People commonly believe that their present selves are sufficiently connected to certain past selves-ones that they remember having been-that they do and should feel responsible for the acts performed by those past selves. They also commonly believe that their present selves are sufficiently connected to certain future selves-ones that they anticipate becoming-that they do and should feel egoistic concern for those future selves. The reason commonly given to support these beliefs is that their present selves are the same persons as their past and future selves.
 

This commonsensical justification for responsibility and self-concern is philosophically puzzling. Even if one once was a certain self, arguably one is no longer that self. And even if one will one day be a certain self, arguably one is not yet that self. So, why should the person who one is now--one's present self of the moment, if you will-feel responsible for the acts of any past self or egoistic concern for the welfare of any future one? Although many philosophers have defended the commonsensical justification, some have claimed that there isn't any reason necessarily why present selves should feel responsible for the acts of their past selves or egoistic concern for the welfare of their future ones. Among contemporary philosophers, Galen Strawson currently espouses this skeptical view.⁽¹⁾
 

Like many philosophical puzzles, this one-over responsibility and egoistic concern-has a long history that goes back to ancient Greece. By the beginning of the fifth century B.C.E., it was common for Greek intellectuals and artists to believe that everything is in a state of constant flux, including human beings. Some of them recognized that if all things are in constant flux, then arguably no one lasts for long and, hence, there is no metaphysical basis for responsibility and self-concern. For instance, in a scene in a play by the comic playwright Epicharmus, which probably dates from the opening decades of the fifth century B.C.E., there is an exchange between two characters, one of whom has no funds but owes the other money. The one without funds--the debtor-claims that just as a pile of pebbles from which one pebble has been removed becomes a different pile of pebbles, anything that undergoes any sort of change thereby becomes a different thing. The one to whom the debtor owes the money--the lender-agrees. 'Well, then', says the debtor, 'aren't people constantly undergoing changes'? 'Yes', agrees the lender. 'So', the debtor concludes triumphantly, 'I owe you nothing since I'm not the same person as the one who owes you money'. Exasperated, the lender hits the debtor, who then protests loudly that he has been physically abused. 'Don't blame me', replies the lender, 'I'm not the same person as the one who hit you moments ago'.⁽²⁾
 

Despite such early recognitions of the intertwined problems of personal identity, responsibility and self-concern, those Greeks whose views were most influential-Plato and Aristotle-did not pursue such issues for long. Instead, they adopted a quick-fix solution. Proposed by Plato, this solution was that our essential selves-our souls-are not in flux but, rather, eternal substances, whose identity and self-concern are unaffected by changes in body or mind. Although not all thinkers of the time accepted this solution, it so simplified the problem, while at the same time securing immortality, an implication most congenial to Judaism, Christianity and Islam, that the individual soul as an immaterial and immortal substance held sway in western thought until late in the 17^th century.⁽³⁾
 

Once John Locke rejected the substance account of personal identity and replaced it with a relational account based on consciousness, the basis for responsibility and especially self-concern once again became a significant philosophical problem for western thinkers.⁽⁴⁾ Locke, of course, was a devout Christian. It is noteworthy that his theory of personal identity had practical implications for what he called 'forensic' concerns-those associated with moral responsibility, both in this life and the next. He held that it would be unsuitable for God, or even civil law, to punish individuals if they could not remember the action for which they were being punished.
 

Locke's memory-based account of the relationship between personal identity and self-concern raised problems that he could not resolve. Nevertheless, it set the agenda for subsequent progressive thinkers. Their task was to explain how temporally distinct selves are sufficiently connected to each other that they feel responsibility and egoistic concern. Throughout the 18^th century, debate over this issue raged between those who continued to insist that an immaterial and immortal soul provided the only solution to the problems of personal identity, responsibility, and self-concern, and those, following Locke, who tried on the basis of a relational account of personal identity to deal satisfactorily with these same issues.
 

Early in the 18^th century, traditionalists had the upper hand. They could point to at least seeming absurdities that followed from a relational approach--absurdities that have reappeared in recent discussions in personal identity theory. In an exchange that in the 18^th century was well known, Samuel Clarke defended the traditional immaterialist view against Anthony Collins' materialist development of Locke's view.⁽⁵⁾ Early in the exchange, Collins had suggested that consciousness was based in brain activity, and had gone so far as to accept a connected consciousness theory of personal identity, whereby each temporally located state of consciousness is accepted as a unique existent that is independent of previous states, but nevertheless connected to them by material transfer of memories from one to the next. Clarke argued that Collins' relational account of personal identity gave the same name to distinct entities, and, as such, could hardly be said to maintain identity in any meaningful sense. Furthermore, he claimed, Collins' account led to a contradiction because of the possibility of multiple fission descendants. He suggested that if God at the resurrection were to create multiple replicas of someone who had lived on earth, on Collins' view all of them would be the same person as the one who had died. But this, Clarke argued, would make the replicas the same person as each other, which he took to be an absurdity. As is well known, comparable arguments developed in the1960's have been used to downplay the importance of personal identity.⁽⁶⁾ But Clarke's intention was not this, but to prove the impossibility of any relational view, thus supporting the traditional substance view of self.
 

Bishop Butler, in his essay on personal identity, started with Clarke's argument against Collins, but went further by highlighting the issue of self-concern.⁽⁷⁾ Butler pointed out that on Collins' view each unique, momentary conscious self, because it is distinctly different from previous and subsequent selves, would have no basis even in this life to have self-concern for its past and future selves. He took this observation to illustrate the absurdity of the relational view.
 

However, in the 18^th century, the relational view was not to be so easily defeated as it had been in ancient Greece and in medieval Christian Europe. Followers of Locke continued to defend his view (e.g., Trotter & Perronet) or to develop a similar relational account based on consciousness (e.g., Collins & Law).⁽⁸⁾ Even those opposed to Locke's view, like Hume and Reid, worked within a relational framework to make sense of how personal identity is to be understood.⁽⁹⁾ Indeed, times had changed so much that by the beginning of the 19^th century, William Hazlitt was able to put forward a theory of personal identity and self-concern that is remarkably similar to Derek Parfit's recent revisionist account. In Hazlitt's first work, An Essay on the Principles of Human Action, he used arguments analogous to those given by Clarke and Butler, not to defend the traditional immaterial substance view, but to go beyond the relational view in a radical new direction.⁽¹⁰⁾
 

Hazlitt relied on his notion that imagination is our only real connection to any future self. But he also argued that there is no absolute identity that necessarily connects us to our future selves. Indeed, it was his reflections on the problem of personal identity in relation to fission examples that first provoked him to develop his theory of action. In these reflections, he considered possible resurrection scenarios, in which multiple copies of himself, or of his continued consciousness, were created by God, and he wondered how he would decide which one among these is really himself and in which he should place egoistic self-concern. The conclusion that he came to from these considerations is that the concept of an 'absolute personal identity' is unfounded and that the belief in personal identity must be an acquired imaginary conception. Ultimately he concluded that since each of us is no more related to his or her future self than to the future self of any other person none of us is 'naturally' self-interested.⁽¹¹⁾
 

How did Hazlitt, at the beginning of the 19^th century, come upon ideas considered radical in the last third of the 20^th century? As we have seen, part of his argument may have derived from discussions throughout the eighteenth century of fission examples involving the resurrection.⁽¹²⁾ However, Clarke and his followers used these sorts of examples not to reject the idea of personal identity but, rather, to support a traditional immaterial substance view of self as soul against Locke's relational view based on consciousness. So, what could have led Hazlitt to turn the argument on its head, and use it to reject the metaphysical notion of personal identity altogether? And what motivated him to connect this issue of personal identity to his novel view of the basis for self-interest?
 

One might suspect that it was Hume's discussion of personal identity in the Treatise, in which he implicitly rejected the notion, that significantly affected later developments, and ultimately set the problem for Hazlitt. But, as we shall see shortly, there was a more likely source of influence on this aspect of Hazlitt's thinking.⁽¹³⁾ Hume did play an important role, however, as the originator of a theory of imagination that connected it to sympathy. This theory, as developed by Adam Smith and then later and especially by Abraham Tucker, culminated in Hazlitt's theory of imagination.⁽¹⁴⁾
 

According to Hazlitt, imagination of the future plays a crucial role in human action, but because he thought that imagination, taken alone, is indifferent to the future of self or other, he viewed it as naturally disinterested. Hence, any special interest we have in our own future over that of another must arise from a different original natural source. One possibility is that humans have an innate belief in their identity over time and innate self-concern for their future selves. But Hazlitt assumed that any special interest in our future selves must be an acquired characteristic in our thinking. He also thought that since it had been learned it could be unlearned.
 

With respect to issues involving personal identity and self-concern, the main influence on Hazlitt seems to have come from an independent discussion that developed toward the end of the century among Unitarian materialists. In this discussion, two original thinkers on the topic, Joseph Priestley and Thomas Cooper, as well as Hazlitt's teacher of divinity at Hackney, Thomas Belsham, were of particular importance. Both Priestley and Cooper were Christian materialists, opposed to the notion of an immaterial, immortal soul as a carrier of personal identity. They thus found it necessary to provide an account of the resurrection that would conform to Christian dogma.
 

Based on the 'new' mechanical philosophy, it was generally thought at the time that the material body alone could not succeed as the basis for maintaining personal identity, even in this life. So, Priestley and Cooper had a special problem to face in dealing with the resurrection. To deal with it, both of them shifted the emphasis away from the issue of personal identity to an alternative conception of what matters in survival. And both suggested that what matters primarily in survival is not maintaining personal identity, but the connectedness of successively related-though not identical-selves. Although these selves are assumed to be connected by a kind of recollective consciousness of previous acts of the living person, neither Priestley nor Cooper supposed that such a connection necessarily implied or constituted identity between the resurrected self and the living person. Rather, all that such a connection provided was a continuity and connectedness of what are now sometimes called continuer selves-that is, selves that maintain psychological continuity and, possibly, causal connectedness, but are not identical with any of their predecessors.
 

When Priestley proposed this sort of shift away from personal identity and toward a conception of what matters in survival as a solution to the problem of accounting for personal identity at the resurrection, he proposed it not as his own theory (following Isaac Watts, Priestley thought that some material core of self is what preserves identity), but to satisfy those who thought that on material grounds identity could not be maintained at the resurrection.⁽¹⁵⁾ Thomas Cooper went further. He held the much more radical view that personal identity is not maintained at all, even in successive stages of earthly life. However, he claimed that it does not matter that personal identity is not maintained because what does matter are the causal consequences of associative connections from each self to temporally successive selves. Until very recently, Cooper has been virtually unknown, even among scholars. Even now, few know about him, and even fewer about his thoughts on identity. So, we will pause to explain who he was and what he said.⁽¹⁶⁾
 

Cooper-lawyer, scientist, philosopher, and political radical-was born in Westminster, England, and educated at Oxford University, where he failed to earn a degree. He was a friend and associate of Priestley, and a member with him of the Manchester Society. Cooper's most important philosophical work, his Tracts, Ethical, Theological and Political (1789), included essays on moral obligation, materialism, and Unitarianism as well as a highly original chapter on personal identity.⁽¹⁷⁾ In 1794, along with Priestley, he emigrated to America in the hope of finding political freedom. And just as Priestley did, Cooper settled in Northumberland, Pennsylvania, where he practiced law and medicine. After an adventurous political career, which included a stay in prison, he eventually accepted a position at South Carolina College (now the University of South Carolina), where initially he taught in the social sciences but subsequently became president of the college.
 

In his chapter on identity, Cooper first surveys the important eighteenth century literature on personal identity, including the views of Locke, Leibniz, Watts, Clarke & Collins, Butler, Priestley, Price, and Bonnet.⁽¹⁸⁾ The heart of his own view is two claims: first, since the notion that an object that changes over time is identical is in a strict philosophical sense incoherent, personal identity over time is impossible; and second, that the non-existence of personal identity is not a problem for us either in this life or the next because--in the language of our own times-personal identity is not what matters primarily in survival.
 

In developing this view, Cooper, in a manner similar to Butler, Hume, and Reid, made a distinction between identity in a 'common and popular' sense, and identity 'strictly and philosophically speaking'.⁽¹⁹⁾ According to the popular sense, all that is required for identity is that the compared objects are sufficiently similar that there is little or no perceptual difference between them. In this popular sense, he claimed, there is no problem with personal identity. People are able to distinguish their own bodies from the bodies of others, and to do this over time, and that is all that it takes to establish identity. He went on to provide a developmental account of how the similarity between our successive perceptions of our own bodies over time and their dissimilarity to our perceptions of the bodies of others, provides the basis for first discovering self and other as separate entities and then subsequently continuing to distinguish between them.⁽²⁰⁾
 

For identity in a philosophical sense, Cooper said, more is required. He defined identity in this stricter sense as 'the continued existence of any being unaltered in substance or in properties'.⁽²¹⁾ He went on to argue that such a definition is never satisfied, either for minerals, vegetables, animals, or persons. His review of developments in chemistry suggested to him that the properties or substance of all material objects is constantly in a process of change. And, he claimed, the situation for living beings is no better, even if life or organization is used as the primary criterion. He considered and rejected the possibility that although the matter out of which a person is composed is constantly replaced, there is a form or structure that remains constant. For, life as an attribute is renewed constantly out of new material, and the organization of all living beings is constantly in flux, along with the organs and matter out of which that organization is constituted. On his view, as the organs change, so must the structure of the whole. So there is no property or form that remains the same.⁽²²⁾
 

The case for personal identity, Cooper claimed, is no better. He said that there is no evidence that people have immaterial souls.⁽²³⁾ He also held that mental activity is a function of brain and body, and that there is ample evidence that all of the matter out of which brains and bodies are composed is constantly being replaced, with nothing remaining constant, and, as previously noted, with the structure and functioning of the whole living system changing in consequence of these changes in its parts. He rejected as too hypothetical the individually unique and unchangeable material 'stamina' theory to which Watts and Priestley subscribed. Of Charles Bonnet's related 'pre-existent germ' theory of personal identity, transmitted in insemination of the embryo, Cooper argued that there is no evidence that these germs exist and that even were they to exist they would induce new properties as well as growth and change in the properties of the embryo, which argue against any notion of maintaining identity. Furthermore these germs, 'while cooped up within the mutable mansion' of the body must undergo 'perpetual alteration' along with the rest of the body, once again transforming possible identity into mere similarity.⁽²⁴⁾
 

With regard to consciousness and the Lockean notion of personal identity, Cooper accepted Collins' idea of sequentially connected but distinct consciousnesses, constituted out of brain activity, as the empirical basis for recollective consciousness; but he rejected the idea that such connections could possibly constitute personal identity. Indeed, his response to the fission examples generated by Clarke was to indicate that fission examples have no force against Collins' view, which already rejects identity in a strict sense of 'same' consciousness. Instead, Cooper wondered why Collins did not just give up the notion of personal identity altogether and go with a theory of the connectedness, or continuation, of similar but non-identical selves, which is all that follows from his view. Indeed, it is this very view of serially connected but distinct selves or persons that Cooper adopted for himself.⁽²⁵⁾
 

In sum, in Cooper's view, no one lasts very long-not even from moment to moment, let alone year to year. Rather, there is a succession of similar persons, each of whom is causally dependent for his or her existence on predecessors in the series. Because persons are similar in this way to previous persons, and causally connected to them, people have supposed mistakenly that personal identity has been preserved over past changes and will be preserved over future ones. Cooper concluded that personal identity is an illusion, or at best merely a pragmatically useful notion with no adequate support in the nature of things.
 

Cooper then replied to the objection that 'the man at the resurrection will, upon this system, be not the same with, but merely similar to the former', by pointing out that even though, 'similarity is all that a reasonable being, deciding according to evidence, can actually predicate in any case of the existence of a living human creature, at any two moments of that existence, . . . we never conceive this to be of the least consequence during our life, nor do we ever suppose that where the particles of which our bodies are to be composed are similar, that identity is of the least consequence'. He concluded, 'Identity then, is not necessary to the phenomena of future existence', even at the resurrection.⁽²⁶⁾
 

In reply to Butler's point, that on the supposition that personal identity 'consists in successive acts of consciousness, the man of to-day need take no care about the man of to-morrow', Cooper argued, first, that there 'is sufficient reason to care' about 'the man of to-morrow, though not in all points the same with' the man of to-day, because he 'depends for his existence on the man of to-day' and, second, that 'the man of to-morrow possessing a reminiscence of the actions of the man of to-day, and knowing that those actions will be referred to him, both by himself and others, they cannot be indifferent to the man of to-day who looks forward to the properties of the man of to-morrow'.⁽²⁷⁾ In these remarks, Cooper showed that he is sensitive to the issue of what practices people will endorse in connection with personal persistence, as well as people's anticipations of 'their own' futures. In the third part of his reply to Butler, Cooper added that 'the approximation to identity, i.e., the high degree of similarity between the two men, is sufficient to make the one care about the other: and in fact they do so'.⁽²⁸⁾
 

So far as considerations of morality are concerned, Cooper replied 'that a good man knowing that a future being, whose existence depends upon his, will therefore be punished or rewarded as the actions of the present man (whose habits and associations will be propagated) deserve, will have a sufficient motive to do right and abstain from wrong'. And, so far as both morality and prudence are concerned:
 

That the man of a twelvemonth hence, or some more indefinitely long period, depending for his existence or properties on the man of to-day, is nearer to the latter considerably, with respect to the interests the latter has in him, than the children of this man of to-day, and yet the children of a person, though at the utmost only half his, furnish very strong motives to care and anxiety concerning them, and a guard upon a man's present conduct, in consideration of the effect it will have upon their future happiness.⁽²⁹⁾

So, in Cooper's view, self-concern is secure.
 

These views of Cooper were subsequently discussed by Thomas Belsham, who would teach theology to William Hazlitt at Hackney College. Ironically, Belsham, who was a friend and colleague of Priestley, also left for America in 1794. In Belsham's Elements of the Philosophy of Mind, which was based on his lectures at Hackney, he first correctly summarized Cooper's view, and then wrote that if 'Cooper's hypothesis were generally admitted and acted upon, it would be very injurious to the cause of virtue: for few would be encouraged to virtue, or deterred from vice, if they had no interest in the reward or punishment consequent upon their moral conduct'. This response suggests that Belsham was not convinced by Cooper's replies to Butler's objection about self-concern. Nevertheless, Belsham reassured his readers, that 'men are so much the creatures of habit, that the most extravagant opinions seldom produce any considerable change in their conduct. And in the present case, the conviction of permanent identity, however acquired, is so firmly fixed in the mind that it is impossible to root it out'.⁽³⁰⁾
 

No doubt Belsham was right that few humans will root out from their minds commitment to the notion of personal identity. But Hazlitt-who, as Belsham's student, was surely aware of Belsham's discussion of Cooper-decided to give it a go anyway. Before considering Hazlitt, however, we want to point out another aspect of Cooper's thought that seems to be in tension with the replies he made to Butler's objection involving self-concern. We believe that Hazlitt may have recognized this tension, and that it may have provided another basis for the novelties in his own view.
 

In Cooper's Essay on Moral Obligation, he distinguished between the causes of moral behaviour and the justification of moral obligation. He explicitly recognized this as a distinction between 'is' and 'ought'.⁽³¹⁾ He also suggested that moral behaviour can be explained simply enough on association principles, whereas moral obligation cannot. He then went on to argue that the only possible justification for moral activity must be our own long-term personal self-interest, since to avoid personal pain and to obtain personal happiness is always the ultimate justification for any action. Further, he claimed that belief in an afterlife is also necessary to virtue, since prudence must always dominate over public virtue, if there is no anticipation of an ultimate reward for virtuous activity. Thus, for instance, he claimed that atheists are not in a position to justify virtuous activity because they do not believe in an afterlife, where virtue will reap ultimate rewards.
 

In this essay of Cooper's, which was written before his essay on identity, it is clear that he does not take into account the theory of identity that he would later develop. For, according to that theory, there can be no prudent activity involving one's own personal future because there is no personal future, only selves in the future who are similar to the present self or depend on it for their existence. Thus, on Cooper's account of moral obligation only a person's present pain or immediate happiness can ever really justify an action.
 

It may be that Belsham recognized this inconsistency in Cooper, since he discussed both Cooper's theory of identity and his theory of moral obligation. Belsham, like Cooper, thought that the Christian belief in an afterlife is the only ultimate justification for virtue. But, unlike Cooper, Belsham optimistically suggests that there may be some material 'germ' that is the basis of identity not only during this life but also at the resurrection, thus providing an ultimate warrant for virtue in this life.
 

Hazlitt may well have puzzled over these issues while studying under Belsham and Priestley at Hackney. Hazlitt claimed that he made his metaphysical discovery during the time that he was at Hackney. And since Cooper is mentioned briefly in the Essay on Action there is independent evidence of his acquaintance with Cooper's views.
 

Although we cannot be certain how Hazlitt came to his own theory, we conclude with what we believe is a reasonable reconstruction of Hazlitt's thoughts, based on evidence presented in this article and elsewhere.⁽³²⁾ As Hazlitt stated in the Essay, he reached his great metaphysical insight shortly after reading in d'Holbach's System of Nature, the speech that the atheist makes to God at the resurrection. In that speech the atheist claims not to have believed in God or in the resurrection, but also claims to have led a virtuous life.⁽³³⁾ This discussion may have led Hazlitt to reflect once again on the puzzle of how one could justify being virtuous, or altruistic, when it might be opposed to long-term self-interest involving an afterlife. In his thoughts, he may have been reminded of Cooper's views both on identity and on moral obligation.
 

When Hazlitt started to imagine the resurrection, presumably he thought personal identity would somehow be maintained. If identity were maintained, then, clearly one ought to act in a manner that would result in one's own ultimate reward. According to association theorists like Hartley and his followers, including Priestley, Cooper and Belsham, this pursuit of final reward is still congruent with general benevolence or altruism. But, Hazlitt probably asked himself, what if I were forced to choose between altruism and my own ultimate reward, why must I choose prudence over altruism? It is at this point that reflections on personal identity may have arisen. Cooper had already argued that personal identity was not maintained either in this life or into the next, but, nevertheless, claimed that we had a special attachment to, and justification to pursue, the interests of our successive selves over any other persons, including our own children. But Hazlitt probably wondered: Is this right? If these future selves are not, strictly speaking, my own self, why should I not prefer to help others over these future similar selves?
 

At this point two ideas may have arisen in Hazlitt's mind: First, that there could be fission descendants at the resurrection; second, that we are connected to our future selves only by use of our imagination. From the idea of fission, it may have become obvious to Hazlitt that God was free to create multiple continuers of any self, and to reward or punish them in different ways. But, if this could occur, on what basis could one decide which of these descendants maintained the identity of the original self? Or, even if these continuers were not, strictly speaking, identical with one's self, which of them should one care about? Wherein was self-concern to be satisfied? At this point, the second idea may have come to Hazlitt's mind, whereby he recognized that one's only real present connection to the future, whether to future continuer selves or to others, was through imagination. Moreover, he recognized that imagination, as the main faculty in the service of action, was indifferent to whether the future was of self or other, hence that choice was not automatically in the service of self-interest.
 

From these two reflections, Hazlitt probably inferred that there could, in fact, be no absolute self with any right to demand the moral obligation of self-interest. Thus, any self-interested motivation had to be acquired rather than be instilled in our nature. And, since there was no actual absolute personal identity there could be no absolute metaphysical justification to pursue our own ultimate interest over the interests of others. Hence, Hazlitt may have concluded that we are morally and prudentially free to act in favor of the future interests of other persons, over the future interests of those successive selves that form the chain of selves typically associated with our own changing human body. In this manner, he may have reached a conclusion, not dissimilar from some recent personal identity theorists, that self-interest cannot be the ultimate justification for moral obligation because there is no relation of absolute identity between successive selves to guarantee that these selves are the same self. Hence, there can be no metaphysical basis for claiming that we have a special obligation to pursue our own future interests over the future interests of others.⁽³⁴⁾

1. Galen Strawson, 'The self', Journal of Consciousness Studies, 41 (1997) No. 5/6: 405-28.

2. David Sedley, 'The stoic criterion of identity', Phronesis, 27 (1982): 255-75, p. 255.

3. The full story is more complicated, particularly as it involved issues associated with the resurrection (see e.g., Catharine W. Bynum The Resurrection of the Body in Western Christianity, 200-1336 (New York 1995); Raymond Martin & John Barresi, 'Personal Identity & What Matters in Survival: A Historical Overview', to be published in Martin & Barresi, Personal Identity and What Matters in Survival (Oxford, in press). Nevertheless, at least until late in the 17th century, the primary concern of most philosophers and theologians was the problem of individuation--that of discriminating one entity or substance from another--rather than the problem of identity--that of identifying the same entity or substance across time. See, e.g., Udo Thiel, 'Individuation' and 'Personal Identity' in Daniel Garber and Michael R. Ayers (eds.), The Cambridge History of Seventeenth-Century Philosophy (Cambridge, 1998), 212-262, and 868-912.

4. John Locke, An Essay Concerning Human Understanding, 2nd ed. 1694, republished by P.H. Nidditch (ed.) (Oxford, 1975)

5. Samuel Clarke, The Works of Samuel Clarke, Vols. I-IV (1738), (reprinted, New York, 1968), vol. 4, 720-913.

6. See, e.g., Derek Parfit, Reasons and Persons, (Oxford, 1984); Raymond Martin, 'Identity's Crisis', Philosophical Studies, 53 (1988) 295-307.

7. Joseph Butler,'Of Personal Identity' (1736), reprinted in John Perry (ed.), Personal Identity, (Berkeley, Los Angeles, & London, 1975), 99-105.

8. Catharine Trotter (Cockburn), A Defense of the Essay of Human Understanding, Written by Mr. Lock, (1702) in T. Birch (ed.), The Works of Mrs. Catharine Cockburn, London, 1751); Vincent Perronet, A Second Vindication of Mr. Locke, (1738), (Reprinted, Bristol 1991); Collins (1707-8), in Clarke, op. cit., V 4; Edmund Law, 'A Defense of Mr. Locke's Opinion concerning Personal Identity', (1769) in The Works of John Locke, 10 VoIs., (Darmstadt, 1963), Vol. 3, 179-201.

9. David Hume, Treatise of Human Nature, (1739), reprinted by L.A. Selby-Bigge (ed.), (Oxford,1888); Thomas Reid, Essay on the Intellectual Powers of Man, (1785), in William Hamilton, (ed.) Philosophical Works of Thomas Reid, Vol 1, (reprinted, Hildesheim, undated), 213-508.

10. William Hazlitt, An Essay on the Principles of Human Action and some Remarks on the Systems of Hartley and Helvetius, (1805; reprinted, Gainesville Fl, 1969). See, Raymond Martin and John Barresi 'Hazlitt on the future of the self', Journal of the History of Ideas, 56 (1995) No. 3, 463-81, for a discussion of Hazlitt's views in this work and their relevance for recent discussions in the philosophy and psychology of personal identity and self. See also, Raymond Martin, Self-Concern: An Experiential Approach to What Matters in Survival, (Cambridge, 1998); John Barresi, 'Extending self-consciousness into the future', in Chris Moore & Karen Lemmon (Eds.), The Self in Time: Developmental Issues, (Hillsdale, NJ, 2001).

11. Hazlitt, ibid, 132-141.

12. Hazlitt must have been aware of fission examples through the debate between Priestley and Price, where on several occasions Price uses Clarke's fission argument involving the resurrection as one of his own main objections to Priestley's account of the resurrection (Joseph Priestley and Richard Price, A Free Discussion of the Doctrines of Materialism, and Philosophical Necessity, In a Correspondence Between Dr. Price and Dr. Priestley, 1778; reprinted, Millwood, NY, 1977, 56-9; 77-8). Priestley was one of Hazlitt's teachers at Hackney College, and Price, who was no longer alive when Hazlitt went there, had created the college. In a brief biographical article on Priestley, Hazlitt mentions this debate and Priestley's 'artful evasion of difficulties' raised by Price (See P.P. Howe (ed.), The Complete Works of William Hazlitt, 22 v's., (New York, 1967), vol. 20, p. 237).

13. In addition to the greater likelihood that Hazlitt was influenced on this issue from sources described below, it is possible that he was not even acquainted with Hume's Treatise at the time that he first formulated his theory, in 1793/4. While describing his first meeting of Coleridge in 1798, Hazlitt mentions that he had just been reading the Treatise, and we suspect that this may have been for the first time (William Hazlitt, 'My first acquaintance with poets', in Winterslow: Essays and Characters Written There, London, 1902, 1-23, p. 10).

14. See David Bromwich, Hazlitt: The Mind of a Critic, (Oxford, 1983), 24-57, for influences on Hazlitt's theory of imagination.

15. Joseph Priestley, Disquisitions Relating to Matter and Spirit and the Doctrine of Philosophical Necessity Illustrated, (1777; reprinted, New York, 1976, 166-7); Joseph Priestley and Richard Price, op. cit. We have provided a fuller account of Priestley's views on identity and what matters in survival in Raymond Martin & John Barresi, Naturalization of the Soul: Self and Personal Identity in the Eighteenth Century (London, 2000). See also, Raymond Martin, John Barresi, & Alessandro Giovannelli, 'Fission examples in the eighteenth and early nineteenth century personal identity debate', History of Philosophy Quarterly, 7 (1998) No. 3: 323-48.

16. But see Udo Thiel, Lockes Theorie der personalen Identität, (Bonn,1983),196-97, 'Locke and Eighteenth-Century Materialist Conceptions of Personal Identity', in The Locke Newsletter, vol. 29, (1998): 59-83, and Thiel (ed.), Philosophical Writings of Thomas Cooper, vol. 1., (Bristol, 2000).

17. Thomas Cooper, Tracts, Ethical, Theological and Political, v. I, (London, 1789). See also, Thiel ed. ibid, vol. 1.

18. Cooper does not mention Hume in this context, and explicitly wrote that only one other account of personal identity--a letter on the topic--has come to his attention. In his other works, he did mention Hume's Essays, but not the Treatise. Because of the similarity between Cooper's views and Hume's, Thiel (op. cit., 1998; p. 72) suggests that Cooper may have left Hume's account out on purpose in order to avoid 'theological controversy'. This seems to us to be out of character for Cooper. We believe it more likely that Cooper had not come across a copy of the very rare first edition Treatise, so he had not read the account of personal identity found only in that work.

19. Cooper, Tracts, v. I, 355-6.

20. Cooper, ibid, 365-8.

21. Cooper, ibid, 308.

22. Cooper, ibid., On minerals, 388ff.; on living organization, 382-8.

23. Cooper, ibid., devotes most of his Essay on Materialism to argue against the notion of an immaterial soul, mainly on empirical grounds.

24. Cooper, ibid., 431-3

25. Cooper, ibid., 338-40; 434;

26. Cooper, ibid, 456-58.

27. Cooper, ibid, 462-63.

28. Cooper, ibid, 463.

29. Cooper, ibid, 463-64.

30. Thomas Belsham, Elements of the Philosophy of Mind, and of Moral Philosophy. (London,1801), 162-63.

31. Cooper, Tracts, vol. 1, 1-122.

32. Lawrence M. Trawick III, 'Sources of Hazlitt's "metaphysical discovery"', Philological Quarterly, 42 (1963) II: 277-282, was the first to try to reconstruct the sources of Hazlitt's insight. His view that Joseph Butler's 'Dissertation on personal identity' was involved was surely right but Trawick didn't go beyond that source on the issue of identity, and he thought that Butler's Sermons on Human Nature also played a role in the discovery. Yet, Hazlitt, according to his own account, was not acquainted with the Sermons at the time of the discovery (William Hazlitt, 'My first acquaintance with poets', op. cit., p. 11). More recently, David Bromwich, op. cit., discusses Tucker's influence on Hazlitt's theory, but the focus here is on Hazlitt's theory of the action and not his theory of identity and what matters in survival. Finally, we have discussed Hazlitt's connection to the fission literature, and to Priestley's discussion of what matters in survival in previous publications (Martin, Barresi, & Giovannelli, op. cit.; Martin & Barresi, op. cit., 2000), but, prior to the present article, no one has mentioned the crucial role that Cooper or Thomas Belsham may have played in the discovery.

33. Hazlitt, ibid, 133-4.

34. Barresi read an earlier version of this paper at the Atlantic Region Philosophers Association Annual Conference, October 14, 2000 at University of Kings College, Halifax, NS. He wishes to thank the Social Science and Humanity Research Council of Canada for providing support for the research upon which the paper is based.