Some statements are entirely about observation. An uncompromising empiricist might say that these statements alone are meaningful; but in that case, theoretical science shares in the downfall that was meant for metaphysics. An uncompromising empiricist might tough it out: science is indeed meaningless, but yields meaningful theorems; or it is entirely about observation, after all; or some of each. But it seems, rather, that science is partly about observation and what we can observe, and partly about the hidden causes and minute parts of what we can observe. And it seems also that science is a package deal, which cannot credibly be split into one part that is meaningful and one part that isn't.

The sensible empiricist, therefore, will retreat. Statements entirely about observation may remain at the core of the meaningful, but scientific statements also will be admitted. Collectively, and even individually, these are at least partly about observation. For an empiricist who wants to be a friend to science, that had better be good enough.

One empiricist who sought to eliminate metaphysics but spare science was A.J. Ayer. Meaningful statements need not be entirely about observation.

Dissatisfied with this use of the notion of evidential relevance, he offers a ‘clearer’ formulation.

In the first edition of Language, Truth and Logic, Ayer proposes that ‘the mark of a genuine factual proposition’ is ‘that some experiential propositions can be deduced from it in conjunction with certain other premises without being deducible from those other premises alone’([1], pp. 38–9).

Berlin objects that the criterion is ‘a good deal too liberal’ and admits patent nonsense. His example, near enough, is that the experiential proposition ‘I dislike this logical problem’ can be deduced from the nonsensical ‘this logical problem is green’ in conjunction with the premise ‘I dislike whatever is green’ without being deducible from the latter premise alone ([2], p. 234). Berlin's point is well taken, if indeed a category mistake is nonsense. But it shows at most that the criterion admits too much. We do not yet know how much too much.

In his introduction to the second edition, Ayer accepts Berlin's point in a greatly extended form. Not only does the criterion admit this or that piece of nonsense; it ‘allows meaning to any statement whatsoever. For, given any statement “S” and an observation-statement “O”, “O” follows from “S” and “if S then O” without following from “if S then O” alone’ ([1], p. 11).

Here Ayer goes wrong. For it may very well happen that the consequent of a conditional ‘if S then O’ does follow from that conditional alone. For instance, O follows from ‘if (P or not P) then O’ and from ‘if not O, then O’. Ayer has just overlooked such cases. In general, O follows from ‘if S then O’ just when ‘S or O’ is analytic.

KNOWING WHETHER AND TELLING WHETHER

Mr. Body lies foully murdered, and the suspects are Green, Mustard, Peacock, Plum, Scarlet, and White. We may take it as settled that one of them did it, and only one. The question is whether Green did it, or Mustard did it, or Peacock, or Plum, or Scarlet, or White. Holmes is on the scene.

If Green did it, then Holmes knows whether Green did it or … or White did it if and only if he knows that Green did it. Likewise if Mustard did it, then Holmes knows whether … if and only if he knows that Mustard did it. Likewise for the other cases. In short, Holmes knows whether … if and only if he knows the true one of the alternatives presented by the ‘whether’-clause, whichever one that is.

Similarly for telling. In at least one principal sense, Holmes tells Watson whether Green did it, or Mustard did it, or Peacock, or Plum, or Scarlet, or White, if and only if Holmes tells Watson the true one of the alternatives presented by the ‘whether’-clause. That is: if and only if either Green did it and Holmes tells Watson that Green did it, or … or White did it and Holmes tells Watson that White did it.

This is a veridical sense of telling whether, in which telling falsely whether does not count as telling whether at all, but only as purporting to tell whether.

INTRODUCTION

This chapter will consider a number of epistemological objections to the moderate rationalism outlined in the previous chapter. What qualifies these objections as distinctively epistemological in character is their underlying concern with whether and why rational insight, as characterized in the preceding chapter, can provide epistemic justification for a belief, in the sense specified in 1.1 above: that is, can yield a compelling reason for thinking that the belief in question is true. There can be little doubt that an apparent rational insight provides some sort of reason for believing the proposition in question. A belief arrived at in this way is certainly not merely arbitrary or capricious and may indeed be psychologically compelling to the point of being inescapable. But none of this shows that the believer in question possesses a genuinely epistemic reason for his belief, and it is this that the objections to be considered attempt to call into question.

I have already remarked that despite the widespread conviction that rationalism is untenable, fully developed and articulated objections to rationalism are difficult to find. This is especially true of the epistemological objections that are the subject of this chapter. Thus, while it is unlikely that anyone who has thought very much about the issue of a priori justification will find the general drift of these objections to be utterly unfamiliar, the specific presentations offered here are largely my own attempts to tease out and develop lines of thought that are usually only briefly hinted at in the literature or, more often, in oral discussion (thus the relative dearth of specific citations).

J. R. Lucas argues in “Minds, Machines, and Gödel” that his potential output of truths of arithmetic cannot be duplicated by any Turing machine, and a fortiori cannot be duplicated by any machine. Given any Turing machine that generates a sequence of truths of arithmetic, Lucas can produce as true some sentence of arithmetic that the machine will never generate. Therefore Lucas is no machine.

I believe Lucas's critics have missed something true and important in his argument. I shall restate the argument in order to show this. Then I shall try to show how we may avoid the anti-mechanistic conclusion of the restated argument.

As I read Lucas, he is rightly defending the soundness of a certain infinitary rule of inference. Let L be some adequate formalization of the language of arithmetic; henceforth when I speak of sentences, I mean sentences of L, and when I call them true, I mean that they are true on the standard interpretation of L. We can define a certain effective function Con from machine tables to sentences, such that we can prove the following by metalinguistic reasoning about L.

1. C1. Whenever M specifies a machine whose potential output is a set S of sentences, Con (M) is true if and only if S is consistent.

2. C2. Whenever M specifies a machine whose potential output is a set S of true sentences, Con (M) is true.

3. C3. Whenever M specifies a machine whose potential output is a set S of sentences including the Peano axioms, Con (M) is provable from S only if S is inconsistent.

INTRODUCTORY COMMENTS

World utilitarianism, as presented in the papers in Part I of this book, is a view in normative ethics. It purports to tell us how we ought to behave. According to that view, at each moment, a person ought to bring about the states of affairs that occur in the best worlds then accessible to him or her. This is my reformulation of what I take to be the core insight of consequentialism. It is my interpretation of the idea that our moral obligation is always to do the best we can – to make the world as good as we can make it.

As stated, the theory is pretty abstract. It purports to tell us something of the structure of the concept of moral obligation, and it purports to tell us something about the connection between normative concepts (moral rightness, wrongness, and obligatoriness) and axiological concepts (good and evil). But it does not tell us what makes one world better than another.

There are many different axiological theories. These include simple hedonism, qualified hedonism, eudaimonism, pluralism, and many others. They generate different rankings of worlds. World utilitarianism might be linked to any of these axiologies – and, of course, we will get different normative results if we link it to one of them rather than to another.

Because hedonism is one of the oldest, most plausible, and simplest of axiological theories, it is reasonable to take it as our starting point.

INTRODUCTORY COMMENTS

The title of this essay suggests that it concerns the controversy about qualified hedonism. However, overt discussion of that controversy takes up just a tiny fraction of the paper. Most of the paper is devoted to a more fundamental issue – how to formulate an axiological theory, with a few versions of hedonism as the examples. I spend a lot of time trying to show that standard formulations of hedonism are so confused that it's pointless to consider whether they are consistent with each other, and so on.

The paper has some clearly “negative” aims. I want to show that typical formulations of hedonism are seriously defective. Thus, for example, I discuss the sort of formulation often encountered in the writings of Plato, Aristotle, Moore, and many others. These philosophers sometimes state the central thesis of hedonism by saying “Pleasure alone is intrinsically good.” The problem here is that “pleasure” seems to be a singular noun – the name of some feeling or some property. But this can't possibly be right. Hedonists do not want to say that there is exactly one intrinsically good thing. They want to say that lots of things are intrinsically good (this life, that world, the other consequence, many different experiences), and they want to say that these things have different intrinsic values and they occur at different times. So, formulations such as the ones cited must have the number wrong.

For as long as I can remember, it has seemed obvious to me that our fundamental moral obligation is to do the best we can – to make the world as good as we can make it. Naturally, I was delighted when I discovered that many others – the utilitarians – maintained approximately the same view. However, my delight turned to consternation when I began to look more closely at the philosophical literature on utilitarianism. One tremendous problem was that the received formulations of the view fail to express the utilitarian insight. Indeed, most received formulations fail to express any insight – they are simply incoherent, as Hector-Neri Castafieda showed in a series of brilliant papers published in the late 1960s and early 1970s.

Utilitarianism has been linked to hedonism since the time of Bentham and Mill. The fundamental insight of hedonism (“pleasure is The Good”) seemed attractive, too. Yet again when I took a look at the literature on hedonism I discovered near chaos. There was enormous confusion about the nature of pleasure, and equally great confusion about precisely what hedonists would want to say about pleasure, assuming that they could reach any agreement about what it is. I could not find a clear statement of the intended theory of value.

Furthermore, even if these underlying conceptual difficulties could be solved, it appeared that the resulting normative theory would surely confront the most troublesome of moral objections.

INTRODUCTORY COMMENTS

If we take our utilitarianism narrowly, we take it simply as a principle about the normative status of actions. If we take it widely, we take it as a family of principles about the normative status of all sorts of things – actions, motives, traits of character, and so on. Because of its more extensive application across the evaluative board, wide utilitarianism is more impressive.

However, it is not clear that one can consistently maintain a utilitarianism of acts and a utilitarianism of, for example, motives. In his important paper “Motive Utilitarianism,” Robert Adams argued that these views are inconsistent in some cases. Thus, the prospects for wide utilitarianism are dim. Adams drew an even more sweeping conclusion: The moral point of view cannot be the utilitarian point of view.

My overt aim in Essay 4 is to show that Adams's argument does not succeed. If we formulate our act utilitarianism and our motive utilitarianism correctly, they are bound to be consistent. Along the way, I try to establish some other small points about Adams's formulation of the doctrines and arguments.

Adams made use of a traditional formulation of act utilitarianism. This is his “AU,” and it is based on the standard concepts of act, alternative, and consequence. Although there is some confusion about this in the essay, it is reasonable to assume that he meant to make use of a similarly traditional formulation of motive utilitarianism.

INTRODUCTORY COMMENTS

Essay 7 has a “negative” aim and a “positive” aim.

In the first several sections of the essay, I explain one of the most influential current views about the nature of pleasure. This is the view of Sidgwick, Brandt, Alston, and others, according to which a feeling is correctly said to be a pleasure if the person who has that feeling likes it for its own sake, or enjoys it, or wants it to continue, or (in Sidgwick's words) “apprehends it as desirable in itself.” In general, according to this view, any sort of feeling might be a pleasure – it doesn't matter how it “feels.” A feeling is a pleasure if the one who feels it has an appropriate attitude toward it when he or she has it.

My negative aim in the essay is to show that it is impossible to combine any such “Sidgwickian” conception of pleasure with the classic Moorean conception of intrinsic value. The problem, in the abstract, is simple: On the Moorean conception of intrinsic value, intrinsically good things are supposed to have their values in virtue of their own intrinsic natures – because of the way they are “in themselves.” Yet on the Sidgwickian conception of pleasure, no feeling is “intrinsically a pleasure.” If a feeling happens to be a pleasure, it is so because of an extrinsic feature – the one who experiences it has the appropriate attitude toward it.

INTRODUCTORY COMMENTS

In Essay 8, I confront the most profound moral objection to act utilitarianism – the objection from justice. It may be useful here to present a particularly stark version of the objection. Suppose there are two sorts of people in the world, the Haves and the Have Nots. Suppose that the Haves have always enjoyed large amounts of pleasure; suppose that they never worked for the goods that provide the pleasure but simply stole them from the Have Nots; suppose, finally, that the Haves are nasty and cruel to the long-suffering Have Nots.

On the other hand, suppose that the Have Nots have struggled and suffered all their lives; they have not enjoyed any pleasures even though they are the rightful owners of the means to pleasure; suppose, finally, that the Have Nots are kind-hearted and decent – they treat each other and the nasty Haves with respect.

Suppose that it is now in your power to distribute some extra goods. You can give the new goods either to the Haves or to the Have Nots – but no other option is available. We stipulate that the amount of pleasure that will result will be exactly the same no matter which you do. The only difference is that if you give the goods to the Haves, they will experience the pleasure and the Have Nots will experience the disappointment of getting the short end of the stick again.