14. Quantum Particles: Identity and Individuality I.

Approaches to the Notion of Individuality

1. Property-Based "Bundle" View An individual = a bundle of properties

Topics: I. Individuality II. Fermions and Bosons III. Classical and Quantum Statistics IV. Quantum Individuals

So: Properties individuate objects. No two individuals can be completely indiscernible (possess all the same properties):

Principle of the Identity of Indiscernibles: If two objects are indiscernible, then they are identical. 2. Haecceitism Every individual possesses a "primitive thisness" (haecceity) that makes it unique. Self-identity formulation: Every individual is identical to itself. What about quantum objects? Can they be considered individuals?

II. Fermions and Bosons

In brief: There are two basic types of quantum particles: fermions and bosons. These behave differently in large groups! (They obey different types of statistics.) And they behave differently from classical particles in large groups. (Classical particle statistics is distinct from the two types of quantum particle statistics.) These differences influence the extent to which quantum particles may be considered individuals. Historical Context: Pauli Exclusion Principle for Atoms: (Pauli 1925) Electrons in an atom are characterized by four properties: energy n, orbital angular momentum , z-component of orbital angular momentum m, and spin ms. Values of these properties are quantized: n = 1, 2, ...; = 0, 1, 2, ... (n 1); m = , ... 0, ..., ; ms = 1/2, +1/2. Exclusion Principle: No two electrons can be in the same state (i.e., no two electrons can have all the same values of these four properties). 1

 The Exclusion Principle dictates how electrons fill up the "orbits" of atoms (it dictates the possible atomic states electrons can be in): n: 1 Z 1 2 3 4 5 6 7 8 9 Element H hydrogen He helium Li lithium Be beryllium B boron C carbon N nitrogen O oxygen F fluorine : 0 1 2 2 2 2 2 2 2 2 2 1 2 2 1 2 2 2 3 2 4 2 5 2 6 orbitals s orbital ( = 0) p orbital ( = 1) d orbital ( = 2) f orbital ( = 3) etc. 2 0 1 3 0 1 2 4 0 1 2 3 energy shells K shell (n = 1) L shell (n = 2) M shell (n = 3) N shell (n = 4) etc.

10 Ne neon

electrons can have the same 4 numbers.

Any electron is described by 4 numbers (n, , m, ms). The Exclusion Principle says no two Example: In a lithium atom, there are three electrons. One has the values (1, 0, 0, +1/2). Another has (1, 0, 0, 1/2). The third has (2, 0, 0, +1/2). Note: Electrons have the property of "Spin-1/2". You can imagine them spinning up or spinning down along any given axis. A Spin-1/2 property has just two values (in general, call them +1/2 and 1/2). (Both Hardness and Color are Spin-1/2 properties.) There are other more complex types of spin properties in which the image of spinning about an axis breaks down: Property Spin-1/2 Spin-3/2 Spin-5/2 . . . Spin-0 Spin-1 Spin-2 . . . #values two four six values 1/2, +1/2 3/2, 1/2, +1/2, +3/2. 5/2, 3/2, 1/2, +1/2, +3/2, +5/2 these are called "half-integer-spin" properties

one three five

0 1, 0, +1 2, 1, 0, +1, +2

these are called "integer-spin" properties

Def. A fermion is a system that possesses a half-integer-spin property. A boson is a system that possesses an integer-spin property. Experimentally: (1) (2) A system cannot be both a fermion and a boson. Fermion multiparticle states and boson multiparticle states are Permutation Invariant: exchanging the places of single particle states results in a multiparticle state that is physically indistinguishable from the original un-permuted multiparticle state. (3) (4) Fermion multiparticle states obey a Generalized Pauli Exclusion Principle: Two or more identical fermion single particle states cannot appear in the same multiparticle state. Boson multiparticle states do not obey such an Exclusion Principle: There can be boson multiparticle states that consist of two or more identical single particle states.

Aside: Permutation Invariance is an experimental result: Permuting the particles (exchanging their places) results in a state that is physically indistinguishable from the original state. To see how it's encoded in the vector space formalism, consider a 2-particle state vector | in a product space H H. Suppose | is a single particle state vector in H and | is a single particle state vector in H. Then here are possible forms | can take: 1. 2. 3. 4. |1 = || |2 = {|| + ||} |3 = {|| ||} |4 = || + {|| ||}

A permutation of | exchanges the places of | and |. Represent it by the following: Def. i. Perm(|) |' = Result of exchanging | for | everywhere in |.

Def. ii. A symmetric state vector is a state vector | such that |' = |. An anti-symmetric state vector is a state state | such that |' = |. Note that |2 is symmetric, |3 is anti-symmetric, while |1 and |4 are neither. To say that | is Permutation Invariant is to say that it represents a state that is physically indistinguishable from the state represented by its permutation |'. Informally this means that the states represented by | and |' agree on all their properties. Formally, this means that, for any operator A representing a property, the expectation value of A is the same for | and |': Def. iii. A multiparticle state vector | is Permutation Invariant just when |A| = '|A|', for any operator A.

Aside, cont. It should be obvious that symmetric and anti-symmetric state vectors are Permutation Invariant. Since state vectors that are neither symmetric nor anti-symmetric in general will be such that | |', these in general will not be Permutation Invariant. So to be safe, we stipulate that all quantum multiparticle state vectors must either be symmetric or anti-symmetric. (Technically, there can be state vectors that are Perm Invariant for which |' is related to | by a phase other than 1. These guys are said to obey "para-statistics".) Now suppose we allow multiparticle states in which the particle states are identical. In the 2-particle system above, let | = |. Then the symmetric state vectors || and || are now possible. These describe two identical particles in the same (2-particle) state! But there will now be no anti-symmetric state vectors! Set | = | in |3 and it vanishes. In general, for identical particle states, the only way |' = | (which defines an anti-symmetric state vector) is if | = 0! This suggests using anti-symmetric state vectors to represent multiparticle states that obey the Pauli Exclusion Principle. This guarantees that the multiparticle states they represent cannot contain particles with all the same (non-spatiotemporal) properties. Upshot: Fermion multiparticle states are represented by anti-symmetric state vectors. Boson multiparticle states are represented by symmetric state vectors.

III. Classical and Quantum Statistics

Group Rule for Fermions: Fermi-Dirac (FD) Statistics. (a) A multiparticle fermion state is invariant under particle permutations. (Permutation Invariance.) (b) Two or more fermions with all the same (non-spatiotemporal) properties cannot be in the same state. (Generalized Pauli Exclusion Principle). Group Rule for Bosons: Bose-Einstein (BE) Statistics. (a) A multiparticle boson state is invariant under particle permutations. (Permutation Invariance.) (b) Two or more bosons with all the same (non-spatiotemporal) properties can be in the same state. (No Exclusion Principle.)
Aside: These experimental rules can be derived in Quantum Field Theory in what's called the "SpinStatistics Theorem". Briefly, if you assume the states of systems obey special relativistic symmetries, and you assume they obey a locality constraint ("micro-causality"), then the group rules follow.

Now: Compare with the statistics for classical particles: Group Rule for Classical Particles: Maxwell-Boltzman (MB) Statistics. (a) A classical multiparticle state is not invariant under particle permutations. (No Permutation Invariance.) (b) Two or more classical particles with all the same (non-spatiotemporal) properties can be in the same state. (No Exclusion Principle.) 4

Example: Suppose we have two particles in a 2-particle state composed of two single-particle states A, B. How can we calculate the probability that one of the particles is in state A? First: Need to determine all the possible forms the 2-particle state can take. Your answer will depend on what type of statistics you use (i.e., what type of particles you're talking about). 1. Classical particles: A (1) B Use MB statistics: There are 4 possible forms for the 2-particle state. (4 possible ways to distribute two classical particles over two states.) Assign each of these possible 2-particle states equal probability of 1/4 (Principle of Indifference). Pr(one particle in A and one particle in B) = Pr(state 3) + Pr(state 4) = 1/4 + 1/4 = 1/2.

(2)

(3)

(4)

2. Bosons: A (1) B Use BE statistics: There are 3 possible forms for the 2-particle state. (3 possible ways to distribute two bosons over two states.) Assign each of these possible 2-particle states equal probability of 1/3 (Principle of Indifference). Pr(one particle in A and one particle in B) = Pr(state 3) = 1/3. (3)

(2)

3. Fermions: A (1) B Use FD statistics: There is only one possible form for the 2-particle state (due to the Exclusion Principle). Pr(one particle in A and one particle in B) = Pr(state 1) = 1.

IV. Quantum Individuals

Question: What does Permutation Invariance of quantum states say about the status of quantum particles as individuals? Knee-Jerk Claim: Classical particle are individuals: switching two of them makes a difference. Quantum particles are not individuals: switching two of them does not make a difference.

But: Permutation Invariance applies to states. It is a constraint on the possible states that a given multiparticle system can be in. Individuality need not depend on constraints placed on the possible states a system can be in.

1. The Haecceity View of Individuals Claim: Classical and quantum particles possess "primitive thisness" (haecceities). The Main Difference: Classical haecceities are physically distinguishable: you can tell them apart based on the states they occupy. Quantum haecceities are physically indistinguishable: no experiment can distinguish between two quantum haecceities.

2. The "Bundle" view of Individuals Claim: Classical and quantum particles consist of bundles of properties. Question: Do classical and quantum particles, so-conceived, satisfy the Principle of the Identity of Indiscernibles (PII)? PII: If two objects are indiscernible, then they are identical.

Depends on what you mean by indiscernible. Suppose two objects are indiscernible just when they have all the same properties. Suppose we identify two types of properties: 6

1. A monadic property = a "single-place" property that an object can possess without reference to other objects. (Example: mass.) 2. A relational property = a "multi-place" property that an object can only possess with respect to one or more other objects. (Example: Being taller than.) PII.v1: If two objects agree on all of their monadic and relational properties, then they are identical. PII.v2: If two objects agree on all of their monadic and relational properties, excluding spatiotemporal properties, then they are identical. PII.v3: If two objects agree on all of their monadic properties, then they are identical.

Now Claim: Classical particles can violate PII.v2 and PII.v3. There can be two classical particles that have the same monadic properties and relational properties, excluding spatiotemporal ones. But classical particles satisfy PII.v1. They can always be discerned by their positions in space (and time). Quantum particles can violate PII.v2, PII.v3, and PII.v1. (Assumption: quantum particles don't always have well-defined positions!) So: Either quantum particles are not individuals, or they must possess haecceities.

But: Do these versions of the PII exhaust all the possible ways two objects can be discerned from each other? Counterexample Consider 2 identical iron spheres one mile apart in an otherwise empty universe.

1 mile

 Spheres agree on all monadic and relational properties, excluding spatiotemporal ones. If spatiotemporal properties are relational with respect to other objects, then they will agree on these. (Ex., They both stand in the spatial relation of being 1 mile from the other.) If spatiotemporal properties are relational with respect to absolute spacetime, then whether or not the spheres are identical will depend on the global topology of the spacetime.

1 mile

1 mile

closed global topology

open global topology

So the spheres violate PII.v1, PII.v2, and PII.v3. But intuitively they are distinct: there are two of them. The spheres are individuated by number alone ("solo numero"). This can be indicated by any two-place irreflexive relational property that they possess. This is a two-place relational property that relates one sphere with the other, but does not relate either of the spheres with itself. Example: "Being 1 mile apart from." This is a two-place relation that either sphere has with the other, but neither has with itself. The latter means that it is irreflexive. Suggests the following ways two objects can be discernible: (a) Two objects are absolutely discernible if they differ in a monadic property.

(b) Two objects are relatively discernible if they differ in a relational property. (c) Two objects are weakly discernible if they differ in an irreflexive relational property.

Putative examples of weakly discernible objects: The two iron spheres. The points in a Euclidean space. Right and left hands in an empty universe.

Claim: Fermions are (minimally) weakly discernible. Due to the Exclusion Principle, two fermions will always be discernible; minimally by their values of spin. Consider a 2-fermion state that is maximally symmetric; i.e., in which the two fermions agree to the maximal extent on all their properties: |1|2 |1|2 ()

Entangled 2-particle state in which both fermions are in superpositions of being spin-up-along-some-axis and spin-down-along-that-axis

In this state, both fermions have exactly the same mass, charge, and all other intrinsic properties. Technically, this state is spherically symmetric (i.e., this is a spatial 2-particle property that the 2-fermion state as a whole possesses), so each fermion has exactly the same spatiotemporal properties and relations. But there is an irreflexive relation that holds between them: "having opposite direction of each component of spin to". So: Fermions satisfy the PII, appropriately construed. Thus, even under the Bundle View, they can be considered individuals. But: Bosons are not constrained by the Exclusion Principle. Hence they can be in states in which they are not discernible, even weakly. An example of a maximally symmetric 2-boson state is |1|2. This is a state in which the two bosons agree on all their properties, monadic, relational, and spatiotemporal. So: Bosons do not satisfy the PII. Thus they cannot be considered individuals under the Bundle View. (They might be considered modes of an appropriate field: photons (EM field), W and Z particles (Weak field), gluons (Strong field), Higgs boson (mass field).)

Aside: A potential problem with this claim about fermions is that it assumes "having opposite direction of each component of spin to" holds for both fermions in the maximally symmetric state (). This suggests that both fermions have definite values of spin when they are in the state (). But this is a matter of interpretation! Under a literal interpretation, we cannot say that either fermion has a definite value of spin in the state ()! This suggests that whether or not quantum particles are individuals depends not only on the theory of individuality you adopt, but also on the interpretation of QM you adopt.