04 Jun Which Is The Most Plausible Account of Personal Identity?
Alin Christoph Cucu
 Internationale Akademie für Philosophie im Fürstentum Liechtenstein (IAP)
Personal identity deals with philosophical questions that arise about ourselves by virtue of our being persons. This essay specifically addresses the question what makes a human person the person he or she is. Extant approaches can be classified into “simple” and “complex” theories. Simple theories take substances (and therefore human substances) to have thisness; complex theories don’t. I shall examine both approaches with respect to their plausibility and come to the conclusion that physicalist (animalist) complex theories are highly implausible and psycho-physicalist complex theories on balance less probable than any simple theory. Following Richard Swinburne (Swinburne 2013, ch. 6) I offer instead a simple theory that takes human persons to be pure mental substances and show that it is logically possible, plausible and coherent, thus making it the best and probably true theory of personal identity.
By a person I understand a substance with a capacity (at least after normal growth) for beliefs and actions typical for what we normally call “humans”. A human is in my definition a person who has or has had a kind of body and an ancestry similar to those persons whom we today call “humans”.
Personal identity can be meant in two senses, of which the latter is the more important one: synchronic unity or diachronic unity. By synchronic unity I mean the state of affairs that fully describes what determines the identity of a human person at a certain point of time. By diachronic unity I mean the entirety of states of affairs over a time that fully describes what determines the identity of a human person over that time. Given those definitions, it follows that if a personal identity theory cannot account for the synchronic unity of human beings, it a fortiori cannot account for their diachronic unity.
A substance is, in my usage, a particular concrete object. Notice that by “concrete” I do not necessarily mean “physical”; thus, souls or angels (if they exist) are substances as well as desks or rocks. A substance can have parts, which are themselves substances. A mental substance is a substance for which the possession of mental properties is essential, i.e. if it did not possess any mental property, it would cease to exist (as a mental substance). Mental substances can be pure mental substances, i.e. whose existence does not involve any physical event (I define pure mental events and properties analogously). A physical substance is one for which the possession of mental properties is non-essential.
A substance has thisness iff there could be another substance instead (or in addition) which has all the same monadic properties (i.e. properties describing only the substance itself, not its relations to other substances).
A rigid designator is a word which “in every possible world, designates the same object” (Kripke 1981, 48). An informative rigid designator is a term such that “anyone who knows what the word means (that is, has the linguistic knowledge of how to use it) knows a certain set of conditions necessary and sufficient (in any possible world) for a thing to be that thing (whether or not he can state those conditions in words.)” (Swinburne 2013, 12). Thus, two things are identical iff their (informative or uninformative) rigid designators are logically equivalent.
Classification of personal identity theories
Theories of personal identity can be divided in two main classes: simple and complex theories. Simple theories assume that the human substance has thisness, whose bearer is commonly assumed to be the human soul (see e.g. Swinburne 2013, 170). According to a simple theory, a person remains the person it is because the thisness of the person-substance remains the same across time. Complex theories take human substances to be without thisness, i.e. as just a bundle of properties. They differ in which properties they hold relevant for personal identity. “Psycho-physical” theories (e.g. Parfit 1986, pt. III, Hume 2011, 1.4.6, Locke 2014, 2.27.19) consider both physical and “psychological” (in my terminology pure mental) properties as relevant, whereas “animalist” (physicalist) approaches regard only physical properties as determining personal identity (not least because they assume that there are only physical substances). As both physical and psychological properties of any person usually change over time, any complex theory depends on spelling out just how much and which kind of discontinuity is allowed for a person S1 at t1 to be identical with person S2 at t2.
Animalist theories cannot account for the synchronic unity
Animalist theories claim that there are only physical substances to some of which mental properties belong non-essentially. Hence, human persons are taken to be mere physical substances with non-essential mental properties. Their personal identity then consists with regard to synchronic unity in having a certain set of mental properties at some point of time t; with regard to diachronic unity it consists in having only a limited amount of change in physical properties over a period of time t1-t2. Their non-essential mental properties may be regarded as being caused by their brains.
It might seem feasible that this suffices to capture personal identity. Animalist theories don’t deny the existence of mental events; they just say they’re not essential to human beings (construed as merely physical substances). But if it could be shown that mental properties are essential to human persons, an animalist theory could not even account for the synchronic unity of a human person. And it seems that exactly this is the case. It is an evident datum of human life that we co-experience different mental events, i.e. whose temporal durations overlap. For example, I might be seeing my hands typing on the keyboard (mental event M1) while I hear Bach’s Christmas Oratio (mental event M2). It seems that co-experienced mental events must belong to the same substance. The easiest way to construe this is to say that this substance is an immaterial one with thisness. In this case, the substance S (me) experiences both mental events M1 and M2 at t. An animalist might reply that it suffices to regard M1 and M2 as caused by my brain and that this is enough for a full description of what constitutes that co-experience. But, first, M1 and M2 are most likely caused by different parts of my brain, and, considering that parts of the brain are substances in their own right, the two co-experienced mental events would be falsely attributed to different substances (which are of course part of one composite substance, my body). Right, but what about the possibility of consciousness disunity, the animalist may retort? There seem to be human persons whose consciousness is “split” in some way or another (see, e.g., Bayne 2012). I have no space here to pursue this interesting topic further, but let me quote Richard Swinburne’s summary after a thorough examination of the possibility of disunified consciousness in detail:
[T]he fact that we humans coexperience conscious events entails that we are mental substances; and that the full history of the world must include the history of mental substances—even if such phenomena as split-brains have the consequence that we cannot always determine how many mental substances there are in some situation. (Swinburne 2013, 148)
Second, it is in virtue of co-experienced mental events that we can (in part) determine the physical properties of a human being. For if M1 and M2 are both caused by the brain belonging to S, their co-experience helps attributing the corresponding brain parts to S. Therefore, what constitutes such a substance which has conscious co-experienced mental events (with mental properties) is determined in part by a mental property. But that entails that the substance is a mental substance, which contradicts the fundamental assumption of animalism.
So far I have been examining only how animalism fares in accounting for synchronic unity. It seems it cannot do that properly; therefore I shall not bother to investigate its ability to account for diachronic unity.
Other complex theories are improbable regarding diachronic unity
Perhaps a psycho-physical theory can yield an appropriate account of what it means to be a person at one moment and further – over time. To be sure, there are views which assume that mental substances do not last longer than a “specious present” (e.g. Strawson 2009, pt. I) and that therefore, in the next moment there is another substance/person. I will not go into that here and take it for granted that in asking about personal identity, we seek a diachronically unified account, i.e. an account which assumes that persons can last much longer than just the specious present.
Remember that complex theories take substances to be just bundles of properties. Now, according to such an account, personal identity depends on the extent to which those properties change over time. In other words: How great must the congruence of property-set P1 of person S1 and property-set P2 of person S2 at t2 be in order for S1 to be identical to S2? First of all, it is crucial to settle which properties are picked among the properties of human persons. Psycho-physical theories rely primarily on psychological features, but usually take the underlying physiology into account as well. Thus, two persons are identical iff their psychological features are closely connected (Hume 2011, 1.4.6, Locke 2014). Derek Parfit (1986) spelled this out in more detail. He analyses identity in terms of psychological connectedness and continuity. Two persons are psychologically connected to the extent to which they have similar memories and character, and the continuity between them consists in the “holding of overlapping chains of strong connectedness”, while it is recognized that psychological features depend on the brain as their cause (at least partly).
Now the plausibility of such views depends much on whether its underlying ontology is physicalist or not. If it is, they run into the same problems as the animalist view. But let’s grant arguendo that they allow for the existence of mental substances. It must first be remarked that such complex theories are more complicated than simple ones and therefore prima facie more improbable. Many psychological and physiological criteria would have to be reviewed and a determining threshold to be found which differentiates identity from distinctness. On a simple theory, there is but one criterion: the thisnesses of the persons need to be identical. Richard Swinburne makes an ingenious shift here by considering the conditions under which a simple account might be false rather than look for conditions which might make a complex account true. He treats several thought experiments of patients having half or parts of their brains cut out and exchanged with halves or parts of another patient’s brain (Swinburne 2013, 151-57). His last thought experiment is of crucial relevance. Suppose a person S has to have 1/10 of his brain cut out each year (say, because of a sickness of his brain matter) and replaced with an equivalent brain mass of another human being. After 10 years, his whole brain matter has been exchanged. Is that person still S or has he become another person S*? Swinburne considers the possibility that during all surgeries, S is conscious. That means, he has overlapping co-experience of conscious mental events. I have above argued that overlapping conscious mental events must belong to one and the same person. Hence, it seems at least possible that S remains the same over the entire 10 years, although his brain is in no part the same. But that entails that S’s identity is completely independent of his body (assuming that other body parts play a far lesser role in identity determination). That means, it is at least possible that a person’s identity hinges on a non-physical thisness.
Of course this does not suffice to refute any complex theory once and for all. But all other things equal, we are reasonable in adopting a theory which makes use of fewer kinds and properties of entities to attain the same explanatory power. Complex theories make use of many psychological and perhaps physical properties to explain why S1 at t1 is identical to S2 at t2. Simple theories employ just one property, namely thisness. Therefore, simple theories, given their at least equal explanatory power, are to be preferred to complex theories.
A simple, substance dualist theory is the best account for personal identity
I suggest that we identify a human person with her soul. I take the soul to be an immaterial, pure mental substance which carries the person’s thisness. In a human being (as they are commonly construed), the soul is connected to a human body. This construal has at least three advantages over rival theories:
- It plausibly accounts for synchronic unity.
- It gives a simple, logically possible and plausible account of diachronic unity.
- It allows for pure mental events as well as for mental events which involve physical events.
The first two have been argued for already in above sections. The third point is important on the plausible assumption that humans have pure mental events as defined above. There seem to be plenty of them, e.g.:
- adding up 1+1
- making the inference from “all men are mortal” and “Socrates is a man” to “Socrates is mortal”
- beliefs about objects which do not exist
Now if they exist (as seems to be the case), pure mental events need to be accounted for. It is hard to see how any physicalist theory can do that, given the fact that mental properties are very different from physical properties. Physical properties include mass, charge, velocity, or length, width and so on. Impure mental events (mental events involving physical events) may or may not have properties equivalent to physical ones; at any rate, pure mental events (e.g. an inference or an intention) do not. Of course one might raise the question how a soul can have (pure) mental properties. My short answer is that if we consider this a problem, we might as well ask how physical objects can have physical properties. No matter how fine-grained we split up the physical world, we end up with objects whose properties we either observe or infer to from observation. If there is an immaterial soul, the property-assignment is quite similar: we observe (mainly from introspection) that it can have pure mental events. From this we infer that it has pure mental properties. We cannot observe the soul as such, but nor can we observe sub-atomic particles like the Higgs boson; what we know about them we know from inference and indirect observation.
A related objection takes issue with souls causing physical events (in the brain). Ansgar Beckermann (2004, 20) writes:
Dass [die Auffassung, daß der Mensch einen Leib und eine Seele hat] unter anderem an den Problemen der kausalen Interaktion von Geist und Körper scheitert, ist schon seit Descartes’ Zeiten bekannt. [Es] stellt sich die Frage, wie es der Geist überhaupt zustande bringen kann, in der körperlichen Welt kausal wirksam zu werden? Wie kann etwas Nichtkörperliches physische Wirkungen haben?
Granted, at first glance physical causation by a non-physical soul might seem mysterious. But on second thought, how is physical causation by physical objects less mysterious? It is hardly because of the fact that we can see those objects; we assume causation by many physical objects of which we can at best see their effects – which would mean to collapse cause and effect. But hardly anyone wishes to draw this conclusion. So, if we assume invisible physical objects to cause something physical, why not allow an immaterial soul to do so?
Let me conclude this defense of the “soul-identity” theory with some remarks about what the soul is not:
- It is not the Aristotelian “form” of the body. Aristotle understood by the soul (i) a universal and (ii) a life-conferring principle. It is of course the very nature of a universal to be non-individualistic, i.e. it can be instantiated in different objects. Souls, as I understand them, individuate people from each other by their thisness, i.e. each soul is only instantiated once.
- It is not made of any “soul-stuff”, as St. Bonaventure (Bonaventure, 17.1.2. Responsio) Any “stuff” (i.e. matter, whether physical or non-physical) should be construed as divisible into sub-parts, while the soul is non-divisible; otherwise it could not individuate human persons.
- It is not an “incomplete substance” in the Thomistic sense. Aquinas thought that souls are incomplete as long as they are disconnected from their bodies and that their respective bodies individuate them. But each soul needs an individuating feature already to fit a certain body, which makes the body-individuation superfluous.
I take it to have made it at least plausible that personal identity can be understood best in terms of an immaterial soul carrying thisness, that this soul is a particular, not made of any stuff and complete even without being connected to a body. This opens up the possibility of life after death and many religious considerations, which I must leave for another time.
Aquinas, Thomas. 2006. Summa Theologica. Translated by Fathers of the English Dominican Province. http://www.documentacatholicaomnia.eu/03d/1225-1274,_Thomas_Aquinas,_Summa_Theologiae_%5B1%5D,_EN.pdf.
Aristotle. n.d. Metaphysics.
Bayne, Tim. 2012. The Unity of Consciousness. Oxford: Oxford University Press, Usa.
Beckermann, Ansgar. 2004. In Schließt Biologische Determiniertheit Freiheit Aus?, edited by F. Hermanni and P. Koslowski, 19–32. Paderborn: Fink Verlag.
Hume, David. 2011. David Hume: A Treatise Of Human Nature: Volume 1: Texts. Edited by David Norton. Oxford: Oxford University Press, Usa.
Bonaventure. n.d. II Sentences.
Kripke, Saul. 1981. Naming and Necessity. New. Oxford, UK ; Cambridge, USA: Wiley-Blackwell.
Locke, John. 2014. An Essay Concerning Human Understanding: Second Treatise of Goverment. UK ed. Wordsworth Classics of World Literature.
Parfit, Derek. 1986. Reasons and Persons. Reprint. Oxford: Oxford University Press, USA.
Strawson, Galen. 2009. Selves. Oxford University Press.
Swinburne, Richard. 1993. “God and Time.” In Reasoned Faith, 204–22.
———. 2004. The Existence of God. 2 Revised edition. Oxford: New York: Clarendon Press.
———. 2013. Mind, Brain, and Free Will. Oxford: Oxford University Press.
 I use the term “point of time” instead of “instant”. An instant is usually defined as the boundary between two periods of time (Swinburne 1993) and hence has an infinitely small temporal extension, which makes this term unfit to apply to anything with temporal duration.
 For a thorough defence of the criterion of simplicity for scientific and philosophical theories, see e.g. (Swinburne 2004, ch. 3)
 My English translation of this quote: „It is known since the time of Descartes that [the notion of human beings having a body and a soul] founders on problems of the causal interaction of mind and body. The question arises how the mind can bring it about to become causally efficacious in the physical world? How can something non-physical have physical effects?”
 “When we have the whole, such and such a form in this flesh and these bones, this is Callias or Socrates; and they are different in virtue of their matter (for that is different), but the same in form (for that is indivisible)” (Aristotle, 1034, 5-7)
 See e.g. Aquinas 2006, Ia.75 and 76