"To adapt a quotation by Kant in a different though related context: philosophy of science without scientific input is empty, while science without philosophical guidance is blind. At any rate it is rendered perilously apt to mistake the seductions of pure hypothetical invention for the business of formulating rationally warranted, metaphysically coherent, and – if only in the fullness of time – empirically testable conjectures."
Hawking is better off writing about the adventures of George.
"Hawking contra Philosophy"
Christopher Norris presents a case for the defence.
Christopher Norris presents a case for the defence.
Stephen Hawking recently fluttered the academic dovecotes by writing in his new book The Grand Design – and repeating to an eager company of interviewers and journalists – that philosophy as practised nowadays is a waste of time and philosophers a waste of space. More precisely, he wrote that philosophy is ‘dead’ since it hasn’t kept up with the latest developments in science, especially theoretical physics. In earlier times – Hawking conceded – philosophers not only tried to keep up but sometimes made significant scientific contributions of their own. However they were now, in so far as they had any influence at all, just an obstacle to progress through their endless going-on about the same old issues of truth, knowledge, the problem of induction, and so forth. Had philosophers just paid a bit more attention to the scientific literature they would have gathered that these were no longer live issues for anyone remotely au fait with the latest thinking. Then their options would be either to shut up shop and cease the charade called ‘philosophy of science’ or else to carry on and invite further ridicule for their head-in-the-sand attitude.
Predictably enough the journalists went off to find themselves media-friendly philosophers – not hard to do nowadays – who would argue the contrary case in a suitably vigorous way. On the whole the responses, or those that I came across, seemed overly anxious to strike a conciliatory note, or to grant Hawking’s thesis some measure of truth as judged by the standards of the natural science community while tactfully dissenting with regard to philosophy and the human sciences. I think the case needs stating more firmly and, perhaps, less tactfully since otherwise it looks like a forced retreat to cover internal disarray. Besides, there is good reason to mount a much sturdier defence on principled grounds. These have to do with the scientists’ need to philosophize and their proneness to philosophize badly or commit certain avoidable errors if they don’t take at least some passing interest in what philosophers have to say.
Science is Philosophical
Professor Hawking has probably been talking to the wrong philosophers, or picked up some wrong ideas about the kinds of discussion that currently go on in philosophy of science. His lofty dismissal of that whole enterprise as a useless, scientifically irrelevant pseudo-discipline fails to reckon with several important facts about the way that science has typically been practised since its early-modern (seventeenth-century) point of departure and, even more, in the wake of twentieth century developments such as quantum mechanics and relativity.
Science has always included a large philosophical component, whether at the level of basic presuppositions concerning evidence, causality, theory-construction, valid inference, hypothesis-testing, and so forth, or at the speculative stage where scientists ignore the guidance offered by well-informed philosophers only at risk of falling into various beguiling fallacies or fictions. Such were those ‘idols of the theatre’ that Bacon warned against in his New Organon of 1620, and such – albeit in a very different philosophic guise – those delusive ideas that, according to Kant, were liable to lead us astray from the path of secure investigation or truth-seeking enquiry. This was sure to happen, he warned, if the exercise of pure (speculative) reason concerning questions outside and beyond the empirical domain were mistakenly supposed to deliver the kind of knowledge that could be achieved only by bringing sensuous intuitions under adequate or answering concepts. While in no way wishing to lumber science with the baggage of Kantian metaphysics I would suggest that this diagnosis, or something like it, applies to a great many of the speculative notions nowadays advanced by theoretical physicists including proponents of string theory (Hawking among them) and some of the more way-out quantum conjectures. These thinkers appear unworried – blithely unfazed, one is tempted to say – by the fact that their theories are incapable of proof or confirmation, or indeed of falsification as required by Karl Popper and his followers. After all, it is the peculiar feature of such theories that they posit the existence of that which at present, and perhaps forever, eludes any form of confirmation by observation or experiment.
True, science has achieved some of its most notable advances precisely by venturing beyond the furthest limits of evidential proof. It has often broken new ground by following some speculative line of thought that involves a readiness, at least for the time being, to make do without the props and securities of ‘good’ scientific method. Indeed, this reliance on theoretical commitments that exceed the utmost scope of empirical testing is something that some philosophers would attribute even to basic physical laws or widely taken-for-granted scientific truths. On their view there is no such thing as plain empirical self-evidence, since observations are always to some degree theoretically informed. By the same token, scientific theories are always ‘underdetermined’ by the best evidence to hand, meaning that the evidence is always open to other, equally rational interpretations given some adjustment of this or that ‘auxiliary hypothesis’ or negotiable element of background belief. All the same, I don’t want to push that line of argument too far, because among some philosophers of science it has now become an article of faith; a dogma maintained just as fixedly as any precept of the old, unreconstructed positivist creed. Moreover it has given rise to a range of relativist or ‘strong’ sociological approaches which use the theory-ladenness and underdetermination theses to cast doubt on any distinction between true and false theories, valid and invalid hypotheses, or science and pseudo-science.
Very likely it is notions of this kind – ideas with their home ground in sociology, or cultural studies, or on the wilder shores of philosophy of science – which provoked Professor Hawking to issue his pronouncement. However they are in no way germane to my point about the speculative element involved in many episodes of major scientific advance and how philosophy has played its jointly enabling and regulative part in that process. By this I mean its role as a source of new ideas or creative hypotheses and also as a source of guiding precepts with respect to such matters as empirical evidence, logical validity, inductive warrant, corroboration, falsification, hypothesis-testing, causal reasoning, probability-weighting, and so forth. These serve to keep science securely on track and prevent it from taking the seductive turn toward pure, evidentially unanchored speculation or sheer science-fiction fantasy. That scientists can mostly do this for themselves is no doubt true enough although, I should add, it is very largely the long-term result of the work of philosophers. Ever since Aristotle there has existed a close though historically fluctuating relationship between the natural sciences and those branches of philosophy that took it as a part of their task to provide science with a clearer grasp of its own methodological bearings. Moreover it has sometimes been primarily a shift of philosophical perspective that has brought about some epochal change of scientific paradigm such as those whereby, in the insouciant phrase of American philosopher W.V. Quine, “Kepler superseded Ptolemy, or Einstein Newton, or Darwin Aristotle.”
I have no quarrel with Hawking’s aversion to philosophy of science in so far as it is provoked by the kind of wholesale paradigm-relativism that Quine was seeking to promote. On Quine’s account (and that of Thomas Kuhn) we should think of scientific theory-change as involving so radical a shift of conceptual schemes as to render the history of science rationally unaccountable and philosophy of science a poor (since entirely dependent) relation of sociology and behavioural psychology. If that were the sole position available to present-day philosophers owing to some large-scale failure of intellectual nerve then Hawking would be fully justified in launching his anti-philosophy salvo. However this ignores the strong turn toward a realist and causal-explanatory approach that has been the single most conspicuous feature of philosophy of science during the past two decades. In place of that earlier relativist drift these thinkers advocate a robust conception of natural kinds along with their essential structures, properties, and causal dispositions. Crucially in the present context their approach offers a critical purchase on the issue of what properly counts as scientific enquiry and what should more aptly be classed as metaphysical conjecture or (at the limit) mere invention.
So philosophy of science now looks set to reoccupy its native ground by getting back in touch with physics. This is not just a relatively trivial semantic point about the physical sciences having been described as so many branches of ‘natural philosophy’ until quite recently. Rather it is the point that scientific theories – especially theories of the ultra-speculative kind that preoccupy theoretical physicists like Hawking – involve a great deal of covert philosophising which may or may not turn out to promote the interests of knowledge and truth. This had better be recognised if we are not to be taken in by a false appeal to the authority of science as if it possessed the kind of sheer self-evidence or indubitable warrant that could rightfully claim to evict ‘philosophy’ as a relic from the pre-scientific past.
Least of all should philosophers carry their justified respect for science and its many impressive achievements to the point of ceding all authority over issues that lie within their own sphere of competence. Thus it is counter-productive for everyone concerned, philosophers and physicists alike, when Quine and others suggest that we should always be willing to change the ground-rules of logic so as to help us find room for certain otherwise puzzling, anomalous, or downright baffling results. Perhaps the seeming quantum paradox of wave/particle dualism can have its sting temporarily removed by lifting the classical rules of bivalence or excluded middle, i.e., those that would require that we accept either the statement ‘light propagates as waves’ or the statement ‘light is a stream of particles’ but surely not both on pain of logical contradiction. However the revisionist ‘solution’ gives rise to yet more intractable problems since it leaves both scientists and philosophers stuck with a huge normative deficit. After all, if they accepted Quine’s proposal then they would lack the most basic conceptual resources for assessing statements, theories or hypotheses in point of their internal (logical) consistency or even concerning the extent to which they hung together properly with other items of scientific lore.
Here again philosophers would do much better to stick to their guns, reject this particular line of least resistance, and hold out for the indispensability (on empirical as well as ‘purely’ rational grounds) of a due respect for the classical rule of bivalent truth/falsehood. Not that it could ever achieve what Hawking seems to envisage in the final paragraph of his book when he marvels at the thought of how ‘abstract logic’ could have thrown up the sheer wondrous profusion of present-day scientific knowledge. Here the point needs making – one to which his own book bears ample witness – that the knowledge in question has resulted from a disciplined yet often highly inventive project of enquiry wherein ‘abstract’ reasoning plays a crucial though far from all-encompassing or self-sufficiently productive role. This project combines the basic procedures of logical, e.g., hypothetico-deductive thought and inductive reasoning on the evidence with a whole range of ancillary resources such as analogy, thought experiments, rational conjecture, and – subsuming all these – inference to the best, most adequate explanation.
Hawking offers numerous examples of the use of each of these philosophical tools in the course of his book, along with other cases where their joint operation is the only thing that could possibly explain how science has been able to achieve some particular advance. All the same he is compelled by the ‘abstract logic’ of his own doctrinaire science-first approach to push that evidence temporarily out of sight when declaring the total irrelevance of philosophy for anyone possessed of an adequate (i.e., scientifically informed) worldview. Indeed it may be good for philosophers occasionally to remind scientists how their most productive thinking very often involves a complex interplay of empirical data, theories, working hypotheses, testable conjectures and even (sometimes) speculative fictions. Likewise absent from Hawking’s account is philosophy’s gatekeeper role in spotting those instances where science strays over without due acknowledgement from one to another mode, or – as frequently happens nowadays – where certain evidential constraints are lifted and empirically informed rational conjecture gives way to pure fabulation.
Besides this, there are supposedly cutting-edge theories which turn out, on closer inspection, to unwittingly replicate bygone notions from the history of thought that have been criticised and eventually laid to rest. Hawking’s book puts forward two such theories. One is his linchpin ‘M-theory’ having to do with the multiple dimensions – eleven at the latest count – that are taken to constitute the ultimate reality beyond appearances despite our sensory perception being limited to the three-plus-one of our familiar spatio-temporal world. On this account there cannot be a single, comprehensive ‘Theory of Everything’ of the kind favoured by sanguine types like Steven Weinberg but we can hope to get a whole range of specially tailored, region-specific theories which between them point toward the nature and structure of ultimate reality. The other, closely related to that, is Hawking’s idea of ‘model-dependent realism’ as an approach that makes allowance (as per orthodox quantum mechanics) for the effect of observation on the item observed but which nonetheless retains an adequate respect for the objectivity of scientific truth.
Here Hawking’s argument shows all the signs of a rudderless drifting between various positions adopted by different philosophers from Kant to the present. He spends a lot of time on what seems to be a largely unwitting rehash of episodes in the history of idealist or crypto-idealist thought, episodes which have cast a long shadow over post-Kantian philosophy of science. That shadow still lies heavy on Hawking’s two central ideas of M-theory and model-dependent realism. They both look set to re-open the old Kantian split between a ‘noumenal’ ultimate reality forever beyond human knowledge and a realm of ‘phenomenal’ appearances to which we are confined by the fact of our perceptual and cognitive limits. So if Hawking is right to charge some philosophers with a culpable ignorance of science then there is room for a polite but firm tu quoque, whether phrased in terms of pots calling kettles black or boots on other feet. For it is equally the case that hostility or indifference toward philosophy can sometimes lead scientists, especially those with a strong speculative bent, not only to reinvent the wheel but to produce wheels that don’t track straight and consequently tend to upset the vehicle.
A firmer grasp of these issues as discussed by philosophers during the past few decades might have moderated Hawking’s scorn and also sharpened his critical focus on certain aspects of current theoretical physics. My point is not so much that a strong dose of philosophic realism might have clipped those speculative wings but rather that philosophers are well practised in steering a course through such choppy waters, or in managing to navigate despite all the swirls induced by a confluence of science, metaphysics, and far-out conjecture. After all, physics has increasingly come to rely on just the kind of disciplined speculative thinking that philosophers have typically invented, developed, and then criticised when they overstepped the limits of rationally accountable conjecture. Such are those ‘armchair’ thought-experiments that claim to establish some substantive, i.e., non-trivial thesis concerning the nature of the physical world by means of a rigorous thinking-through that establishes the truth (or, just as often, the demonstrable falsehood) of any statement affirming or denying it.
No doubt there is room to debate whether these are really (and remarkably) instances of scientific discovery achieved through an exercise of a priori reasoning or whether they amount, as sceptics would have it, to a species of disguised tautology. However there are just too many impressive examples in the history of science – from Galileo’s marvellous thought-experiment showing that Aristotle must have been wrong about falling bodies to a number of crucial quantum-related results – for anyone to argue convincingly that results obtained in the ‘laboratory of the mind’ can only impress philosophers keen to defend their patch. Indeed, there is a sense in which the scientific enterprise stands or falls on the validity of counterfactual-conditional reasoning, that is to say, reasoning from what necessarily would be the case should certain conditions obtain or certain hypotheses hold. In its negative guise, this kind of thinking involves reasoning to what would have been the outcome if certain causally or materially relevant factors had not been operative in some given instance. Hawking constantly relies on such philosophical principles in order to present and justify his claims about the current and likely future course of developments in physics. Of course he is very welcome to them but he might do better to acknowledge their source in ways of thinking and protocols of valid argumentation that involve distinctly philosophical as well as scientific grounds.
This brings us back to the point likely to provoke the most resistance from those scientists – chiefly theoretical physicists – who actually have the most to gain from any assertion of philosophy’s claim to a hearing in such matters. It is that scientists tend to go astray when they start to speculate on issues that exceed not only the current-best observational evidence but even the scope of what is presently conceivable in terms of testability. To speak plainly: one useful job for the philosopher of science is to sort out the errors and confusions that scientists – especially theoretical physicists – sometimes fall into when they give free rein to a speculative turn of mind. My book Quantum Theory and the Flight from Realism found numerous cases to illustrate the point in the statements of quantum theorists all the way from Niels Bohr – a pioneering figure but a leading source of metaphysical mystification – to the current advocates (Hawking among them) of a many-worlds or ‘multiverse’ theory. To adapt the economist Keynes’ famous saying: those scientists who claim to have no use for philosophy are most likely in the grip of a bad old philosophy or an insufficiently thought-out new one that they don’t fully acknowledge.
There is a large supply of present-day (quasi-)scientific thinking at the more – let us say – creative or imaginative end of the scale that falls into just this hybrid category of high-flown metaphysical conjecture tenuously linked to certain puzzling, contested, or at any rate far from decisive empirical results. Nor is it mere hubris for philosophers to claim a special competence in judging when thought has crossed that line from the realm of rational, scientifically informed but so far unproven conjecture to the realm of unanchored speculation or outright science fiction fantasy. One has only to pick up a copy of New Scientist or Scientific American to see how much of the latest thinking inhabits that shadowy border-zone where the three intermingle in ways that a suitably trained philosopher would be best equipped to point out. Nowhere is this more evident than in the past hundred years of debate on and around the seemingly paradoxical implications of quantum mechanics. Those paradoxes include wave/particle dualism, the so-called ‘collapse of the wave-packet’, the observer’s role in causing or inducing said collapse, and – above all since it appears the only way of reconciling these phenomena within anything like a coherent ontology – faster-than-light interaction between widely separated particles.
I shall risk the charge of shameless self-advertisement and suggest that readers take a look at my book for the case that these are pseudo-dilemmas brought about by a mixture of shaky evidence, dubious reasoning on it, fanciful extrapolation, and a flat refusal to entertain alternative theories (such as that of the physicist David Bohm) which considerably lighten the burden of unresolved paradox. At any rate we are better off trusting to the kinds of advice supplied by scientifically-informed philosophers with a well-developed sense of how speculative thinking can sometimes go off the rails than the kinds – including the advice ‘let’s put a stop to philosophy’ – issued by philosophically under-informed scientists.
No doubt there is a fair amount of ill-informed, obtuse, or ideologically angled philosophy that either refuses or tries but fails to engage with the concerns of present-day science. One can understand Hawking’s impatience – or downright exasperation – with some of the half-baked notions put around by refuseniks and would-be engageniks alike. All the same he would do well to consider the historically attested and nowadays more vital than ever role of philosophy as a critical discipline. It continues to offer the sorts of argument that science requires in order to dispel not only the illusions of na ïve sense-certainty or intuitive self-evidence but also the confusions that speculative thought runs into when decoupled from any restraining appeal to regulative principles such as that of inference to the best explanation. To adapt a quotation by Kant in a different though related context: philosophy of science without scientific input is empty, while science without philosophical guidance is blind. At any rate it is rendered perilously apt to mistake the seductions of pure hypothetical invention for the business of formulating rationally warranted, metaphysically coherent, and – if only in the fullness of time – empirically testable conjectures.
[Christopher Norris is Professor of Philosophy at Cardiff University.]
Stephen Hawking with Leonard Mlodinow, The Grand Design: new answers to the ultimate questions of life (Bantam Press, 2010)
Christopher Norris, Quantum Theory and the Flight from Realism: philosophical responses to quantum mechanics (Routledge, 2000)
David Papineau (ed.), The Philosophy of Science (O.U.P., 1996)
Thanks to stringer Tim for the article.