Opting for the Best
“This is a beautiful and insightful work on an important set of topics. It systematically brings together a number of important and interrelated issues.” —Peter Vallentyne, Florence G. Kline Professor of Philosophy, University of Missouri
Abstract: The book concerns what is, perhaps, the least controversial normative principle concerning action: you ought to perform your best option—best, that is, in terms of whatever ultimately matters. The book sets aside the question of what ultimately matters so as to focus on the following questions. What are our options? Which options do we assess directly in terms of their own goodness and which do we assess in terms of the goodness of the more encompassing options of which they’re a proper part? What do we hold fixed when assessing how good an option is? Do we, for instance, hold fixed the agent’s present beliefs, desires, and intentions? And do we hold fixed the agent’s predictable future misbehavior? The book argues that addressing these sorts of questions is the key to solving certain puzzles concerning what we ought to do, including those involving supererogation, indeterminate outcomes, overdetermined outcomes, and predictable future misbehavior.
Purchase from Amazon: https://www.amazon.com/Opting-Best-Oughts-Options-Oxford/dp/0190945354/ref=sr_1_1?crid=3HLAQ89YILAWC&keywords=opting+for+the+best&qid=1560281446&s=gateway&sprefix=opting+for+the+%2Caps%2C270&sr=8-1
Upcoming Event: There will be an author-meets-critics session on the book at the Pacific Division Meeting of the American Philosophical Association in April 8–11, 2020, San Francisco, CA. The critics are Chrisoula Andreou, Peter A. Graham, and Ralph Wedgwood. And G. Shyam Nair will be chairing the session.
CHAPTER 1: Opting for the Best
This chapter argues for the opting-for-the-best view: the view that, for any subject S and any member φ of a certain subset of S’s options, S ought to φ if and only if φ is the best member of this subset in terms of whatever ultimately matters. It’s argued that we need to restrict ourselves to a subset of the subject’s options because of the problem of act versions. This problem arises in the following sort of case. One’s best option is to kiss one’s partner passionately. Kissing her nonpassionately would be disastrous. Indeed, it would be better not to kiss her at all. But, unfortunately, if you were to kiss her, you would do so nonpassionately. Should you kiss her? Of course, you have to kiss her to kiss her passionately, which is your best option. But kissing her would result in your kissing her nonpassionately, which is your worst option.
1.1 The Opting-for-the-Best View
1.1.1. Perhaps, some options that ought to be performed are suboptimal
1.1.2. Not all possible events are eligible for deontic status
1.1.3. Alternatives matter
1.1.4. Oughts versus obligations
1.1.5. Best option versus best outcome
1.1.6. The best must be sufficiently good
1.1.7. ‘Option’ versus ‘can’
1.1.8. Objective oughts versus subjective oughts
1.1.9. Directive oughts versus evaluative oughts
1.2 The Ecumenical Nature of the View
1.3 A Potential Objection to the Opting-for-the-Best View
1.4 Remaining Controversies and the Plan for the Rest of the Book
CHAPTER 2: What are our options?
This chapter argues that whereas any event can be good or bad, only certain events can be obligatory. This, it’s argued, is because obligations have two additional roles to play beyond mere evaluation. One of these additional roles is a directive one. Obligations, unlike mere evaluations, must direct us to respond in certain ways, and for these directives to have any point at all, they must direct us to respond only in those ways that we have the option of responding. The other role that obligations play is an inculpatory one. For there is a conceptual connection between failing to abide by an obligation and being accountable for so failing—at least, absent some suitable excuse. And since it’s inappropriate to hold subjects to account for events over which they lacked control, we must restrict a subject’s options to those events over which she exerted the relevant sort of control.
2.1 The Need to Restrict What Can Count as an Option
2.2 The Necessity of Control
2.3 The Sufficiency of Control
CHAPTER 3: What’s the relevant sort of control?
Our options include all and only those events over which we exert the relevant sort of control. This chapter argues that the relevant sort of control must be complete as opposed to partial and synchronic as opposed to diachronic (i.e., the sort of control that we exercise at a moment in time rather than over a span of time). And it’s argued that the relevant sort of control is personal as opposed to subpersonal (i.e., the sort of control that makes our actions intelligible in terms of reasons and not just in terms of cause and effect). Last, it’s argued that the relevant sort of control is not the sort that we exert directly over our intentional acts by forming certain volitions, but the sort that we exert directly over our reasons-responsive attitudes (e.g., our beliefs, desires, and intentions) by being both receptive and reactive to reasons.
3.1 Complete and Synchronic Control
3.2 Personal Control versus Sub-Personal Control
3.3 Rational Control versus Volitional Control
3.3.1. Formations of attitudes
3.3.2. Mixed acts
3.3.3. Automatic, unthinking acts
3.3.4. Acts stemming from volitions that weren’t under the subject’s rational control
3.3.5. Acts that manifested a poor quality of will and were expressive of one’s deep self but were not under one’s rational control
3.4 Volitional Control and the Trouble with Insisting on Always Tracing Back to Some Intentional Act
CHAPTER 4: Which options have their deontic statuses in virtue of their own goodness?
The problem of act versions arises because one’s best option can be a version of a bad option. For instance, kissing passionately is a version of kissing. But it may be that although kissing passionately is one’s best option, kissing is a bad option. For it could be that, as a matter of fact, one would kiss nonpassionately if one were to kiss. This chapter argues that the best solution to this problem lies with adopting maximalism. On this view, the only options that have their deontic status in virtue of their own goodness are maximal options—options that are entailed only by evaluatively equivalent options (those being options that are identical in terms of whatever ultimately matters). According to maximalism, then, one ought to kiss even if one would, as a matter of fact, kiss nonpassionately. On this view, one ought to kiss, because one ought to kiss passionately, and kissing passionately entails kissing.
4.1 Omnism (All Options) and the Problem of Act Versions
4.2 Nonnullusism (Only Some Options)
4.3 Nullusism (No Option)
4.4 Supererogation and the Latitude Problem
4.5 Supererogation and Evaluative Inheritance
4.6 Three Objections to Maximalism
4.6.1. Ross’s paradox
4.6.2. The arbitrariness objection
4.6.3. The implausible grounds objection
4.7 The implications of maximalism
CHAPTER 5: Rationalist Maximalism
Rationalism holds that our current options consist in all and only those events over which we presently exert rational control. Maximalism holds that the only options that have their deontic status in virtue of their own goodness are maximal options. When we combine these two, we get rationalist maximalism. This chapter argues that there are reasons to accept rationalist maximalism apart from the reasons that were already given for accepting each of its two components. First, it avoids the sorts of objections to which other versions of maximalism are susceptible. Second, it provides us with a plausible alternative to both actualism and possibilism. And, third, it is uniquely well situated to accommodate the idea that a moral theory ought to be such that the agents who satisfy it, whoever and however numerous they may be, are guaranteed to produce the morally best world that they could (in the relevant sense) together produce.
5.1 Three More Objections to Maximalism
5.1.1. The ‘did φ’-implies-‘had the option to φ’ objection
5.1.2. Gustafsson’s objection
5.1.3. The Professor Procrastinate objection
5.2 The Actualism-Possibilism Debate versus the Omnism-Maximalism Debate
5.3 The Principle of Moral Harmony and the Problem of Overdetermination
CHAPTER 6: Maximalism and the Ought-Most-Reason View
Maximalism forces us to deny either that we ought to do whatever we have most reason to do or that how much reason there is to perform a given option is directly proportional to how good that option is. In this chapter, I argue that we should deny the latter. We should instead hold that how much reason there is to perform a given option is directly proportional to how good the best version of that option is. Thus, how much reason I have to kiss my partner is not a function of, say, the bad consequences that would result from my doing so (for assume that I would in fact kiss her nonpassionately if I were to kiss her), but a function of the good consequences that would result from my kissing her passionately.
6.1 Omnism and Maximalism about Reasons
6.2 The Inheritance Problem for Omnism
6.3 The Intuition Problem for Omnism
6.4 The All or Nothing Problem for Omnism
6.5 Maximalism and the Basic Belief
6.6 Objections to Maximalism about Reasons
6.6.1. Incorrect explanation
6.6.2. Incorrect weights
6.6.3. Incorrect account of instrumental reasons
6.6.4. Incorrect account of reasons to perform acts with side effects
CHAPTER 7: Which, if either, are we to assess directly in terms of what ultimately matters: our options or their prospects?
This chapter argues that our uncertainty with respect to whether a subject has violated a constraint in φ-ing can stem not only from our limited knowledge of the actual world but also from the indeterminacy that’s inherent in certain possible worlds. And this means that constraint-accepting theorists will need to find a way to deal with such uncertainty even when it comes to giving an account of objective rightness. And it’s argued that the best way for constraint-accepting theorists to deal with this issue is to adopt a teleological theory. In doing so, she needn’t give up on any of our common-sense moral intuitions. Nor would she need to give up the deontological idea that constraints are ultimately grounded in our duty to respect people and their capacity for rational, autonomous decision making. Thus, it’s argued that we should accept a teleological version of rationalist maximalism.
7.1 A Case Study: Teleology, Deontology, Nonteleology, and Agent-Centered Constraints
7.2 Ignorance versus Indeterminacy
7.3 Indeterminacy and Subjective Rightness
7.4 Indeterminacy and Objective Rightness
7.5 A Teleological Approach
7.6 Rationalist Teleological Maximalism
CHAPTER 8: Rationalist Teleological Maximalism
This chapter demonstrates how rationalism, maximalism, and teleology operate together to determine the deontic statuses of our options. And the chapter explains what rationalist teleological maximalism’s main virtues are. Last, the chapter explains why it’s important to work out the structure of our normative theories.
8.1 An Illustration of How Rationalist Teleological Maximalism Works
8.2 Rationalist Teleological Maximalism’s Many Virtues
8.3 On What Matters and the Importance of Structure
It’s often unclear what we ought to do. Much of the time this is because it’s unclear what matters. Suppose, for instance, that we’re poultry farmers wondering whether we ought to put our chickens in cages or let them roam free. We know that we’ll make more money if we put them in cages but also that they’ll suffer more if we do. Still, if we want to know what we ought to do, we need to know whether their suffering matters, and, if so, how much it matters as compared to our profits.
Sometimes, though, we know exactly what matters and by how much, and it’s still unclear what we ought to do. To illustrate, suppose that what matters is my health and that what would be best for my health is my quitting smoking and eating less (JACKSON & PARGETTER 1986). Assume, though, that if I were to quit smoking, I would eat more, not less. It’s not that I couldn’t quit smoking and eat less; it’s just that I wouldn’t. And assume that my weight is such a concern that I would be better off maintaining the status quo than quitting smoking and eating more. So, should I quit smoking? On the one hand, I must quit smoking to do what’s best, which is to quit smoking and eat less. On the other hand, my quitting smoking would result in my eating more, which is worse than the status quo. So, figuring out what I ought to do can require knowing more than just what matters. I must also know how what matters determines what I ought to do. Should I, for instance, do what’s entailed by my doing what’s best in terms of what matters even if this would, as it happens, result in something worse than the status quo?
This book seeks to address the issues that remain after setting aside the question of what matters. These issues include, What are our options? Do I, for instance, have the option of typing out the cure for cancer if that’s what I would in fact do if I were to have the right intentions at the right times (e.g., the intention to type the letter T at t1, the intention to type the letter H at t2, the intention to type the letter E at t3, etc.)? Which options do we assess in terms of their own goodness and which do we assess in terms of the goodness of the options that entail them? How do the things that matter determine which options we ought to perform? If, for instance, one of the things that matters is my not violating anyone’s rights, do we assess a given option in terms of whether it entails my violating someone’s rights or do we first assess its prospect (i.e., its possible outcomes and their associated probabilities) in terms of this and then assess the option in terms of how its prospect compares to those of its alternatives? And what if there is uncertainty or indeterminacy concerning whether an option would involve my violating someone’s rights?
In this book, I argue that we should set aside the issue of what matters if we want to solve certain puzzles concerning what we ought to do, including puzzles concerning supererogation, indeterminate outcomes, overdetermined outcomes, predictable future misbehavior, and good acts that entail bad acts.