Ash Catton

PhD Candidate, Clinical Psychology Student

Personal Identity as Fiction in Hume’s Treatise

October 19, 2018

Hume’s initial conception of diachronic identity in the Treatise concludes with the claim that personal identity is a construct of the imagination and is thus a fiction. However, a reading of the subsequent worries he expresses in the appendix may suggest that Hume is forced to retreat into an agnostic position on the matter. In this essay, I aim to illustrate how Hume’s worry arises by analysing the feasibility of Hume’s argument that he cannot find a self through introspection. In §1 I will provide a sketch of the original argument followed by a summary and interpretation of the appendix. In §2 I will outline Hume’s argument from introspection but conclude that it is not possible to discover a self through introspection alone. §3 contains my attempt to resolve the limitation exposed in §2 by suggesting that Hume’s account of sympathy resolves the inconsistencies while managing to remain consistent with the core aspect of the rest of his thesis.

1       Personal Identity as Fiction

David Hume’s Treatise of Human Nature is notable for its characteristic themes of empiricism and naturalism. His modus operandi was to investigate philosophical puzzles by unpacking the mental activities behind the puzzle rather than conceptually analysing the content of issue itself. His investigation of the issue of diachronic identity, or personal identity over time, is a product of this same treatment.

1.1     Hume’s First Thoughts
Hume begins his analysis by deconstructing the assumptions behind ascriptions of personal identity and concludes that it must rest on one or two properties.[1] The first is perfect simplicity which refers to a substance that cannot reduced to constituent parts. The second is perfect identity, or the property of something that remains unchanged over time. However, Hume claims that personal identity fails to exhibit either of these properties. The self lacks simplicity as it can be reduced to constituent perceptions and it can change over time. Thus, Hume’s first principle in this thesis is that the self is composed of a diversity of perceptions that are constantly in flux and apt to change. 

Furthermore, Hume claims that our perceptions are not tied to a persisting substance. His second principle is that our perceptions exist independently and lack any necessary connection among them. Given that we merely observe a continual chain of causation from one perception to another, coupled with our faculty of memory, personal identity is an inference generated by the imagination and nothing more. This results in his famous claim that persons are nothing but a bundle of different perceptions in flux.[2] The tendency to make this inference generates a habit-forming belief that results in judgements of diachronic identity.

Despite his argument, Hume does not take issue with the fact that personal identity possesses practical value. His discussions of morality later in the Treatise assume the construct of personal identity as it is commonly understood. Hume understands personal identity is a necessary fiction essential for the daily life of social creatures such as humans. However, if there is nothing that unites the bundle necessarily, how do we distinguish between different bundles for social purposes? Hume notes that our perceptions are linked by causation however this is insufficient as a criterion for our purposes. The ease with which people express offence at statements made by other people illustrates how one person can cause a perception in another. Hume’s argument from introspection suggests that he may be relying on accessibility of perceptions as the necessary criterion. In other words, Egbert is different from Gertrude because he does not have direct experiential access to Gertrude’s perceptions. However, this reliance on the first-person access may do considerable harm to Hume’s position.

1.2     Hume’s Second Thoughts
In the appendix to the Treatise, Hume restates his two principles but despairs that he cannot render them consistent.[3] It is not clear how the two principles are inconsistent however it has been noted that such an admission forces Hume to retreat to a position of agnosticism on the matter.[4] For instance, Swain suggests that the inconsistency lies not in the content of the principles but in the implications of them if they are true.[5] This leads to the view expressed by some that Hume cannot attribute any principles to the mind if he cannot offer a necessary criterion of individuation to guarantee the existence of a mind in the first place.[6] In other words, Hume’s own claim that there is nothing to unite the bundle requires that he posit something to unite it. How this problem arises is perhaps best exemplified by the argument from introspection.

2       The Problem of Introspection

In his argument against the Self[7] possessing the property of perfect identity, Hume illustrates the point by introspecting. He claims that when he turns his gaze inward, all that he sees are perceptions. As he says, “I never can catch myself at any time without a perception…”.[8] The point of this line of argument is to show the reader that if they performed the same exercise that they will not see some uniting immaterial substance like a soul, but the series of successive perceptions that Hume has argued for. This assumes that if the Self was not a fiction, that it could be found by introspection. However, it is not clear that this is the case.

2.1     Finding the Self in First-Person
Before addressing the question of whether it is possible to introspect the Self, there are some preliminary matters to address before we proceed. Firstly, before he presented the argument from introspection, Hume confessed that he has no idea of Self.[9] This is problematic if he has any chance of introspecting a Self, given that he has no idea what he is looking for. Consider the case of a fish swimming around a fish tank looking for this mysterious substance called ‘water’ without knowing what it is. This is expressed well in an objection commonly associated with Wittgenstein where he argued that nothing meaningful can be extracted from phenomenal experience alone.[10] In other words, without stating exactly what one is looking for beforehand, one could reply to Hume that they have found themselves via introspection. Without any idea of self to verify the claim there is no way to settle the matter. However, while it is not clear what Hume is looking for, he has already told us what function this potential object possesses. Either we are to find a necessary connection among our perceptions, or a substance that unites them. 

 Secondly, we could take issue with Hume’s use of “I” in his argument. It could be argued that Hume is not entitled to refer to himself in the first person given his negative thesis. However, this objection would only work if Hume denied the existence of the Self as a subject of experience. As Strawson notes, Hume could maintain that his negative thesis only applies to the persisting subject of the experience, and not the subject of experience as it relates to the perception at any one time.[11] However, it could be also be argued that “I” cannot be both subject and object of experience in the same time. For instance, it could be compared to the task of looking at your own eyes without a reflective surface. Ryle’s answer to this objection is to posit the level of awareness that Hume is engaging in as a higher-order processes, and Shoemaker distinguishes between objects of thought and their relation to mental states.[12] Thus, we can conclude that the “I” is a grammatical convenience that refers to a higher order process that looks inwardly at perceptions, as one may do when engaging in psychological mindfulness or meditation.

Despite these preliminary objections, it is still not clear whether it is possible to introspect the Self. Perhaps this is what Ryle refers to when describes the “systematic elusiveness of the ‘I’”.[13] Shoemaker makes the point that knowledge via introspection presupposes some knowledge of the self, claiming that even to recognise oneself in a mirror requires some knowledge that the object in the mirror is the self.[14] This is important for our purposes for two reasons. Firstly, it renders it doubtful that one can find the Self by introspection alone.[15] Secondly, it may account for Hume’s claim of inconsistency in the appendix. Hume needs to be able to individuate the bundle without severely compromising the groundwork he has already laid out. Given that the purpose of personal identity is largely a social or moral one, it is possible that the missing component lies in those contexts. Following from Hume’s empirical and naturalist tendencies, perhaps there is something to be gleaned from modern cognitive science.

3       Finding the Self in Third-Person

Social psychologists regard the concept of Self as being the product of social influences.[16] There are at least two possible explanations in the psychology literature to explain this. Firstly, Self could be an innate concept bestowed by evolution that enables us to navigate social structures. For instance, the social brain hypothesis argues that the brain size is a function of social group size.[17] Thus, this line of reasoning may entail that Self is a mere mental heuristic built into the social brain. However, there is some research to suggest that self-awareness can be taught. A recent study found that rhesus monkeys can be taught to recognise themselves in a mirror, challenging previous assumptions that their brain was not capable of doing so.[18] Whether innate or learned, it is likely that a sense of self relies in part upon interaction with others. Now we shall see if we can subsume these findings under Hume’s own work to find some criterion of individuation.

3.1     Mirrors of Sympathy
In his discussion of the passions in Book II of the Treatise, Hume suggests that sympathy acts as a channel of communication in allowing us to feel what other people are feeling, and this forms a crucial part of his account of moral psychology.[19] But it has been noted that this account of sympathy serves both as a source of belief in the existence of other minds and as a source of moral motivation.[20] Crucially, this may be where we can find our criterion of individuation. Earlier in Book II, Hume has established that passions such as pride and humility have Self as their object, whereas Hume conceives of sympathy as being other-regarding.[21] Thus, sympathy may satisfy our requirement for an external component which enables a sense of self to obtain. If we consider sympathy as a mirror upon which other people reflect our emotional states back at us, it seems inevitable that Self will eventually result. This is not troubled by Shoemaker’s mirror objection as Hume’s mirrors of sympathy provide feedback that links first-person experience to third-person agency. Hume is further able to individuate Self by his claim that we can only infer emotional states in others from their behaviour, while still only having access to one set of raw phenomenological content.[22] Relating back to our mirror, we can understand this in the way we can see the expression of our own emotions in a mirror but cannot see the experience of them. Thus, we can still affirm Hume’s claim that personal identity is a fiction, albeit a fiction generated by social interaction.

There is at least one significant problem for this conception. Hume explicitly distinguished between practical and theoretical contexts of personal identity. [23] The account presented here violates this distinction by supplementing Hume own theoretical account in Book I of the Treatise with his practical accounts in Books II and III. However, the distinction is not quite clear given that Hume’s naturalist tendencies meant that he investigated theoretical issues in a practical context. Although attempts have been made to elaborate on Hume’s distinction to make it clearer, admittedly this account is not helped by those contributions.[24]

4       Concluding Remarks

It is unclear precisely what Hume was worried about in the appendix. However, we have been able to find an inconsistency in the implications of the principles Hume invokes in his treatment of personal identity. Hume has been forced to deny the existence of any individuating object to guarantee personal identity, yet his own argument requires that one exist. His argument from introspection suggests that the over-reliance on his own mental processes means that he conceives Self in very individualistic terms, contrary to the context in which he treats Self for the purposes of moral evaluation. By complementing his bundle theory with sympathy, we have been able to provide the missing sufficient condition for individuation that Hume requires. Despite violating Hume’s distinction between contexts, this account upholds his claim that personal identity is a fiction.


Works Cited

Chang, L., Q. Fang, S. Zhang, M. M. Poo, and N. Gong. "Mirror-Induced Self-Directed Behaviors in Rhesus Monkeys after Visual-Somatosensory Training." Curr Biol 25, no. 2 (Jan 19 2015): 212-17.
Dunbar, R. I. "The Social Brain Hypothesis and Its Implications for Social Evolution." Ann Hum Biol 36, no. 5 (Sep-Oct 2009): 562-72.
Garrett, Don. Cognition and Commitment in Hume's Philosophy. New York: Oxford University Press, 1998.
Greco, Lorenzo. "The Self as Narrative in Hume." Journal of the History of Philosophy 53, no. 4 (2015): 699-722.
Hume, D., L.A. Selby-Bigge, and P.H. Nidditch. Enquiries Concerning the Human Understanding. Read Books, 1978.
Oyserman, Daphna, Kristen Elmore, and George Smith. "Self, Self-Concept, and Identity." In Handbook of Self and Identity, edited by Mark R. Leary and June Price Tangney. New York: Guildford Press, 2011.
Pitson, Tony. Hume's Philosophy of the Self. London; New York: Routledge, 2002. doi:10.1093/mind/113.450.384.
Ryle, Gilbert. The Concept of Mind. Hammondsworth, Middlesex: Penguin, 1963.
Shoemaker, Sydney. The First-Person Perspective and Other Essays. Cambridge; New York: Cambridge University Press, 1996.
Shoemaker, Sydney, and Richard Swinburne. Personal Identity. Oxford: Blackwell, 1984.
Strawson, Galen. The Evident Connexion. Oxford: Oxford University Press, 2011.
Swain, Corliss Gayda. "Personal Identity and the Skeptical System of Philosophy." In The Blackwell Guide to Hume's Treatise, edited by Saul Traiger. Malden, MA: Blackwell, 2008.
Vitz, Rico. "The Nature and Functions of Sympathy in Hume's Philosophy." In The Oxford Handbook of Hume, edited by Paul Russell. New York: Oxford University Press, 2016.
Wittgenstein, Ludwig. Philosophical Investigations. Translated by G. E. M. Anscombe, P. M. S. Hacker and Joachim Schulte. Malden, MA: Wiley-Blackwell, 2009. 1953.


[1] D. Hume, L.A. Selby-Bigge, and P.H. Nidditch, Enquiries Concerning the Human Understanding (Read Books, 1978), 252-63.[2] Ibid., 252.[3] Ibid., 635.[4] Galen Strawson, The Evident Connexion (Oxford: Oxford University Press, 2011), 54.[5] Corliss Gayda Swain, "Personal Identity and the Skeptical System of Philosophy," in The Blackwell Guide to Hume's Treatise, ed. Saul Traiger (Malden, MA: Blackwell, 2008), 144-45.[6] Tony Pitson, Hume's Philosophy of the Self (London; New York: Routledge, 2002), 69; Strawson, The Evident Connexion, 46, 107.[7] Self with capital S used to distinguish the concept from the pronoun.[8] Hume, Selby-Bigge, and Nidditch, Enquiries Concerning the Human Understanding, 252.[9] Ibid.[10] Ludwig Wittgenstein, Philosophical Investigations, trans. G. E. M. Anscombe, P. M. S. Hacker, and Joachim Schulte (Malden, MA: Wiley-Blackwell, 2009), 95e.[11] Strawson, The Evident Connexion, 41-42.[12] Gilbert Ryle, The Concept of Mind (Hammondsworth, Middlesex: Penguin, 1963), 186-89; Sydney Shoemaker, The First-Person Perspective and Other Essays (Cambridge; New York: Cambridge University Press, 1996), 5-8.[13] Ryle, The Concept of Mind, 178.[14] Shoemaker, The First-Person Perspective and Other Essays, 13; Sydney Shoemaker and Richard Swinburne, Personal Identity (Oxford: Blackwell, 1984), 105.[15] And consequently, it means that the Wittgensteinian objection has not been successfully addressed.[16] Daphna Oyserman, Kristen Elmore, and George Smith, "Self, Self-Concept, and Identity," in Handbook of Self and Identity, ed. Mark R. Leary and June Price Tangney (New York: Guildford Press, 2011).[17] R. I. Dunbar, "The Social Brain Hypothesis and Its Implications for Social Evolution," Ann Hum Biol 36, no. 5 (2009).[18] L. Chang et al., "Mirror-Induced Self-Directed Behaviors in Rhesus Monkeys after Visual-Somatosensory Training," Curr Biol 25, no. 2 (2015).[19] Hume, Selby-Bigge, and Nidditch, Enquiries Concerning the Human Understanding, 385.[20] Rico Vitz, "The Nature and Functions of Sympathy in Hume's Philosophy," in The Oxford Handbook of Hume, ed. Paul Russell (New York: Oxford University Press, 2016).[21] Hume, Selby-Bigge, and Nidditch, Enquiries Concerning the Human Understanding, 277.[22] Ibid., 576.[23] Ibid., 253.[24] Don Garrett, Cognition and Commitment in Hume's Philosophy (New York: Oxford University Press, 1998), 168; Lorenzo Greco, "The Self as Narrative in Hume," Journal of the History of Philosophy 53, no. 4 (2015).