TY - JOUR
AU - Silverstein, Matthew
AB - Andreas Müller’s goal in Constructing Practical Reasons is to show that constructivism about practical reasons is a distinctive and plausible metanormative theory. Wisely avoiding the question of whether other theories that go by the name ‘constructivism’ (such as those defended by Christine Korsgaard and Sharon Street) really deserve the label, Müller focuses instead on articulating a view that (a) lives up to the metaphor that ethical facts are made rather than discovered and that (b) is different from realism and reductionism on the one hand and expressivism on the other. He acknowledges that he cannot provide a complete defence of constructivism, since this would involve discussing the virtues and vices of these competing theories. His aim instead is merely to show that there is a version of constructivism about reasons that is ‘coherent and prima facie attractive’ (p. 3; subsequent page references are to Müller 2020 unless otherwise specified). At the beginning of the opening chapter, Müller sketches his constructivist theory as follows: Facts about what is a reason for doing what obtain in virtue of the activity of practical reasoning and the role that the corresponding reasons judgments play in it. More specifically, such a fact obtains because the corresponding reason judgment is true, and the judgment is true because the episode of reasoning that it is apt to guide is sound. (p. 6) The subsequent chapters are largely devoted to filling in the details of this sketch. Chapters 2 and 3 articulate a novel account of the attitude of judging that something is a reason, one according to which what distinguishes this attitude is the role it plays in practical reasoning. Chapter 4 presents a non-representationalist but also non-deflationist account of the truth of judgments about reasons. Chapter 5 introduces an account of what makes practical reasoning sound. These are the three main components of Müller’s view. The final chapter, which feels like something of an afterthought, adds a discussion of different conceptions of mind dependence and attempts to show that constructivism can capture at least one important version of the idea that practical reasons are objective. My review will proceed by describing the three components of Müller’s view in greater detail, registering my reservations about the first two along the way. As will become clear, I have some doubts about whether Müller’s view is distinct from various prominent forms of expressivism, as well as more serious doubts about whether what is distinctive in Müller’s view is plausible. Müller’s account of judgments about reasons for action begins with an account of practical reasoning. What is distinctive about practical reasoning, on Müller’s view – what distinguishes it from other processes that might also produce actions or intentions – is that it is always guided by a reason judgment. As Müller puts it, every episode of practical reasoning necessarily ‘involves a reason judgment in support of its conclusion’ (p. 44). Suppose that I intend to get to Chicago as soon as I can, I believe that flying is the fastest way to get there, and then I reason my way to the intention to fly to Chicago. Müller’s view is that this reasoning must proceed via a reason judgment – in this case, presumably the judgment that the fact that flying is the fastest way to get there is a reason for me to fly to Chicago (rather than drive or walk), given my intention to get to Chicago as soon as I can. The guiding role of reason judgments is then spelled out in dispositional terms, but in a way that distinguishes the work of what Müller calls ‘guiding-attitudes’ from that of the other inputs of practical reasoning. This ends up being especially important, since it helps Müller’s theory separate cases of bad or incorrect reasoning from cases that do not involve reasoning at all. In cases of bad reasoning, although we are guided by reason judgments in the normal fashion, those judgments are false and so guide us to conclusions that are not in fact supported by our starting points. Müller observes, however, that there might be other psychological processes – emotional or otherwise nonrational in nature – that could begin from the very same starting attitudes and arrive at the very same intention but that would clearly fail to count as any sort of reasoning. He then argues, convincingly in my view, that because various other dispositionalist accounts of reasoning do not reserve a special role for guiding-attitudes, they have trouble distinguishing these nonrational psychological processes from instances of bad reasoning. Müller’s central insight in this part of the book is that if our account of reasoning does not include an attitude or process that constitutes our responsiveness to the putative normative force of the reasons for which we act, then it will be difficult for that account to differentiate reasoning from various other psychological processes with similar inputs and outputs. Müller assumes, though, that the only kind of attitude that could constitute our responsiveness to the putative normative force of some consideration is a reason judgment – a judgment to the effect that this consideration is a reason. He assumes, in other words, that the only way to respond to or be guided by the normative force of a reason is to be guided de dicto, or under that description. That is why he concludes that practical reasoning must be carried out under the guise of reasons. If, however, we can respond to the apparent normative force of a consideration without doing so under that description – without deploying or even possessing the concept of a normative reason – then Müller’s case will turn out to be weaker than it seems. Our theory of practical reasoning will still have to include an attitude that constitutes our being guided (and not merely caused) by our reasons, but there will be attitudes other than reason judgments available to play this role. Adding to the pressure on Müller’s claim that practical reasoning is always guided by reason judgments is his recognition that we sometimes act contrary to our normative judgments: Why deny that there can be two processes of reasoning involved in akratic action: one that concludes with a normative judgment in support of one action, and one that concludes with the intention to perform another? … Suppose that you are on a diet and think that you ought not to eat any ice cream. Yet, watching your favourite TV show after a long week of work, you nevertheless form the intention to eat some ice cream. … You realize that in order to eat ice cream now, you first have to buy some at the corner shop. So, you form the intention to go to the shop and walk out the door. (pp. 38–39) This is a striking passage in a chapter defending the view that practical reasoning is carried out under the guise of reasons. If practical reasoning is always guided by reason judgments, how could it fail to be guided by the judgment that the reasons in favour of eating ice cream are outweighed by the reasons in favour of abstaining? Müller attempts to make room for the possibility of akratic actions like this one by distinguishing ‘first-order’ practical reasoning from what he calls ‘metareasoning’ (pp. 80–81). Run-of-the-mill, first-order practical reasoning simply involves the transition from various premise-attitudes to an intention via a guiding reason judgment. It need not involve any higher-order judgments comparing the strengths or weights of some reasons with others. Müller argues that since first-order reasoning can proceed without higher-order metareasoning, his view that practical reasoning is always guided by reason judgments does not entail that practical reasoning is always guided by judgments about what we have most reason to do. He is committed merely to the claim that one cannot reason one’s way to the intention to perform some action without being guided by the judgment that there is at least some reason – however weak – in favour of that action. That seems right: there is indeed logical space for the view that although practical reasoning must be guided by a reason judgment, it need not be guided by one’s judgment about what one has most reason to do. This logical space does not seem very accommodating, though. In other words, although the possibility of clear-eyed akrasia is indeed consistent with the letter of Müller’s account, it is hardly in keeping with its spirit. Recall that the insight underlying this account is the idea that in order for a psychological process to count as reasoning, it must involve some attitude that constitutes one’s sensitivity to the normative or guiding force of reasons. But how sensitive to the normative force of reasons am I if I judge that I have most reason to abstain but end up choosing to indulge anyway? If, as Müller assumes, the attitudes that constitute my responsiveness to reasons must be reason judgments, then my opting for the ice cream is going to look like the product of something other than normative guidance. After all, most of my reason judgments are pointing me away from ice cream. If, on the other hand, the attitudes that constitute my responsiveness to reasons could be something other than reason judgments, then we can see how I might choose to eat ice cream on the basis of practical reasoning even while judging that I have much better reasons to abstain. Together, then, Müller’s recognition that reasoning must involve an attitude or process that constitutes guidance by reasons and his acknowledgment that we sometimes reason our way to akratic actions seem to tell against the view that reasoning must always be guided by reason judgments. Let me turn now from Müller’s account of practical reasoning to his account of the judgments that guide such reasoning. Although reason judgments have normative content – Müller tells us that they guide practical reasoning in virtue of the way ‘their content normatively connects the premise-attitudes to the conclusion attitude’ (p. 48) – they are not beliefs. In other words, the role these judgments play in our psychology is not fundamentally representational. Here Müller’s constructivism clearly diverges from both nonnaturalist and naturalist varieties of realism. It is less clear whether and to what extent his constructivism differs from various forms of expressivism. Müller insists that it does differ. He claims that although reason judgments are not beliefs, neither are they desires or ‘desire-like pro-attitudes’ (p. 101): their characteristic function is not to produce motivation but rather to play ‘the guiding role in episodes of practical reasoning’ (p. 90). The distinction between desire-like pro-attitudes and attitudes that guide practical reasoning is an important one for Müller, given that one of his goals is to show that the constructivist metaethic he is sketching is different from prevailing forms of expressivism. He attempts to shore up this distinction – to show that there could not be an attitude that both produces motivation and guides practical reasoning – by contrasting his account of reason judgments with Allan Gibbard’s expressivist theory of normative judgments in Wise Choices, Apt Feelings. That is a strange choice, given that in 2003 Gibbard published Thinking How to Live, which refined and expanded his view in ways that make it look more like Müller’s. Gibbard’s view in the later book is that judgments about what we ought to do are Bratman-style intentions or plans. These planning states are distinctive practical attitudes: they produce motivation but are not desires. They are also conclusions of practical reasoning. Gibbard devotes relatively little attention to reason judgments, focusing largely on ought judgments, yet he does propose at one point that to judge that some consideration R is a reason to do X ‘is to weigh consideration R toward doing X’ (Gibbard 2003 pp. 189–90). On Gibbard’s view, then, the psychological role of reason judgments is to weigh considerations for and against practical conclusions. This sounds quite a bit like the role such judgments are supposed to play on Müller’s view. This does not show that there is anything wrong with Müller’s view, of course. But it does suggest that his account of the attitude of judging that something is a reason is not quite as novel as he suggests. Genuine novelty does, however, enter the picture with Müller’s account of the truth and falsity of reason judgments. Having unambiguously rejected the idea that the functional role of reason judgments is to represent the world, he also rejects the accompanying idea that such judgments are true (when they are true) because they correspond to the way the world is. As a constructivist, Müller thinks that reason facts are made rather than discovered, and so he takes the grounding here to operate in the opposite direction: ‘reason facts obtain because of the truth of the corresponding reason judgments’ (p. 110). But since the truth of reason judgments plays an important explanatory role (by grounding reason facts) on Müller’s view, he also rejects deflationary accounts of their truth. As Müller observes, ‘a deflated notion of truth is too lightweight, as it were, to play the role that it is assigned in the constructivist account of practical reasons’ (pp. 112–13). What his constructivism needs, then, is a substantive (that is, non-deflationary) account of truth that is different from the familiar, representationalist version of the correspondence theory of truth. Müller suggests that Crispin Wright’s neo-pragmatist conception of truth fits the bill perfectly. Very roughly, Wright’s proposal is that, in some domains of judgment, we can understand a judgment’s truth as a function of that judgment’s justificatory status. Müller refines that rough idea into his constructivist thesis that we can ‘characterize the ground of a reason judgment’s truth in terms of sound reasoning’ (p. 131): a reason judgment is true if and only if – and because – the episode of reasoning this judgment is apt to guide is sound. Müller’s non-representational but also non-deflationary account of the truth of reason judgments is certainly novel. Whether it is plausible or appealing is another matter. On Müller’s view, although reason judgments are made true by facts about sound reasoning, they are not themselves about sound reasoning. Instead, reason judgments are about reason facts. What are reason facts? They are the things that obtain in virtue of true normative judgments. There is at least a hint of circularity here. Müller argues – correctly, I think – that if we can understand what makes reasoning sound without appealing to reason facts, there will be no circularity in his account of what makes reason judgments true. But what about his account of the content of reason judgments? My worry is that Müller’s constructivism does not tell us much of anything about what reason judgments are about. Of course, to expect illumination on this front might be to slide back into a representational theory of reason judgments, which is precisely what Müller rejects. Like expressivists, Müller thinks that what is distinctive about reason judgments is their unique functional role in our psychology. By explaining that role, we thereby explain the content of those judgments ‘obliquely’ (Gibbard 2003, p. 190). But Müller cannot simply leave things there. After all, he repeatedly asserts that reason judgments are able to play their psychological role – that of guiding practical reasoning – in virtue of their content. Recall that Müller tells us that reason judgments guide practical reasoning in virtue of the way ‘their content normatively connects the premise-attitudes to the conclusion-attitude’ (p. 48). This claim now looks frustratingly empty – or at least opaque. This opacity is all the more problematic given the way in which Müller objects to competing accounts of reasoning. He complains that because Paul Boghossian’s rule-following account of reasoning leaves the crucial notion of rule-following unanalysed, that account does not provide the sort of explanation of reasoning we are after (pp. 63–64). Moreover, as we have seen, Müller criticizes dispositionalist theories of reasoning on the grounds that they cannot distinguish cases of bad or incorrect reasoning from processes that are not reasoning at all. He rightly rejects the manoeuvre of saying that in cases of reasoning our dispositions are engaged in some ‘special’ way – where the specialness is left unanalysed – as empty and uninformative (p. 59). But what, then, should we make of his own account of practical reasoning – and in particular his claim that reason judgments guide practical reasoning in virtue of their content? He tells us that reason judgments are ‘concerned with the normative relation between the starting point (the content of the premise-attitudes) and the end point (the content of the conclusion-attitude) of [practical] reasoning’ (p. 45, emphasis in original). But what is this relation? According to Müller, reason judgments play a particular psychological role, and they are able to play this role in virtue of their content. Yet all we are ever told about that content is that it is normative: reason judgments are about reason facts. And so we are left wondering how reason judgments guide our reasoning. Müller could always abandon the claim that reason judgments guide practical reasoning in virtue of their content, but this would leave him without an informative alternative to the dispositionalist and rule-following theories of reasoning he finds inadequate. Without any story about how reason judgments guide our reasoning, Müller has merely named what is distinctive about reasoning rather than explained it. We are left simply with the claim that the difference between practical reasoning and other psychological processes that bring about action is that the former is guided by reason judgments, where all we know about reason judgments is that they are attitudes that guide practical reasoning. Müller opts for his Wright-inspired neo-pragmatist theory of truth because he wants to say that the truth of reason judgments is grounded in facts about sound reasoning without saying that reason judgments are about sound reasoning. It is not immediately clear why he wants to avoid this further claim, though. The resulting view would have exactly the same normative profile as Müller’s: we would have all of the reasons we have on Müller’s view, and those reasons would have the same degree of objectivity. Yet instead of sitting atop Müller’s problematic theory of truth, this normative profile would depend on a reductionist metaethic, one according to which reason facts are reducible to facts about sound reasoning, where the soundness of reasoning can be cashed out in nonnormative terms. This reductionist spin on Müller’s view would preserve the central constructivist idea that ‘reason facts ultimately obtain because of or in virtue of the soundness of certain episodes of practical reasoning’ (p. 21, emphasis in original). It would also, of course, be committed to the idea that reason judgments are representational, and so it would not have the novelty of Müller’s view. But that is not necessarily a bad thing, especially when novelty comes at an unattractively high price. Reductionism obviously has costs of its own, but I worry that Müller does not do nearly enough to help us see why he thinks the costs of his preferred approach are worth bearing. Now let me turn to the final component of Müller’s constructivism: his account of sound reasoning. Müller recognizes that if reason facts are grounded in facts about sound reasoning, sound reasoning cannot – on pain of circularity – be explained in terms of good reasons, and so he presents a series of arguments against the view that sound reasoning is a matter of reasoning in accordance with one’s reasons. These arguments are forceful, but the discussion here is impeded by a frustrating ambiguity. Müller frequently describes the view he is criticizing – which he calls the ‘buck-passing account’ of sound reasoning – as being committed to the thought that the soundness of some pattern of reasoning depends ‘on the reasons for or against engaging in that kind of reasoning’ (p. 147). But there are two rather different sets of reasons this phrase might pick out. On the one hand, there are the reasons for or against engaging in the activity of deliberation – reasons of the sort that are salient in thought experiments in which an evil demon threatens us unless we deliberate a certain way. On the other hand, there are the reasons for or against drawing a particular conclusion or arriving at a particular intention, given a particular set of premise-attitudes. Müller’s criticisms of the buck-passing view all presuppose that it must be framed in terms of the first sort of reason. I think he is almost certainly correct that this version of the view will issue all sorts of counterintuitive verdicts about when reasoning is and is not sound. But the central idea of the buck-passing account – that the soundness of some episode of reasoning depends on the reasons for or against engaging in that kind of reasoning – can also be understood in terms of the second sort of reason. Framed in this way, the buck-passing account becomes the view that the soundness of reasoning from premise-attitudes a, b, and c to conclusion-attitude d is determined, at least in part, by the reasons for and against arriving at that conclusion on the basis of those premises. As far as I can tell, Müller’s criticisms of the buck-passing account leave this rather more plausible version of the view largely unscathed. Müller does, nevertheless, advance the philosophical conversation about sound reasoning when he turns to the task of delineating and defending a viable alternative to the buck-passing view. According to the account he proposes, the soundness of a particular course of practical reasoning is a function of whether it follows the constitutive rules of practical reasoning. This account obviously presupposes that reasoning is the sort of activity that is constitutively governed by certain rules, but Müller does an excellent job motivating and developing that idea. He also makes some illuminating observations about how we might discover the constitutive rules of reasoning – observations that turn into a fascinating discussion of what kind of normative epistemology we should expect his constructivism about practical reasons to yield. This is one of the strongest parts of the book. Here Müller points out that although his constructivism explains reasons in terms of the rules of reasoning, it does not follow that we will discover what our reasons for action are by first discovering the rules of practical reasoning. As Müller puts it, ‘in the theory of practical reasons, as well as elsewhere, explanation and justification can proceed in opposite directions’ (p. 177). So, although constructivists can opt for the sort of constitutivist view that seeks to derive normative conclusions from nonnormative premises about agency or practical reasoning, they may instead opt for a more traditional normative epistemology, one that proceeds via the familiar method of reflective equilibrium. The only difference Müller’s constructivism will make to this epistemology is to ensure that the materials with which we begin our search for equilibrium include not only our various judgments and theories about what we have reason to do but also our judgments and theories about what makes for sound practical reasoning. Having now discussed the three primary components of Müller’s view, let me conclude by taking a step back and assessing his constructivism more holistically. The view’s central normative commitment is that a reason judgment is true if and only if – and because – the episode of reasoning this judgment is apt to guide is sound. This idea strikes me as more than a little plausible, but one need not be a Müller-style constructivist in order to accept it. As I noted above, a reductionist view according to which reason facts are reducible to facts about sound reasoning will have precisely the same normative upshot. Even a quasi-realist expressivist can accept that reason facts obtain in virtue of facts about sound reasoning. Of course, for an expressivist, the ‘in virtue of’ here will have to be deflated, but that is just to say that the grounding will have to be purely normative rather than metaphysical. Müller clearly thinks that neither of these alternatives is satisfying, but when I reached the end of the book, I had a hard time seeing what we gain by avoiding them. Müller makes a compelling case that there is logical space for a constructivist view of the sort he has sketched – a view that is neither expressivist nor reductionist. I think he needs to do more, though, to convince us that this space is worth occupying – or even ‘prima facie attractive’ (p. 3). References Gibbard Allan 2003 , Thinking How to Live ( Cambridge, MA : Harvard University Press) . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Müller Andreas 2020 , Constructing Practical Reasons ( Oxford : Oxford University Press) . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC © Mind Association 2022 This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Constructing Practical Reasons, by Andreas Müller
JF - Mind
DO - 10.1093/mind/fzab089
DA - 2022-01-06
UR - https://www.deepdyve.com/lp/oxford-university-press/constructing-practical-reasons-by-andreas-m-ller-hIvPi4xcum
SP - 531
EP - 539
VL - 132
IS - 526
DP - DeepDyve
ER -