Naturalism and Normativity

INTRODUCTION

SCIENCE, NATURALISM, AND THE PROBLEM OF NORMATIVITY

Mario De Caro and David Macarthur

Normativity is ubiquitous

—Hilary Putnam

NORMATIVITY AND THE SCIENTIFIC NATURE

Normativity concerns what we should or ought to do and our evaluations of things or states of affairs. We normally say, for example, that one ought to keep one’s promises, that if one accepts p and If p, then q, one ought to accept q, or that Mozart was a better musician than Salieri. Plausibly, the sciences describe how things are, particularly the causal powers or causal regularities that exist in the world, lawlike or otherwise. Consequently, if one follows modern Scientific Naturalism in supposing that natural science, and only natural science, tells us what there is in the world, then there seems to be no room for the existence of normative facts—or at least this will be so insofar as they cannot be reduced to the kinds of objective, causal facts with which natural science deals. Such considerations set the stage for one of the fundamental issues confronting philosophers today: Are there any indispensable, irreducible normative facts involving, say, reasons, meanings, and values that are not, or cannot, be accommodated within the scientific image of the world? John McDowell provides a trenchant expression of the sense that this is indeed the case:

Modern science understands its subject matter in a way that threatens, at least, to leave it disenchanted, as Weber put the point in an image that has become a commonplace. The image marks a contrast between two kinds of intelligibility: the kind that is sought by (as we call it) natural science [the kind we find in a phenomenon when we see it as governed by natural law] and the kind we find in something when we place it in relation to other occupants of the logical space of reasons [the kind of intelligibility that is proper to meaning].¹

The issue has been in the air since at least the late nineteenth century when there arose in the German-speaking world debates about the aims, subject matter, and methods of the social sciences: the so-called Methodenstreit.² Are human beings, their societies, and their cultural institutions open to the same kind of explanations as the natural objects studied by the natural sciences? If so, can they be fully explained in such terms? Does the fact that human beings engage with and respond to normative items such as meanings, values, and reasons require methods of investigation and interpretation (e.g., empathetic understanding, intentional explanation, introspection) that have no counterparts in the natural or, more broadly, social sciences?

In contemporary philosophy these issues often find expression in terms of the problem of placing or locating normative phenomena in the scientific image of the world. According to the most common form of naturalism, the image of the world provided by the natural sciences is all the world there is. Since this image seems, prima facie, not to include normative phenomena, the following question arises: What place can we find for the normative in the natural world? The question becomes urgent if, as seems highly plausible, we suppose that central normative phenomena are not going to be explained away or eliminated. Call this, for present purposes, the placement problem. Similar placement problems arise with respect to intentional attitudes, consciousness, abstract entities, and so forth.

Although the placement problem is, in the first instance, posed in ontological terms, it has a semantic analogue or counterpart that is at least equally pressing. The aim of a great many programs of naturalization is to explain the role or function of normative concepts, given the supposed scientific discovery that the world lacks any autonomous normative facts. Theoretical options at this point include error theories, according to which normative concepts, although representational in intent, are true of nothing in the actual world; reductionist theories, according to which normative concepts are ultimately reducible to respectable scientific concepts, which themselves straightforwardly represent features of the world; and nonfactualist theories (e.g., classical expressivism), according to which normative concepts are not in the business of representing the world at all. Nonfactualist theories typically go on to offer alternative accounts of the roles normative concepts play in our lives, for example, being expressive of some noncognitive state such as desire.³

THE DOCTRINES OF SCIENTIFIC NATURALISM

In order to understand better why naturalism is thought to give rise to problems in accounting for normativity, we must enter into the difficult business of trying to provide at least a preliminary definition of the term naturalism. It is a commonplace that this term has a long history and a variety of uses. The idea of philosophy having some important relation to, or conception of, nature has been a recurrent theme since at least the time of Aristotle. Our main concerns in this volume, however, will be with two varieties of naturalism that are of particular relevance to contemporary philosophy: Scientific Naturalism, which on a certain narrow understanding represents the current orthodoxy among the large majority of Anglo-American philosophers who fly the banner of naturalism; and Liberal Naturalism, a position or set of positions that, according to the authors of this volume, hold the promise of overcoming the impasses associated with scientific naturalist orthodoxy. These include the apparent intractability of the placement problem and the alleged unattractiveness of the alternatives open to a naturalist treatment of normative and evaluative concepts. Moreover, since orthodox Scientific Naturalism seems at odds with the idea of a distinctive philosophical method, Liberal Naturalism promises to allow for a more realistic and fruitful conception of the relation between philosophy and science.

What makes Scientific Naturalism and Liberal Naturalism both versions of naturalism is that neither countenances the supernatural, whether in the form of entities (such as God, spirits, entelechies, or Cartesian minds), events (such as miracles or magic), or epistemic faculties (such as mystical insight or spiritual intuition). The importance of this for the philosophical approach to normativity is that any form of naturalism will be opposed to Platonism about norms, where this is understood as the view that normative facts hold wholly independently of human practices (say, of reason giving) and are, as it were, simply there anyway waiting to be discovered. For similar reasons it will be opposed to a Moorean non-naturalism that holds that our access to normative facts is by way of a sui generis epistemic faculty of intuition directed at just this kind of fact. And of course it will be opposed to any theistic foundation for normative facts or our access to them.

However the very category of the supernatural is controversial since, for one thing, not everything that we take to be nonsupernatural, such as numbers or propositions, can be understood in unproblematically causal terms. One way of thinking about the difference between Scientific and Liberal Naturalism is as a dispute about how the complementary categories of the natural and the supernatural should best be understood. Scientific Naturalism interprets the natural strictly in terms of the scientific image of the world, narrowly or broadly conceived, whereas Liberal Naturalism, or some versions of it, offers a broader, more expansive conception of nature that makes room for a class of nonscientific, but nonetheless nonsupernatural, entities.

Let us now turn to consider the general category of Scientific Naturalism in some more detail. Scientific Naturalism can be seen as a cluster of positions that arise from an ontological doctrine and/or a methodological doctrine.⁴ The ontological doctrine of scientific naturalism is famously, if vaguely, expressed by Wilfrid Sellars’s remark that science is the measure of all things, of what is that it is, and of what is not that it is not.⁵ Somewhat more precisely, the view is the following:

Ontological doctrine of Scientific Naturalism. The world consists of nothing but the entities to which successful scientific explanations commit us.

Scientific Naturalism is also definable in terms of a methodological commitment, often held in tandem with the ontological commitment above. Arthur Danto, some time ago, wrote that naturalism is a species of philosophical monism according to which whatever exists or happens is natural in the sense of being susceptible to explanation through methods… paradigmatically exemplified by the natural sciences.⁶ From a survey of the literature it seems that according to the majority of scientific naturalists, these explanations and methods are broadly empirical in character.⁷ Consequently, another tenet of Scientific Naturalism is the following:

Methodological doctrine of Scientific Naturalism. Scientific inquiry is, in principle, our only genuine source of knowledge or understanding. All other alleged forms of knowledge (e.g., a priori knowledge) or understanding are either illegitimate or are reducible in principle to scientific knowledge or understanding.⁸

Some immediate comments about these two doctrines are in order. First, the ontological doctrine is a metaphysical thesis not a scientific one and it plays a primary role in the way scientific naturalists give content to the otherwise rather vague antisupernaturalist claim shared by all contemporary forms of naturalism, scientific or nonscientific. Second, that scientific naturalists tend to have an ontology based on causal facts is evident from the way in which the ontological status of mathematical and abstract entities is somewhat of an embarrassment, and so generally avoided or quietly passed over. Arguably, these entities have to be accepted because of their indispensability in the explanations of the natural sciences, but their nature is not easily explained in scientific terms—especially if we think of science as primarily in the business of providing causal explanations.⁹ Third, the interpretation of the ontological doctrine is importantly sensitive to how broadly the scope of the term science is conceived and, consequently, sensitive to how broadly one interprets the scientific image of the world and our methods for investigating it.

A further set of issues is raised by the fact that the philosophical method is traditionally supposed to be a priori, the discovery of necessary truths about the world by the use of reason alone. The scientific naturalists’ methodological doctrine thus has strongly antitraditional implications for our conception of the method and role of philosophy in relation to the sciences. In particular, following Quine, many scientific naturalists claim that philosophy, far from being the final tribunal or foundation for scientific claims, must be methodologically continuous with science. David Papineau, for example, writes that philosophy and science are engaged in essentially the same enterprise, pursuing similar ends and using similar methods. In Papineau’s opinion, the relatively superficial differences between the two disciplines include philosophy’s greater generality, differences in the ways the two disciplines gather their data, and the fact that the philosophical issues tend to generate some kind of theoretical tangle.¹⁰

In Naturalism and Quietism, which was originally written for this volume,¹¹ Richard Rorty follows Brian Leiter in dividing the Anglo-American philosophical world into Naturalists and Wittgensteinian Quietists, a division that Rorty describes as the deepest and most intractable difference of opinion within contemporary Anglophone philosophy. Rorty sees this split as, among other things, a battle between a resurgent (scientific) metaphysical approach to philosophy associated with Saul Kripke, David Lewis, Kit Fine, Peter Van Inwagen, and David Armstrong and the alternative, typically nonmetaphysical, approaches proposed by Ludwig Wittgenstein, Wilfrid Sellars, Donald Davidson, John McDowell, and Robert Brandom. In further articulating this division, Rorty discusses three debates: between Huw Price and Frank Jackson over representationalism; between John McDowell and Timothy Williamson over whether there can be elusive objects; and between Robert Brandom and Jerry Fodor over inferentialist semantics.

Hilary Putnam in his chapter in this collection takes up the question of the relation between science and philosophy and casts a critical eye over the naturalistic attempt to absorb the latter into the former. In Putnam’s reconstruction the growing secularization of philosophy in the past century or so has coincided with the rise of Scientific Naturalism, which involves a new conception of the relation between philosophy and science. According to the scientific naturalist, philosophy not only loses its traditionally preeminent role as a final court of appeal regarding matters of existence and knowledge; indeed, philosophy survives in name only, for it has no autonomous role to play. In Putnam’s alternative image, philosophy is neither absolutely autonomous nor totally lacking in autonomy. In an age of science, Putnam urges us not to forget that philosophy has, and needs, both a theoretical face and a moral face. With regard to the former Putnam argues that philosophy retains a role as a source of (revisable) interpretations of the meanings of our theories. For example, in the case of fundamental physical theories there is, on Putnam’s view, a fruitful interpenetration of science and metaphysics. Part of the problem in seeing this matter clearly, however, is that there are myths of what science is in both popular culture and philosophy, for example, the myth of value-free facts produced according to canons of induction.

In the background of the debate about whether philosophy has any legitimate and autonomous role to play distinct from that of the sciences is a wider ideologically driven debate about how we should think about the natural world that we inhabit. In The Wider Significance of Naturalism: A Genealogical Essay, Akeel Bilgrami attempts to uncover the genealogical basis of the deep division among philosophers today over naturalism, which he traces, ultimately, to a deep disagreement over whether irreducible values are part of the world or not. Bilgrami first gives an argument and motivation for the view that values are properties in the world, appealing to considerations of agency and the first-person point of view and exploiting some remarks in Gareth Evans about the relation between self-knowledge and intentionality, to show that seeing values as part of the world is essential to the fact of our possessing agency. He then traces genealogically why such a picture of value and the world we inhabit as agents came to be undermined by a metaphysics that grew around the new science in the late seventeenth century, making it seem inevitable to us that such a worldview was incompatible with that of science. Focusing on England, Bilgrami argues that this metaphysics of what is often too summarily described as disenchantment seems inevitable only because we write out of intellectual history; the vocal and lively dissenting scientific voices of the late seventeenth century, the deracination of value from nature and the world around us were neither compulsory nor the inevitable consequences of the new science and its laws. These voices were silenced not because of any scientific or philosophical superiority of the orthodox metaphysics they opposed, which was forming among the ideologues of the Royal Society around the new science. Rather, the dissenters lost out because of powerful political alliances that these ideologues forged with established Anglican interests, as well as with the newly emerging commercial interests that had a worldly interest in a vision of nature as brute and stripped of value. And such a vision of nature, according to Bilgrami, was realized in, and supported by, a transformation of the very conception of nature from the concept of an environment in which we merely live into the concept of natural resources to be systematically controlled and conquered for extractive economies. Bilgrami goes to extend his argument to trace and expose how later such a disenchanted worldview was modified to make it seem not merely a consequence of a commitment to the methods and metaphysics of the natural sciences, but also intrinsic to a dominant picture of the social sciences that prevails to this day.

In Taking the Human Sciences Seriously, David Macarthur argues that three families of Scientific Naturalism can be usefully distinguished, depending on how they define what the legitimate ontology, methods, and scope of science are. The first, Extreme Scientific Naturalism, identifies science with physics alone. The second, Narrow Scientific Naturalism, allows that some of the other sciences of nature may be irreducible to physics but at the same time claims that science proper is to be limited to the natural sciences as opposed to the human sciences. Grouped together, these two views represent the naturalist orthodoxy in contemporary philosophy. However, it is important to see that Scientific Naturalism may also come in a third form, Broad Scientific Naturalism, which identifies science with both the natural sciences and the social and human sciences where at least some of the human sciences are understood as methodologically and ontologically irreducible to the natural sciences.

In light of these distinctions, Macarthur argues that the idea that there is a dualism of norm and nature is an overstatement—even from the limited perspective of a scientific naturalist. From the point of view of Broad Scientific Naturalism—a position that is not clearly visible in contemporary philosophy—it is plausible that certain indispensable and irreducible normative items that appear in, or are presupposed by, social scientific explanations can be admitted into the scientific image of the world.¹² Reasons and cognitive values are examples. This more liberal version of Scientific Naturalism promises not a solution to but a dissolution of the placement problem. That alone provides a strong motive for orthodox scientific naturalists to move toward a more liberal understanding of their own commitment to naturalism. However, Macarthur goes on to argue that Broad Scientific Naturalism faces the apparently intractable problem that it cannot scientifically explain rational normativity. This problem is particularly embarrassing if, as Macarthur supposes, the doctrines of Scientific Naturalism are themselves claims about what is rationally normative.

Macarthur’s chapter raises a more general question. Is a Broad Scientific Naturalism that admits the irreducibility of the social sciences, with their peculiar ontologies and epistemologies, liberal enough? Many of the authors of this volume want to argue that it is not. It might be claimed, for example, that there are normative phenomena to which we have access only from the perspective of fully engaged agents in the world. And this is a perspective, it seems, that is incompatible with the perspective of a scientific observer, however empathetic. To admit this would be to accept the existence of nonscientific nonsupernatural phenomena. Or, it might be claimed that there are nonscientific modes of understanding or knowing (e.g., aesthetic or moral reasoning) that are incapable of being properly acknowledged or represented by scientific inquiries and discoveries. Both of these directions of thought fall within the category of what we have called Liberal Naturalism (more on this later).

In Why Scientific Realism May Invite Relativism, Carol Rovane discusses another problem facing scientific naturalists. She argues that scientific realism (which is her label for the ontological component of Scientific Naturalism) is undermined by unnoticed liaisons dangereuses with relativism. In Rovane’s view, in order to rule out relativism, scientific realists should establish a view from everywhere, that is, the possibility of a super-knower who could know all the truths potentially known by every other possible knower. However, once one has assumed the realist conception of the facts as mind-independent (which strongly suggests that there may be knowers very different from us), it is doubtful that all the truths knowable in principle could be conjoined and embraced together by a single possible knower. In this light, Rovane claims that, far from possessing the conceptual resources that would be necessary to rule out relativism, scientific realism paradoxically encourages it.

Peter Menzies’s chapter, Reasons and Causes Revisited, is a reconsideration of the relation between reasons and causes in ordinary intentional explanation. Scientific Naturalism holds that intentional psychology is akin to a theory for explaining human behavior in terms of intentional states, but, according to Menzies, it has trouble acknowledging the normativity of intentional psychology. Kantian rationalism accepts this kind of normativity but mysteriously treats reasons as radically distinct from causes. For Menzies the problem is that both parties to this debate are wedded to the deductivenomological account of scientific theorizing according to which causal explanations are one thing and rational explanations another. In Menzies’s model-based conception of intentional psychology, we appeal to intentional states to predict and explain behavior because they both rationalize and cause behavior. Reasons are causes but, in contrast to Davidson’s version of this position, only one set of generalizations is needed to provide both rationalizing and causal explanations.

THE DOCTRINE OF LIBERAL NATURALISM

Apart from raising questions about whether orthodox versions of Scientific Naturalism are viable, the papers collected here raise the question whether an alternative form (or forms) of naturalism—which fall within the field of what we have called Liberal Naturalism—are available and potentially more attractive than any version of Scientific Naturalism. One of the primary motivations for the present volume is the thought that the debate over which form of naturalism is best will depend to a considerable extent on which provides the best account of core normative phenomena such as reasons and values. As Putnam says, normativity is ubiquitous in our thought and talk. So it has some claim to being the central area for which philosophy must provide an account if we are to achieve the sort of self-understanding that Socrates, and many other philosophers since, have promised us.¹³

Liberal Naturalism, as we understand it, is not a precisely defined credo. It is better seen as a range of attempts to articulate a new form of naturalism that wants to do justice to the range and diversity of the sciences, including the social and human sciences (freed of positivist misconceptions), and to the plurality of forms of understanding, including the possibility of nonscientific, nonsupernatural forms of understanding (whether or not these also count as forms of knowledge). In addition, some of our authors want to allow for the possibility of nonscientific, nonsupernatural entities. Most of the papers in the current volume, therefore, can be thought of as investigations into the question of how best to characterize an alternative conception of naturalism along these lines. Rather than try to adjudicate this debate by prematurely deciding for a single alternative, we prefer to present these papers as contributions to a fruitful controversy that we want to recommend to the reader’s attention.

If Liberal Naturalism cannot at this stage be characterized definitively, the question arises whether there is any useful way of characterizing what the authors of this volume have in common? Liberal Naturalism, as we have seen, is best thought of as occupying the typically overlooked conceptual space between Scientific Naturalism and Supernaturalism. A necessary condition for a view’s being a version of Liberal Naturalism is that it rejects Scientific Naturalism, hence that it rejects the ontological doctrine or the methodological doctrine, or both. There are different ways of liberalizing naturalism beyond its orthodox understanding and also different ways of conceiving of, hence rejecting, the supernatural.

Let us start, then, with those who are liberal naturalists solely in virtue of rejecting the methodological doctrine. Consider Huw Price who rejects standard Scientific Naturalism (which he calls object naturalism) on the grounds that it makes substantial representationalist assumptions about language prior to empirical study of the functions language actually serves. His idea of a subject naturalism starts by rejecting representationalism across the board and then showing how one can explain the functions of linguistic concepts or expressions in other terms. On one reading, Price holds to the orthodox naturalist ontology of the natural sciences in thinking that a commitment to the reality of values, say, does not require us to suppose that they are a special kind of entity. The vindication of our commonsense realist talk of values only requires the availability, from an anthropological point of view, of a nonrepresentational function for such talk. But even if that is so, Price departs from any kind of Scientific Naturalism in arguing that a science of linguistic functions can itself show us that there are nonscientific modes of knowing and understanding.

In Truth as Convenient Friction, reproduced in the present volume,¹⁴ Price considers the functional role of the normative concept of truth. He primarily takes issue with Rorty’s pragmatist conviction that there is no significant practical distinction between justification and truth. Rorty claims that obedience to a… commandment to seek the truth… will produce no behaviour not produced by the need to offer justification. In Price’s view, this claim is false on the grounds that there is a widespread behavioral pattern that results from the fact that speakers take themselves to be subject to a commitment to seek the truth. Moreover, it is a behavioral pattern that Rorty, of all people, cannot afford to dismiss as a by-product of bad philosophy: it is conversation itself—something he champions. In order to account for ordinary conversational practice, in Price’s view, we need to recognize that speakers take themselves to be governed by a norm stronger than that of justification—a norm that speakers acknowledge they may fail to meet, even if their claims are well justified. This norm provides the automatic and quite unconscious sense of engagement in common purpose that distinguishes conversation from a roll call of individual opinion. Truth, on Price’s view, is the grit that makes our individual opinions engage with one another.

A debate between Rorty and Price over these matters is also included in this anthology. This debate is interesting for a number of reasons, not the least of which is that they are two of the leading neopragmatists in the world. Both Price and Rorty, for example, claim that truth is not a substantial property about which there could be an interesting philosophical theory. Their disagreement centers rather around the question of whether there is any need for a norm of truth over and above a norm of intersubjective agreement and the form that epistemic norms might take.

In Two Directions for Analytic Kantianism: Naturalism and Idealism, Paul Redding criticizes Price’s version of Liberal Naturalism—at least on the narrow reading sketched above—for a residual scientism, which threatens nihilism, a philosophical stance that overlooks or undermines our commitment to norms and values. On the positive side, Redding adopts Bernard Williams’s idea that a view that avoids both scientism and Supernaturalism can be achieved only if philosophy is conceived as a humanistic discipline—one whose main goal is to contribute to our self-understanding through a critical reflection on the contingent, historical genesis of our ideas. From this perspective, Redding endorses a conception rooted in the lessons of Kant’s German idealist successors. He suggests that we can only do justice to our normative practices if we follow Brandom’s neo-Hegelian attempt to explain norms in terms of reciprocal recognition.

It is arguable, however, that Price does not restrict himself to a narrowly scientific ontology because he rejects the whole enterprise of ontology as a distinct philosophical explanatory project. By talking in certain ways we commit ourselves to the reality of certain kinds of objects—which ones being a matter of interpretation—but this is not a philosophical discovery about science; it is a truth about assertoric discourse in general. Indeed, from a subject naturalist perspective science itself can discover that science is just one thing among many that we do with the linguistic tools of ontological commitment.¹⁵ We cannot start (prior to empirical enquiry) with the metaphysically weighty view that only science can tell us anything about the range, function, and possible plurality of our ontological commitments. So perhaps Price is best read as rejecting both the ontological and methodological doctrines of Scientific Naturalism.

Another liberal naturalist position is articulated by John Dupré who, in addition to accepting nonscientific forms of knowing and understanding, holds to a scientific ontology of both the natural and human sciences.¹⁶ He shares Price’s view that in order to understand the human world, an irreducible plurality of insights, including more traditionally humanistic ones, is indispensable. In How to be Naturalistic Without Being Simplistic in the Study of Human Nature, Dupré’s rigorous empiricism assumes that if something exists and is made of something, then, in contrast to abstract entities, it is made of physical stuff. Nonabstract entities or properties must be accountable by a recognizable extension of the empirical methods of investigation employed by the sciences. But, on Dupré’s view, the empirical evidence also shows us that some things have irreducible nonphysical properties and causal powers that do not figure in the explanations offered by physics. The moral is a pluralistic conception of science both in epistemic and ontological terms. According to Dupré, scientific naturalists are committed to a questionable metaphysical monism that we have no good empirical reason to accept. This questionable monism includes the reducibility of all natural sciences to physics, the "principle of the causal closure of the physical world," and the idea that nonphysical properties are reducible to, or at least supervene (globally if not locally) on, physical properties.

In Is Liberal Naturalism Possible? Mario De Caro and Alberto Voltolini approach the issue of Liberal Naturalism by considering a dilemma posed by Ram Neta.¹⁷ As Neta sees it, either Liberal Naturalism implies that controversial items (say, normative items) are in principle reducible to the ontology of the natural sciences or they are irreducible to it. If the former, then this view is simply orthodox Scientific Naturalism; if the latter, then the view becomes ipso facto a form of Supernaturalism. Thus, it seems, there is no logical space for Liberal Naturalism.

In order to contest this argument, De Caro and Voltolini offer the following characterization of Liberal Naturalism. First of all, like the vast majority of contemporary naturalisms, this view is committed to the claim (the constitutive claim of naturalism) that no entity or explanation may be accepted whose existence or truth would contradict the laws of nature insofar as we know them. More specifically, Liberal Naturalism is characterized by two provisos—one epistemological and one ontological—that complement this claim. As to the first, even if some of the controversial entities were actually reduced or shown to be ontologically dependent on scientific entities, in order to account fully for the features of these entities, one might still have to turn to forms of understanding (such as conceptual analysis, imaginative speculation, or introspection) that are neither reducible to scientific understanding nor supernatural.¹⁸ As to the ontological proviso of Liberal Naturalism, there may be entities (e.g., numbers) that do not and cannot causally affect the world investigated by the sciences and are both irreducible to and ontologically independent of entities accountable by science but are not supernatural either, since they cannot violate any laws of nature.¹⁹

By analyzing these two provisos, De Caro and Voltolini argue that Liberal Naturalism does not collapse into Scientific Naturalism, as the first horn of Neta’s dilemma claims, because of its broader epistemological and ontological attitudes. The second horn of Neta’s dilemma, the threat that Liberal Naturalism is Supernaturalism in disguise, is not serious, either. Supernaturalism is ruled out since no object, property, or event can be causally efficacious in the natural world and yet fail to be an object of scientific investigation (in principle, at least). In this light, liberal naturalists have no problem in ruling out, on scientific grounds, supernatural entities such as immaterial gods, infinite and perfect divine attributes, irreducibly miraculous events, or Cartesian minds—that is, causally efficacious immaterial particulars that cannot in principle be investigated scientifically. Moreover, supernatural entities (both causally efficacious and noncausally efficacious) would require special modes of understanding that would be irreconcilable with scientific explanation—and would thereby violate the fundamental claim of naturalism.

In Metaphysics and Morals, reproduced here,²⁰ Thomas Scanlon articulates a liberal naturalist position that sees no clash between the world posited by the sciences and the normative phenomena that are registered in our accepting the truth of various moral and practical judgments. The latter are simply not in the business of purporting to refer to causal matters instantiated in space, according to Scanlon. Consequently, the obtaining of moral truths (or facts) is not to be thought of as part of or in competition with scientific accounts of the causal regularities of the world. One should not take the fact that, say, moral judgments are often based on causal claims concerning physical and psychological events as evidence that the claims made by moral judgments are themselves causal. Nor should one treat moral knowledge (supposing there is such) as analogous to perceptual knowledge, since the latter, in contrast to the former, involves the right kind of causal connections to items located in space. On Scanlon’s view, moral judgments can be true without being about things investigated by the sciences and without needing any metaphysical support.

In The Naturalist Gap in Ethics, Erin I. Kelly and Lionel K. McPherson are concerned with how to understand the authority or bindingness of moral reasons. Their chapter can be read as steering a middle course between the extreme positions of moral naturalism, on the one hand, and moral cognitivism (including that defended by Scanlon), on the other. According to Kelly and McPherson, the cognitivist may be right that moral judgments are reason-sensitive but is mistaken in thinking these judgments must have authority for all rational agents. At the other extreme, the naturalist may be right to challenge the rational authority of moral judgments as binding for all rational agents but wrong to suppose that the reason sensitivity in question can be reductively explained by appeal to psychology or biology. At the heart of their account, Kelly and McPherson draw a distinction between acknowledging and accepting a reason. Structural accounts of the normativity of reasons that appeal to presuppositions of reason, communication, or agency can at best show that we must acknowledge moral reasons in the sense of appreciating their normative significance relative to certain specific pursuits, goals, and attachments. They do not show that we must accept moral reasons in the sense of having to be guided or motivated by them ourselves on pain of irrationality. On this view, there is a gap between being aware of the normative significance of moral imperatives and accepting these same reasons in one’s own deliberations. Frankfurt and Blackburn are right to think that a person’s cares and commitments are needed to account for the rational force or bindingness of moral reasons. But their mistake is a failure to see that this concession is compatible with a cognitivist approach to the content of moral reasons. As Kelly and McPherson see it, moral judgments are not simply expressions of a person’s sentiments or attitudes; rather, having certain sentiments or attitudes is a condition for having moral concern at all.

Another problem with moral naturalism concerns its account of moral psychology. In Phenomenology and the Normativity of Practical Reason, Stephen L. White challenges the widely influential Humean moral psychology that supposes that practical reason only plays an instrumental role in finding the means to antecedently established ends (given by one’s desires) and that what explains why the agent did what she did is simply some belief-desire pair. Although a sophisticated version of this doctrine can overcome various problems associated with traditional Humean moral psychology (extreme imprudence, pathological indifference, normativity), it still faces the problem of explaining what makes an action an action as opposed to something that merely happens to one. White argues that the basic problem plaguing the Humean account is that desires, given as objects, simply cannot explain action. For example, the Humean cannot adequately characterize the difference between what White calls the passive subject—one who agrees about all the objective facts stated in nonagential vocabulary but who finds the language of agency unintelligible—and ordinary agents like ourselves.

The Humean, on White’s view, has no account of how a desire is given subjectively (at the personal level) in the normal way. White concludes that the representation of things as desirable (that is, where the objects of one’s desires are given as desirable) is more basic and more primitive than the representation of things as desired by me. This requires a phenomenology that is both inflationary and deflationary relative to sense-data theories: inflationary, since seeing things as desirable (hence valuable) is the normal way desires are given to us; and deflationary, since we are often given external objects and their desirability directly without there being something else at the personal level (e.g., sense data) by virtue of which we have this access. White goes on to show that this perspective allows for a solution to the moral problem of Michael Smith and for a vindication of Aristotle’s claim that the virtuous person is superior to the merely continent person.

Another version of Liberal Naturalism is that of John McDowell, whose naturalism of second nature involves a rejection of both the ontological and methodological doctrines of Scientific Naturalism. In this volume, both Peter Godfrey Smith and Marie McGinn want to retain something of the spirit or motivation of McDowell’s version of Liberal Naturalism but take issue with its formulation, in particular, the lack of any substantial conception of the relation between first nature, the picture of the human provided by the natural sciences (particularly biology), and second nature, the picture of the human provided by our language and culture, including philosophy itself. McDowell seems to be content to say simply that our first nature (say, our physical, chemical, and biological nature) constrains the development and shape of our second nature: the innate endowment of human beings must put limits on the shapings of second nature that are possible for them.²¹

In Dewey, Continuity, and McDowell, Peter Godfrey-Smith sees Dewey, like McDowell, as attempting to overcome dualistic conceptions of the relation between normative thought and the natural world. In Dewey’s account this dualistic inheritance of traditional philosophy can be traced all the way back to the Greeks and their prioritizing, in the realm of Being, the permanent and self-possessed and, in the realm of Knowing, the process of copying. If we do justice to the changing and relational from the perspective of a problem-solving epistemology, Dewey, on Godfrey-Smith’s reading, shows us how we can accept the reality of normative phenomena left out of the natural scientific description of the world. Moreover, Dewey makes room for a philosophical project of relating normative phenomena to the world as described by the sciences, a sideways-on project that McDowell decisively eschews. From Dewey’s perspective this is not a matter of reducing and then subsuming such phenomena under natural laws. As Godfrey-Smith puts it, What the philosopher wants to do is ask general questions about how the ’habits of thought and action’ involved in our use of normative concepts [especially the concept of a reason] relate to other facts about us, and how these habits function as human cognitive tools.

In Wittgenstein and Naturalism, Marie McGinn raises related concerns by way of a discussion of two ways of reading the bearing of Wittgenstein’s later philosophy on rule following: Wright’s reading, according to which our responses to rules are grounded in primitive dispositions described in the language of disenchanted nature; and McDowell’s naturalized Platonist reading, according to which when reasons come to an end or when we act blindly we are still able to conceive ourselves as within the space of reasons. Although McGinn comes down on the side of McDowell in this debate, she wants to press further the question of how we are to understand the relation of our natural history to our capacity to follow rules. McGinn is at pains to provide a conception of Wittgenstein’s naturalism that avoids seeing him as claiming that the truth of our judgments depends on human agreements while at the same time being alive to the idea that the sense of our judgments does depend on certain general facts of nature, including certain constancies in human reactions, responses, and actions—including, in contrast to McDowell’s naturalism, those that are preconceptual and untrained. In McGinn’s view, what McDowell misses in assimilating his naturalism to Wittgenstein’s is the way, in Wittgenstein’s hands, a naturalism that involves attending to the details of things—that is, to the specific and variegated ways things act and work—can be fundamental to his method of philosophizing.

The essays collected here open up the territory between, on the one hand, an exclusively scientific ontology and scientific modes of knowing and, on the other, all forms of Supernaturalism (including, of course, a Platonism that treats norms as intangible objects).²² This is the contested terrain of what we are calling Liberal Naturalism, and it is an open question what the best way of conceiving this new form of naturalism is. Should we suppose that science, broadly conceived, settles all questions of ontology so that Liberal Naturalism simply takes the form of an epistemological commitment to nonscientific modes of knowing or understanding rationality, other minds, oneself, ethics, aesthetics, etc.? Or is there, in addition, a nonscientific, non-supernatural ontology (of, say, actions, agents, values) that must be acknowledged if we are to make best sense of our practical and social lives? Furthermore, what kinds or categories of irreducible and categorical norms are there, and how should we go about conceiving and studying them if we are to avoid both a reductive scientism and an otherworldly Platonism? Is there a social phenomenology of normative phenomena such that the norms and values are simply those that we can mutually acknowledge as such?

Another important matter is the relation between what McDowell calls first and second nature, the puzzling relationship between the normatively structured space of reasons, on the one hand—or perhaps we should say spaces of reasons if we suppose that reasons come in importantly different kinds—and the domain of naturalistic lawlike or causal explanations characteristic of modern science, on the other. The hope is that a Liberal Naturalism could bring the relations between these two senses of nature into better focus, thereby either endorsing McDowell’s thesis of the autonomy of the space of reasons and the manifest image or revising it on the basis of more nuanced conceptions of the scientific image and the manifest image, thus, of the relation between the two images.

ACKNOWLEDGMENTS

The outlook of Liberal Naturalism was originally inspired by the writings of John McDowell and Hilary Putnam and we want to acknowledge a debt of gratitude to them for having blazed the trail. In the development of the outlook and in helping to gain a sense of its internal variety we have benefited from reviews of a previous collection of ours, Naturalism in Question (Harvard University Press; 2004/2008) and discussions over several years with Akeel Bilgrami, John Dupré, Huw Price, Hilary Putnam, and Stephen White. We would also like to gratefully acknowledge the encouragement and invaluable advise of Akeel and Stephen in the detailed conception and preparation of this work. Stephen is to be thanked, moreover, for providing very helpful comments on previous drafts of the introduction.

This book is dedicated with love to Giulia and Talia.

NOTES

1. John McDowell, Mind and World (Cambridge, Mass.: Harvard University Press, 1994), 70. The two inserted quotes are from p. 71.

2. Max Weber, The Methodology of the Social Sciences, trans. E. A. Shils and H. A. Finch (New York: Free Press, 1949). See also Richard Swedberg, Max Weber and the Idea of Economic Sociology (Princeton, N.J.: Princeton University Press, 1998).

3. For a discussion of the difficulties associated with these alternatives, particularly the second and third, see Michael Smith, The Moral Problem (Oxford: Blackwell, 1994).

4. Arguably, most scientific naturalists are committed to both the Ontological Doc trine and the Methodological Doctrines. However, one might argue for the Onto logical Doctrine on purely a priori grounds which, depending on one’s view of the a priori, might involve denying the Methodological Doctrines. Whether one could hold the Methodological Doctrine whilst denying the Ontological Doctrine will depend on one’s understanding of the relation between scientific explanations and ontological commitment.

5. Wilfrid Sellars, Empiricism and the Philosophy of Mind, in The Foundations of Science and the Concepts of Psychoanalysis, ed. H. Feigl and M. Scriven, Minnesota Studies in the Philosophy of Science 1 (Minneapolis: University of Minnesota Press, 1956), 173.

6. Arthur Danto, Naturalism, Encyclopedia of Philosophy, ed. Paul Edwards (New York: Macmillan, 1967), 5:448.

7. It should be noticed that, in a very different spirit, some liberal naturalists also endorse radical empiricist views of knowledge. John Dupré, for example, endorses empiricism as the answerability of our belief to what we actually experience in the world (Human Nature and the Limits of Science [Oxford: Oxford University Press, 2001], 163–64).

8. Some scientific naturalists, such as Frank Jackson, defend the idea of revisable a priori knowledge. See his From Metaphysics to Ethics: A Defence of Conceptual Analysis (Oxford: Clarendon Press, 1998).

9. An exception to this attitude is Hartry Field’s mathematical fictionalism. See his Science Without Numbers (Oxford: Blackwell, 1980).

10. David Papineau, Naturalism, Stanford Encyclopedia of Philosophy, http://plato.stanford.edu/entries/naturalism/.

11. Even though this paper was written especially for inclusion in the present volume, because of Rorty’s untimely death in 2007 it first appeared in his Philosophical Papers, vol. 4: Philosophy as Cultural Politics (Cambridge: Cambridge University Press, 2007), 147–59.

12. There may be various reasons the human and social sciences are thought to be irreducible to the natural