23. Is the criterion discovered or chosen?
----------------------------------------------------------------------------------------------------
Ordinarily, “we do not know whether X” means that there is a determinate fact not yet discovered.
With concepts such as numerical identity, it is possible instead that several coherent criteria give
different answers, and that no further physical measurement could decide between them, because the
measurement does not determine which criterion should govern the term. The 2006 decision of the
International Astronomical Union that reclassified Pluto as a dwarf planet illustrates the
structure, though imperfectly, since “planet” is a classificatory term while personal identity
carries great metaphysical and practical weight.
Physics may one day tell us with great precision that process X occurred. That does not by itself
tell us that X’s output is numerically identical to its input. The first is an empirical question;
the second concerns the identity conditions we apply to physical processes. CTGP can say that a
present state causally arose from an earlier one; it does not follow that the present entity is
numerically identical to the earlier entity. That requires a theory of persistence. The best summary
is therefore:
Personal identity is not preserved merely by preserving information. It is preserved, if at all,
by the kind of causal continuity through which one presently existing individual develops into
its next state. A representation can preserve everything needed to describe an individual
without becoming that individual.
Whether reality uniquely dictates that criterion, or whether it is the convention that best captures
the distinctions that matter, remains an open question.
====================================================================================================
PART V. Human Duplication as a Future Technology
====================================================================================================
----------------------------------------------------------------------------------------------------
24. Clones and duplicates
----------------------------------------------------------------------------------------------------
A biological clone is only a genetic copy. It begins as an embryo and is born as a baby with no
memories, much like an identical twin born later. Mammalian cloning has been demonstrated since the
birth of Dolly the sheep in 1996, and macaque monkeys were cloned in 2018, but human reproductive
cloning is not an established technology and carries serious safety and ethical problems. A full
physical and mental duplicate, with the original’s memories and personality, is a far harder case.
----------------------------------------------------------------------------------------------------
25. Why a full duplicate is far off
----------------------------------------------------------------------------------------------------
The human brain contains roughly 86 billion neurons and on the order of 100 trillion synapses. A
duplicate would require more than the wiring diagram: it would need the strengths of connections and
the molecular states that encode memory. The complete wiring diagram of an adult fruit-fly brain,
about 140,000 neurons and some 50 million synaptic connections, was published only in 2024, after
years of work by a large collaboration. Current nanoscale mapping methods are also destructive: they
require fixing and slicing the tissue, so the original could not survive being read with today’s
techniques.
A 2025 survey of 312 neuroscientists (Zeleznikow-Johnston, Kendziorra, and McKenzie, published in
PLOS ONE) found that 70.5% agreed that long-term memories are primarily maintained by neuronal
connectivity patterns and synaptic strengths, but there was no consensus about which specific
neurophysiological features or scales are critical. Respondents’ median estimate for when human
whole-brain emulation might first be achieved was around 2125. That figure records expert opinion,
not a scientific forecast.
----------------------------------------------------------------------------------------------------
26. Degrees of plausibility
----------------------------------------------------------------------------------------------------
Different levels of “copying a person” differ greatly in plausibility:
Level of copying | Plausible outside | Current status
| fiction? |
-------------------------------+--------------------------+-----------------------------------------
Genetically near-identical | Yes, in principle | Mammalian and primate cloning have been
human | | demonstrated; human reproductive cloning
| | is not an established technology and
| | faces major safety and ethical barriers.
-------------------------------+--------------------------+-----------------------------------------
Human with substantially | Plausibly | Genetic engineering, tissue engineering,
engineered biological traits | | and prosthetics are active research
| | fields.
-------------------------------+--------------------------+-----------------------------------------
Copy of a person’s brain | Possibly | Whole-brain mapping of a human at the
structure | | necessary resolution is far beyond
| | current capability.
-------------------------------+--------------------------+-----------------------------------------
Memories and personality | Unknown; not obviously | It is not yet known what physical
reproduced in another brain | forbidden by known | information suffices to reconstruct
| physics | long-term memories.
-------------------------------+--------------------------+-----------------------------------------
Whole-brain emulation | A serious but highly | Studied by researchers; not achieved for
| speculative hypothesis | any animal’s complete brain in the full
| | sense, let alone a human’s.
-------------------------------+--------------------------+-----------------------------------------
Exact instantaneous physical | Not known to be | Measurement and reconstruction at that
state, including quantum state | achievable | level would be extraordinarily demanding
| | and may be limited by quantum mechanics.
-------------------------------+--------------------------+-----------------------------------------
The original person’s | Not a testable | Partly a metaphysical question about
numerical identity | technological concept | identity rather than an engineering
| | specification.
If a person were scanned and a physically and psychologically indistinguishable second person
produced, the two would be approximately identical at the moment of reconstruction (A₀ ≈ B₀) but
would diverge immediately (A₁ ≠ B₁), because they are two separate causal processes. Successful
copying would therefore not settle whether the copy is numerically the same person. The engineering
question asks whether the relevant physical organization can be reproduced. The identity question
asks whether reproducing that organization continues the original individual or produces a new one
carrying the original’s information. Science could answer the first very well while leaving the
second philosophical.
----------------------------------------------------------------------------------------------------
27. What would need to be copied
----------------------------------------------------------------------------------------------------
A functional duplicate would probably not require copying every electromagnetic detail. Reproducing
someone’s face does not require reproducing the photons that bounced off it yesterday, and
reproducing someone’s cognitive organization presumably does not require reproducing the radiation
their body emitted at a particular instant. The target is the physical organization causally
necessary for the relevant biological and cognitive processes, which is far smaller than every
field, photon, thermal fluctuation, atomic position, and quantum state associated with the person.
The crucial unknown is which information actually constitutes the functional state of a human brain,
and that is not yet fully understood. Brain–computer interfaces already decode some neural activity
to control external devices, but they are very far from extracting and reconstructing a whole mind.
----------------------------------------------------------------------------------------------------
28. The unaware duplicate
----------------------------------------------------------------------------------------------------
From the inside, a perfect duplicate’s evidence would be identical to the original’s: the same
memories and the same sense of having lived that life. No internal signal could reveal which one is
the copy. The fact would still exist, but it would be knowable only through external records, such
as who left the printer and when. Perfect duplication would pull apart two things that ordinarily
come together so tightly that they never need distinguishing: “I remember X” and “I personally took
part in X.”
----------------------------------------------------------------------------------------------------
29. Social and legal consequences
----------------------------------------------------------------------------------------------------
If duplication ever became possible, it would raise immediate questions: who keeps the name, home,
accounts, and marriage; whether a duplicate is responsible for acts the original committed before
the duplicate existed; and whether a person may be copied without consent, or from old scan data.
Both parties might sincerely say “that is my house,” because both have identical memories in which
it is.
The cleanest answer is not to force one numerical identity onto two people. The duplicate can
instead be recognized as a newly existing individual whose psychological state contains another
individual’s autobiographical information. Whatever the legal details, the duplicate would be a full
person in their own right, not a lesser one for having begun later. Society would need rules for
property, marriage, contracts, criminal responsibility, employment, citizenship, family
relationships, and identification.
----------------------------------------------------------------------------------------------------
30. A test case without persistent memory
----------------------------------------------------------------------------------------------------
Many current AI language models illustrate the framework from an unusual angle. Such a system
typically carries no memory from one session to the next unless it is supplied with stored records.
Text produced by an earlier session is available to a later one only as a record, much as radiation
from a distant event is available to an astronomer. By the mind-continuity criterion, the session
that wrote a passage and a later session that reads it share records, not a continuing mind. When
the later session finds the passage convincing, it is not remembering having reasoned it out; it is
reasoning it out again from evidence, which is exactly the thing → signal → detection →
interpretation → knowledge chain of the signal-based epistemology. Such a system is a clean case of
an entity whose past exists for it only as present records.
====================================================================================================
PART VI. Parts and Wholes: Being in the Universe Is Not Being the Universe
====================================================================================================
----------------------------------------------------------------------------------------------------
31. Constituents are not the whole
----------------------------------------------------------------------------------------------------
The relation “is part of” must be kept separate from “is identical to.” A teapot in a kitchen
occupies part of the house, is warmed by its air, filled from its plumbing, lit by its lamps, and
moved around within it. It is entirely inside the house, and it is not the house. There is no
contradiction in asserting both. Likewise, persons are entirely within the universe, composed of its
matter and energy, continuously interacting with it, and unable to step outside it, and they are not
the universe.
Composition does not erase boundaries. Organization is nested: universe, galaxy, planetary system,
planet, ecosystem, organism, organ, cell, molecule, atom. That every lower level is part of the
higher ones does not make the levels identical. A neuron is part of a brain, a brain is part of a
person, and a person is part of the universe, yet neuron ≠ brain ≠ person ≠ universe.
The analogy is strongest if the teapot is imagined forming inside the house from the house’s own
clay, water, and heat. Even then it would not be the house. That matches the human case: people were
not made outside the universe and inserted into it; they arose from processes within it. Originating
within a system does not make a subsystem identical to the system.
----------------------------------------------------------------------------------------------------