DIII Boundary Anomaly vs Matter-Content Constraints

Purpose

This note advances the roadmap item:

• compare the DIII boundary anomaly data with the matter-content anomaly constraints.

The main outcome is a clarification of scope. The comparison is real, but the strongest possible phrasing is too strong in its current form.

Starting point

Two statements are already live elsewhere in the corpus:

1. the corrected Spin(2,3) topological reading gives a DIII-type bulk/boundary story with unit winding magnitude |W_3| = 1, so the m = 0 transition surface carries a protected massless T1 boundary mode and a corresponding T_0 / parity-type anomaly;
2. the static-consistency side treats T1 \otimes (3 \oplus 1) as the left-handed matter seed and uses anomaly cancellation to constrain what additional right-handed states are required.

The tempting bridge claim is:

the DIII inflow condition and the matter-content anomaly cancellation are literally the same constraint.

This note checks how far that can safely be pushed.

The 4d matter-content anomaly constraints

Take one Standard-Model-shaped generation written in the usual left/right language:

• Q_L : (3,2)_{y_Q}
• L_L : (1,2)_{y_L}
• u_R : (3,1)_{y_u}
• d_R : (3,1)_{y_d}
• e_R : (1,1)_{y_e}
• optional nu_R : (1,1)_{y_nu}

Then the familiar four-dimensional anomaly-cancellation conditions are:

\(A_{SU(3)^2U(1)} = 2y_Q - y_u - y_d = 0,\) \(A_{SU(2)^2U(1)} = 3y_Q + y_L = 0,\) \(A_{\mathrm{grav}^2U(1)} = 6y_Q - 3y_u - 3y_d + 2y_L - y_e - y_{nu} = 0,\) \(A_{U(1)^3} = 6y_Q^3 - 3y_u^3 - 3y_d^3 + 2y_L^3 - y_e^3 - y_{nu}^3 = 0.\)

There is also the mod-2 SU(2) global condition:

• the number of left-handed SU(2) doublets must be even.

These constraints are the ones that tell the consistency layer which right-handed completion is allowed once the left-handed T1 seed is chosen.

Two structural features matter here:

• the perturbative gauge anomalies depend on linear and cubic hypercharge data;
• the SU(2) global obstruction is mod 2.

The 3d DIII boundary anomaly data

On the topological side, the m = 0 transition surface is being read as the effective boundary of a DIII bulk. The relevant anomaly is the inability to regularize the boundary while keeping the protecting symmetry and gauge invariance manifest at the same time.

For a 2+1-dimensional boundary fermion system, the induced parity-anomaly data is encoded by Chern-Simons counterterms. Up to convention-dependent normalizations, the gauge pieces are controlled by quadratic indices:

\(\Delta k_G \propto \frac12 \sum_i T(R_i),\) \(\Delta k_{U(1)} \propto \frac12 \sum_i q_i^2,\)

with an analogous gravitational piece controlled by the net boundary fermion content. The DIII bulk inflow cancels this boundary obstruction.

Three points are crucial:

1. the boundary coefficients are quadratic in charges and representation indices;
2. the DIII invariant fixes a unit of protected boundary anomaly data, with sign/orientation tied to the chosen convention for W_3;
3. this data lives naturally on a 2+1-dimensional boundary/transition surface, not directly in the 3+1-dimensional perturbative anomaly polynomial.

Direct comparison

The comparison is informative, but it does not support the literal statement that the two analyses are already “the same constraint.”

What does not match directly

The perturbative four-dimensional anomaly constraints and the three-dimensional parity anomaly do not use the same invariants:

• SU(3)^2U(1), SU(2)^2U(1), and U(1)^3 anomalies are linear/cubic in hypercharge;
• the parity anomaly is quadratic in the boundary charges and Dynkin indices.

So a bare coefficient match of the form

DIII inflow coefficient = full 4d gauge-anomaly coefficient

is too strong as written. They are different anomaly polynomials in different dimensions.

What can match

There is still a meaningful lower-dimensional shadow:

1. the SU(2) mod-2 condition is the clearest common remnant. The T1 matter seed gives 3 + 1 = 4 weak doublets per generation, which is even. That is exactly the kind of datum that survives both as the four-dimensional Witten condition and as the boundary parity/global obstruction.

2. the boundary parity anomaly can constrain the reduced transition-surface spectrum. If one dimensionally reduces the candidate T1 matter content to the m = 0 surface, then the sums of quadratic indices must be compatible with the DIII inflow supplied by the bulk.

3. the DIII story can therefore control a parity/global shadow of the matter content. What it does not yet do is replace the full perturbative SU(3)^2U(1), SU(2)^2U(1), U(1)^3, and mixed gravitational-U(1) cancellation problem.

What the bridge should now mean

The safe reformulation is:

DIII inflow may encode the 2+1-dimensional parity/global shadow of the T1 matter-content constraints after reduction to the transition surface.

That is a real and potentially valuable bridge. It would mean:

• the topological side checks whether the reduced T1 boundary spectrum is globally/parity consistent;
• the consistency side checks whether the four-dimensional hypercharge completion is perturbatively anomaly free;
• a successful bridge would show that these are compatible reductions of one common matter sector, not that one calculation simply replaces the other.

Refined next calculation

The right next step is no longer “match one coefficient and declare success.”

It is:

1. specify the reduced T1 boundary spectrum at the m = 0 transition surface;
2. compute its 2+1-dimensional parity-anomaly data for SU(2), the selected U(1), and any retained color sector;
3. isolate the mod-2/global piece and compare it with the four-dimensional weak-doublet counting condition;
4. only after that ask whether the quadratic boundary data can be obtained from a controlled reduction of the higher-dimensional anomaly polynomial.

That would turn the bridge from a slogan into a finite calculation.

Minimal boundary-spectrum bookkeeping

One can already push the reduced bookkeeping a little further if one makes the minimal localization assumption:

at the m = 0 transition surface, the protected boundary matter inherited from the static side is the left-handed seed Q_L \oplus L_L coming from T1 \otimes (3 \oplus 1).

This is only a working assumption. It is not yet derived that the full static matter seed localizes on the DIII transition surface in this way. But under this assumption, the parity/global bookkeeping is explicit.

Weak/global piece

Per generation, the reduced boundary seed contains:

• one quark doublet Q_L : (3,2)
• one lepton doublet L_L : (1,2)

So the number of SU(2) doublets is

which is even. This matches the four-dimensional mod-2 SU(2) consistency condition.

For the perturbative 2+1-dimensional parity bookkeeping, the induced weak Chern-Simons level shift is proportional to

with color multiplicity counted explicitly. Since T_2(\mathbf 2)=1/2,

\(\sum_i T_2(R_i) = 3 \cdot \frac12 + 1 \cdot \frac12 = 2,\) \(\Delta k_{SU(2)} = 1.\)

So the weak-sector parity shift is integral. In other words:

• the global/mod-2 obstruction vanishes;
• the perturbative weak parity shift is not half-integral.

This is the cleanest point of contact between the DIII boundary story and the matter-content constraints.

Color piece

For the same left-handed seed, the color contribution is

Only the quark doublet contributes, with weak multiplicity 2. Since T_3(\mathbf 3)=1/2,

\(\sum_i T_3(R_i) = 2 \cdot \frac12 = 1,\) \(\Delta k_{SU(3)} = \frac12.\)

That is half-integral rather than integral. So if one literally localizes only the left-handed T1 seed on the transition surface, the color parity bookkeeping is nontrivial and would require either:

• additional boundary states,
• an inflow term carrying the corresponding color response,
• or a weaker reading in which the DIII boundary mode is not the full Q_L \oplus L_L matter seed.

U(1) piece

The U(1) part is normalization-dependent, so the safest statement is symbolic:

for the left-handed seed alone.

If one uses the Standard-Model target values only as an illustrative check,

then

So in the illustrative Standard-Model normalization, the U(1) parity shift is again nontrivial.

What this already tells us

The reduced bookkeeping now separates into two qualitatively different pieces:

1. the weak/global piece lines up neatly with the even-doublet consistency condition;
2. the color and U(1) parity data do not collapse to that same simple statement.

So the comparison has become sharper:

• the DIII bridge is already credible for the weak/global shadow;
• it is not yet credible as a full one-to-one identification of the entire gauge-anomaly system.

There is also a counting tension with the topological side. Even under the sharpened reading, the DIII discussion in kernels/topological.md concerns one protected massless T1 channel in the minimal reduced block at the transition surface. By contrast, the literal left-handed T1 matter seed already contains, per generation,

• three quark-doublet copies, and
• one lepton-doublet copy.

So if the boundary story were read particle-by-particle, the static seed would look like a multi-multiplet boundary spectrum, not like a single protected mode. This strongly suggests that at least one of the following must be true:

1. the DIII boundary mode is a collective reduced channel rather than the full localized matter seed;
2. only a proper subchannel of the static T1 matter seed is topologically protected at the transition;
3. additional degeneracy, flavor, or localization structure still has to be specified before the topological and static counts can be compared directly.

Minimal-block reading already suggested by the corpus

The existing corpus already points to the most conservative reading of that counting tension.

On the topological side, the DIII calculation in papers/paper2-diii-winding-number/ is performed on the reduced chiral Hamiltonian built from the bare Spin(2,3) gamma-matrix block. Its direct input is the four-component spinor alone.

On the static/reduction side, the corpus already says that this four-component block decomposes as

so the visible weak-doublet multiplicity is already part of the minimal topological block. By contrast, the color and generation structures are introduced elsewhere:

• color through the (\mathbf 3 \oplus \mathbf 1) factor in the static matter ansatz;
• generation through J_3(O) or the higher 2/4/6 access-ladder discussion.

So the safest current interpretation is:

W_3 = 1 is first a statement about one protected reduced T1 channel in the minimal Spin(2,3) block, not yet a statement about the full color-and-generation-dressed matter spectrum.

This weakens the apparent contradiction substantially. The current topological result is prior to tensoring in any extra spectator structure.

Spectator-copy consequence

If the extra internal sectors enter only as uncoupled spectators, then the total Hamiltonian takes the block form

and all additive topological/parity data scale with the number of uncoupled copies. In particular, one would expect:

• the total protected-mode count to multiply by the number of spectator copies;
• the parity-anomaly coefficients to add copy by copy.

But the current DIII paper computes only the invariant of the minimal block H_{\mathrm{Spin}(2,3)}. It does not yet specify whether color, generation, or right-handed completion appear as:

1. spectator tensor factors;
2. boundary-localized degrees of freedom inside the same protected channel;
3. bulk-only or non-topological sectors not counted by the minimal DIII invariant.

So the remaining question is not simply “why does one not equal many?” It is sharper:

which parts of the static matter structure are actually present in the boundary Hamiltonian whose DIII invariant was computed?

Scenario matrix

The localization problem can now be stated as a short list of explicit Hamiltonian scenarios.

Scenario A: minimal reduced block only

Implications:

• the computed claim |W_3| = 1 is the whole topological statement;
• the protected object is one reduced T1 channel in the bare four-component block;
• color, generation, and right-handed completion are not yet part of the boundary Hamiltonian.

This is the most conservative reading, and it is fully compatible with the current DIII paper.

Scenario B: uncoupled spectator copies

Implications:

• the protected-mode count multiplies by N_{\mathrm{spec}};
• additive parity data also multiply by N_{\mathrm{spec}};
• if W_3(H_{\mathrm{Spin}(2,3)}) = \pm 1, then the total invariant is expected to be \(W_3(H_{\partial}) = N_{\mathrm{spec}}\,W_3(H_{\mathrm{Spin}(2,3)}).\)

So if the full boundary Hamiltonian literally carried, for example, one independent copy for each color/lepton slot, then one would expect a multiple of the minimal invariant rather than the bare unit result. This means Scenario B is not the current calculated statement; it is a possible extension of it.

Scenario C: full left-handed T1 matter seed localized on the boundary

Here one takes the boundary spectrum to contain the whole Q_L \oplus L_L seed.

Implications:

• the weak/global SU(2) shadow works cleanly;
• the color and illustrative U(1) parity data are nontrivial, as computed above;
• the protected-channel count no longer looks minimal unless the localized seed is further organized into collective channels.

So Scenario C requires additional structure beyond the present topological computation.

Scenario D: only a reduced protected subchannel localizes

Here the DIII invariant protects one collective channel, while the rest of the static matter structure is:

• bulk-only,
• non-topological,
• or localized but not symmetry-protected by the same minimal invariant.

This is currently the best candidate if one wants to keep:

• the exact minimal-block |W_3| = 1 result,
• the weak/global SU(2) shadow,
• and the possibility of richer color/generation structure elsewhere in the framework.

Immediate consequence

Scenarios A and D are compatible with the present state of the corpus without further topological input.

Scenarios B and C are not ruled out, but they are not what has been computed. They would require a new Hamiltonian-level extension in which the spectator or matter sectors are explicitly included in the boundary operator before any W_3 or inflow claim is upgraded.

Corpus default and upgrade rule

For the current corpus, the clean default reading should now be fixed explicitly:

• established reading: Scenario A;
• working extension for the anomaly bridge: Scenario D;
• upgrade-required alternatives: Scenarios B and C.

This means:

1. whenever the repo cites the present DIII calculation as an established result, it should be read in Scenario A form: one protected reduced T1 channel in the minimal four-component Spin(2,3) block;
2. whenever the repo speculates about connecting that topological statement to richer matter structure, the safe working hypothesis is Scenario D: one protected reduced channel plus additional static structure that is not yet proved to sit in the same protected boundary Hamiltonian;
3. any attempt to promote spectator multiplication or full Q_L \oplus L_L localization into the live topological claim must first write down the enlarged boundary Hamiltonian and recompute W_3 and the induced parity data there.

So the practical rule is simple:

until a larger boundary Hamiltonian is explicitly built and analyzed, the topological corpus should speak in Scenario A language and treat Scenario D as the leading bridge hypothesis.

This suggests that the next calculation must answer a localization question before it can answer a matching question:

does the DIII boundary mode represent the whole static T1 matter seed, or only a reduced protected channel inside it?

Until that is fixed, any stronger anomaly-inflow identification remains underdetermined.

Bottom line

The DIII/topological comparison is still worth pursuing, but it sharpens the project in a more disciplined way than the strongest bridge claim suggested.

• The strongest current statement is too strong: the DIII inflow condition is not literally identical to the full four-dimensional matter-content anomaly constraints.
• The promising weaker statement survives: DIII inflow may capture a lower-dimensional parity/global shadow of those constraints, especially the even-doublet obstruction.
• The next useful task is therefore a reduced boundary-spectrum calculation, not a direct identification of the entire gauge-anomaly system with the DIII inflow coefficient.