⚛️ lattice

Bosonization versus the Nielsen-Ninomiya theorem

This paper demonstrates that chiral lattice fermions can be constructed from an ultra-local bosonic model via bosonization, yielding a doubler-free Dirac operator that circumvents the Nielsen-Ninomiya theorem by being non-local while still allowing for the gauging of non-anomalous symmetries.

Original authors: Saif Ullah Baig, Shi Chen, Aleksey Cherman, Maria Neuzil

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Saif Ullah Baig, Shi Chen, Aleksey Cherman, Maria Neuzil

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to build a video game world where the characters are "chiral fermions." In the language of physics, these are massless particles that only spin one way, like a screw that only turns clockwise. For decades, physicists have hit a massive roadblock when trying to put these characters onto a computer grid (a "lattice").

There's a famous rule in this field called the Nielsen-Ninomiya theorem. Think of it as a "No Free Lunch" law for grid-based physics. It says: If you want your grid to look like smooth, real-world physics, you can't have just one type of particle spinning one way. You will always accidentally create "ghost" copies of your particles (called doublers) spinning the wrong way, or you'll have to break the rules of symmetry.

It's like trying to comb a hairy ball: no matter how hard you try, you'll always end up with a cowlick (a problem spot) somewhere. You can't make the hair lie perfectly flat everywhere.

The Big Idea: The Magic of Bosonization
The authors of this paper, Saif Ullah Baig and his team, decided to try a different trick. Instead of trying to force the "hairy ball" to comb itself flat using the usual rules, they asked: What if we stop using "fermion" hair entirely and use "boson" hair instead?

They used a mathematical magic trick called bosonization. This is like saying, "We don't need to build the character out of fermion blocks; we can build it out of boson blocks, and the math says they will act exactly the same." They used a specific model called the 2D modified Villain model, which is a grid of bosonic fields.

The Result: A Ghost-Free, But Wobbly, World
Here is the exciting part: They successfully built a grid where the "ghost" copies (the doublers) do not exist. They managed to create a single, clean chiral fermion without the annoying extra copies.

But wait, there's a catch.
The Nielsen-Ninomiya theorem is a very strict bouncer; it doesn't let you off the hook that easily. The theorem says you can't have a local grid (where particles only talk to their immediate neighbors) that is also ghost-free.

So, how did the authors get around this? They didn't break the theorem; they just changed the rules of the game.

The Microscopic Level (The Ingredients): The underlying model they built is ultra-local. This means the basic building blocks (the bosons) only talk to their immediate neighbors. It's a perfectly tidy, local neighborhood.
The Macroscopic Level (The Result): When they looked at the "reconstructed" fermion (the character made from those bosons), they found it is non-local.

The Analogy: The Telepathic Neighbor
Imagine you live in a neighborhood where every house only talks to the houses right next to it (that's the ultra-local boson model). But, if you ask a specific question to a "fermion" character in that neighborhood, the answer doesn't come from the next-door neighbor. Instead, the answer seems to come from a house three blocks away, or even across the street, instantly.

The "Dirac operator" (the mathematical machine that tells the fermion how to move) is non-local. It has long-range connections. The paper proves that this non-locality is the price you pay to avoid the "ghost" particles. The theorem is satisfied because the "machine" isn't local anymore, even though the "ingredients" are.

What They Actually Found (and What They Didn't)
The team didn't just guess this; they did the math and the simulations.

They proved that the "ghost" particles are gone. There is only one zero (one particle type) in their momentum space.
They calculated that the "machine" connecting these particles falls off slowly, like $1/|x-y|$ . This confirms it is non-local.
They showed that even though the underlying model is simple and Gaussian (like a perfect bell curve), the resulting fermions actually interact with each other in a tiny way at the grid scale. However, they calculated that this interaction is "irrelevant," meaning if you zoom out to the real world (the continuum limit), these interactions vanish, and you get a perfect, free particle.

What They Explicitly Rule Out
The paper is very clear about what this is not:

It is not a way to make a local, ghost-free fermion. The theorem still holds; you can't have both.
It is not a microscopic theory of fermions. The fundamental fields are bosons. The fermions are "reconstructed" or "emergent" objects.
It does not solve the problem for 3D or 4D worlds yet. The authors admit that while they hope this works in higher dimensions (like 3D), bosonization in those dimensions is still poorly understood. They are only sure about the 2D case.

The Bottom Line
The authors have shown that you can build a lattice model that has no "ghost" particles and preserves chiral symmetry, but only if you accept that the resulting particle moves in a "non-local" way. The microscopic world is tidy and local, but the emergent particle is a bit of a wanderer, connecting to distant parts of the grid.

They haven't "solved" the Nielsen-Ninomiya theorem; they've shown a clever way to dance around it by giving up locality for the reconstructed particle, while keeping the underlying theory perfectly local. It's a valid, calculated, and exact solution for 2D, offering a fresh perspective on how to think about these stubborn particles.

Technical Summary: Bosonization versus the Nielsen-Ninomiya Theorem

Problem Statement
The central challenge addressed in this work is the formulation of massless chiral fermions on a Euclidean spacetime lattice while preserving chiral symmetries. The Nielsen-Ninomiya theorem establishes a fundamental obstruction: any lattice Dirac operator that is translation-invariant, local, and possesses the standard chiral structure must inevitably produce unphysical "doublers" (extra massless modes) or violate chiral symmetry. Standard lattice approaches typically resolve this by either breaking chiral symmetry (e.g., Wilson fermions) or accepting non-ultra-local symmetries (e.g., overlap/domain-wall fermions). An alternative strategy involves abandoning Grassmann fields entirely, utilizing bosonic lattice models to realize the desired symmetries and anomalies, and identifying a critical point where chiral fermions emerge. While the 2D modified Villain model has been proposed as a lattice regularization for a massless Dirac fermion via bosonization, the specific properties of the resulting chiral fermion operators at finite lattice spacing—particularly regarding locality and the evasion of the Nielsen-Ninomiya theorem—remained unclear.

Methodology
The authors investigate the 2D modified Villain scalar model, a lattice regularization of a compact scalar field that preserves the mixed 't Hooft anomaly between shift and winding symmetries. The methodology proceeds as follows:

Model Definition: The study utilizes the 2D modified Villain model defined by a lattice path integral over a real scalar field $\phi$ , an integer-valued link field $n$ , and a compact scalar field $\theta$ on plaquettes. The coupling is fixed at the self-dual radius $R = 1/\sqrt{2}$ , where the bosonic theory is dual to a massless Dirac fermion.
Operator Construction: To define lattice Weyl fermion operators, the authors adapt the continuum Coleman boson-fermion dictionary. They construct composite operators involving the scalar field $\phi$ and the vortex defect operator $e^{i\theta}$ . Crucially, to ensure gauge invariance under the lattice gauge redundancy, these operators are attached to topological lines (rays) emanating from the lattice sites.
Correlation Function Calculation: The authors compute the exact two-point correlation functions of these lattice Weyl operators. This involves evaluating the path integral, which reduces to calculations involving the 2D lattice Green function $G(x)$ .
Reconstruction of the Dirac Operator: By inverting the computed two-point correlation functions in momentum space, the authors reconstruct the effective lattice Dirac operator ( $D$ ) that governs the dynamics of these fermionic operators.
Analysis of Locality and Doublers: The reconstructed Dirac operator is analyzed in momentum space to determine its zero structure (doublers) and its spatial decay properties (locality).

Key Contributions and Results

Exact Correlation Functions: The paper derives the exact two-point correlation functions $S_{\pm}(x-y)$ for the lattice Weyl operators. These functions exhibit the correct fermionic statistics (anticommutativity) and asymptotic behavior $S_{\pm}(x) \sim (x_1 \pm i x_2)^{-1}$ in the continuum limit, matching the correlation functions of a massless Dirac fermion.
Doubler-Free Spectrum: The reconstructed lattice Dirac operator is shown to be free of doublers. In momentum space, the inverse propagator possesses a single zero at $p=0$ (corresponding to the physical fermion). Depending on the choice of contact terms (renormalization), additional zeros may appear at the Brillouin zone boundaries (e.g., $(\pi, 0), (0, \pi), (\pi, \pi)$ ), but their total topological index cancels out, ensuring no extra physical massless modes exist.
Non-Locality of the Dirac Operator: The primary result is that while the underlying microscopic bosonic model is ultra-local (the action and symmetries involve only nearest-neighbor interactions), the reconstructed lattice Dirac operator is non-local. The operator kernel decays as $O(1/|x-y|)$ at large distances.
Multi-point Functions: The authors demonstrate that the lattice Weyl operators possess non-vanishing connected four-point functions at finite lattice spacing. This indicates a non-trivial two-body interaction at the lattice scale, which is irrelevant in the continuum limit, consistent with the theory flowing to a free massless Dirac fermion.

Significance and Evasion of the Nielsen-Ninomiya Theorem
The paper clarifies how the bosonized model evades the Nielsen-Ninomiya theorem. The theorem forbids a local, translation-invariant, doubler-free Dirac kernel with standard chiral structure. The modified Villain model respects translation invariance and is doubler-free, but it evades the theorem by sacrificing locality in the reconstructed Dirac operator.

The authors emphasize a critical distinction: the non-locality resides in the effective Dirac operator derived from the bosonic theory, not in the microscopic degrees of freedom. The underlying bosonic lattice model remains ultra-local. Consequently, the non-locality of the Dirac operator does not obstruct the gauging of the model's symmetries (such as $U(1)_V$ and $U(1)_A$ ), as these are gauged directly on the local bosonic variables.

Conclusion
This work provides a microscopic demonstration that chiral fermions can be realized on a lattice without doublers using purely bosonic variables. The price for this realization is that the effective Dirac operator is non-local. The paper concludes that while this non-locality is a necessary consequence of the Nielsen-Ninomiya theorem for a doubler-free theory, it does not render the theory pathological, as the fundamental bosonic action remains ultra-local and well-defined. The authors suggest that similar mechanisms may apply to higher-dimensional dualities, such as 3D bosonization involving Chern-Simons terms, though this remains an open area for future investigation.

More like this