Professor and Director, Dianoia Institute of Philosophy at Australian Catholic University
& Adjunct Professor of Philosophy, University of Southern California
Can normative words like "good," "ought," and "reason" be defined in entirely non-normative terms? Confusion of Tongues argues that they can, advancing a new End-Relational theory of the meaning of this language as providing the best explanation of the many different ways it is ordinarily used. Philosophers widely maintain that analyzing normative language as describing facts about relations cannot account for special features of particularly moral and deliberative uses of normative language, but I argue that the End-Relational theory systematically explains these on the basis of a single fundamental principle of conversational pragmatics. These challenges comprise the central problems of metaethics, including the connection between normative judgment and motivation, the categorical character of morality, the nature of intrinsic value, and the possibility of normative disagreement. This linguistic analysis has deep implications for the metaphysics, epistemology, and psychology of morality, as well as for the nature and possibility of normative ethical theory. Most significantly it supplies a nuanced answer to the ancient Euthyphro Question of whether we desire things because we judge them good, or vice versa. Normative speech and thought may ultimately be just a manifestation of our nature as intelligent animals motivated by contingent desires for various conflicting ends.
Review: “This is one of the richest, most sophisticated, and most impressive books on metaethics to have been published in my lifetime. Everyone with an interest in normative language ought to read it. Those who would seek to defend naturalist views of the sort Finlay develops here will find it a treasure trove of dialectical resources that they will want to plunder repeatedly. Those who seek to attack such views or to defend rival views will find it a challenge it would be shameful to ignore.” James Lenman, Language.
More Reviews: Matthew Chrisman, Notre Dame Philosophical Reviews, Alex Gregory, Analysis, Russell Blackford, The Philosopher’s Magazine, Janice Dowell, Mind, Dan Fogal, Ethics. (See also Analysis symposium and my "Reply" below.)
Papers in preparation (comments welcome!):
What is the future of normativity? What should it be? I distinguish a narrow, recent inquiry conducted under the rubric ‘normativity’ (meta-‘normative’ theory) from a broad, timeless inquiry (metaethics), characterizing meta-‘normative’ theory as merely the latest epicycle in a futile circle of metaethical debate by analogy to the doctrine of Samsara’s Wheel. This futility is attributed to a four-way ambiguity running systematically through all “normative” terminology. I present a two-dimensionalist (or perspectivist) interpretation of paradigmatically and uncontroversially “normative” thoughts which enables us to explain the three major camps in metaethics (noncognitivism, objectivism, and subjectivism) as generated by the three different ways of collapsing the two-dimensional phenomenon into a single dimension (either of content or perspective). This reveals the characteristic errors made by each camp, and finds their claims about “normativity” to be addressing different subject-matters with diverging extensions. Four obstacles to metaethical enlightenment are identified: illusion/delusion and attachment (especially in objectivists), charity (in their opponents), and forgetfulness. I close with practical proposals about what to do next. (Updated 7/28/2022)
Papers published/ forthcoming:
We defend the Quantifier Analysis of modal terms (QA) against an objection based on a problem arising from certain epistemic 'might p' sentences, which appear felicitous although according to QA are necessarily false, such as '899 might be prime.' According to our solution, utterances of these sentences are felicitous despite being false because of the appropriateness of their conversational implicature, that the speaker takes p to be a serious option for reasoning. (Posted 4/16/2020)
Contribution to a symposium on my Confusion of Tongues (OUP 2014). I reply to commentaries generously provided by Alex Worsnip, Janice Dowell, and Glen Koehn. Main topics include whether contextualism about normative terms should be "ecumenical" or go all the way down, COT's implications for normative ethics, the suitability of an analytic method in metaethics, the semantics of 'good', conflicts between ends, and the analysis of final value. (Posted 4/16/2020)
This paper champions the view (REG) that the concept of a normative reason for an agent S to perform an action A is that of an explanation why it would be good (in some way, to some degree) for S to do A. REG has numerous virtues, but faces some significant challenges which prompt many philosophers to be skeptical that it can correctly account for all our reasons. I demonstrate how five different puzzles about normative reasons can be solved by attention to the concept of goodness, and in particular observing the ways in which it—and consequently, talk about reasons—is sensitive to context (ends and information). Rather than asking simply whether or not certain facts are reasons for S to do A, we need to explore the contexts in which it is and is not correct to describe a certain fact as “a reason” for S to do A. These five puzzles concern: (1) reasons for attitudes of the “right kind”, (2) evidence as reasons, (3) normative facts as reasons, (4) subjective reasons, and (5) attitudes as reasons. (Updated 4/30/2020)
This paper investigates whether different philosophers’ claims about “normativity” are about the same subject or (as recently argued by Derek Parfit) theorists who appear to disagree are really using the term with different meanings, in order to cast disambiguating light on the debates over at least the nature, existence, extension, and analyzability of normativity. While I suggest the term may be multiply ambiguous, I also find reasons for optimism about a common subject-matter for metanormative theory. This is supported partly by sketching a special kind of hybrid view of normative judgment, perspectivism, that occupies a position between cognitivism and noncognitivism, naturalism and nonnaturalism, objectivism and subjectivism, making it more plausible that radically different metanormative theories could be about the same thing. I explore three main fissures: between (i) the “normativity” of language/thought versus that of facts and properties, (ii) abstract versus substantive senses, and (iii) formal versus robust senses. (Updated 11/25/18)
Speech and thought about what the law is commonly function in practical ways, to guide or assess behavior. These functions have often been seen as problematic for legal positivism in the tradition of H.L.A. Hart. One recent response is to advance an expressivist analysis of legal statements (Toh), which faces its own, familiar problems. This paper advances a rival, positivist-friendly account of legal statements which we call “quasi-expressivist”, explicitly modeled after Finlay’s metaethical theory of moral statements. This consists in a descriptivist, “rule-relational” semantics combined with a pragmatic account of the expressive and practical functions of legal discourse. We argue that this approach is at least as well-equipped as expressivism to explain the motivational and prescriptive features of “internal” legal statements, as well as a fundamental kind of legal disagreement, while being better positioned to account for various “external” uses of the same language. We develop this theory in a Hartian framework, and in the final part of the paper argue (particularly against Toh’s expressivist interpretation) that Hart’s own views in The Concept of Law are best reconstructed along such quasi-expressivist lines. (Updated 4/16/20)
According to content-relativist theories of moral language, different speakers use the same moral sentences to say different things. Content-relativism faces a well-known problem of lost disagreement. Recently, numerous content-relativists (including the author) have proposed to solve this problem by appeal to various kinds of non-content-based, or broadly pragmatic, disagreement. This presents content-relativists with a new problem—of found agreement. Which (if any) of these newly identified kinds of conflict is correctly identified as the lost moral disagreement we were looking for? This paper offers a critical comparison of different content-relativist proposals. It divides them into two broad categories, quasi-expressivist theories (QED) and metalinguistic theories (MLD). Objections to each are considered, and QED is tentatively found to be superior. (Updated 10/1/17)
A critical survey of various positions on the nature, use, possession, and analysis of normative concepts. We frame our treatment around G.E. Moore’s Open Question Argument, and the ways metaethicists have responded by departing from a Classical Theory of concepts. In addition to the Classical Theory, we discuss synthetic naturalism, noncognitivism (expressivist and inferentialist), prototype theory, network theory, and empirical linguistic approaches. Although written for a general philosophical audience, we attempt to provide a new perspective and highlight some underappreciated problems about normative concepts. (Updated 5/9/16)
This paper argues that the innovation of an ordering source parameter in the standard Lewis-Kratzer semantics for modals was a mistake, at least for English auxiliaries like ‘ought’, and that a simpler dyadic semantics (as proposed in my earlier work) provides a superior account of normative uses of modals. I programmatically investigate problems arising from (i) instrumental conditionals, (ii) gradability and “weak necessity”, (iii) information-sensitivity, and (iv) conflicts, and show how the simpler semantics provides intuitive solutions given three basic moves: (1) an end-relational analysis of normative modality, (2) a “most” analysis of “weak necessity”, and (3) well-motivated appeals to pragmatics. I conclude with metasemantic observations about the desiderata for a semantics for ‘ought’. (Updated 10/29/14)
Relational theories of normative language allegedly face special problems in accounting for the extent of disagreement, but this is everybody’s problem because normative sentences are relativized to different information in contexts of deliberation and advice. This paper argues that a relational theory provides a pragmatic solution that accounts for some disagreements as involving inconsistent preferences rather than beliefs. This is shown to be superior to the semantic solution offered by expressivists like Allan Gibbard, as it accounts for a wider range of disagreements, explains a puzzling asymmetry, and avoids the expressivist’s problem with negation. This pragmatic account extends to fundamental disagreements involving preferences for different ends. Three different kinds of normative disagreement are distinguished: instrumental, rational, and outright. [NOTE: this paper is based on Chapter 8 of Confusion of Tongues. It expands on ideas in ‘Metaethical Contextualism Defended’, 2010.] (Posted 7/17/14)
Some philosophers hold that ‘ought’ is ambiguous between a sense expressing a propositional operator and a sense expressing a relation between an agent and an action. We defend the opposing view that ‘ought’ always expresses a propositional operator against objections that it cannot adequately accommodate an ambiguity in ‘ought’ sentences between evaluative and deliberative readings, predicting readings of sentences that are not actually available. We show how adopting an independently well-motivated contrastivist semantics for ‘ought’ according to which ‘ought’ is always relativized to a contrast set of relevant alternatives enables us to explain the evaluative-deliberative ambiguity and why the availability of these readings depends on sentential grammar. (Updated 9/13/12)
What does it mean to call something a “reason”? This paper offers a unifying semantics for the word ‘reason’, challenging three ideas that are popular in contemporary philosophy; (i) that ‘reason’ is semantically ambiguous, (ii) that the concept of a normative reason is the basic normative concept, and (iii) that basic normative concepts are unanalyzable. Nonnormative uses of ‘reason’ are taken as basic, and as meaning explanation why. Talk about normative reasons for action is analyzed in terms of explanations why acting would be good in some way. I show how a number of obstacles for this idea—including extending the analysis to normative reasons for attitudes—can be overcome by adopting a reductive, end-relational analysis of the meaning of ‘good’ which I have defended elsewhere. Finally, I analyze talk of “motivating” reasons in terms of (supposed) normative reasons for which agents act. [NOTE: this paper is an abbreviated version of a draft Chapter 5 of Confusion of Tongues, and builds on some ideas in ‘The Reasons that Matter’, 2006.]
In his contribution to the 2010 Spindel Conference, David Shoemaker argues from (A) psychopaths’ emotional deficiency, to (B) their insensitivity to moral reasons, to (C) their lack of criminal responsibility. This response observes three important ambiguities in this argument, involving the interpretation of (1) psychopaths’ emotional deficit, (2) their insensitivity to reasons, and (3) their moral judgements. Resolving these ambiguities presents Shoemaker with a dilemma: his argument either equivocates or it is falsified by the empirical evidence. An alternative perspective on psychopaths’ moral and criminal responsibility is proposed.
In his response to my paper ‘The Error in the Error Theory’ criticizing his and J. L. Mackie’s moral error theory, Richard Joyce finds my treatment of his position inaccurate and my interpretation of morality implausible. In this reply I clarify my objection, showing that it retains its force against their error theory, and I clarify my interpretation of morality, showing that Joyce’s objections miss their mark.
We defend a contextualist account of normative judgments as relativized both to (i) information and to (ii) standards or ends, against recent objections that turn on practices of disagreement. Niko Kolodny & John MacFarlane argue that information-relative contextualism cannot accommodate the connection between deliberation and advice. In response, we suggest that they misidentify the basic concerns of deliberating agents, which are not to settle the truth of particular propositions, but to promote certain values. For pragmatic reasons, semantic assessments of normative claims sometimes are evaluations of propositions other than those asserted. Other writers have raised parallel objections to standard-relative contextualism, particularly about moral claims; we argue for a parallel solution.
Survey of some recent literature on normativity, including nonreductionist, neo-Aristotelian, neo-Humean, expressivist, and constructivist views.
Some intuitive normative principles raise vexing ‘detaching problems’ by their failure to license modus ponens. I examine three such principles (a self-reliance principle and two different instrumental principles) and recent stategies employed to resolve their detaching problems. I show that solving these problems necessitates postulating an indefinitely large number of senses for ‘ought’. The semantics for ‘ought’ that is standard in linguistics offers a unifying strategy for solving these problems, but I argue that an alternative approach combining an end-relational theory of normativity with a comparative probabilistic semantics for ‘ought’ provides a more satisfactory solution.
Normative concepts have a special taste, which many consider to be proof that they cannot be reductively analyzed into entirely nonnormative components. This paper demonstrates that at least some intuitively normative concepts can be reductively analyzed. I focus on so-called ‘hypothetical imperatives’ or ‘anankastic conditionals’, and show that the availability of normative readings of conditionals is determined by features of grammar, specifically features of tense. Properly interpreted, these grammatical features suggest that these deontic modals are analyzable in terms of conditional necessity with a certain temporal structure.
Since its publication in 1979, Bernard Williams’ ‘Internal and External Reasons’ has been one of the most influential and widely discussed papers in ethics. I suggest here that the paper’s central argument has nevertheless been universally misinterpreted. On the standard interpretation, his argument is perplexingly weak. In the first section I sketch this Standard argument, and detail just how terrible it is. The badness of the argument itself may not be a very strong reason not to ascribe it even to a great philosopher, but Williams himself seems to point out the very flaws that make it so terrible. The second part of the paper proposes and defends an interpretation on which he offers an Alternative argument, one which is immune to the objections that seem fatal to the Standard argument. On this interpretation, better supported by the textual evidence and the principle of charity, Williams’ conclusion seems to follow validly from defensible premises, including a substantive and interesting analysis of the concept of a normative reason.
This paper advances a reductive semantics for ‘ought’ and a naturalistic theory of normativity. It gives a unified analysis of predictive, instrumental, and categorical uses of ‘ought’: the predictive ‘ought’ is basic, and is interpreted in terms of probability. Instrumental ‘oughts’ are analyzed as predictive ‘oughts’ occurring under an ‘in order that’ modifer (the end-relational theory). The theory is then extended to categorical uses of ‘ought’: it is argued that they are special rhetorical uses of the instrumental ‘ought’. Plausible conversational principles explain how this end-relational ‘ought’ can perform the expressive functions of the moral ‘ought’. The notion of an ‘ought-simpliciter’ is also discussed.
Some of the opponents of desire-based views of normativity seek to undermine them by arguing that even the existence of instrumental normativity (reasons to pursue the means to your ends) entails the existence of a desire-independent rational norm, the instrumental norm. Once we grant the existence of one such norm, there seems to be no principled reason for not allowing others. I clarify this alleged norm, identifying two criteria that any satisfactory candidate must meet: reasonable expectation and possible violation. Some interpretations meet the first criterion and others meet the second, but there are no interpretations that meet both. After surveying the interpretations of Sidgwick, Hampton, and Korsgaard, I suggest that there is no instrumental norm of reason. The final section offers an alternative, desire-based account of instrumental normativity, on which individual normative requirements to pursue means derives from each individual desire for an end. [Note: accepted for publication in 2003]
Encyclopedia article on internal and external reasons.
“The Error in the Error Theory,” Australasian Journal of Philosophy 86(3), 2008: 347-69 [open access] (AJP Best Paper Award, 2009)
Moral error theory of the kind defended by J.L. Mackie and Richard Joyce is premised on two claims: (1) that moral judgements essentially presuppose that moral value has absolute authority, and (2) that this presupposition is false, because nothing has absolute authority. This paper accepts (2) but rejects (1). It is argued first that (1) is not the best explanation of the evidence from moral practice, and second that even if it were, the error theory would still be mistaken, because the assumption does not contaminate the meaning or truth-conditions of moral claims. These are determined by the essential application conditions for moral concepts, which are relational rather than absolute. An analogy is drawn between moral judgements and motion judgements.
Rationalists including Nagel and Korsgaard argue that motivation to the means to our desired ends cannot be explained by appeal to the desire for the end. They claim that a satisfactory explanation of this motivational connection must appeal to a faculty of practical reason motivated in response to desire-independent norms of reason. This paper builds on ideas in the work of Hume and Donald Davidson to demonstrate how the desire for the end is sufficient for explaining motivation to the means. Desiring is analyzed as having motivation towards making the end so, which is analyzed as engaging in mental activity aimed at facilitating that end. I conclude that it is constitutive of an agent’s desiring an end that he is motivated towards what he believes to be means.
This essay explains for a general philosophical audience the central issues and strategies in the contemporary moral realism debate. It critically surveys the contribution of some recent scholarship, representing expressivist and pragmatist nondescriptivism (Mark Timmons, Hilary Putnam), subjectivist and nonsubjectivist naturalism (Michael Smith, Paul Bloomfield, Philippa Foot), nonnaturalism (Russ Shafer-Landau, T. M. Scanlon) and error theory (Richard Joyce). Four different faces of ‘moral realism’ are distinguished: semantic, ontological, metaphysical, and normative. The debate is presented as taking shape under dialectical pressure from the demands of (i) capturing the moral appearances and (ii) reconciling morality with our understanding of the mind and world.
I believe that normative force depends on desire. This view faces serious difficulties, however, and has yet to be vindicated. This paper sketches an Argument from Voluntary Response, attempting to establish this dependence of normativity on desire by appeal to the autonomous character of our experience of normative authority, and the voluntary character of our responses to it. I first offer an account of desiring as mentally aiming intrinsically at some end. I then argue that behaviour is only voluntary if it results from such aiming; hence all voluntary behaviour is produced by desire. Full-blooded responses to normativity, I then argue, are voluntary actions: motivation to act arises voluntarily from perception of reasons to act. This fits the desire-based model of normativity but not its rivals. However this argument concludes merely that our responses to normativity are desire-based. I end with some observations about how I think we can bridge the gap from the nature of response to normativity to the nature of normativity itself.
This paper addresses the nature and relationship of morality and self-interest, arguing that what we morally ought to do almost always conflicts with what we self-interestedly ought to do. The concept of morality is analyzed as being essentially and radically other-regarding, and the category of the supererogatory is explained as consisting in what we morally ought to do but are not socially expected to do. I express skepticism about whether there is a coherent question, ‘Which ought I all things considered to obey?’ and suggest that the best substitute is a question about which is more important for me. Importance for a person, in turn, is explained as dependent upon what a person is disposed to care about. I suggest that morality and self-interest are both relatively unimportant for us when compared with our other ends.
Motivational reasons-internalism (Bernard Williams) fails to capture our first-order reasons judgements, while nonnaturalistic reasons-externalism (Derek Parfit) cannot explain the nature or normative authority of reasons. This paper offers an intermediary view, reformulating skepticism about external reasons as the claim not that they don’t exist but rather that they don’t matter. The end-relational theory of normative reasons is proposed, according to which a reason for an action is a fact that explains why the action would be good relative to some end, where the relevant end for any ascription of reasons is determined by the speaker’s conversational context. Because these ends need not be the agent’s ends, Williams is wrong to reject the existence of external reasons. But contra Parfit, a reason for action is only important for an agent if it is motivationally internal to that agent.
Moral assertions express attitudes, but it is unclear how. This paper examines proposals by David Copp, Stephen Barker, and myself that moral attitudes are expressed as implicature (Grice), and Copp’s and Barker’s claim that this supports expressivism about moral speech acts. I reject this claim on the ground that implicatures of attitude are more plausibly conversational than conventional. I argue that Copp’s and my own relational theory of moral assertions is superior to the indexical theory offered by Barker and Jamie Dreier, and that since the relational theory supports conversational implicatures of attitude, expressive conventions would be redundant. Furthermore, moral expressions of attitude behave like conversational and not conventional implicatures, and there are reasons for doubting that conventions of the suggested kind could exist.
Analyses of moral value judgements must meet a practicality requirement: moral speech acts characteristically express pro- or con-attitudes, indicate that speakers are motivated in certain ways, and exert influence on others’ motivations. Nondescriptivists including Simon Blackburn and Allan Gibbard claim that no descriptivist analysis can satisfy this requirement. I argue first that while the practicality requirement is defeasible, it indeed demands a connection between value judgement and motivation that resembles a semantic or conceptual rather than merely contingent psychological link. I then show how a form of descriptivism, the interest-relational theory, satisfies the requirement as a pragmatic and conversational feature of value judgement – thereby also accommodating its defeasibility. The word ‘good’ is always indexed to some set of motivations: when this index is unarticulated in many contexts the speaker conversationally implicates possession of those motivations. [Note: this ‘interest-relational theory’ is superceded by the ‘end-relational theory’ in my subsequent work.]
2001 Ph.D. dissertation, directed by James D. Wallace.
Page Last Updated 1/17/2022