pith. sign in

arxiv: 2606.20637 · v1 · pith:XMSWMDB3new · submitted 2026-06-02 · 💻 cs.AI · cond-mat.stat-mech· cs.CY· physics.soc-ph

Constituency Optimisation Through Hamiltonian Representation Of Mandates (COTHROM): Algorithmic Redistricting of Irish Election Boundaries

Ruaidhr\'i Campion , Matthew Fenlon , Joshua Cooney Mercedal , Casey Farren-Colloty , Eliza Somerville , Michael A.J. Mitchell This is my paper

Pith reviewed 2026-06-28 10:21 UTC · model grok-4.3

classification 💻 cs.AI cond-mat.stat-mechcs.CYphysics.soc-ph
keywords redistrictingIrish electionsPotts HamiltonianMCMC optimizationproportional representationconstituency boundariessimulated annealingPareto optimality
0
0 comments X

The pith

A Potts Hamiltonian model of Irish constitutional rules, minimized by MCMC and selected by Pareto analysis, produces constituency boundaries that score higher than existing legal maps on proportionality and compactness in County Cork.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces COTHROM as a framework that encodes the multiple, often conflicting constitutional objectives for Irish electoral boundaries as separate terms in a Potts Hamiltonian. Coupling constants in the Hamiltonian act as adjustable weights for each objective, allowing the problem to be treated as an energy minimization task. Markov Chain Monte Carlo sampling combined with simulated annealing explores the enormous space of possible boundary configurations, after which multi-criterion decision analysis and Pareto optimality identify non-dominated solutions across different weightings. When applied to County Cork, the resulting boundaries improve on the current legal constituencies for the modeled objectives across a range of weight combinations.

Core claim

COTHROM parses the redistricting problem using statistical physics, where constitutional objectives are considered as terms in a Potts Hamiltonian. Markov Chain Monte Carlo methods and simulated annealing are employed to minimise this objective function, systematically exploring this configuration space, with coupling constants as proxies for objective weightings. Multi Criterion Decision Analysis (MCDA) and Pareto Optimality is then utilised to remedy the ambiguity in choosing a certain objective weighting combination over others. With respect to proportional representation and compactness objectives evaluated in County Cork, COTHROM consistently improves on the existing legal constituency

What carries the argument

The Potts Hamiltonian whose additive terms represent distinct constitutional objectives and whose coupling constants serve as tunable proxies for their relative weightings; the Hamiltonian is minimized by MCMC and simulated annealing, after which MCDA and Pareto selection extracts the final boundaries.

If this is right

  • Trade-offs between objectives become explicit and quantifiable through the choice of coupling constants.
  • The configuration space of possible boundaries can be searched systematically rather than by manual adjustment.
  • MCDA and Pareto optimality provide a way to compare solutions without committing to a single weighting in advance.
  • The method yields boundaries that improve measured scores on proportional representation and compactness relative to the existing legal map in County Cork.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same Hamiltonian construction could be adapted to other countries that use single-transferable-vote systems by redefining the energy terms to match their specific legal rules.
  • If the unmodeled constraints turn out to be binding, the Pareto front produced by the current method would require an additional post-processing filter before any boundary could be adopted.
  • Scaling the MCMC search to the full national level would depend on whether the computational cost grows acceptably with the number of constituencies.

Load-bearing premise

Constitutional objectives can be faithfully represented as additive energy terms in a Potts Hamiltonian whose coupling constants act as independent, meaningful proxies for legal weightings, and that MCDA/Pareto selection will produce boundaries that remain valid under the complete set of constitutional constraints even when some of those constraints are left out of the model.

What would settle it

Apply COTHROM to County Cork at the same objective weightings used for the reported results, then check whether any generated boundary set violates a constitutional rule not included in the Hamiltonian or scores worse than the legal map when evaluated on the full unmodeled legal criteria.

Figures

Figures reproduced from arXiv: 2606.20637 by Casey Farren-Colloty, Eliza Somerville, Joshua Cooney Mercedal, Matthew Fenlon, Michael A.J. Mitchell, Ruaidhr\'i Campion.

Figure 1
Figure 1. Figure 1: Exponential distribution (Equation 7) for [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: A temperature-dependent algorithm maximising a one-dimensional function. At fixed [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Plotted are the values for the individual terms in a typical COTHROM Hamiltonian [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: A plot capturing the convergence behaviour of the algorithm as a function of tem [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The Pareto front (denoted by the grey line) arising from considering only proportional [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The 2023 electoral boundaries for Cork. The values in the legend indicate the departure [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
Figure 5
Figure 5. Figure 5: Figure 6 and Figure 7 show the 2023 Cork constituency boundaries and a Pareto [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: The electoral boundaries generated by COTHROM, using the objective weightings [PITH_FULL_IMAGE:figures/full_fig_p015_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Electoral boundaries generated by COTHROM for midland counties with two 3-seater [PITH_FULL_IMAGE:figures/full_fig_p016_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Examples of constituency configurations, illustrating situations where county bound [PITH_FULL_IMAGE:figures/full_fig_p022_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: The area of each constituency-county overlap as a fraction of the total constituency or [PITH_FULL_IMAGE:figures/full_fig_p024_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: The area of each constituency-county overlap as a fraction of the total constituency or [PITH_FULL_IMAGE:figures/full_fig_p025_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: The area of each constituency-county overlap as a fraction of the total constituency or [PITH_FULL_IMAGE:figures/full_fig_p026_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: The area of each constituency-county overlap as a fraction of the total constituency or [PITH_FULL_IMAGE:figures/full_fig_p027_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Area fractions of overlapping constituency-county regions for 2023, 2017, 2013, and [PITH_FULL_IMAGE:figures/full_fig_p028_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Area fractions of overlapping constituency-county regions for 2023, 2017, 2013, [PITH_FULL_IMAGE:figures/full_fig_p029_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Statutory sub-ED splits illustrating weakness of assumption of ED indivisibility. In [PITH_FULL_IMAGE:figures/full_fig_p031_16.png] view at source ↗
read the original abstract

Electoral redistricting in Ireland's Proportional Representation Single Transferable Vote (PR-STV) system faces the challenge of selecting an optimally representative set of electoral boundaries from an enormous set of possible configurations, and where ``representative'' is a delicate balance of constitutional objectives that are often in tension with one another. We present the first computational framework for Irish electoral redistricting that systematically optimises across multiple constitutional requirements while making trade-offs explicit and quantifiable. The electoral redistricting problem is parsed using statistical physics, where constitutional objectives are considered as terms in a Potts Hamiltonian. Markov Chain Monte Carlo (MCMC) methods and simulated annealing are employed to minimise this objective function, systematically exploring this configuration space, with coupling constants as proxies for objective weightings. Multi Criterion Decision Analysis (MCDA) and Pareto Optimality is then utilised to remedy the ambiguity in choosing a certain objective weighting combination over others. With respect to proportional representation and compactness objectives evaluated in County Cork, COTHROM consistently improves on the existing legal constituency boundaries for a range of objective weightings.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 0 minor

Summary. The paper presents COTHROM, the first computational framework for Irish PR-STV redistricting. Constitutional objectives are encoded as additive terms in a Potts Hamiltonian; MCMC with simulated annealing minimizes the energy using tunable coupling constants as objective weightings; MCDA and Pareto optimality then select among the resulting maps. The central claim is that, evaluated on proportional representation and compactness in County Cork, the method consistently improves on existing legal boundaries across a range of weightings.

Significance. If the Hamiltonian faithfully encodes the legal criteria and the sampled maps satisfy all unmodeled constraints, the work would supply a novel, transparent, physics-based method for quantifying trade-offs in a domain that has lacked systematic algorithmic treatment. The explicit use of MCDA to handle weighting ambiguity and the attempt to produce falsifiable, multi-objective improvements are strengths. At present the absence of any reported quantitative metrics, validation protocol, or explicit term construction limits the assessed significance.

major comments (3)
  1. [Abstract] Abstract: the claim that COTHROM 'consistently improves' on legal boundaries rests on an evaluation performed in only one county; no quantitative metrics, baselines, error bars, number of weightings tested, or validation against held-out constitutional criteria are supplied, which is load-bearing for the central empirical claim.
  2. [Abstract] Abstract (Hamiltonian construction): coupling constants are introduced as free proxies for objective weightings and then tuned; the subsequent MCDA step is invoked precisely to resolve the ambiguity created by those free choices, creating moderate dependence of the final maps on modeling assumptions rather than on external benchmarks.
  3. [Abstract] Abstract (Potts terms): it is not shown whether the proportionality term computes exact PR-STV quota and transfer effects or only approximates them via additive penalties; if the latter, minima may violate the full set of constitutional constraints (population equality, contiguity, county integrity), undermining the validity of any reported improvement.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the detailed and constructive report. We address each major comment below and indicate planned revisions to the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that COTHROM 'consistently improves' on legal boundaries rests on an evaluation performed in only one county; no quantitative metrics, baselines, error bars, number of weightings tested, or validation against held-out constitutional criteria are supplied, which is load-bearing for the central empirical claim.

    Authors: We agree the abstract claim requires supporting detail. County Cork serves as the initial case study due to data availability and computational scale. The full manuscript contains figures comparing optimized maps to legal boundaries on proportionality and compactness, but these lack the explicit numerical values, baselines, error bars, and weighting counts requested. In revision we will expand the abstract, add a results section reporting specific metrics (e.g., proportionality deviation and Polsby-Popper scores), baselines, standard deviations from repeated MCMC runs, the exact number of weightings evaluated, and a limitations paragraph noting the single-county scope and absence of held-out validation. The central claim will be qualified accordingly. revision: yes

  2. Referee: [Abstract] Abstract (Hamiltonian construction): coupling constants are introduced as free proxies for objective weightings and then tuned; the subsequent MCDA step is invoked precisely to resolve the ambiguity created by those free choices, creating moderate dependence of the final maps on modeling assumptions rather than on external benchmarks.

    Authors: The tunable coupling constants are intentionally introduced as proxies to sample the multi-objective space; MCDA and Pareto analysis are then applied to identify maps that remain competitive across weightings. This design choice makes trade-offs transparent rather than fixing a single weighting a priori. We will revise the abstract and methods to state explicitly that MCDA selection mitigates single-assumption dependence and that the legal boundaries serve as the external benchmark for improvement. No change to the underlying approach is required. revision: partial

  3. Referee: [Abstract] Abstract (Potts terms): it is not shown whether the proportionality term computes exact PR-STV quota and transfer effects or only approximates them via additive penalties; if the latter, minima may violate the full set of constitutional constraints (population equality, contiguity, county integrity), undermining the validity of any reported improvement.

    Authors: The proportionality term employs additive penalties on quota deviation to approximate PR-STV effects, as exact transfer simulation at every MCMC step is computationally intractable. Hard constraints on population equality, contiguity, and county integrity are enforced separately via the configuration space definition and post-optimization checks. In revision we will add an explicit subsection describing term construction, state the approximation, report that all presented maps satisfy the hard constraints, and discuss the approximation's limitations. revision: yes

Circularity Check

0 steps flagged

No circularity: optimization result is independent of inputs

full rationale

The provided abstract and context describe a standard multi-objective optimization pipeline: constitutional goals are encoded as additive Potts terms, couplings act as explicit tunable weights, MCMC/annealing minimizes the resulting Hamiltonian, and MCDA/Pareto filtering selects among the resulting trade-off surfaces. The reported improvement is obtained by direct comparison of the minimized objective values against the fixed legal boundaries on the same metrics; this is a computational search outcome, not a quantity that equals its own modeling choices by construction. No equations, self-citations, uniqueness theorems, or fitted-parameter-as-prediction steps are present in the given text that would collapse the final maps back to the input ansatz or couplings. The derivation therefore remains self-contained against external benchmarks (the existing legal map).

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on treating constitutional mandates as additive Potts terms whose couplings are free parameters, plus the assumption that MCDA can resolve the resulting weighting ambiguity without introducing new circularity.

free parameters (1)
  • coupling constants
    Introduced as proxies for the relative importance of each constitutional objective; their values determine the energy landscape explored by MCMC.
axioms (1)
  • domain assumption Constitutional objectives can be represented as additive terms in a Potts Hamiltonian
    The paper parses the redistricting problem using statistical physics this way in the abstract.

pith-pipeline@v0.9.1-grok · 5761 in / 1374 out tokens · 26250 ms · 2026-06-28T10:21:43.935749+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

47 extracted references · 22 canonical work pages

  1. [1]

    1937 , note =

    Constitution of. 1937 , note =

    1937

  2. [2]

    Constituency Review Report 2023: Dáil and European Parliament Constituencies , year =

    2023

  3. [3]

    Electoral Redistricting in Ireland

    Coakley, John. Electoral Redistricting in Ireland. Redistricting in Comparative Perspective. 2008

    2008

  4. [4]

    Electoral Reform Act 2022 , year =

    2022

  5. [5]

    Electoral Reform Act 2022, Section 57: Constituency Review Reports , year =

    2022

  6. [6]

    Census of Population 2022 Profile 1 -- Population Distribution and Movements: Background Notes , year =

    2022

  7. [7]

    1964 , note =

    Constitution of. 1964 , note =

    1964

  8. [8]

    Book Review: David M

    Catt, Helena , year=. Book Review: David M. Farrell and lan McAllister, The Australian Electoral System: Origins, Variations and Consequences , volume=. Political Science , publisher=. doi:10.1177/003231870605800209 , number=

    work page doi:10.1177/003231870605800209

  9. [9]

    The Gregory Fractional Transfer , howpublished =

  10. [10]

    The Politics of Electoral Systems , publisher =

    , editor =. The Politics of Electoral Systems , publisher =. 2005 , month =. doi:10.1093/0199257566.001.0001 , url =

    work page doi:10.1093/0199257566.001.0001 2005

  11. [11]

    Droop, H. R. , year=. On Methods of Electing Representatives , volume=. Journal of the Statistical Society of London , publisher=. doi:10.2307/2339223 , number=

    work page doi:10.2307/2339223

  12. [12]

    McBride, James P. , year=. Positive aspects of PR‐STV for the Irish political system , volume=. Representation , publisher=. doi:10.1080/00344899608522980 , number=

    work page doi:10.1080/00344899608522980

  13. [13]

    Nagel , journal =

    Stuart S. Nagel , journal =. Simplified Bipartisan Computer Redistricting , urldate =

  14. [14]

    and Hess, Sidney W

    Weaver, James B. and Hess, Sidney W. , year=. A Procedure for Nonpartisan Districting: Development of Computer Techniques , volume=. The Yale Law Journal , publisher=. doi:10.2307/794769 , number=

    work page doi:10.2307/794769

  15. [15]

    Spin systems and Political Districting Problem , volume=

    Chou, Chung-I and Li, Sai-Ping , year=. Spin systems and Political Districting Problem , volume=. Journal of Magnetism and Magnetic Materials , publisher=. doi:10.1016/j.jmmm.2006.11.086 , number=

    work page doi:10.1016/j.jmmm.2006.11.086 2006

  16. [16]

    Chou, Chung-I and Li, S.P. , year=. Taming the Gerrymander—Statistical physics approach to Political Districting Problem , volume=. Physica A: Statistical Mechanics and its Applications , publisher=. doi:10.1016/j.physa.2006.01.082 , number=

    work page doi:10.1016/j.physa.2006.01.082 2006

  17. [17]

    Garfinkel, R. S. and Nemhauser, G. L. , year=. Optimal Political Districting by Implicit Enumeration Techniques , volume=. Management Science , publisher=. doi:10.1287/mnsc.16.8.b495 , number=

    work page doi:10.1287/mnsc.16.8.b495

  18. [18]

    Evolutionary Strategy for Political Districting Problem Using Genetic Algorithm , ISBN =

    Chou, Chung-I and Chu, You-ling and Li, Sai-Ping , year =. Evolutionary Strategy for Political Districting Problem Using Genetic Algorithm , ISBN =. doi:10.1007/978-3-540-72590-9\_174 , booktitle =

    work page doi:10.1007/978-3-540-72590-9

  19. [19]

    The Potts model , author =. Rev. Mod. Phys. , volume =. 1982 , month =. doi:10.1103/RevModPhys.54.235 , url =

    work page doi:10.1103/revmodphys.54.235 1982

  20. [20]

    Local linear forests

    Fifield, Benjamin and Higgins, , Michael and Imai, Kosuke and Tarr, Alexander , year=. Automated Redistricting Simulation Using Markov Chain Monte Carlo , volume=. Journal of Computational and Graphical Statistics , publisher=. doi:10.1080/10618600.2020.1739532 , number=

    work page doi:10.1080/10618600.2020.1739532 2020

  21. [21]

    and Jacobson, Sheldon H

    Swamy, Rahul and King, Douglas M. and Jacobson, Sheldon H. , year=. Multiobjective Optimization for Politically Fair Districting: A Scalable Multilevel Approach , volume=. Operations Research , publisher=. doi:10.1287/opre.2022.2311 , number=

    work page doi:10.1287/opre.2022.2311 2022

  22. [22]

    IEEE Transactions on Evolutionary Computation6(2), 182–197 (2002).https://doi.org/10.1109/4235.996017

    Deb, K. and Pratap, A. and Agarwal, S. and Meyarivan, T. , year=. A fast and elitist multiobjective genetic algorithm: NSGA-II , volume=. IEEE Transactions on Evolutionary Computation , publisher=. doi:10.1109/4235.996017 , number=

    work page doi:10.1109/4235.996017

  23. [23]

    Multi-objective optimization using evolutionary algorithms: Multi-objective optimization using evolutionary algorithms, by Kalyanmoy Deb, Pp

    Griffel, David , year=. Multi-objective optimization using evolutionary algorithms: Multi-objective optimization using evolutionary algorithms, by Kalyanmoy Deb, Pp. 487. £60. 2001. ISBN 0 471 87339 X (Wiley). , volume=. The Mathematical Gazette , publisher=. doi:10.1017/s0025557200173498 , number=

    work page doi:10.1017/s0025557200173498 2001

  24. [24]

    2018 , howpublished =

    2018

  25. [25]

    Recombination: A Family of Markov Chains for Redistricting , url=

    DeFord, Daryl and Duchin, Moon and Solomon, Justin , year=. Recombination: A Family of Markov Chains for Redistricting , url=. doi:10.1162/99608f92.eb30390f , journal=

    work page doi:10.1162/99608f92.eb30390f

  26. [26]

    Correction to: Political Geometry – Rethinking Redistricting in the US with Math, Law, and Everything In Between , ISBN=

    Duchin, Moon and Walch, Olivia , year=. Correction to: Political Geometry – Rethinking Redistricting in the US with Math, Law, and Everything In Between , ISBN=. doi:10.1007/978-3-319-69161-9_25 , booktitle=

    work page doi:10.1007/978-3-319-69161-9_25

  27. [27]

    Sequential Monte Carlo for sampling balanced and compact redistricting plans , volume=

    McCartan, Cory and Imai, Kosuke , year=. Sequential Monte Carlo for sampling balanced and compact redistricting plans , volume=. The Annals of Applied Statistics , publisher=. doi:10.1214/23-aoas1763 , number=

    work page doi:10.1214/23-aoas1763

  28. [28]

    2023 , howpublished =

    Eliza Somerville , title =. 2023 , howpublished =

    2023

  29. [29]

    Election Law Journal: Rules, Politics, and Policy , volume =

    Amariah Becker and Moon Duchin and Dara Gold and Sam Hirsch , title =. Election Law Journal: Rules, Politics, and Policy , volume =. 2021 , doi =

    2021

  30. [30]

    Discrete geometry for electoral geography , volume=

    Duchin, Moon and Tenner, Bridget Eileen , year=. Discrete geometry for electoral geography , volume=. doi:10.1016/j.polgeo.2023.103040 , journal=

    work page doi:10.1016/j.polgeo.2023.103040 2023

  31. [31]

    Gurnee, Wes and Shmoys, David B. , year=. Fairmandering: A column generation heuristic for fairness-optimized political districting , ISBN=. doi:10.1137/1.9781611976830.9 , booktitle=

    work page doi:10.1137/1.9781611976830.9

  32. [32]

    Political districting: from classical models to recent approaches , volume=

    Ricca, Federica and Scozzari, Andrea and Simeone, Bruno , year=. Political districting: from classical models to recent approaches , volume=. 4OR , publisher=. doi:10.1007/s10288-011-0177-5 , number=

    work page doi:10.1007/s10288-011-0177-5

  33. [33]

    1990 , howpublished =

    1990

  34. [34]

    1995 , howpublished =

    1995

  35. [35]

    2004 , howpublished =

    2004

  36. [36]

    2007 , howpublished =

    2007

  37. [37]

    2012 , howpublished =

    2012

  38. [38]

    2017 , howpublished =

    2017

  39. [39]

    2023 , howpublished =

    2023

  40. [40]

    Daniel and Vecchi, Mario P

    Kirkpatrick, Scott and Gelatt, C. Daniel and Vecchi, Mario P. , title=. Science , year=

  41. [41]

    Equation of State Calcula- tions by Fast Computing Machines,

    Metropolis, Nicholas and Rosenbluth, Arianna W. and Rosenbluth, Marshall N. and Teller, Augusta H. and Teller, Edward , title =. Journal of Chemical Physics , volume =. 1953 , type =. doi:10.1063/1.1699114 , url =

    work page doi:10.1063/1.1699114 1953

  42. [42]

    W. K. Hastings , journal =. Monte Carlo Sampling Methods Using Markov Chains and Their Applications , urldate =

  43. [43]

    Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , ISBN=

    GEMAN, STUART and GEMAN, DONALD , year=. Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , ISBN=. doi:10.1016/b978-0-08-051581-6.50057-x , booktitle=

    work page doi:10.1016/b978-0-08-051581-6.50057-x

  44. [44]

    , title =

    Ross, Sheldon M. , title =. Simulation , publisher =. 2005 , type =

    2005

  45. [45]

    2023 , url =

    Constituency Review 2023 Working Papers , author =. 2023 , url =

    2023

  46. [46]

    Electoral Act 2017 , year =

    2017

  47. [47]

    Electoral (Amendment) Act 2023 , year =

    2023