Those who disturb the hornets’ nest better be thick skinned. Keeping that in mind, thus far I kept my silence in the face of the adverse reactions my essay “On What One Knows” has provoked in the literature. As to the merits of the various objections, moreover, I trusted the judgement of the philosophical community to sort them out, and either recognize their futility, or make the necessary adjustments in my arguments without the need to reject the conclusions.

In the case of Dunn and Suter, however, I decided to make an exception. The reason is that they indeed succeeded in pointing out certain difficulties in my position, which are not due to misunderstandings, conflicting linguistic intuitions, or other superficial sources. And, second, meeting their objections gives me the opportunity of proposing a significant refinement of my theory, the need for which I increasingly felt in the meanwhile myself.

A little-discussed aspect of the “conservative” position on abortion involves the claim that human beings exist at all stages of gestation — that even the human fertilized ovum (zygote) is a human being. There are good reasons of both a practical and theoretical nature why this particular idea has received so little attention in the abortion controversy. On the practical side there is the fact that zygotes are rarely if ever the subjects of abortion decisions; and on the theoretical side, there is the prima facie implausibility of the idea itself — an implausibility stemming from the fact that human zygotes possess none of the features that distinguish clear-cut examples of human beings from other living things.

I would like to suggest, however, that there are good reasons — again both practical and theoretical — for examining with some care the idea that human zygotes are human beings. For even if this idea is not central to the abortion controversy, it has clear relevance to the morality of using intrauterine devices (I.U.D.'s) as a means of procreation control.

According to its proponents, eliminative materialism is a promising alternative to the problem-beset identity theory. A popular materialist strategy for handling thoughts, sensations, beliefs, and intentions has been to identify such mental states with physical states. The identity theorist, however, must confront difficult questions concerning identity criteria, essential properties and category mistakes. The eliminative theorist wants to bypass these problems by maintaining that we will someday discover that mental entities simply do not exist. If there are no mental entities, then we need not worry about determining their essential properties or determining the conditions under which they would be identical with physical entities. ‘Category mistake’ objections, which might apply to an identity thesis, would clearly be irrelevant.

As formulated by its chief proponent, Richard Rorty, the eliminative theory is suspiciously straightforward. Indeed, a closer look at its underpinnings reveals a highly questionable assumption. I will propose a reformulation of eliminative materialism which avoids this assumption.

Might God have a body? The overwhelming answer from within Christian orthodoxy is a resounding “No”. A concept of God adequate for sophisticated theism must, it is held, involve the notion of incorporeality: any being which had a body would, on that ground alone, be disqualified as a contender for the title “God” irrespective of other considerations.

Part of the reason forth is insistence on God's incorporeality is that God is held to be the being who is supremely worthy of worship. Now, if God were embodied in the manner that the Greek gods were conceived to be, it is alleged that such a “Zeus-like” deity would not be worthy of worship. Therefore either we must dismiss all thought of an embodied God, it is urged, or else we must cease to worship him, thus in effect dismissing Christianity. And there is an additional ingredient: if we choose the former course, and declare the doctrine of the incorporeality of God, then although we preserve the concept of a God who is worthy of worship, we preserve it at a very great cost.

Berkeley's idealism consists of the following claims. Objects such as chairs, apples, mountains, and our bodies are combinations of sensible qualities. Sensible qualities and combinations of such are ideas or sensations. In the philosophical sense of ‘substance’ there is no such entity as a substance. There are minds which perceive and will: When a mind perceives it has sensations or ideas; and when a mind wills it produces or causes sensations or ideas. These claims are grounded in the ontological and epistemological framework of assumptions, distinctions and arguments informing Berkeley's doctrine that esse est percipi aut percipere. The purpose of this paper is to explore and expose the main features of that framework.

What is the ontological force and import on Berkeley's claim that esse est percipi? One influential recent suggestion1 is this: Berkeley holds that (1) material substances do not exist, (2) the analysis of sensible things yields nothing but sensible qualities, (3) minds are mental substances and (4) no sensible quality or collection of sensible qualities can exist without inhering in some substance.

Alice Ambrose once criticized Moore for treating the proposition ‘There are external objects’ as an empirical one. She said that those who denied that we could know this proposition to be true would not accept any evidence as going against their denial of it, and were not regarding the issue of its truth as empirical. She also maintained that one could not point out an external object in the way in which one could point out a dime or nickel and alleged on these grounds that saying that there are external objects is not the same sort of thing as saying that there are coins. The issue arose concerning Moore's paper, “Proof of an External World.“

In “Reply to My Critics,” Moore pointed out that he had been concerned in “Proof of an External World” not to prove that we know that there are external objects, but rather to prove that there are external objects. But he applied Ambrose's remarks to that proposition. Moore rightly asserted that the fact that someone denies that there are external objects and treats all evidence as irrelevant to the issue does not show that the proposition he denies fails to be empirical.

The problem of induction has long been one of the central problems of empiricist epistemology. There are two main versions of this problem: to justify a strictly universal statement on the basis of a finite set of singular statements and to justify a new singular statement on the basis of some finite set of accepted singular statements. In both cases it is assumed that we have a set of singular statements with which to begin and that these singular statements are, somehow, provided by observation; neither of these assumptions will be disputed in this paper. But there is another assumption without which there might well be no problem of induction:

It is only because philosophers assume A that they are faced with the problem of showing how we can infer from statements which have been justified by observation to others which have not. It is this assumption which will concern us here, for if there are no adequate grounds for accepting A, the problem of induction may not arise at all.

Many recent discussions of the identity and individuation of actions focus on attempts to find satisfactory answers to questions like, “When, if ever, is a shooting (stabbing, poisoning, etc.) a killing?” Those who attempt to answer such questions divide themselves, on the whole, into two opposing groups. I. Thalberg has conveniently labelled the members of one group ‘unifiers’, and the members of the other group ‘multipliers’.

The unifier account is commonly attributed to philosophers such as G. E. M. Anscombe and Donald Davidson. Proponents of this account characteristically believe that many non-basic actions (where ‘basic’ is defined in terms of bodily movements) such as shootings, killings, etc., are identical with those basic actions by means of which the non-basic ones are performed. This position is indicated in the following remarks:

A great deal of philosophical consideration has been given in recent years to the issue of justice. In large measure this effort has focused upon justice in . relation to social institutions, to the distributive question of disbursing social benefits on the one hand or punishments and burdens on the other hand. Essentially we may view justice as having two basic requirements. In the first place we need conditions assuring impartial treatment of individuals and, in the second place, conditions for reasonable treatment of individuals — as injustice may arise either if relevantly similar persons are accorded dissimilar treatment or if they receive similar but unreasonable treatment.

In this article I shall examine a number of devices which have been proposed to meet the first requirement — that of impartiality. Social theorists have pursued impartiality with strategies ranging from the disinterested Ideal Observer of many utilitarians to the ‘veil of ignorance’ conditions in John Rawls's revitalized contract theory. I shall demonstrate that the preoccupation with impartiality in social institututions has had an unfortunate consequence. It has obscured the original motive for impartiality and led to a misunderstanding of the concept.

Suppose that a scientific theory X is well confirmed, and either states or (in conjunction with certain specified systems of logic, mathematics, and semantics) implies a statement like “classes exist”. Suppose, if the statement is not explicit but implied, that the specified systems are ‘accepted’ by two individuals, R and F, as the ‘best available'. Suppose, finally, that both R and F, in some yet to be explained sense of the word, ‘accept’ and use theory X to regulate their experience. Then we have something very like the situation discussed by Putnam and van Fraassen in their debate over ‘fictionalism'. I will argue that, in this situation, there is a great mystery over what would separate a fictionalist F from a realist R. Neither Putnam nor van Fraassen seems to be conscious of the problems involved.

Fictionalism, according to its opponent, Putnam, states

Alvin Plantinga has recently argued that there are certain propositions which are necessary but known only a posteriori. If Plantinga is correct then he has shown that the traditional view that all necessary truths are knowable a priori is false. Plantinga's examples deserve special attention because they differ in important respects from other proposed examples of necessary a posteriori truths. His examples depend on a certain conception of possible worlds and in particular on his conception of the actual world. It will be argued that these examples of necessary a posteriori propositions can be understood in two different ways. According to one way of understanding Plantinga, the propositions turn out to be contingent a posteriori truths, and according to the other way they turn out to be necessary a priori truths. The plausibility of Plantinga's position is due to a confusion between the two possible interpretations.

It is immensely difficult to give a philosophically adequate account of mental imagery. Peter F.R. Haynes (Mental Imagery, this Journal 6 [1976), pp. 709–19) objects to the standard accounts, and offers one of his own which avoids the standard difficulties. Unfortunately it in turn seems to lapse into incoherence.

Haynes rejects Cartesian accounts which would make images private objects in non-physical space. He also rejects current alternative views: both Rylean or behaviourist ones; and also intentionally complex ones, which assert that the relevant terms change their meaning (so that “red” means something different when applied to red balls and red images). He thinks that J. J. C. Smart's “topic neutral” approach may supply “the basis for a correct, intentionally simple, intuitively satisfying account of mental imagery” (p. 707). But an adequate account must also be neutral, as Smart's is not, between different theories of mind (pp. 707–8). In addition to avoiding these difficulties an adequate theory has certain other desiderata; in particular Haynes places valuable emphasis on the need to explain the fact that we can sometimes mistake images for real perceptions or, more real perceptions for images.

Philosophers, psychiatrists, and social scientists would welcome clarification of the distinction between rational and irrational desires. It may be proper to say that rational desires are those which manifest rationality (or which at least do not conflict with its manifestations). But since this seems a rather unilluminating characterization, philosophers sometimes offer definitions of what constitute such manifestations of rationality. I shall consider definitions provided by John Rawls and Richard Brandt. Their definitions are unsatisfactory mainly because they include subjunctive conditionals. An alternative approach, which avoids conditionals, is attractive. But it encounters so many additional problems that I shall conclude that we are not now in a position to define rationality in this area and must treat it as a state or disposition which to date has only been partially characterized. Thus, if we want a definition of the difference between rational and irrational desires, we must at present settle for the rather unexciting one mentioned above.

On 5 Nivôse of the Year II (Christmas Day, 1793), addressing the National Assembly on behalf of the Committee of Public Safety, Robespierre declared: “The theory of revolutionary government is as new as the revolution that has brought it about. It should not be sought in the books of political writers, who have not foreseen this revolution, nor in the laws of tyrants, who content to abuse their power, are little concerned to investigate its legitimacy.” It is tempting to suppose Robespierre is exaggerating. Whatever the relation of revolutionary government to “the laws of tyrants,” Robespierre's ties to the republican tradition in political philosophy, and especially to Rousseau, are striking and nearly everywhere acknowledged, not least by Robespierre himself. However, I am not concerned here with the question of Rousseau's influence upon Robespierre, but with the rather different question of conceptual affinity. Even if Robespierre's intent were just to put The Social Contract into practice (if need be, by despotic means), it is still appropriate to ask whether Robespierre is a Rousseauean.

Not so long ago it was fashionable to claim that it is not the moral philosopher's business to say what things are good or what actions we should perform. This view is succinctly stated by A. J. Ayer:

On the other hand, in direct opposition to this, recently many philosophers actively moralize, in the sense that they argue for substantive normative ethical positions. In doing this they tend to assume but not to explore seriously two views: (1) that the notion of moral expertise is unproblematic, and (2) that moral philosophers in particular are moral experts. My aim in this paper is to promote the exploration of these questions. (1) is logically prior to (2). Nevertheless I begin with discussing (2); for puzzles about expertise in morals naturally emerge when one examines the credentials of candidates for the job.

Philosophical logic has its problem-children; and among these the Principle of Substitutivity of codesignating expressions — the linguistic spawn of Leibniz's law—has achieved a place of prominence. It has become increasingly apparent that a certain style of linguistic analysis, which seeks to impose formal regimentation ruled by the constraints of classical quantification theory, does not yield results with the kind of uniformity and elegance one should hope for from a satisfyi.ng theory. The root of the difficulty, I believe, bears upon the answers to fundamental questions concerning the nature of cognitive agents and the purposes which their use of language is designed to further. If this is the case, one might expect to find some of the trouble surfacing at precisely those points where the language is used to convey information about the cognitive activities associated with the use of language by such agents. So it is.

The central problem of this essay is the interpretation of referential expressions in statements involving propositional attitudes, with the focus on belief contexts. But I shall begin by placing the discussion within a broader theoretical framework.

In a recent article entitled “Scepticism and Knowing That One Knows” (Canadian Journal of Philosophy 6, pp. 655–63), Michael Hall poses a sceptical argument based upon two rules of inference:

where ‘Kp’ stands for ‘it is known that p'. The argument goes:

Hall (rightly, in my opinion) attacks R2, but he says little about R1. I wish to add a few words of warning about R1 as well.

Perhaps it is best to begin on a semi-autobiographical note. In my liberal days, Mill's arguments in On Liberty for freedom of speech struck me as a paradigm of rationality: the force and eloquence of his presentation, I then thought, could not fail to impress themselves on any mature member of our species. But I am a Marxist now, and more and more of my former political beliefs now strike me as less and less tenable. It was considerations such as the one that follows which have recently led me to modify drastically, and in some cases abandon entirely, my former political views vis à vis the question of freedom of speech.

1. What is the Problem of Personal Identity?

Locke posed the problem of personal identity in one brief question, “What makes the same person?” This formulation is deceptively simple. My aim is to offer a new interpretation of the problem and to suggest a method for finding a solution.

Investigations of personal identity are usually cast in terms of finding the criterion for personal identity. Yet talk of criteria is ambiguous. In one sense of the term, the criterion of personal identity would be special evidence for personal identity. According to the second usage, the criterion of personal identity would be that state of affairs for which evidence for personal identity is evidence. We will not be concerned with the first construal. On that reading, the problem of personal identity would be an epistemological issue. But the basic question concerning personal identity, “What is one person?” does not appear to be an epistemological question.