Psychological Continuity and Personal Identity in Parfit's Thought Experiment — Epoche C1
The case, and exactly what it is a case of In section 89 of Reasons and Persons , published in 1984, Derek Parfit asks his reader to suppose that his brain is divided down the middle and that each half is transplanted into the body of one of his two triplet brothers, whose own brains have been destroyed. Two people wake up. Each has half of Parfit's brain, each remembers his life from the inside, each has his intentions and his character, and neither has any way of telling from the inside that anything has happened. This is the fission case, and everything that follows in this essay concerns what should be said about it. Parfit does not present the case as science fiction, and the reason matters for how much weight it can carry. He rests its coherence on two bodies of clinical work he discusses in the preceding section. The first is commissurotomy: from the middle of the twentieth century, surgeons treated intractable epilepsy by cutting the corpus callosum, the bundle of fibres joining the two cerebral hemispheres, and Roger Sperry and his collaborators showed in the 1960s that the separated hemispheres could then respond independently to stimuli confined to one half of the visual field — one hemisphere could answer a question the other could not. The second is hemispherectomy, in which one hemisphere is removed or disconnected and the patient survives with the remaining one. Together these establish, as far as any actual evidence can, that half a brain is enough for a mental life and that the two halves are not so integrated that dividing them destroys both. The thought experiment extrapolates from clinical fact; it does not invent a capacity. Why identity cannot branch, proved rather than asserted The original version of this essay stated that if identity is "strictly transitive and one-to-one", then the pre-fission person cannot be identical with both survivors. That is true, and it deserves to be shown rather than announced, because the whole problem rests on it and the demonstration takes three lines. Identity here is numerical identity — the relation each thing bears to itself and to nothing else — not qualitative similarity. Two mass-produced coins may be exactly alike and are still two coins. Numerical identity obeys two principles that no one in this debate disputes. The first is that it is an equivalence relation: reflexive, symmetric and transitive, so that $$ (a = b) \wedge (a = c) \;\Rightarrow\; (b = c). $$ The second is Leibniz's Law, the indiscernibility of identicals: if $b$ and $c$ are one thing, then every property of $b$ is a property of $c$. Now call the original person $A$ and the two survivors $B$ and $C$. Suppose $A = B$ and $A = C$. By the displayed principle, $B = C$. But $B$ and $C$ are in different rooms, have different experiences from the first moment after the operation, and can be pointed at separately; by Leibniz's Law, $B \neq C$. The supposition is therefore false: $A$ cannot be identical with both. Since nothing whatever distinguishes the two survivors' claims — each has half a brain, each is continuous in exactly the same way — there is no ground for saying $A$ is one and not the other. The conclusion is forced: $A$ is identical with neither. This is the "one-to-many problem". It is not a discovery about persons. It is a consequence of what identity is, and it would follow just as fast for a divided amoeba or a split river. Connectedness and continuity are two relations, and the essay conflated them The original text defined psychological continuity as "memories, intentions, beliefs, and character traits". That is a definition of a different relation, and the difference is the one piece of technical apparatus the argument cannot do without. The distinction has a history worth two paragraphs, because it explains why the apparatus exists. In the second edition of An Essay Concerning Human Understanding , published in 1694 — the chapter on identity was not in the first edition of 1689 — John Locke proposed that a person at one time is the same as a person at an earlier time just so far as the later one is conscious of the earlier one's actions and thoughts. Personal identity, for Locke, extends exactly as far back as memory reaches. Thomas Reid demolished the simple version in 1785, in the third of his Essays on the Intellectual Powers of Man , with the case of the brave officer. A boy is flogged for robbing an orchard. Years later a young officer takes a standard from the enemy, and at that moment remembers the flogging. Years later still, the man is a general, and remembers taking the standard but has entirely forgotten the flogging. On Locke's criterion the officer is the boy, and the general is the officer, but the general is not the boy — which contradicts the transitivity that identity must have. The repair, which is the modern psychological criterion, is to distinguish two relations. Psychological connectedness is the holding of direct psychological links: remembering a particular experience, carrying out an intention formed earlier, retaining a belief or a trait. Connectedness is what Locke had, it holds in degrees, and — as Reid showed — it is not transitive. Psychological continuity is defined as the ancestral of connectedness: the holding of overlapping chains of strong connectedness, so that the general is continuous with the boy by way of the officer even with no direct link between the endpoints. Continuity is transitive by construction, which is exactly why it, and not connectedness, is the relation a criterion of identity can use. Parfit adds a threshold for "strong": enough direct connections to match the number that hold over a day in the life of nearly any actual person. Keeping the two apart is not pedantry here, because Parfit's positive claim depends on it. What he thinks matters in survival is connectedness as well as continuity, and connectedness admits of degrees. If he is right, survival is a matter of degree, and the all-or-no