Thanks to Richard Davies and an anonymous referee for useful comments on earlier versions of this paper.
1Very often, philosophers ask questions of the form ‘What is X?’ – for example, ‘what is the good?’, ‘What is justice?’, ‘What is truth?’. There is nothing surprising about that, for philosophy is, in a sense, just the rational effort to clarify the deep nature of things. But questions of the form ‘What does “X” mean?’ – for example, ‘what does “good” mean?’ ‘what does “justice” mean?’, ‘what does “truth” mean?’ – are also common in philosophy. And one might wonder whether, in what cases and to what extent questions of the former kind can be reduced to questions of the latter. This is an important meta-philosophical issue, about which I shall say little if anything here. Let me just observe that the distinction between questions about nature and questions about meaning appears clear-cut and well founded in a large number of contexts. It would be crazy, for example, to try to discover the nature of pain by reading in a dictionary what ‘pain’ means. In other cases, however, the distinction appears less clear. A famous case concerns ships. Plutarch writes:
‘The ship on which Theseus sailed with the youths and returned in safety, the thirty-oared galley, was preserved by the Athenians down to the time of Demetrius Phalereus. They took away the old timbers from time to time, and put new and sound ones in their places, so that the vessel became a standing illustration for the philosophers in the mooted question of growth, some declaring that it remained the same, others that it was not the same vessel’ (Plutarch, 1914, 23, 1).
- 2 See also Hobbes, 1976: ‘If some part of the first material has been removed or another part has bee (...)
2In more recent centuries, following an ancient suggestion, Thomas Hobbes imagined in his De Corpore that the ship’s old planks had been collected on land, and reassembled just as they were arranged in the original ship (Hobbes, 1999, 2, 11, 7). If this were to happen, he asked, which ship would be the original one? The one preserved by the Athenians or the reassembled one? Although Hobbes argued that the reassembled ship is the true original,2 this may depend more on language than on reality. Consider that you can know everything there is to know about each particular wooden plank of the ships in question and still have no idea of which ship is the original one. So one might say: look, it just depends on what you mean by “ship”! If you conceive of a ship as something invariable in its material composition, then the reassembled ship is the true original. If you conceive of a ship as something spatio-temporally continuous, then the original ship is the preserved one.
- 3 It is natural to distinguish between two questions. 1) What necessary and sufficient conditions mus (...)
3It has been suggested that the reason why questions about nature and questions about meaning tend to merge in the case of ships is that ships are reducible to other things. They ultimately consist in things that are not ships (wooden planks and the like). Some philosophers are as reductionist about persons as they are about ships. They believe that, ultimately, persons are nothing over and above things that are not persons.3 In their view, a complete catalogue of what there is in heaven and earth can fail to mention persons, and a complete description of reality can be formulated in impersonal terms. And they think that questions about the nature of reducible entities are more about meaning than about reality.
- 4 On the distinction between descriptive and revisionary metaphysics, see Strawson, 1959.
4The very idea of a reduction is widely discussed and highly controversial, and the relations among reduction, emergence, supervenience, ontological dependence and so on are far from clear. This is an intricate and slippery matter, and it is not among my aims to take discussion of such kind of issues into consideration. Let me just assume that reducing things of kind F to things of kind G is not claiming that the Fs do not exist, so that reductionism about the Fs is not eliminativism about them. Thus, a one-category ontology, which reduces everything that exists to things of just one category (say, particulars, universals, or tropes) need to be revisionary4 – i.e., to deny the existence of a wide number of commonsensical categories of objects – but aims only to explain their nature in purportedly more basic and fundamental terms. On the one hand, reduction can work as a tool for explanation, taxonomical simplification and ontological deflation, which explains why reductionism is a very natural ontological attitude so long as ontological economy is seen as a theoretical desideratum. To reduce things of one category to things of another helps to decrease the global number of categories, enabling us to improve what David Lewis calls “qualitative parsimony”. On the other hand, reductionism mitigates the impact of metaphysical analysis on our ordinary ontological commitments, helping to reconcile the metaphysical image of the world with what Sellars called “the manifest image” of it.
- 5 The alternative is often seen as an indispensable key for grasping the ultimate meaning of the prob (...)
5The contrast between reductionist and anti-reductionist views of persons has shaped the discussion about the nature of persons over the last decades. Starting from Parfit (1984), the contrast between so-called “complex” (or reductionist) and so-called “simple” (or anti-reductionist) views of persons has been taken as a fundamental divide between two kinds of deeply different conceptions of persons: on the one hand, the reductionist (or complex) views, according to which persons are nothing over and above things that are not persons and facts about personal identity across time are nothing over and above impersonal facts; on the other hand, the anti-reductionist (or simple) view, according to which persons consist in nothing but themselves and there is nothing impersonal in which facts about personal identity across time can consist. In this view, the alternative between simple and complex theories of persons is an extremely important one on which much depends, and any other difference between theories of persons, however important, is less basic and fundamental.5
6I want to argue that the substantive importance of this contrast has been overestimated. Philosophical conceptions of persons can differ in a number of respects that are more important and fundamental than their attitude to reducing (or resistance to reducing) persons to purportedly more basic entities. Philosophers give different answers to substantive questions such as: are persons material or not? Do they belong to the natural domain or not? Are they indeterminate or not? Do they endure, perdure or exdure? But, as I shall argue, neither reductionism nor anti-reductionism about persons can dictate on their own any precise answers to these questions. However important the contrast between simple and complex views of persons may be on a meta-metaphysical level, it is on its own of surprisingly little substantive import. In particular, there is no logical or semantical tie between reductionism about persons as such and the idea that personal identity across time is indeterminate (or even that it is indeterminate whether some entity is a person or not). Likewise, reductionism about persons entails neither materialism nor naturalism about ourselves (both in a weak and in a strong sense of ‘materialism’ and ‘naturalism’). And, conversely, neither materialism nor naturalism about ourselves entails a reductionist view of our nature. In addition, in no way does reductionism about persons entail that we are dispensable and unimportant, or that, so to say, we should not be taken seriously, ontologically speaking. Finally, neither reductionism about persons nor its denial can dictate how persons persist across time (i.e., whether they perdure, endure or exdure).
- 6 Some of the main ideas of the book were partially developed by Parfit in the seventies in some pape (...)
7A paradigmatic reductionist approach to personal identity was offered in the 1980s by Derek Parfit, in his famous book Reasons and Persons.6 In this book, Parfit defends reductionism about persons and their identity, which he calls “the Complex View”, against anti-reductionism about persons and their identity, which he calls “the simple view”. He formulates the complex view as the thesis that facts about personal identity consist in other impersonal facts. And he argues, more specifically, that facts about personal identity are nothing over and above facts about psychological connectedness and continuity. As is well known, the core of his psychological reductionism is a certain inversion of the dependence relation between a subject and her mental states. The existence of a mental state is no longer said to require the existence of a subject, it is rather the existence of a subject that is said to require the existence of a number of adequately related mental states. Roughly, the idea goes as follows. First, mental states existing at the same time are unified as a maximal group of simultaneous mental states by their co-consciousness (where the notion of co-consciousness of two or more mental states must be assumed as primitive and intuitively clear, and not defined as consciousness of the same subject, to avoid circularity). Second, two maximal groups of simultaneous co-conscious mental states existing at different times are of the same subject just in case a sufficient number of mental states of the former group are causes or concauses of a sufficient number of mental states of the latter group - and, in addition, there is enough qualitative similarity between mental states of the former group and mental states of the latter. Since no precise threshold is given beyond which qualitative similarity-cum-causal connectedness of two groups of mental states is “sufficient” in order for them to be of the same person, personal identity across time can only be indeterminate. But this indeterminacy has nothing ontic about it, it is just the indeterminacy of a certain way of describing some phenomena – in personal rather than impersonal terms. Ultimately, an ‘empty question’ is just one whose answer depends on semantics rather than on reality.
8No doubt, Parfit’s reductionism treat persons as no less indeterminate, strictly material and ontologically dispensable than ships, but it is not qua reductionist that it does so. Let me begin by showing why reductionism about persons does not on its own imply that persons are indeterminate and, conversely, antireductionism about persons does not on its own imply that persons are determinate.
9Suppose one supplements the meaning of “psychologically connected” with a precise criterion for establishing to what degree two persons are psychologically connected, and Parfit’s criterion of personal identity with a precise specification of the minimum degree to which two persons must be psychologically connected in order to be the same person. This supplemented quasi-Parfitian view would still be complex, but would no longer entail that persons are indeterminate. Conversely, imagine one believes at the same time that there is something as ontic vagueness, that persons are simple immaterial beings and that simple immaterial beings are vague entities. Or else, suppose one believes that persons are some sorts of fundamental, ontologically indeterminate quantum objects. In both cases one would believe that persons are simple and indeterminate.
10Likewise, reductionism about persons as such does not imply materialism about them and, conversely, anti-reductionism about persons as such does not imply anti-materialism about them. To begin with, consider a Berkeleyan version of the psychological criterion of personal identity conceiving of mental states and events as purely immaterial ideas and thoughts rather than neurophysiological states and events. This unorthodox version of the psychological criterion would plainly be no less reductionist than the standard familiar version, but it would be radically anti-materialist and anti-naturalist as much as the standard version is radically materialist and naturalist. It is far from clear that this imaginary anti-materialist version of the psychological criterion of personal identity would be paradoxical or incoherent, even though its coherence with contemporary sciences may of course be put in doubt.
11Again, imagine a theory according to which (i) persons are complexes of (say) four very thin immaterial atoms, (ii) they have the same components at any moment of their life, and (iii) the identity over time of their atomic components is brute and non-analysable (suppose the four atomic parts are conceived of as something like self-subsistent pure forms, souls or the like). Although, as far as I know, nobody has ever embraced (i)-(iii), a theory incorporating these theses would entail that persons are nothing over and above things that are not persons, and their identity across time consists in the diachronic identity of things that are not persons. But still, that reductionist theory would treat persons as purely immaterial entities.
12Examples of this sort clarify a point that should be evident a priori. The idea that persons are nothing over and above things that are not persons need not be either materialist or anti-materialist per se, nor either naturalist or anti-naturalist per se. This just depends on the nature of the non-personal entities to which persons are purportedly reduced. Since, in principle, they can be either material or immaterial, they can be members of the orthodox ontology of natural sciences or not, and they can belong, in general, to a wide number of heterogeneous domains, it is completely unmotivated to think that reductionism can be provided on its own of any substantive value.
13The case of a materialist anti-reductionist theory of persons is less exotic. To make just a couple of examples, both Lynne Rudder Baker and Jonathan Lowe, have proposed in recent years approaches to the nature of persons that are anti-reductionist but mildly materialist. The same may also be true, at least in a possible version, of some narrativist approaches to diachronic personal identity.
- 7 Baker, 1997, 2000.
- 8 See especially Baker, 2013.
14According to Baker, I bear to my living body the same relation that a bronze statue bear to the matter of which it is made.7 A human animal constitutes me, but I am not identical with it, just as the statue is not identical with the bronze, even if there is a sense of ‘is’ (the so-called ‘is’ of composition) in which the statue can correctly be said to be identical with the bronze. The statue is not numerically identical with the bronze because it has essential properties that the bronze lacks – in particular, the shape that the sculptor has given to the bronze, making the statue just the statue it is. Likewise, I have essential properties that my body lacks – according to Baker, in particular, my ability to refer to myself as “I”, to perceive the experiences and perceptions that I happen to have as mine, to refer to any other people as “you”, and so on. In short, what Baker calls “first-person perspective”. In Baker’s view, a person existing at t1 is identical with a person existing at t2 just in case they have the same first-person perspective, but there is nothing more basic in which having the same first-person perspective consists. This makes her approach at least similar to what Parfit calls “the simple view”, for in Baker’s view there is ultimately nothing in which facts about personal continuity across time can consist. So, leaving aside its possible difficulties, Baker’s approach is anti-reductionist, but it is perfectly compatible with materialism, for it has no need to appeal to immaterial souls and assumes that we are essentially embodied, i.e. necessarily constituted by something material. However, it is anti-naturalistic in orientation.8 In fact, as Baker herself remarks the very idea of a first-person perspective is an impossible challenge to naturalism, for naturalistic theories are rentlessly third-personal in character.
- 9 See in particular Lowe, 2001.
15A different version of a materialist but non-reductionist approach to the nature of persons, sometimes called ‘Brobdingnagian atomism’, was also defended by Jonathan Lowe since the nineties in a number of works.9 According to Lowe, I am a material entity located in the extended region in which my body is also located, but I am not identical with my body, for my body has parts while I do not. The idea of an extended atom, indivisible even though located in a divisible spatial region and locally differentiated in the various parts of the occupied region is extremely difficult to be grasped. How can I have a colour where there are my eyes and another colour where there is my neck if not by having different parts of different colours? And how can I weigh 86 kgms if not by having exactly 86 proper disjoint parts weighing 1 kg each?
16One tentative way to clarify the point is by means of a temporal analogy. Endurantists treat ordinary objects as enduring, i.e. temporally atomic even though temporally persistent. In their view, I have no temporal parts, so I am entirely present at each of the different moments at which I exist, even though I have different qualities at those moments. I may be conceived of likewise, as entirely located in every point of an extended region, having different qualities in many of those points. Just as endurantists deny that I divide across time, having different parts at different moments, Lowe denies that I divide in any of the three ordinary spatial dimensions. In short, I am a 0-dimensional brobdingnagian extended atom. However, this way to clarify Lowe’s view might be flawed after all, and a Brobdingnagian atom, contrary to an enduring thing, might instead fail to be multi-located. It might uniquely exist in the entire spatiotemporal region where it extends and in no proper part of it, and inherit its local properties from the parts of the co-located hunk of matter.
17Be that as it may, Lowe offers an argument to the effect that we are Brobdingnagian atoms. The argument appears formally correct, even though its premises are extremely controversial. Let me skip this. What matters in this context is that Lowe’s Brobdingnagian atomism is an anti-reductionist but materialist account of our own nature. In this respect, Lowe’s approach converges with Baker’s, but diverges from it inasmuch as it denies that I have parts. Also, Lowe’s approach is not so clearly anti-naturalistic as Baker’s, for both multi-located entities and extended simples may be not so clearly excluded by our best scientific theories (in particular by quantum-mechanics), however unpalatable they may be at the macro-level.
- 10 See for example Stump, 1995.
18One might ask: is this kind of mildly materialist anti-reductionism about persons also a fundamental aspect of hylomorphic conceptions of our own nature - i.e., the general idea that the relation between a person and her body is a particular case of the matter-form relation?10 One might tentatively answer: only inasmuch as forms are not treated as ontologically independent from matter in the case of persons, i.e., only inasmuch as they are not treated as self-subsistent. If, on the contrary, forms come to be seen as some sort of self-subsistent souls capable of disembodied existence, as in Aquinas’ view, in no way can hylomorphism be a sort of mild materialism. As is well known, however, hylomorphism is as variable as the very notion of form, and can accordingly admit a great number of substantially divergent versions.
19On the other hand, the inclusion of narrative ties in the domain of personal continuity relations across time (see for example Slors, 1998), is prima facie likely to weaken the hard reductionism implicit in psychological criteria. Since narration develops through intentional and semantic rather than merely causal ties, it can in principle be interpreted as narration of a subject, who has therefore to be imagined as already given ab initio. But ties of that sort might also be conceived of as naturalizable in a third-person perspective, as in Dennett’s idea of the self as the centre of narrative identity.
20Even granted all this, one may nonetheless insist that there is at least one substantive consequence about ourselves that reductionism about persons must have as such. If persons are nothing over and above things that are not persons, a complete inventory of what exists in heaven and earth will not include me, and a complete description of reality can fail to mention me. If so, however, I am so irrelevant and unimportant as to be absent from the layer of what really matters in the world. But this may be false, for the catalogue can include me even if persons are nothing over and above things that are not persons.
- 11 Van Inwagen, 1990, 2007.
21Consider in this connection Peter van Inwagen’s idea that necessarily, if x is a person existing at time t1 and y exists at time t2, then x = y just in case x’s biological life at t1 is identical to y’s biological life at t2, where a biological life is a self-sustaining chemical process that always coincides with a living organism.11 If this is true, I am a human animal who happens to be now (but not some decades ago) a person. This account of my nature says nothing fundamental about what it is to be a person. It says that human persons are identical to human animals but does not say that being a human person consists in being a human animal. If being persons requires being conscious, having desires, aims and so on, there are human animals that are not persons (for example, a three month human embryo), as well as human animals that have never been, and will never be, persons (for example, a human embryo miscarried at three months). So, some human animals are not persons. And perhaps, conversely, some persons are not human animals, if there are or could be non-human or non-organic persons – e.g., disembodied minds, angelic and divine persons, artificial minds (if there are any). Therefore, being a human animal is neither necessary nor sufficient for being a person and, at least in principle, the animalist can completely endorse Parfit’s answer to the question “In what does it consist to be a person?”. So, he can be no less reductionist about persons than the friend of the psychological criterion. Since I am a human animal, however, I continue to be included in the catalogue, and a complete description of the universe cannot fail to mention me. This shows once again how a reductionist theory of persons can fail to have any substantive consequences about our own nature.
22It may seem natural to believe that, if persons are fundamental irreducible entities, they can only persist by enduring, i.e., by being entirely present at any time at which they exist. The purported reason is that perduring entities are mereological sums of temporal parts and parts are most naturally seen as more fundamental than the whole they are parts of (after all, the existence of a whole may seem to depend on the existence of each of its parts in a way in which the existence of any of its parts does not seem to depend on the existence of the whole).
23If this is true, the alternative between simple and complex views of persons profoundly affects the substantive issue of how persons persist across time. But there are doubts that this is so. On the one hand, persons might fail to endure by exduring instead of by perduring. An exduring thing is a stage, i.e., an instantaneous object existing on its own at just one time (at any other time it can only exist by proxies – instantaneous representatives existing on their own at just that time).12 Since an object of this sort is a simple, non-reducible entity, simple persons might exdure, and so persist without enduring.
24On the other hand, the idea that the parts are prior to the whole is highly controversial, insofar as friends of priority monism believe exactly the contrary: it is the whole that is prior to its parts, not the reverse (see for example Schaffer, 2010). If priority monists are right, a person might have temporal parts while being a basic, irreducible thing – and so a simple entity at the fundamental level. But even if priority monists were wrong and nothing irriducible could have any parts, simple persons could perhaps perdure in a non-mereological sense, in which something can perdure without having any temporal parts. Of course, if a perduring object is conceived of as a mereological sum of its temporal parts, nothing simple can perdure. But the very idea of a perduring object can be characterized in locational rather than mereological terms (i.e., in terms of the relations that persisting objects bear to the spacetime region(s) where they are located rather than in terms of part-whole relationships).
- 13 See Gilmore, 2006; Parsons, 2008.
25Say that an entity x is exactly located in a (spacetime) region r if and only if x has exactly the same shape and size as r (in other words, just in case r is x’s ‘shadow’ in spacetime). And say that x is totally located in a region r if and only if r is the fusion (i.e., the sum) of all the regions r’ such that x is exactly located in r’. The region r in which x is totally located is often called “x’s path”.13
26Given these definitions, the very concept of a perduring object can be rephrased in locative terms as follows: a persisting object perdures just in case its total location and its exact location are identical (i.e., just in case it is exactly located at its path and nowhere else). As can easily be seen, something can perdure in this sense even if it is an extended simple, i.e. no part of it is located in any proper part of its path. Since what perdures in this way while having no temporal parts is an extended atom, it is simple and does not consist in anything else. Therefore simple persons might perdure in this way.
27I’ve argued that, contrary to the mainstream, the contrast between so-called simple and so-called complex views has little bearing on the substantive problem of our own nature. This is not to say that it has no bearing at all. For example, if one rejects priority monism and conceives of perdurance in mereological terms, the ‘simple’ thesis (unlike the ‘complex’ one) entails that persons do not perdure. If one doesn’t believe in ontic vagueness, the ‘simple’ thesis (unlike the ‘complex’ one) entails that they are determinate. And so on and so forth. Therefore, the complex and the simple theses do not inferentially behave in the same way – i.e., they do not always yield the same substantive conclusions in conjunction with the same premisses. Taken by themselves, however, they say very little about our own nature. However important their contrast may be on a meta-metaphysical level, it is of surprisingly little substantive import.
RSS feed
