Contribution
The entry is the field’s standard map of personal identity, and its value is in the distinctions. It splits “personal identity” into seven questions and shows that the persistence question is regularly confused with characterization (“the same person” in the sense of character) and with a narrower question that assumes we are persons essentially. It then gives the persistence debate in its current form: the psychological view as the majority view, fission as its central problem, and the too-many-thinkers objection as the animalist’s main argument.
Key insight
“What makes you the same person?” is three different questions, and most of the debate’s confusion comes from answering one while asking another. The numerical question (whether a past being is you at all) is about what kind of thing you are, and so may not be answerable from the concept of a person alone.
Assumptions
- Persistence is numerical identity, which is transitive and one–one. The arguments from fission (§5) and from memory chains (§4) turn on this.
- Human organisms think if anything does. The too-many-thinkers objection (§6) depends on it, and the entry says the denial is “hard to defend”.
- Organisms have brute-physical persistence conditions: a transplanted brain does not take its organism with it. “New animalism” (§6), which denies this, is noted.
- Thought experiments are evidence about our persistence conditions. Brain transplants, fission and duplicating machines carry the argument.
Key results
There are no formal results. The entry’s structure, in order:
- §1, seven questions. Characterization (“Who am I?”, a “small subset” of one’s properties, contingent and changeable; “the characterization question (Schechtman 1996: 1)”), personhood (Locke’s “thinking intelligent being” definition as the commonest answer), persistence, evidence (memory versus physical continuity; Williams’s Guy Fawkes case), population (split brains, multiple personality, partial twinning), personal ontology (seven candidate answers from animalism to “we don’t really exist”, with “no consensus or even a dominant view”) and what matters in survival.
- §2, the persistence question. Not a question about meaning, which would make the answer a priori and the same for all possible people. Not “remaining the same person”, which “presupposes” numerical identity. Framing it as “a person at one time and a person at another” assumes person essentialism, which with a Lockean definition of person implies you were never an embryo and could not be in a vegetative state. The memory criterion, so framed, is “compatible with your surviving with no memory continuity at all, as long as this happens when you are not a person”.
- §3, three answers. Psychological continuity, brute-physical continuity, anticriterialism.
- §4, psychological continuity. The memory criterion fails transitivity (the student, the lawyer and the old woman) and is circular (genuine memory presupposes identity). Quasi-memory answers the circularity. Causal dependence (Shoemaker) answers dreamless sleep: beliefs “have caused themselves to continue existing”. Whether brain-state transfer counts is left open (Shoemaker yes, Unger no).
- §5, fission. Lefty and Righty cannot both be you. Multiple occupancy (two people coincide before the operation, usually with temporal parts) or non-branching (fission is death). Non-branching makes survival depend on whether a second hemisphere is destroyed, which the entry calls “mysterious”, and it is why Parfit and others say identity is not what matters.
- §6, too many thinkers. If you are not your organism, two beings think your thoughts and you cannot know which you are, as after a duplicating machine. Reply one: organisms cannot think (threatens to make them zombies). Reply two: only persons, defined to persist psychologically, are referred to by “I”. Reply three: new animalism.
- §7, animalism. That we are organisms. Not an account of personhood, and consistent with “wholly inorganic people such as gods or intelligent robots”. It fits every actual case of survival and our having been foetuses. The transplant intuition is the main cost, weakened by Snowdon’s tumour case.
- §8, wider themes. Temporal-parts ontology yields many candidate referents for “I” (a continuity-aggregate, a connectedness-aggregate, an organism, a stage), so reference and therefore what we are may be indeterminate (Sider).
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | ”Personal identity” names several largely independent questions, often confused | strong, as a conceptual map | §1 definitions and examples |
| C2 | The persistence question concerns numerical identity and is not the “same person” (characterization) question | strong | §2 argument that the latter presupposes the former |
| C3 | Framing persistence as person-to-person assumes person essentialism, which is controversial | strong as a point about framing | §2; the memory-criterion and embryo examples |
| C4 | Memory continuity cannot be what persistence consists in; causal psychological continuity avoids the obvious objections | moderate | §4: transitivity, circularity, sleep; Shoemaker’s causal-dependence repair |
| C5 | Fission forces psychological views into multiple occupancy or non-branching, and the latter makes identity not what matters | strong as a dilemma | §5 |
| C6 | Psychological-continuity views imply two thinkers per human, with an epistemic problem about which one is you | moderate; depends on organisms thinking | §6 argument and duplicating-machine analogy |
| C7 | The transplant objection to animalism is “less compelling than it first appears” | weak to moderate; author’s side | §7, Snowdon’s tumour case |
| C8 | On a temporal-parts ontology it may be indeterminate what we are | moderate, conditional on temporal parts | §8, after Hudson and Sider |
Concepts
- persistence question — what is necessary and sufficient for a past or future being to be someone existing now (numerical identity over time).
- characterization question — what makes someone the person they are, in the “identity crisis” sense; the term is Schechtman’s.
- person essentialism — nothing that is in fact a person could exist without being a person.
- psychological connectedness / continuity — causal dependence of later psychological states on earlier ones / an overlapping chain of such connections.
- brute-physical view — persistence consists in a physical relation not involving psychology; with animalism, in being the same organism.
- anticriterialism — there are no informative necessary and sufficient conditions for persistence.
- multiple occupancy / non-branching — the two responses to fission.
- too-many-thinkers objection — if you are not your organism, both think your thoughts.
Connections
- Dufner, “Personal Identity and Ethics” (LIT-130, NOTE-166). The sister SEP entry. Dufner surveys the ethical uses of the positions this entry sets out, and finds “the relation that fixes persistence and the relation that matters ethically keep coming apart”. This entry’s §5 is where that split starts: non-branching makes identity not what matters. Read the two together.
- Olson, The Human Animal (LIT-448, Deferred). §§6–7 compress its argument: the “fetus problem” and the transplant cases are its chapter titles. The entry cites “Olson 1997: 22–26” for the claim that the memory criterion, framed person-to-person, says nothing about non-persons.
- Schechtman, The Constitution of Selves (LIT-468, Deferred). The entry takes the term “characterization question” from her and treats that question as not the persistence question. Her publisher says she “takes issue with analytic philosophy’s emphasis on the first sort of question [the amnesiac’s] to the exclusion of the second [the adolescent’s]”. So the entry sets aside the question her book says analytic philosophy wrongly sets aside. Olson and Witt’s “Narrative and Persistence” (2019), in the bibliography, is where the author takes narrative views on directly.
- Floridi (LIT-136, NOTE-130). Floridi’s deflation of the diachronic question to purpose-relative levels of abstraction is a version of the view, noted in §2, that the persistence question is answerable by conceptual means. The entry rejects that view.
- Dennett, “Where Am I?” (LIT-405, NOTE-338). The duplicated narrator is the entry’s brain-state-transfer and duplicating-machine cases in story form. Dennett answers by point of view; the entry would ask which being the point of view belongs to.
- Dennett, “The Self as a Center of Narrative Gravity” (LIT-442, NOTE-364). It answers a question the entry declines to ask, what a self is, and its “two selves in one body” bears on the population question.
- Meincke’s process view of pregnancy (LIT-142, NOTE-115) is a counting problem of exactly the population kind, and the entry’s §1 cites partial twinning as one. Meincke’s answer (something between one and two) is the kind of indeterminacy §8 reaches by another route.
- Parallel filings. The classical sources the entry reports (Locke’s definition and memory criterion, Hume’s bundle, Parfit’s fission and “what matters”, Williams’s Guy Fawkes and body-swap cases, Lewis on survival and multiple occupancy) are being filed by another reader at the same time. This entry is the frame they fit into.
Bearing on the record
- The record’s self cluster. Most of what the record holds on the self (Dennett, Minsky’s self-models in LIT-432, Levin’s scale-free selves in LIT-439, the extended self of LIT-097) is about what a self is, which this entry explicitly puts aside. So the entry does not adjudicate them. It shows which question each is answering: Levin and Clark and Chalmers answer the personal-ontology question about boundaries, Dennett the characterization and population questions, none the persistence question.
- THEORY-043 (the bridge between a society of mind and a conscious United States). The too-many-thinkers objection is a counting argument of the kind THEORY-043 meets: whether a part (an organism, a hemisphere) and a whole that shares its thinking can both be thinkers. The pressure the entry relies on, that there should be one thinker per chair, is an anti-duplication intuition like the anti-nesting principles that document lists, applied to persons and organisms rather than to groups. The pairing is mine; the entry does not discuss groups.
- Machines. Personhood is asked of computers and animalism allows robot persons. The entry does not discuss AI, uploading or model copies, but its §4 (brain-state transfer) and §5 (fission) are the standard frame for them, and LIT-206’s per-conversation instance is a population-question answer.
- No ML instruction. Nothing here belongs in the anthology.
Limitations
- A survey with a side. §6 presents the objections to the psychological view at length and their replies as hard to defend. §7 presents the main objection to animalism and then argues it is “less compelling than it first appears”. The author is the leading animalist.
- No self. The narrative self, the minimal self and self-models are outside it by decision, so it is not a map of the self literature this batch files.
- Citation slip. §2 cites “Olson and Witt 2020” for the claim that person essentialism is controversial. The bibliography’s only Olson and Witt entry is 2019, “Narrative and Persistence”, Canadian Journal of Philosophy 49: 419–434. Either the in-text year or the bibliography is wrong, or an entry is missing. I did not check which.
Open questions
- Does the characterization question have a metaphysics of its own, as Schechtman claims, or is it psychology? The entry treats it as the latter by placing it outside persistence.
- What does the population question say about a system that runs many simultaneous copies of one set of weights? The entry’s tools apply; the entry does not apply them.