Stephen Hawking recently declared that “Philosophy is dead”. In the opening chapter of his book (co-authored with Leonard Mlodinow) The Grand Design: new answers to the ultimate questions of life, he writes:
…people have always asked a multitude of questions: How can we understand the world in which we find ourselves? How does the universe behave? What is the nature of reality? Where did all this come from? … Traditionally these are questions for philosophy, but philosophy is dead. Philosophy has not kept up with modern developments in science, particularly physics. Scientists have become the bearers of the torch of discovery in our quest for knowledge. (5)
Hawking’s challenge is one we philosophers must take seriously. This is crucial not just to uphold our public pride or our funding or our place in universities. Challenges like those can be answered by appeal to the usefulness of teaching philosophy in preserving cultural heritage, instilling critical thinking, and educating thoughtful citizens. But the real challenge the question presents is to say what it is that philosophical thought – philosophical research – can contribute. And answering this question is crucial not just for external reasons to do with how we appear to others, but internally to us as philosophers, to have a sense of what our legitimate quarry can be, where we can hope to make a contribution, and of what methods might best enable us to do that.
Some philosophers have expressed outrage at Hawking’s comments. But Hawking only articulates a problem – a philosophical problem – we have known about for a long time. Here I want first to look back at how the problem arose within philosophy, and at what responses have been made in the past. Then I will go on to defend a new response: that philosophy should be understood as a normative discipline – one concerned not with how things are, but how they ought to be.
The problem of determining the proper role of philosophy emerged quite naturally and inevitably given the history of the discipline. Philosophy began simply as a “love of wisdom”. Philosophers from Aristotle through Descartes and Kant made contributions not just to topics that would now be discussed in philosophy departments, but also to biology, physiology, physics, mathematics, and astronomy. Gradually, beginning with physics around the seventeenth century, specialised scientific disciplines began to diversify and separate, acquiring their own distinctive subject matter and methodology.
This left philosophers with a pressing question: What is left for philosophers to do? And how should they do it?
The question becomes even more worrying when we contrast the progress and successes in the sciences with the ever-increasing divergences in philosophical theories. Is there anything left for philosophers to do – or is philosophy either to be incorporated with the diverse sciences, or discarded as “dead”?
One early attempt to distinguish the work of philosophy and the sciences was to say that philosophy involved the study of internal, mental phenomena, while the sciences studied external, physicalphenomena. This led to the view that became known as psychologism: that logic and epistemology are really parts of psychology – studies of the laws of thought. But psychology, too, would separate itself from philosophy around the end of the nineteenth century, and the doctrine of psychologism fell on hard times. A central problem, emphasised by Edmund Husserl, is that psychological studies of how we actually think or reason cannot tell us how we ought to reason. But logic, as Husserl argued, is a “practical-normative” discipline, about how we ought to reason, and so logical truths cannot be identified with empirical generalisations about how people actually think.
Questions about the proper role of philosophy were an obsessive concern for philosophers in the early twentieth century. As Gilbert Ryle put it in his “Autobiographical” reflections, “We philosophers were in for a near-lifetime of enquiry into our own title to be enquirers”.
In this stage of the discussion, a broad consensus emerged about the answer. Early twentieth century philosophers as diverse as Ryle, Wittgenstein, Husserl and Carnap embraced a relatively cohesive view of the relation between philosophical and scientific work. Philosophy, on this conception, is conceptual work, while the sciences are engaged in empirical work. Despite internal variations in terms of the role of language in these analyses and the form the analyses should take, all were agreed that the role of philosophy was fundamentally conceptual, not aiming at factual knowledge or empirical explanations. This seemed, at least, to provide a clear division of labour for philosophy and the sciences, and a distinctive (if limited) role for the former -whether in distinguishing sense from nonsense and dissolving pseudo-problems that way, in examining the concepts used in the sciences, or analysing the concepts of ordinary language.
The idea that what philosophy can legitimately do is conceptual rather than empirical work had its heyday in the positivism of the early twentieth century and the ordinary language philosophy of the mid twentieth century. But it didn’t last.
Why not? First, to many it seemed – and still seems – disappointing to think that philosophy can do no more than limn the structure of our ordinary language or concepts, abandoning its traditional grander ambitions. As Ted Sider puts it in his recent book, Writing the Book of the World: “Who would prefer exploring our perhaps parochial conceptual scheme to exploring the fundamental features of reality?” (xxiv). Philosophers working in other areas raised similar worries: many ethicists, for example, would say that they aren’t just interested in our concepts of right and wrong: they care about working out what’s really right and wrong. So the conceptualist move provoked an inevitable backlash from those who hoped to preserve the idea that philosophy could do something more.
Secondly, Quine raised worries about the very idea that there is a distinction between what is analytic – a matter of truth in virtue of meaning (or conceptual truth) – and what is synthetic (a matter of empirical investigation). These were widely, though prematurely, taken on board as reasons for rejecting the division of labour view. Instead, many followed Quine in thinking that philosophical questions in ontology, about what exists, are “on a par with the questions of natural science”.
This in turn led to a great revival of metaphysics and an ever more diverging range of answers to questions about what exists. Some have argued that the only composite objects that exist are organisms (so that, in the memorable criticism of Eli Hirsch, one ends up accepting that apple trees but not apples exist). Others have argued that we should accept only that certain basic, simple objects (“simples”) exist – but not any composite objects like artefacts or organisms. Still others have denied that there are any objects at all (accepting only undifferentiated “stuff”, not divided into objects). In the hands of such metaphysicians, philosophy has come again to look like a quasi-science – with philosophers pronouncing from the armchair about the constituents and structure of reality, and offering a huge diversity of theories with no empirical means to choose among them – indeed nothing more than appeals to competing “theoretical virtues” that seem to simply trade places from theory to theory.
The increasing proliferation of often wildly revisionary and empirically indistinguishable metaphysical “theories” has led to new doubts about the role and limits of metaphysics, and more broadly, about what philosophy can do and how we can do it. We have reached the stage of a new crisis about the methods and role of metaphysics, and of philosophy more generally. As the plethora of recent papers, conferences, and volumes on philosophical methodology and metametaphysics makes clear, the problem of what philosophy can legitimately do is all too familiar to contemporary philosophers. Hawking may have brought the problem into recent public discussion, but he cannot be credited with identifying it or bringing it to the attention of philosophers themselves.
The recent history of the debate about the proper role of philosophy apparently leaves us stuck in a dilemma. Hold that there is nothing distinctive that philosophy can do, that it is “on a par” with the natural sciences, and you end up with an untethered proliferation of fanciful views that seem like either bad science or wild speculation, and no idea how to choose among them. Adopt the idea that there is a division of labour, and that philosophy’s role lies in conceptual work, and you end up with a rather parochial and limited conception of philosophy that doesn’t seem to leave proper room for important areas like ethics.
Let’s look again at Hawking’s challenge. The questions he lists as best handed over to science are questions like “How does the universe behave?”, “What is the nature of reality?”, and “Where did all this come from?” These I (admittedly unlike some of my brethren) agree cannot be addressed without the serious empirical work of the sciences.
But other core questions of philosophy are notably absent from his list: How ought I to live my life? What should we treat as fundamental human rights? What form of government would be the most just? How should we reason in order to best ensure that we reach a true conclusion, if we start from true premises? How should we judge the value of a work of art?
Philosophers working in such core areas as ethics, political philosophy, social philosophy, logic, and aesthetics might be irritated by the generality of Hawking’s remarks, but not at all worried.
What do these remaining questions have in common? They are all normative questions: questions about what we ought to do, how we ought to behave, be governed, reason, or evaluate art. The questions Hawking focused on, by contrast (and not surprisingly), are factual questions about the history, structure, or nature of the universe (what exists, what it’s like, what laws it follows, where it came from).
This distinction was already present in the standard criticism of psychologism – by those who argued that we must distinguish the descriptive question of how people actuallyreason from the normative question about how we ought to reason. While the former question might be best farmed out to psychology (or ultimately perhaps to neuroscience), questions about how we should reason remain for logic.
This idea can be turned into a general proposal: that philosophy is, at bottom, a normative discipline. This has a certain advantage at the start, given the familiar distinction between is and ought: scientific, empirical methods are the best methods by far of telling us what is the case. But, taken on its own, this does not tell us anything about what ought to be the case – about how we ought to speak, think, reason, or act. Putting it in this form already allays some of the objections to old-fashioned conceptual analysis. For philosophy on this model is acknowledged to be concerned with norms generally – including but leaving room for more than just the norms governing reasoning with our concepts.
While this view clearly fits well with much work in ethics and political philosophy, logic and aesthetics, one might still worry that it builds a fortress that leaves large portions of philosophy – including philosophy of science, epistemology, and metaphysics – outside the wall and unprotected.
But it is not difficult to see at least large portions of philosophy of science and epistemology in a normative light. Epistemology investigates such overtly normative questions as how we should acquire evidence and draw conclusions, when we should count someone as having knowledge, and how we should reason under conditions of uncertainty.
Philosophers of science investigate such issues as how we should reason based on evidence, what methodologies should be employed or cautions should be heeded in scientific reasoning, what conclusions about the world should be drawn from the empirical results of quantum mechanics, or what conclusions about the mind should be drawn from results in cognitive science and psychology. This of course does not – and should not – make philosophy of science entirely distinct from science. Scientists, too, are concerned to work out what conclusions about the world we should draw from the evidence. Moreover, philosophers of science and scientists alike may be concerned with normative issues about what conceptual scheme we should adopt to do work in biology, ecology, or quantum mechanics. Such choices may have to be empirically informed and responsive to the practices and needs of working scientists – and to that extent cannot be done without intimate knowledge of the sciences. But it may also involve the kind of broad conceptual work that involves considering alternative conceptual schemes and integrating conceptual schemes into a wider picture – tasks we might hope philosophers will be particularly well suited to help with. In any case, the goal is not and should not be to make science and philosophy disjoint, but rather to show a role for philosophical work, whether in the hands of scientists or philosophers, or both working together.
But what about good old metaphysics? Is it simply to be left to the slaughter, so that we can retreat and save the rest of philosophy? (It was always first to be suspected, last to be saved.) I actually think that this doesn’t leave metaphysics behind as a sacrifice, but instead gives a healthy way of reconceiving what much work in metaphysics has been, and can (with undeniable legitimacy) purport to be up to.
Think first about ordinary language philosophy and traditional conceptual analysis. Those who work to analyse our concept of freedom are, at least in part, concerned with determining what (given the contours of that concept) we should conclude about freedom – whether it precludes determinism, entails responsibility, etc. Those who work on personal identity are concerned, at least in part, with what (given the contours of our concept of “person”, or given what we want or need that concept to do for us) we should conclude about when a person survives or perishes. A great many metaphysical puzzles, such as puzzles about material constitution, are best construed as puzzles about which of various premises we should drop, or reinterpret, when we have an apparent contradiction. (We think there is a statue there, and we think there is a lump of clay there, and we think that the statue isn’t identical to the lump since it couldn’t survive being squashed while the lump could. But we also think there is only one object there. What should we do in the face of this apparently inconsistent set of beliefs?) The project of conceptual analysis – which so many took to be the legitimate role for philosophy, in the wake of the first phase of worries – can be seen as engaged in figuring out what conclusions we should draw about various topics, given the contours of our concepts. As Husserl would have put it, it is concerned with rules of material inference (given the structure of our concepts) rather than the merely formal rules of reasoning logic addresses.
But does it then inevitably leave metaphysics to be just a parochial game of figuring out how to reason from our extant concepts?
Not necessarily. Carnap acknowledged a useful role for philosophy not just in conceptual analysis, but also in what he called “conceptual engineering” – working out what system of concepts is, or would be, most useful for science. In normative terms, these are questions about what concepts we ought to adopt and work with – for a given purpose, in a given context.
The project of “conceptual engineering” can be understood more broadly, concerning more purposes than just those of the sciences, and more concepts than the logical and scientific. Even in standard metaphysics, we need not ask simply “What is freedom?” or “What is a person?” and try to work out what follows from our standard concept – whether by traditional analysis or experimental philosophy (though that may be interesting enough). We may do better to step back to see what function these concepts need to serve for us – why we want such concepts, what role they are to play in our attributions of praise and blame, of legal and moral responsibility, of rights and duties. But we also do better at that if we are clear-headed about the nature of our project and don’t deceive ourselves into thinking we are discovering some metaphysical reality that others were mistaken about.
More interesting projects also arise, which are far from parochial dredging of our conceptual floor. Sally Haslanger, for example, argues powerfully in her book Resisting Reality that what we should really be doing in philosophy of race and gender is not to ask what race and gender are – in the sense of analysing how our extant concepts work, or what we should conclude based on them. Instead, she argues, we should ask what they should be – that is, what categories we should use to try to achieve social justice (365).
On this view, we retain something of the traditional idea that metaphysics is engaged in conceptual work, but broaden it. Philosophy, on this model, needn’t be engaged just in conceptual analysis, for it can take on normative conceptual work of all kinds – in working out what follows from our extant concepts, how we should revise vague or problematic concepts (given our purposes), which concepts should be kept or adapted, ditched or invented given a variety of purposes from the scientific to the ethical.
To be clear, this proposal – that philosophy be seen as a normative discipline – should itself be seen as normative. The aim here is not to describe all that philosophy has done or conceives of itself as doing, but rather to propose a self-conception that gives philosophy a clear and useful role, answers the challenge, and still has a way to capture much of what is interesting, exciting, and useful in what philosophy has done and can do.
Seeing metaphysics, and philosophy more generally, in that light does more than just defend it against the doubters and naysayers. It may also make a difference to how we philosophers think of what we are doing, and to what methods we employ and standards we use when evaluating competing views. It might even make the prevalence of competing views less of an embarrassment than it seemed. For if we see the competing views as tacitly answering questions about what we should do, we might expect some legitimate variation in the answers wherever the purposes are different. Similarly, where different norms are at stake – between those engaged, say, in determining what we should say given our extant conceptual structure, and those engaged in figuring out what our conceptual structure ought to be – we might dissolve old disputes and make new progress.
This proposal has many advantages. First, on this view, philosophy is not a rival to the sciences, and has a clear role. This is, of course, not to suggest that the roles of philosophy and the sciences are totally distinct – scientists themselves, certainly, may be concerned with how we should reason given the evidence, with what species concept one should adopt for one or another purpose in biology, and so on. But these are questions, too, that (while empirical data may be relevant and impactful), we should not fool ourselves into thinking can be determined purely empirically, without bringing in normative work. Here we may overlap and work fruitfully together without the risk of philosophical work being rendered redundant or pseudo-scientific.
The proposal won’t satisfy everyone, of course. Those who think of metaphysics as discovering the basic structure of reality, or figuring out what exists, or what is fundamental (and who resist reinterpretation of these projects in normative/conceptualist terms) will not like it. That is their prerogative – but I will have to leave them to face Hawking’s challenge alone.
For myself, I find that answering Hawking’s challenge in this normative way not only saves our self-respect, but also, more importantly, can leave us with a far clearer picture of what we can do and how to do it. It also leaves us with a very rich picture indeed of what philosophy can do.