have not yet offered, in our account of the relation between particular and property, a description of the start of the process. This is to describe what is happening when a particular first begins to 7 D. Baxter Australasian Journal of Philosophy 79/4 p. 460 ibid. 9 ibid. p. 457 8 - 73 - BJUP - 2(1) - April 2007 instantiate a property. It is important to note the chain of command here. It is the particular, the token, which is doing the instantiating. The universal, the type, is that which is instantiated. So, where can we find an example of the process of instantiation taking place? Maybe we cannot. Perhaps a particular instantiates a property once and that is the end of the story. Or perhaps instantiation is an ongoing process; a particular is constantly instantiating all of its properties. If the latter were the case, and it seems quite reasonable that it might be if we accept Baxter’s view of partial identity, then it would be very interesting to investigate the notion of changing properties in particulars. This would be, after all, the only time when properties start and stop being instantiated. According to Baxter, a change in properties is simply different aspects of the same particular coming to the fore. Different temporal locations are different aspects of the particular. Instantiated properties can then be second order aspects of the temporal location aspects. Baxter describes a leaf changing its colour from green to red: ‘[T]he leaf insofar as it is green is an aspect of the leaf insofar as it exists at one time, but is not an aspect of the leaf insofar as it exists at another time.’10 This is clearly a helpful description, but it digs up some problems that have been slowly spreading their roots since the beginning of Baxter’s theory of aspects. I am worried that although this theory may provide a very satisfactory account of the one over many problem and thus instantiation, it may end up exhausting any theory of properties we may have. Particulars’ temporal parts are merely aspects of particulars; particulars’ instantiation of properties are merely aspects of particulars; properties’ instantiation in particulars are merely aspects of properties. What then, are the particular and property that all these things are aspects of? We seem to be presented with yet another bundle theory. Moreover, if we accept a bundle theory of either substance or property or both, we may end up with a bundle theory upon a bundle theory upon a bundle theory. Of course, we do not have to accept any of these theories. But if we do, we must constantly be asking the question, ‘what holds the bundle together?’. In the case of aspects, Baxter would say 10 D. Baxter Australasian Journal of Philosophy 79/4 p. 459 - 74 - BJUP - 2(1) - April 2007 that the aspects do not need to be held together, since they are already partially identical and thus one and the same. It is simply a way of counting. But, as we said before, accepting this would be tantamount to Nominalism. We may have solved the problem of instantiation, but by doing so we have created a new, perhaps even larger, problem for any theory of properties. Bibliography Armstrong, D., A World of States of Affairs, Cambridge, 1997 Armstrong, D., Truth and Truthmakers, Cambridge, 2004 Baxter, D., Instantiation as Partial Identity, Australasian Journal of Philosophy 79/4, Routledge, 2001 Goldfain, A., On the Nature of the Instantiation Relation, Buffalo, 2005, http://www.cse.buffalo.edu/~ag33/Instantiation.pdf Kapitan, T., Bradley’s Regress, Independently published, http://www.niu.edu/phil/~kapitan/Bradley's%20Regress.pdf Lewis, D., Papers in Metaphysics and Epistemology, Cambridge, 1999 Loux and Zimmerman, The Oxford Handbook of Metaphysics, Oxford, 2003 Mumford, S., Laws in Nature, Routledge, 2004 Strawson, P. F., Individuals, Routledge, 1959 Van Inwagen, P., Metaphysics, Oxford, 1993 - 75 - British Journal of Undergraduate Philosophy, 2(1), April2007 © Jeremy Thomas; all rights reserved. Gewirth, Darwin, eugenics, and solutions to the problem of future agency Jeremy Thomas University of Sheffield lwa04jpt@shef.ac.uk I intend to show that the Gewirthian principle of Generic Consistency (1978) combined with an extension of the Beyleveld/Brownsword principle of Precautionary Reasoning (2001) may provide a dialectically necessary requirement to engage in compulsory germline gene manipulation on current agents as a eugenic technique to preserve the dignity of future agents. Gewirth’s Principle of Generic Consistency (hereafter, PGC) requires agents to act in accordance with the generic rights of other agents, lest they contradict that they are agents themselves (a rational impossibility, as will be logically demonstrated). Needs of agency are generic if they are prerequisites of an ability to act at all, or with any general chances of success. Generic needs of agency are ordered hierarchically according to certain criteria. Certain requirements impinge on the ability to be able to act autonomously – freedom and well-being are therefore judged to be qualitatively high in the said hierarchy. Needs required to be able to act at all are labelled as ‘basic’ needs (e.g. life or the possession of sufficient mental capacity to be able to translate preferences into positive action). Needs required for successful action are divided by Gewirth into ‘Non-subtractive needs’ – interference with which diminishes an agent’s chances of attaining his/her purposes) – and ‘Additive needs’ – interference with which affects an agents chances of improving their capacities for successful action. In essence, all rights are qualitatively assessable. Qualitatively, basic needs are more necessary for action than non-subtractive needs, which are more necessary than additive needs; that is, it is more serious to breach the PGC by denying an agent a basic right, than any other right. - 76 - BJUP - 2(1) - April 2007 The rational requirement for respecting an agent’s dignity in this way is to be found within justifications for the PGC, as follows. By claiming to be an agent, I claim: (1) I do (or intend to do) X voluntarily for a purpose E that I have chosen. Because E is my freely chosen purpose, I must accept: (2) E is good This means only that I attach sufficient value to E to motivate me to pursue E (i.e., that I value E proactively). If I do not accept (2), then I deny that I am an agent, which is just to say that it is dialectically necessary for me to accept (2). (3) There are generic needs of agency. Therefore, I must accept: (4) My having the generic needs is good for my achieving E whatever E might be. Which is just to accept: (5) My having the generic needs is categorically instrumentally good for me. And because I value my purposes proactively, this is equivalent to my having to accept: (6) I categorically instrumentally ought to pursue my having the generic needs. Because my having the generic needs is necessary for me to pursue my having the generic needs, I must hold: - 77 - BJUP - 2(1) - April 2007 (7) Other agents categorically ought not to interfere with my having the generic needs against my will, and ought to aid me to secure the generic needs when I cannot do so by my own unaided efforts if I so wish. Which is just to say I must hold: (8) I have both negative and positive claim rights to have the generic needs. And this is equivalent to: (9) I have the generic rights. It follows (purely logically) that I must hold not only (9), but also: (9`) I am an agent, therefore I have the generic rights. Since I deny that I am an agent by denying (9`), every agent denies that it is an agent by denying (9`). Thus, (9`) (which is the PGC) is dialectically necessary for every agent. (Beyleveld and Brownsword, 2001: 72- 74). Having outlined the logical reasoning behind the PGC, it is necessary to make apparent the nature of Gewirth’s requirement of a moral duty to what he labels ‘future agents’. The concept of Precautionary Reasoning, as introduced by Beyleveld and Pattinson (1998), is sufficiently described by first asserting that it is categorically necessary under Gewirthian theory (that is, it is morally required) that agents must not violate the PGC. The next stage is the certainty that presuming x is an agent would leave no possibility of violating the PGC. On the other hand, presuming that x is not an agent leaves open the possibility that one may behave in such a way towards x as to violate the PGC.1 Therefore, if x is an ‘ostensible agent’ (Beyleveld and 1 Here, one may be able to identify an obvious parallel with the concept of Pascal’s Wager, and also the ideology underpinning the ‘innocent until proven guilty’ principle in legal theory. - 78 - BJUP - 2(1) - April 2007 Brownsword, 2001: 121), or indeed, a ‘partial agent’2 (Beyleveld and Pattinson, 1998: 43) it is dialectically necessary that one must treat him as an agent. Under the PGC, agents owe duties to future agents (i.e., agents which do not currently physically exist) as agents equal to those that they owe to present agents. If an agent has the choice of whether or not to engage in a behaviour that may damage a future agent, then he is under a prima facie duty to not engage in such behaviour (subject to qualitatively countervailing duties) (Beyleveld and Brownsword, 2001: 154). It has been contended that this is a problem – that it is too abstract a concept to propose a duty to agents which may exist far into the future. I therefore submit the following ‘chain of duty’ solution, as an effective solution to this problem: (1) Agent x1 has a categorically binding duty under the concept of precautionary reasoning to future agent x2 not to violate the rights of x2 established under the PGC. (2) Agent x1 has a duty to ensure, to the best of his abilities, that x2 is facilitated in carrying out his purposive action and therefore his duties under the PGC. (3) Agent x2 has a categorically binding duty not to violate the rights of x3 and facilitate his purposive action for the same reason, x3 has just such a duty to x4, x4 has just such a duty to x5 and so on to xn. Therefore: 4) 2 x1 has an effective duty to protect the agency status of xn . One which appears to possess only some agency bearing characteristics. - 79 - BJUP - 2(1) - April 2007 I will now be referring specifically to eugenic techniques by drawing together the mentioned topics of future agency and precautionary reasoning via an analogy that will become apparent. In applying the PGC in relation to future agents (hereafter, FAs), it can be argued that there is a duty to employ gene manipulation to, for example, cystic fibrosis at the foetal stage – if not to do so would mean treating him/her differently from other agents in this respect, then this would therefore be a breach of FA’s dignity (Beyleveld and Brownsword, 2001: 155). Of course, this would not apply to the concept of a FA ‘desiring’ the characteristics of, for example, a virtuoso pianist, since a FA does not ostensibly require these characteristics in order to act with general chances of success. My argument, again, relates to the categorically necessary moral requirement of using eugenic techniques whenever they become available (now, or in the future) to help create agents which are best suited to the environment in which the eugenic techniques apply (both now, and any ostensibly likely future environment). This is where a direct analogy with the Precautionary Reasoning principle aforementioned becomes interesting. The most widely accepted present theory on that which ‘governs’ life (that is, the way in which a particular species develops), it would seem (without engaging in biological or theological debate), is the principle of evolution via natural selection. This is the principle which describes the manner in which life (and therefore humanity3) is ‘controlled’. It seems ill-conceived, therefore, that a deontological theory would not adapt to live within it? Under Precautionary Reasoning, we have a duty to agent x if x appears ostensibly to possess characteristics of agency (as we have seen). All varied forces which are believed by individuals to govern life (e.g., some form of deity, nothing, or whatever else one may place one’s faith in) are inherently subjective. However, life, death, reproduction and all factors which affect human existence – be them positively (promoting species survival within an environment) or negatively (threatening 3 Although Gewirthian theory does not assert that only humans may be agents. Indeed, the definition of Gewirth’s ‘agent’ is equivalent to Kant’s concept of ‘a rational being with a will’ (Beyleveld and Brownsword, 2001: 72). - 80 - BJUP - 2(1) - April 2007 species survival within an environment) – can be most objectively (or at least, more objectively than a theological belief) verified by the principle of natural selection (again, as the leading theory in its field, at present). This is not to say the decision to place a belief in evolution is not subjective in itself, but I merely wish to assert that evolutionary theory is the most scientifically (and therefore for these purposes, objectively) verifiable relevant belief system at present. And because of the notion that this principle is ostensibly the most objectively verifiable of all principles that ‘control’ life, my contention is that: We have a natural duty under an extension of the principle of Precautionary Reasoning (and the PGC’s hierarchical requirement of a moral duty to future agents) to ensure that the dignity of future agents is not threatened by the possibility of inheriting characteristics which may (due to the pathologically threatening nature to an organism within its particular environment that disease may entail) have debilitating effects on their capacities for agency. Indeed, whilst it is important to recognise that natural selection is a principle which operates passively, it illustrates in principle to us that organism a is more likely to survive that organism b if organism a is better adapted to surviving in environment e. If one is more likely to survive, then one is less likely to not survive or to not exist in the future (in the case of future agents). We therefore have an obligation to judge which traits would ostensibly benefit organism a, over whom we have some form of control, within environment e, which we can perceive. But there may still be a problem of countervailing duties to current agents ‘trumping’ the rights of future agents. An obvious argument in relation to the dialectical necessity of considering future agents is that a current agent is ostensibly more likely to be an agent than a future agent who presently does not exist, and therefore ostensibly has rights which ‘trump’ the rights of a future agent. We might, however, consider: - 81 - BJUP - 2(1) - April 2007 (1) It is ostensibly very likely that at some point in the future, an individual subjected to compulsory somatic gene manipulation will reproduce. (2) This breach of autonomy (due to compulsory nature of gene manipulation) would be, in the hierarchical rights established by the PGC, not such a serious violation of the latter than, for example, removing their life or mental capacity to translate preferences into positive action. (3) Reproducing with inheritable life-threatening characteristics (which often have a higher chance of being agency-debilitating characteristics, especially if they result in premature death or birth with no chances of gaining capacity for agency whatsoever) threaten the capacities for agency (i.e., through threatening life or mental composition) of any FA. (4) These potential (ostensibly likely) and avoidable threats breach the PGC to any FA. (5) This avoidable breach of the PGC is hierarchically more serious than the breach owing to the compulsory nature of somatic gene manipulation, since it is ostensibly likely to threaten the FA’s life and/or mental composition (‘basic’ needs under the PGC). It has also been contended that to engage in eugenic techniques as an overall concept is ‘unnatural’ and so somehow contrary to natural selection. Therefore, it is maintained, such an engagement is in some way inherently dangerous. Indeed, Ho (1998) argues that such action is too dangerous and may ‘bring humans to the brink of the apocalypse.’ I however contend: - 82 - BJUP - 2(1) - April 2007 (1) Through natural selection, we have evolved to be capable of sufficient capacities4 to ‘discover’ certain eugenic techniques (including what we label as ‘rationality’). (2) ‘Using’ these capacities, we have discovered eugenic techniques conceptually in exactly the same way as chimpanzees discovered the use of tools to extract termites. (3) It is therefore ‘natural’ that we do what we will with what we have discovered by virtue of our ‘naturegifted’ abilities. My argument in relation to FAs may prima facie appear to be utilitarian in nature. It may then be falsely contended that asserting an overriding duty to the future of the species over current agents is in inherently favouring a collective over individuals. I believe, however, that this argument is extended from considerations of individual agency of future agents, and therefore merely an extension of Gewirth’s (1978) categorically (and rationally) binding Principle of Generic Consistency with indirect reference to the Precautionary Reasoning principle of Beyleveld and Brownsword (2001) and what I label as the ‘chain of duty’ principle. I consider this to be an interesting idea worthy of provoking future debate, due to its conformity with the widely held belief that moral systems exist purely to promote the survival of a group (see, for example, Bohlin, 1992). 4 Some may use the term ‘intelligence’. - 83 - BJUP - 2(1) - April 2007 Bibliography Beyleveld, D., and Brownsword, R., (2001), Human Dignity in Bioethics and Biolaw (OUP: New York) Beyleveld, D., and Pattinson, S., (1998), ‘Precautionary Reason as the Link to Moral Action’, in Boylan M. (ed.), Medical Ethics (Upper Saddle River, NJ: Prentice-Hall), 39-53 Bohlin, R., (1992), ‘Sociobiology: Evolution, Genes and Morality’ in Probe Ministries, as part of the Telling the Truth Project Gewirth, A., (1978), Reason and Morality (Chicago, University of Chicago Press) Gewirth, A., (1996), The Community of Rights (Chicago: University of Chicago Press) Ho, M., (1998) Genetic Engineering- Dream of Nightmare? The Brave New World of Bad Science and Big Business, Bath: Gateway Kant, I., (1948), Groundwork of the Metaphysics of Morals (1785), trans. Lindsay, A. D., (London: Dent Dutton) - 84 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © Guy Bennett-Hunter; all rights reserved. Interpretation and mystery: religious beliefs and Heidegger’s account of the experience of art Guy Bennett-Hunter Durham University g.a.bennett-hunter@dur.ac.uk My purpose in this paper is to challenge two major assumptions upon which the enterprise of philosophy of religion rests, as this enterprise has traditionally been conceived. (I’ll restrict myself to philosophy of the Christian religion since it’s Christianity that has had the most interaction with Western philosophy and it’s this religion that traditional philosophical discussion has most often taken as its point of departure.) It is my suggestion that two different claims are more plausible premises for a philosophy of religion and that these allow for a conception of religion that is intimately related to an account of art provided by the existential phenomenology of Heidegger. By the term ‘traditional philosophy of religion’, I mean what the theologian John Macquarrie has in mind when he speaks of ‘natural theology’ whose purpose, he suggests, is ‘to supply rational proof of the reality of those matters with which theology deals.’1 If the definition is broadened to include less sympathetic and perhaps more philosophical discussions which do not, by default, result in ‘proof’, we find ourselves in the familiar territory of, for instance, philosophical arguments about God’s existence – teleological, cosmological, ontological and so on. If these arguments succeed, the traditional philosopher of religion holds, religious expression is valuable; if not, then it isn’t. But it seems to me that this conception of the enterprise of a philosophy of religion is based on two main premises. First that religion makes factual claims about entities of various kinds, some of which are presumably ‘metaphysical’ or ‘spiritual’ since they can’t be directly experienced. And second that these claims can be philosophically tested by attempting to 1 Principles of Christian Theology (London: SCM, 1966), p. 41. - 85 - BJUP - 2(1) - April 2007 support them by empirical claims about the material world. For instance, the teleological argument assumes that religion makes the claim that a super-sensible being exists and attempts to support this claim by the observation that there are processes in nature that are apparently the products of a designing intelligence. It’s my suggestion that these two premises rather severely misinterpret the nature of religion and the claims it makes, certainly as these are understood by theologians. A better interpretation rests on premises that are broadly the opposite of those just outlined. It is clear that there is an important sense in which religious beliefs and claims depend upon, and interpret the meaning of, religious texts. However theologians have long accepted that these texts themselves are not, and cannot be, factual accounts of historical events but are, rather, later interpretations of, and reflections upon, these events. Macquarrie gives the example of the Christian claim of the resurrection of Christ. The most that an historical empirical enquiry could ever confirm, he argues, (even if one were there in person) is that a person who had been presumed dead was later seen, apparently alive. This is not the Biblical claim (made at Acts 2.24) that ‘God raised him up’. Since God is believed to be invisible, this claim is not open to empirical investigation. As Macquarrie puts it, ‘this…judgment…goes beyond bare happening to offer an interpretation of the happening – and no ordinary interpretation, either.’2 It follows from this that, far from being factual and empirical, religious beliefs and the claims which express them are interpretative of texts which are themselves interpretations of happenings which in turn might be literary fabrications in some cases. We might, with some theologians, go further still and claim that the happening need never have ‘happened’ at all. Even if they are literary fabrications, their interpretative value still holds, just as it does for the Greek Myths.3 Indeed, ‘mythology’ is a word used among theologians for much of the Biblical material. Religious claims, 2 Essays in Christian Existentialism (Montreal: McGill University Press, 1965), p. 142. This may be said to be a controversial claim. Cf., e.g. S. Paul’s claim at 1Cor. 15.17: ‘If Christ has not been raised, your faith is futile’. But I have been arguing that Scriptural claims interpret ‘bare happening’. The meaning of S. Paul’s claim depends, most importantly, on the meaning of ‘raised’ and ‘faith’ – meanings which are themselves interpretative and, it is generally agreed, far from clear. 3 - 86 - BJUP - 2(1) - April 2007 then, do not necessarily describe a metaphysical realm of spiritual beings and events but do, of necessity, interpret this world, as we experience it, in the existential significance that it has for religious people or for people when they look at the world in a religious light. To turn to our second premise, that religious beliefs can be inferred from empirical beliefs about the world. This can be challenged along the Logical Positivist line of argument (already intimated) which points out that religious expressions are not of a kind that could, in principle, be verified or falsified. If, for example, God’s existence were proven tomorrow, there would seem little point in religious expression since a statement expressing belief in God would become the kind of statement that it would be irrational to reject. As Wittgenstein said in one of his later writings, evidence would ‘destroy the whole business.’4 One way of insulating the claim of God’s existence from the destructive influence of ‘evidence’ has been to posit the concept of ‘necessary existence’. In the attempt to avoid the pitfall of accepting that God’s existence is a fact like any other, theologians have claimed that God’s existence could not, logically, be otherwise – it is a logical necessity rather than a mere contingent fact. But this concept turns out to be meaningless. The function of an existential statement ‘x exists’ is to rule out x’s nonexistence and nothing more. If the non-existence of x is (logically) impossible, as when God is conceived as a necessary being, the claim ‘x exists’ is robbed of its logical function and becomes meaningless. Another way of resolving the difficulty of verifiability (or falsifiability) in the case of religious statements has been to distinguish the Pascalian ‘God of the philosophers’ from the God of religious devotion. In Macquarrie’s words, ‘the God who is the conclusion of an argument…is not the God who is worshipped in religion.’5 Such a distinction has often been made with reference to the notion that the ‘God of religion’ is essentially mysterious in nature, a notion that owes much to Christianity’s interaction with Platonist systems in its formative period. The Christian Platonist of the Sixth Century, pseudoDionysius (known as Dionysius the Areopagite) whose writings were 4 Lectures and Conversations on Aesthetics, Psychology and Religious Belief (Oxford: Blackwell, 1966), p. 56. 5 Principles of Christian Theology, p. 45. - 87 - BJUP - 2(1) - April 2007 regarded as orthodox until the 15th or 16th century, emphasized God’s transcendence and mystery to the point where he could be described as hyperousia, ‘beyond being’. ‘Mystery’ was understood in this context, not as what is opaque or only partially knowable but in the strong sense of that which is necessarily resistant to conceptual formulation and linguistic articulation – the ineffable. This strong sense of mystery, obviously prevalent in the so-called mysticism of Dionysius and others, has penetrated to the thought of the most ‘orthodox’ of thinkers since Aquinas, who pointed out that, owing to the mystery of God, religious statements are separated from divine reality by an infinite distance. The notion of mystery, then, is more pervasive in religion than the traditional philosophers of religion have been ready to allow. It could thus be argued that claims about the Divine (conceived as the ineffable in the strong sense just outlined) cannot be inferred, logically or otherwise, from observations about the nature of the observable world. I now want to argue that it is in respect of these two observations on the nature of religion (that religion interprets the world in its existential significance and that its claims can’t be logically inferred from the world owing to divine mystery) that religious experience can and should be viewed as intimately related to the experience of art as phenomenologically interpreted by Heidegger. In his famous essay, The Origin of the Work of Art, Heidegger’s focus is on the experience, rather than the definition, of art and the way in which that experience relates to and conveys the meaning which, in both cognitive and non-cognitive forms, always mediates humanity’s engagement with the world. In Heidegger’s view the role of phenomenological analysis (or description) is to interpret this meaning, and the experience of art has a similar function which he calls the opening up of a world. Insofar as it fulfils this function, art is phenomenology by other means. Heidegger brings out the meaning of this phrase (‘the opening up of a world’) with an interpretation of Van Gogh’s painting of a peasant’s shoes. Heidegger illustrates the importance of the distinction between the everyday experience of a pair of shoes, where their function is of the greatest importance, and the phenomenologically illuminated experience of them, in which their meaning and significance is more palpable, a meaning and significance - 88 - BJUP - 2(1) - April 2007 which is brought to light by art. For Heidegger, the painting of the shoes intimates, in his words, ‘the toilsome tread of the worker,’ ‘the fallow desolation of the wintry field,’ the ‘worry as to the certainty of bread;’ whereas ‘the peasant woman [whose actual shoes the painting depicts]…simply wears them.’6 The full significance of the shoes, the intimation of their owner’s world with all its nuances of meaning, affective as well as factual, only comes to light in the art-work since, when they are being worn, none of this is consciously recognized. But it is owing to the fact that they are worn and used in an everyday context that the shoes have the meaning that they do. It is only in and through an intimate relation of engagement between the peasant and the peasant’s world (indeed, the peasant’s shoes) that ‘meaning’, in Heidegger’s broad phenomenological sense, is possible. ‘World’, then, is something familiar constituted by us and for us. It is a function of art to bring it to light (‘open it up’) but also to ‘set it up’ – were it not for art, Heidegger argues, our social world would not be what it is. The ancient Greeks would not have been the people they were (historically, culturally, politically) were it not for the Greek Temple. The Greek Temple is the work of art that, for Heidegger, defined the ancient Greek existence. On the other hand, Heidegger claims, art also ‘sets forth the earth’. In contrast to ‘world’, which is familiar and revealed, ‘earth’ is concealed, opaque and mysterious. The ‘world’, as we have seen, is constituted by and for us in the interpretative activity in which we are ‘always already’ engaged. What Heidegger calls ‘earth’, on the other hand, is that upon which all our interpretations depend – it is the mysterious ‘ground’, the condition of there being anything to interpret. Truth, for Heidegger, at once reveals and occludes, and this duality in his account of truth corresponds to the duality between ‘world’ and ‘earth’ which characterizes his account of art, which he then describes as ‘the happening of truth’. Julian Young illuminates Heidegger’s rather abstruse arguments by interpreting them from a more traditional philosophical perspective. He 6 ‘The Origin of the Work of Art’ in D. F. Krell (ed.), Basic Writing: Martin Heidegger (London: Routledge, 1978), pp. 143-206 (p. 159-60). - 89 - BJUP - 2(1) - April 2007 begins with the observation that the correlation of words and things is inadequate to determine reference, and thus truth cannot reside in a correspondence between a statement and a fact. If I point to a table which I have, for some reason, decided to call ‘Smith’ and say ‘Smith is colourless’ it’s not clear whether my statement refers to the table which is brown, the region of space the table exactly occupies which is colourless, or the invisible, non-spatial demigod which, I also believe occupies the same space as the table. The point is that whenever we make predications about a ‘region of Being’, we always presuppose what Young calls a ‘horizon of disclosure’ which renders intelligible the kind of predication we make.7 I can view the table from within an everyday horizon (as something for eating off) or a scientific horizon (as a lump of atoms) or even a religious horizon (as the house of a demigod). It is not inconceivable that, one day, a Martian physicist might find an entirely new way of disclosing the region of Being that is, for us, now, the table. These possibilities are infinite and essentially mysterious. But some horizons are mutually exclusive. As Young writes ‘Being, disclosed in the manner of atomic physics cannot be disclosed as everyday objects.’8 One such horizon of our own making is our ‘world’ in which we have chosen to live, eating off tables and sitting on chairs. But in choosing ‘world’ as our interpretation, we occlude the infinite penumbra of other possible horizons that, together, constitute ‘earth’, which mysteriously grounds our existence in the world and about which we know, and can know, nothing. Art in its key functions with respect to ‘world’ and ‘earth’ has a foot in both camps – both the revealed and familiar, and the concealed and mysterious. Its function as the ‘happening of truth’ is, in Heidegger’s words, that of making ‘world’ transparent to ‘earth’. In other words it makes us consciously aware of two things. First the meaning which our world has as a result of our own making (through our interpretation). And second that this meaning depends upon an infinite and mysterious plethora of possible horizons of meaning (or, to borrow Nietzsche’s word, ‘perspectives’) which is not at all of our making. 7 8 Heidegger’s Philosophy of Art (Cambridge: CUP, 2001), p. 29. Ibid., p. 39. - 90 - BJUP - 2(1) - April 2007 To return to the subject of religion with which I began, it becomes clear that Heidegger’s analysis of art allows for an intimate relationship between it and religion when understood according to my two observations. The ‘world’ element of art for Heidegger, like one element of religion for Macquarrie, helps to interpret the world in the existential significance it has for religious believers. But more interesting is the notion of essential mystery that the ‘earth’ element of Heidegger’s interpretation of art brings out. Since, as I have argued, the notion of mystery is essential to an adequate understanding of religion, an obvious parallel can be drawn between the ‘earth’ element in art and what I have argued is this central religious concept. But there is an objection to such an idea. If we agree with William James’s claim that it is a feature of the mysterious to resist conceptual formulation and articulation in language (if we take seriously his notion of ‘ineffability’), should not the appropriate response be to say nothing at all? This would appear to be Wittgenstein’s proposal, since ‘what we cannot speak about we must pass over in silence.’9 According to Young, however, Heidegger’s converse response is ‘to become a poet’; that is, to channel the creation and reception of works of art, especially literature.10 This response may be justified by appeal to Heidegger’s conception of truth (in the ‘happening’ of which art consists) as something which is prior to correspondence with facts, a relation which is always dependent on cognitive forms of meaning. The experience of art, according to Heidegger, ‘attunes’ us to the mysterious truth of Being, it effects cultivation of a certain comportment to the world and to mystery. David Cooper explains that ‘comportment’ doesn’t involve only cognitive concepts and propositions but is a ‘way of seeing or experiencing Being.’11 Such cultivations (like metaphor), Cooper argues, ‘hint at’, ‘intimate’, ‘inspire’ or ‘evoke’; they are not assertoric and so don’t fall prey to the inconsistency of making assertions about what cannot be asserted – or, as someone once put it, of trying to ‘‘eff’ the ineffable’. 9 Tractatus Logico-Philosophicus, (London: Routledge, 2002), p. 89. Heidegger’s Later Philosophy, (Cambridge: CUP, 2002), 20. 11 The Measure of Things: Humanism, Humility and Mystery (Oxford: Clarendon Press, 2002), 294 10 - 91 - BJUP - 2(1) - April 2007 This idea of meaningful expression which is non-cognitive and in some cases non-linguistic can be perceived in the religious notion of epiphany. This notion is expressed in the context of art in the Kantian ‘aesthetic ideas’, according to which something that language and conception cannot get on ‘level terms’ with is not only expressed but also ‘bodied forth’. In religious terms, Cooper writes, God ‘epiphanizes as human or other worldly forms, or perhaps as the world as a whole.’12 In religious thought, these ‘worldly forms’ have also included religious texts, liturgies and, of course, works of religious art – notably icons. Of course not all art is religious, but if, as I’ve been suggesting, mystery ought to be viewed as a common core to religious expression, all art in its intimation and bodying forth of the mysterious ground of Being will have some ‘religious’ dimension in a broad sense of the term. Heidegger gives a concrete example of this ‘bodying forth’ and I conclude with his words written with a work of religious art in mind: ‘by means of the temple, the god is present in the temple…[and] it is the same with the sculpture of the god…it is a work that lets the god be present and this is the god himself.’13 12 13 A Philosophy of Gardens, (Oxford: Clarendon Press, 2006), 130. Op. cit., 168. - 92 - BJUP - 2(1) - April 2007 Bibliography Cooper, D. E., The Measure of Things: Humanism, Humility and Mystery (Oxford: Clarendon Press, 2002) Cooper, D. E., A Philosophy of Gardens (Oxford: Clarendon Press, 2006) Heidegger, M., ‘The Origin of the Work of Art’ in D. F. Krell (ed.), Basic Writings: Martin Heidegger (London: Routledge, 1978), pp.143206 Kant, I., Critique of Judgment, trans. Pluhar, W. S. (Indianapolis: Hackett, 1987) Macquarrie, J., Essays in Christian Existentialism (Montreal: McGill University Press, 1965) Macquarrie, J., Principles of Christian Theology (London: SCM, 1966) Wittgenstein, L., Lectures and Conversations on Aesthetics, Psychology and Religious Belief (Oxford: Blackwell, 1966) Young, J., Heidegger’s Philosophy of Art (Cambridge: CUP, 2001) Young, J., Heidegger’s Later Philosophy (Cambridge: CUP, 2002) - 93 - British Journal of Undergraduate Philosophy, 2(1), April 2007 © The reviewers; all rights reserved. Book reviews Feminism: issues and arguments Jennifer Saul, Oxford University Press Heather Arnold University of Sheffield pip06hea@shef.ac.uk I approached Jennifer Saul’s book on feminism with enthusiasm, inspired by the thought of learning more about a topic I had previously only dabbled with. It wasn’t long, however, before my enthusiasm had waned somewhat, a waning that was not just the inevitable reduction in the vigour with which one approaches a new book after the first crack in the spine appears. Rather, my enthusiasm disappeared as I read Saul’s Introduction, where readers are warned not to expect any discussion of the very concept of ‘woman’ and whether such a concept makes sense. This issue, to me, forms the very basis of any discussion of feminist issues and cannot be counted as a mere omission, but rather can only leave a gaping hole in the text. Despite this initial disappointment, however, I carried on into chapter one: ‘The Politics of Work and Family’. Now, the very title of the first chapter gives a clue as to the most crucial choice made by Saul in writing her book: the text is structured by topic, rather than by schools of feminist thought or any other theoretical approach. Consequently, Saul presents chapters titled, ‘Sexual Harassment’, ‘Pornography’, ‘Abortion’, ‘Feminine Appearance’, ‘Feminism and language Change’, ‘Women’s ‘Different Voice’’, ‘Feminism, Science, and Bias’ and ‘Feminism and ‘Respect for Cultures’. Each of these titles grabs the attention of the reader, who may be unsurprised to see some of the more classic feminist issues included in the list, but also intrigued and interested to read about - 94 - BJUP - 2(1) - April 2007 some of the more unusual and supremely relevant topics such as language change and feminine appearance. I felt that this choice of structure resulted in something akin to a graph presenting discrete data. In other words, the book presents each topic as separate from the last, with links or recognition of possible links between chapters and ideas sorely lacking. However, the structure also has its benefits. It makes the text very welcoming to those new to the subject, in that they can feel that they can connect in some way with the issues that are being presented; and it appeals by way of a challenge to those who are sceptical about the contribution that philosophy can make to ‘real-world’ issues. Further, the structure means that a casual reader can simply dip in and out of the book and learn something without having to read a dense fifty pages or so of theoretical explanation before they even get a whiff of the issue of ‘respect for cultures’. However, I found that the immediate appeal of the structure lost its gloss when, by the fifth or sixth chapter, I was somewhat puzzled by some of the conclusions that were being reached by Saul, and particularly by their apparent incompatibility. The first puzzle was centred on the issue of language and its arguable power to influence our sense of reality. Saul rejects the claim that men have created language, and therefore, that they have created reality. She argues that such a view entails the ‘ridiculous’ idea that men have created amoebas, clouds and even other men. In rejecting this view, Saul seems to have forgotten that in the chapter on sexual harassment, a crucial idea was that women were not taken seriously until there was a term for what they were experiencing at work, namely, ‘sexual harassment.’ In other words, until the term was brought into being, sexual harassment as we now understand it was not a part of reality, or at least was not a part of legal reality. A further puzzle emerged when I read the final chapter of the book, where Saul relies on the idea that there is nothing essential to any culture (that is, there is nothing that can be considered a necessary condition for a particular culture to remain that particular culture, as indeed there is no such thing as a ‘particular’ culture). This antiessentialist view about culture enables her to argue that no single custom is immune from criticism by others on the grounds that it is a - 95 - BJUP - 2(1) - April 2007 necessary part of a particular culture. However, throughout the book Saul seems to speak about women as a group necessarily sharing particular characteristics; in other words, as sharing some kind of essential nature. It is not clear why ‘women’ should be any more of a fixed idea than ‘culture’, and for this she does not argue. In both of these cases, my real problem is that the book’s structure allows for claims made in one chapter to be seemingly swept under the carpet in order to round off the next chapter in a neat and tidy way. Although Saul is clear in her reluctance to make monumental conclusions about each topic, I still found that some of the claims that were made were somewhat unsubstantiated, or even sometimes in conflict with ideas used in other chapters. But Feminism Issues and Arguments is clearly presented as an introductory text on some of the problems in the field. Saul is honest in her introduction about the intentions and inherent limitations of the book. No reader expecting to find a detailed analysis of Marxist feminism will be hoodwinked by the back page, and nowhere does Saul claim that all her arguments will be watertight and cohere with one another in some kind of fluid harmony. The point of the book seems to be to provide a springboard for students new to feminism; it is supposed to be a catalyst for discussion, debate, essays and personal reflection. In this regard, I think that the book is successful. There are enough dubitable points in each chapter to spur students on in discussion, and the issues always take centre stage rather than particular thinkers, which ensures that plenty of space is left for personal responses that need not fit in with a theoretical viewpoint in order to be deemed valuable. For those seeking more detailed accounts or interpretations of the issues, Saul provides comprehensive ‘Suggested Further Reading’ lists for each chapter. Despite the notable absence in these lists of some of the key figures in both historical and contemporary feminism, such as Simone de Beauvoir, Anne Phillips and Chantal Mouffe, they do provide something for interested readers to get their teeth into. One fundamental quality of Saul’s book is its clarity. She manages to express complex issues in simple language, clarifying potentially unfamiliar words along the way, whilst maintaining the feeling that she - 96 - BJUP - 2(1) - April 2007 is writing for an intelligent audience. This is no mean feat when you consider the finicky nature of some of the classic feminist issues. Often, Saul’s aim does not seem to be to provide further enlightenment on an issue. Rather, in some cases she simply expresses the importance of taking a new angle on an old debate (as in the chapter on Abortion), whilst in others she elucidates not our understanding of an issue, but rather that we are all still groping in the dark as regards to what we mean by certain terms (as in the chapter on Sexual Harassment). At times I think that Saul fails to dig topics out of triviality (Language Change) and at others that she seems overly despairing at our prospects (Feminine Appearance). On the whole, however, I think that what Saul gives the reader is a suitably comprehensive introduction to some feminist topics, and this in a language that is just the right side of simple in a highly readable format. Feminism Issues and Arguments will not have you feeling horribly overwhelmed and unsure of where to start in the same way that de Beauvoir’s The Second Sex almost certainly will. Neither will it make you want to stand at your window and shout to the whole world how unfair the labour market is and that you don’t really like wearing makeup in the same way that Naomi Wolf’s The Beauty Myth might. But it should catch your interest, introduce you to areas of old debates that you hadn’t really considered before and leave you feeling like you want to find someone clever to argue with over the idea that women have a ‘different voice’. - 97 - BJUP - 2(1) - April 2007 Four Shorts Paul Murphy Independent paulaustinmurphy@yahoo.co.uk Key philosophers in conversation: the cogito interviews Andrew Pyle (ed.), Routledge This is an interesting and, at times, entertaining book – if only because analytic philosophers are rarely caught off-guard or asked about their personal views and experiences. There are twenty philosophers interviewed here. They include Dummett, Warnock, Quine, Putnam, Scruton, Dawkins (sic), Dennett and so on. The book, as one would expect, concentrates primarily on the philosophical ideas of the philosophers interviewed. Nevertheless, as hinted at, there are little oases of light relief here and there, as well as illuminating insights into the philosophers’ personal motivations. In the Dummett interview the philosopher says: ‘You have to follow the argument and see where it leads.’ Nonetheless, a little later he continues: ‘My religious belief would tell me I must have made a mistake somewhere.’ That last admission is a reference to the ‘mistake’ of his anti-realism leading to ‘atheistic conclusions’. Dummett, as is well known within the realm of analytic philosophy, is a staunch Catholic. Did he mean by what he said that he thinks that, almost by definition, he must have made a mistake of some kind because atheism simply cannot be true? Or does he mean that whilst it looks like atheism, it’s just not? The interview with Quine, who died five years ago, shows the American philosopher at his tight-lipped best. For example, Quine is asked about his time in Germany (in 1932, with the Vienna Circle). Cogito enquires about the German philosophers who were the ‘vacillating exceptions’ to the logical positivists who opposed and then fled the Nazi regime. - 98 - BJUP - 2(1) - April 2007 Quine simply replies, ‘De mortuis nihil nisi bonum’, which means ‘never speak ill of the dead’. Is this a thinly veiled reference to Heidegger (amongst others)? Maybe, but the fact remains that if such philosophers had not fled Germany in the 1930s, being mainly logical positivists and logical empiricists, they might never have taken over the many American philosophy departments that they did take over. And thereby they might never have established the hegemony of analytic philosophy within them. It is also interesting to note how strongly Anglo-American analytic philosophers really feel about their Continental rivals. Take the case of Hugh Mellor. When quizzed about Derrida, Mellor says that the infamous French deconstructor is ‘wilfully obscure’, before going on to accuse him of ‘triviality’ and ‘mystery-mongering’. He also claims that Derrida’s ideas are ‘trivial truisms’, plain ‘nonsense’, or worse. And, finally and perhaps most tellingly, Mellor takes those who follow Derrida to be members of ‘the chattering classes’. Putnam offers a good riposte to Mellor’s vituperative generalisations. He says: Just as there are different sorts of poet and different sorts of scientist, so there are different sorts of philosopher. What made Kierkegaard a great philosopher was not the same thing that made Carnap a great philosopher. Putnam, in any case, is part of that small but growing movement of analytic philosophers that is reassessing its relationship with Continental philosophy and beginning to see things about this Great Schism in a less black and white (or Mellor-like) way. Finally, I must mention the rather strange inclusion of Richard Dawkins in this book. To state the obvious, Dawkins is far from being a philosopher of any description, no matter how much one agrees with much that he says. Andrew Pyle, the editor, does not tell us, in his introduction, why he included Dawkins. Despite that, Dawkins’ spoken prose is very clear and precise, as it always is, and I regard him simply as a welcome inclusion in the book. - 99 - BJUP - 2(1) - April 2007 Ludwig Wittgenstein: the duty of genius Ray Monk, Vintage Here is yet another addition to the huge, and still-growing, Wittgenstein industry. This industry, on the last count, produced around two thousand books, essays, papers, articles, etc., on Wittgenstein in one year. So I will tread carefully in this review because I am neither a Wittgenstein expert nor, should I say, a Wittgenstein acolyte. One can take it as given that many Wittgensteinians will say that Ray Monk’s biography has ‘got Wittgenstein wrong’ on many issues. I found Monk’s book a very rewarding read. After all, one can never find out too much about Wittgenstein. And, in any case, after one reads this book one will feel a little more comfortable when the name ‘Wittgenstein’ is brought up, or at the conferences at which this name is bound to be uttered more often than, say, that other favourite analytic philosophy phrase, ‘if and only if’. Monk succeeds quite well in fusing the ‘life and times’ of Wittgenstein with his actual philosophy. This is not a complete surprise because Monk is trained in philosophy. More relevantly and specifically, Monk did his M.Litt on Wittgenstein’s philosophy of mathematics, and Wittgenstein’s philosophy of mathematics, one can argue, shows us the Austrian philosopher at both his most pure and most radical. The book is divided into four temporal parts: (I) ‘1889-1919’, (II) ‘1919-28’, (III) ‘1929-41’ and (IV) ‘1941-51’. In the first part we find out about the extremely wealthy late 19th century Wittgenstein family, of Vienna, that had nine grand pianos and had recently converted to Christianity. Monk then discusses Wittgenstein’s time at Manchester University, where he studied mathematics and engineering. As one would expect, Monk then covers the period at the University of Cambridge when Wittgenstein met Bertrand Russell. Monk calls the Wittgenstein of this time Russell’s ‘protégé’. Following this, as is well - 100 - BJUP - 2(1) - April 2007 known, it didn’t take long for Wittgenstein to become Russell’s ‘master’. (It is interesting to note that Russell didn’t really rate Wittgenstein’s later work. He saw it, as one can see from a passage quoted in Monk’s book, as lacking ‘seriousness’.) Then we cover Wittgenstein’s period as a soldier in the First World War, after which he spent a little time as a schoolteacher and then as a gardener. More interestingly, we see that Wittgenstein sought information, from the monks who owned the garden he laboured at, about joining the monastery. Needless to say he didn’t become a monk. Nevertheless the very fact that he enquired displays to us something of Wittgenstein’s character. This period also covers the publication of Wittgenstein’s masterpiece – the Tractatus Logico-Philosophicus (1921). This esoteric but profoundly brilliant and influential work began to become a little better received in the Germany and Austria of the early 1920s. Indeed, the work became many philosophers’ Bible. Not only that, but Wittgenstein’s every utterance began to be scrutinised by his disciples (as was also the case later at the University of Cambridge). Many philosophers, if only in retrospect, can be seen to have interpreted Wittgenstein’s idea that the meaning of a sentence is its mode of verification in a simplistic manner. Later, Wittgenstein scoffed at the naiveté of some of these interpretations and even renounced them entirely. The last section, ‘1941-51’, deals with Wittgenstein’s writing of the Philosophical Investigations (amongst other things). Many philosophers take this work to renounce all, or at least most, of the specific doctrines that Wittgenstein upheld in the 1920s and before. However, just as certain philosophers have stressed the cleavage between early and late Wittgenstein, others have emphasised the continuity. The overall impression one gets from Monk’s book is of the essential complexity and humanity of Wittgenstein the man. His complexity goes without saying and is well noted. But the word ‘humanity’ is taken here to refer less to Wittgenstein’s benign or altruistic natures, but to his deeply emotional and discursive temperament. In any case, the very many people whom Wittgenstein despised and criticised (not only philosophers) show us the darker side of the great philosopher. I think - 101 - BJUP - 2(1) - April 2007 that if there is a general fault with Monk’s biography, it is its unswervingly uncritical and positive account of not only Wittgenstein’s philosophy, but also of the man himself. As one would expect, Monk is at his best when discussing Wittgenstein’s philosophy of mathematics. He captures well Wittgenstein’s later approach to the philosophy of mathematics – something that can be called, and often has been called, ‘anthropological’ in nature. Monk writes that Wittgenstein ‘wanted to show that the inexorability of mathematics does not consist in certain knowledge of mathematical truths, but rather in the fact that mathematical propositions are grammatical.’ There is so much one can say about this book because there is so much one can say about Wittgenstein himself. Nevertheless, I shall close this review with a passage Wittgenstein wrote in a letter to Norman Malcolm (quoted by Monk): what is the use of studying philosophy if all that it does for you is enable you to talk with some plausibility about some abstruse questions of logic, etc., and if it does not improve your thinking about the important questions of everyday life? Philosophy of Mind: a Guide and Anthology John Heil (ed.), Oxford University Press John Heil, the editor of this anthology, is himself a notable philosopher of mind. So one can rightly expect that this anthology would be a worthwhile contribution to the many already-existing anthologies we have on the philosophy of mind. Heil, of course, includes many classics in his anthology. For example: ‘Identity and Necessity’ by Kripke; ‘Minds, brains, and programs’ by Searle; ‘Eliminative materialism and the propositional attitudes’ by Paul Churchland; and ‘Individualism and the mental’ by Tyler Burge. There are also a few new works and novelties included too. For example: Lynne Rudder Baker’s riposte to eliminative materialism, ‘Cognitive - 102 - BJUP - 2(1) - April 2007 suicide’; ‘Could love be like a heatwave’, Janet Levin’s contribution to the mental subjectivity debate; and E. J. Lowe’s ‘Non-Cartesian dualism’, a rare example of a work that goes against the prevailing hegemony of non-reductive physicalism. The anthology is dived into twelve parts, which, in itself, shows us that it is a pretty big book. Some of the parts go by the titles ‘Behaviourism and mind-body identity’, ‘Functionalism’, ‘Artificial Intelligence’, ‘Eliminativism’, ‘Consciousness’, ‘Reduction’ and so on. A couple of the less familiar and unexpected sections include ‘Interpretationalism’, and the most interesting, ‘Challenges to contemporary materialism’ (which includes a piece written by the only non-philosopher of the anthology – the physicist, Eugene Wigner). The book also includes an extensive and long introduction, notes on the contributors, and various suggested readings. I found Heil’s own sub-introductions to the twelve parts most enlightening. They are extremely easy to read and yet informative at the same time. Even though a few of the papers included in the anthology are rather advanced in nature, all of Heil’s introductions are primarily aimed at the beginner. Some of them are quite short and others quite long. The introductions themselves are divided into sections. For example, Part I’s introduction includes the sub-sections ‘Philosophical behaviourism’, ‘Privileged access’, ‘Pain and pain behaviour’ and so on. Each introduction is written, as I said, in a non-academic style and precludes the analytic philosopher’s usual tone of seriousness. My own favourites from this anthology include Fodor’s ‘The mindbody problem’, ‘Thought and Talk’ by Davidson, and the aforementioned contributions from Searle and Lowe. The work by Fodor is basically a brief historical survey of the most recent debates on the mind-body problem as well as the contemporary responses (including Fodor’s own) to them. The whole paper is very easy to read and will be of use to both student beginners and the more advanced. I have included Searle’s paper among simply because it is so well known and influential, or at least its subject matter is. This too is written in a very clear prose style. In it Searle doesn’t even attempt to hide his negative attitudes towards cognitive science, computationalism and - 103 - BJUP - 2(1) - April 2007 those philosophers who believe in what he calls ‘strong-AI’ (i.e., artificial intelligence). Davidson provides an interesting account of the distinctions that can be made, or denied, between thought and talk. Is thought silent talk? Can one think without words? Do animals think at all? These are some of the questions addressed by Davidson. I am also including Lowe’s ‘Non-Cartesian dualism’ here. It happens that I strongly disagree with the general tenor of this piece. And so too do the majority of philosophers of mind. Nevertheless, Lowe’s iconoclastic status within this anthology, and the philosophy of mind generally, is reason enough to single out his contribution to the many debates on the nature of mind that are also captured in this excellent and very extensive anthology. Contemporary philosophy of thought: truth, world, content Michael Luntley, Blackwell At first glance, or even at second glance, I thought that Michael Luntley’s book was a bit of a motley assortment of philosophical topics. On reflection I thought the contrary. Luntley discusses, amongst other things, Frege’s theory of sense and reference, Russell’s theory of descriptions, Tarski’s semantics, mathematical intuitionism, Quine’s theory of radical translation, Davidson’s radical interpretation, Kripke’s rigid designators, Putnam on natural kinds, and so on and so on. In other words, this book includes work on semantics, the philosophy of mind, the philosophy of mathematics, the philosophy of logic, metaphysics, and, to a much smaller extent, even epistemology. Throughout all this, however, there are central themes, or even one central theme, running through the book. The book is subtitled Truth, World, Content. The word ‘truth’ refers to Luntley’s primarily Fregean position on truth (that it must be tied to both sense and reference, i.e., to meaning). This standpoint on truth is also mediated by Tarskian semantics and the work of Donald Davidson. The word ‘content’ represents the philosophy of mind component of this book. What is content? In this work it is any thing or state the mind ‘contains’. For example, a belief can count as content, and so too can an actual experiential mental event or state. Luntley’s - 104 - BJUP - 2(1) - April 2007 notion of content again owes much to Frege and specifically it relies heavily on Frege’s notion of a ‘singular sense’. As for the ‘world’ of the title. This is a reference to Luntley’s attempt to provide a general theory that will enable us to capture the world and all its contents. It does so, however, by saying that the world is captured, as it were, from the very beginning. Luntley’s theory is neither naively realist nor sceptical. We have and keep the world because of the fact that the mind is indeed a thoroughly natural entity. And through the work of Davidson, who, according to Luntley, rejects the offending representationalism of our philosophical tradition, we find that we are not representers of the world at all, nor are we external to it. We are, instead, of the world and part of the world. We don’t rely on representations of any kind, whether they are Lockean ‘ideas’, Kantian ‘phenomena’, the phenomenalist’s ‘sensedata’, or even Quine’s ‘sense-events’. We do not ‘posit’ or ‘infer’ from such representations the ‘true nature’ of the world and its objects. We get the world straight, not indirectly. And we do so, Luntley argues, because the scheme/content distinction is bogus. That is, there is no real epistemic division between our conceptual scheme, or conceptual schemes, and content (again, whether they are ‘sense impressions’ or Carnap’s ‘cross-sections of experience’). This conclusion of Luntley’s is heavily indebted to Davidson’s work. The main title of this book is Contemporary Philosophy of Thought, which is to say that it is not a contemporary philosophy ‘of Mind’. Luntley uses the word ‘thought’, instead of ‘mind’, primarily because of his allegiance to Frege’s theory of a ‘singular sense’ and the kind of mental content that implies. Mental content can often be characterised as something over and above our various linguistic expressions of it. For example, there is something about our object-involving mental states that is not fully ‘codifiable’, as Luntley puts it, by our linguistic characterisations. Thus we can say that in this work, and therefore also in Frege’s, the ‘linguistic turn’ in philosophy (which began with Frege, early Wittgenstein and Russell in the late 19th and early 20th century) has itself led to another philosophical turn. In this case, that turn is to a philosophy of thought that is ‘externalist’, as philosophers put it, and fully committed to the idea that the mind is both embodied and embedded in the world. It is not a Cartesian thing that simply inspects the world and is therefore both outside of it and of a non-natural kind. This non- 105 - BJUP - 2(1) - April 2007 linguistic turn can be said to have begun with Kant’s fusion of ‘internal’ and ‘external sense’, which was then followed by Hegel’s substance/object monism. After that, we then have the work of Heidegger. And later, in the analytic tradition, that of Davidson, Brandon and Tyler Burge (amongst others) – Luntley himself follows this non-linguistic tradition. Luntley is Senior Lecturer in Philosophy at the University of Warwick. And he has written books on both the Continental and the analytic tradition – sometimes on both traditions in the same work (e.g., his Reason, Truth, and Self). As for Luntley’s actual writing in this book, he successfully makes the majority of ideas and problems he tackles comprehensible, and he does so at the same time as sustaining a coolheaded and impersonal prose style. Some of the sections are evidently more difficult than others. For example, the logical intuitionist digression into formal logic may seem a little beside the point when put in the context of the whole book. However, Luntley’s work is strengthened and made more palatable by the continuous aforementioned thread that runs through it. That thread can be summed-up with a negation of a phrase from Richard Rorty: the world is not well lost. Or we can say: the world cannot be ‘well lost’ because, in a way, it was never found. - 106 - Upcoming BUPS and BJUP events Philosophy is of course much, much better if you’re with people who are passionate about the subject and know what they’re talking about. BUPS and the BJUP exist to bring together undergrads who love philosophy. Our events offer opportunities to give or discuss really great papers, to meet and mix with other undergrads who think worrying about ethics or the fundamental structure of mind and world is kinda cool. To build an understanding of how philosophy is done across the country. To meet other students who like this stuff as much as you do, have done their reading and want to talk. BUPS also organises the UK’s only big, annual national undergraduate philosophy conference, and the BJUP is Britain’s only national undergraduate philosophy journal. Interested? Good, then you should be at the BUPS and BJUP events. You can get to information about all of these from our website – www.bups.org. Here you can also see a typical programme, browse past conference info, and even download a sample copy of an issue of the BJUP. If you’re not already on the BUPS-L mailing list for announcements, or the BUPS-Dis forum for discussion, you can subscribe through the site. Don’t worry – BUPS membership is free and our conferences are all tailored to fit a student budget. Submit a paper or come along when you can – we’d love to meet you! Latest details of all our activities, profiles of the committee and a continually updated list of upcoming events are always available at: www.bups.org Any enquiries can be addressed to: info@bups.org - 107 - The British Journal of Undergraduate Philosophy presents… The BJUP 2007 Essay Contest Submission deadline: Friday 18th May, noon The BJUP welcomes essays related to one or more of the following topics: MIND LANGUAGE TRUTH KNOWLEDGE Prizes! First prize: £200 and a year’s subscription to The Philosopher’s Magazine Second prize: £50 and £50 in book tokens Third prize: £30 in book tokens & a year’s subscription to the BJUP Outstanding papers will be offered publication in the BJUP and a presentation slot at a BUPS conference This has all been made possible by The Royal Institute of Philosophy, The Philosopher’s Magazine, and Edinburgh University Press www.royalinstitutephilosophy.org www.philosophersnet.com www.eup.ed.ac.uk - 108 - Further details of the BJUP 2007 essay contest Aim The BJUP essay contest has been set up in order to provide an extra forum in which undergraduates can excel, to celebrate the role of undergraduates in academic philosophy, and to encourage the brightest of budding philosophers. For those aspiring to an academic career, this should be incentive enough to enter, although there is also money to be won, as well as reputation. Authors successful in the competition will be offered publication in the BJUP, as well as the opportunity to speak at one of BUPS’s events. Entrance Essays related to at least one of the following topics will be accepted for review: mind, language, truth, knowledge. Submissions should be of no more than 4,000 words in length and should be handled using the journal’s usual policy for ensuring anonymous reviewing. (Essays should be emailed to contest@bups.org in a word document that does not contain the author’s name anywhere. A second document should be sent with the essay title, word count, and author’s name, so that we can re-match essays to authors after reviewing.) Authors will be notified of the competition panel’s decision within a month of the deadline. Only one piece of work per author may be submitted. For this reason, and to regulate the volume of submissions, we can only accept submissions from authors currently enrolled at a UK higher-education institution. Members of the BUPS committee are excluded from the competition. Judging will be handled by our team of professional academics, who act as reviewers on a voluntary basis. - 109 - Subscribing and submitting papers to the BJUP BJUP Subscriptions The BJUP is the Britain’s only national undergraduate philosophy journal. We publish the best papers from BUPS’ conferences, but also accept high-quality essays by direct submission. Our non-profit status keeps the cost of subscription to our print version down, and all BUPS members receive the electronic version of the journal for free. New issues go out quarterly. We offer three levels of subscription: BUPS Member Subscription (Electronic) Becoming a member of BUPS is really, really easy – all you need to do is join the BUPS-L mailing list. The electronic version of the journal is distributed to all BUPS members. We hope you enjoy it! Individual Subscription (Print) An annual subscription to the print version of the journal costs £40 in the UK, and a little more for international postage. Printed in A5 size on 80gsm paper with a 250gsm card cover. Institutional Subscription (Print + Electronic) Institutions (libraries, schools, universities) wishing to subscribe to the journal receive both a print copy and a personalised electronic copy licensed for unlimited distribution to, and printing by, current students of the institution. This package costs £60 per year for UK delivery, slightly more for overseas postage. Subscriptions run for a single academic year, a current subscription covering the print version of issues 2(1)–2(4). Full details of how to subscribe, and methods of payment we accept are available at the journal’s webpage: www.bups.org/BJUP - 110 - Submitting a paper to the BJUP Most papers we publish will be 2,000 – 2,500 words in length. However we will consider papers of any length. We would suggest that you limit your submission to a maximum of 5,000 words, though, since papers longer than this are often better dealt with as a series of shorter, tighter, more focused essays. What we’re looking for in papers that we publish is actually quite simple. We like work that is: • • • • • • carefully structured argumentative rather than merely descriptive clearly written knowledgeable about a given subject area offering a new argument or point of view not just written for area specialists As a general tip, don’t write with ‘This is for a journal, I must be technical, formal and use lots of jargon to show I know my subject...’ running through your mind. Explanation to others who may not have read the same authors as you, clear laying out of thoughts and a good, well-worked-out and -offered argument that says something a bit different and interesting – these are the key characteristics of the best papers we’ve received. Don’t be afraid to tackle difficult or technical subjects – we’re all keen philosophers here – but do so as carefully and clearly as possible and you have a much better chance of being published. Most of our papers are analytic, but we are delighted to accept and publish good papers in both the analytic and continental traditions. We accept papers electronically as Microsoft Word .DOC. If you have problems sending in this format, please contact us and we will try to find another mutually acceptable file format. Papers should be submitted via email to bjup@bups.org and should be - 111 - prepared for blind review with a separate cover sheet giving name, affiliation, contact details and paper title. Don’t worry about following the journal’s house style before submission. The only requirement we have in advance is that you follow English spelling conventions. Any other requirements will be made clear if your paper is accepted for publication. Please do not submit papers for a BUPS conference and the journal at the same time. We’ll make suggestions for rewriting or restructuring papers we think could be publishable with a bit of work. Please do not re-submit a particular paper if it has been rejected for a BUPS conference or the BJUP and has not been reworked. Reviewing papers fairly is a difficult and time-consuming job – please give us a month or so and do not submit your paper elsewhere in the meantime. We run the journal on the minimum copyright requirements possible. By submitting work you license BUPS and the BJUP to publish your work in the print and electronic versions of our journal, and agree to credit the journal as the original point of publication if the paper is later published as part of a collection or book. That’s all – you are not giving us copyright over your work, or granting a licence to reprint your work in the future. We’re budding philosophers not lawyers, so we hope that’s pretty clear and fair. - 112 -