| paper | d4-conf-aldt-2021-0014 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | yes |
The Condorcet Amendment theorem derives a rule axiomatically and gives no named computational result to carry into a continuous setting. The proposed population version is precise and plausible, but computing it would be a new task rather than a mirror of a computational result in the paper.
fails bit a — no named computational result to mirror
The proposed extension covers the Condorcet Amendment rule; it leaves aside the founding-rule characterizations, Conservative Amendment, and the iterative-equivalence result.
I can’t honestly give an anchor-compliant case: the supplied paper has no named computational result asserting tractability or hardness, and its theorem titles have no printed theorem numbers. Its propositions characterize rules; they do not classify the complexity of a problem. The closest model source is “Theorem [Condorcet Amendment],” but it is an axiomatic characterization, not an eligible computational anchor.
The strongest positive case is that this theorem gives an unusually direct basis for a new high-multiplicity computation. Its content is: under the stated axioms and single-peaked preferences, change the current threshold \(\delta\) upward to the largest \(x>\delta\) for which more than \(xn\) agents have ideal points at least \(x\), or downward to the smallest \(x<\delta\) for which more than \(\delta n\) agents have ideal points at most \(x\); if neither change qualifies, keep \(\delta\).
A plausible regime is a large membership community whose members’ preferences over constitutional thresholds cluster at a modest number of ideal points—for example, many members aligned with a small number of caucus positions. A type is an ideal point \(\theta\in[1/2,1)\), which completely determines an agent’s preferences in the paper’s model. Its mass \(\mu_\theta\) is the fraction of members with that ideal point. The decision variable is the new threshold; the selection criterion is the paper’s Condorcet and posterior-consistency requirements, rather than a monetary objective.
A precise candidate question is High-Multiplicity Condorcet Amendment. An instance gives a rational current threshold \(\delta\), a finite support \(S\subseteq[1/2,1)\) of rational ideal points, and binary-rational masses \(\mu_\theta>0\) summing to one. Define
\[ U=\{x\in(\delta,1):\mu([x,1))>x\},\qquad D=\{x\in[1/2,\delta):\mu([1/2,x])>\delta\}. \]
Return the extremal qualifying upward and downward thresholds, if they exist; if neither direction qualifies, return “retain \(\delta\).” If a qualifying set has no required extremum, report its exact endpoint and that it is unattained, since then the literal continuous rule selects no amendment in that direction. Types are indistinguishable exactly as the paper’s model requires, and replacing counts by population mass leaves its amendment test intact. The paper’s authors could plausibly recognize this as computing their amendment rule on a continuous population.
For this candidate problem, I expect tractability: sort the \(\tau=|S|\) ideal points, scan the cumulative masses, and compare rational values. That takes polynomial time in \(\tau\) and the input bit length \(L\). This is a prospective classification, not a result proved by the paper. It also generates questions about sensitivity to small changes in \(\mu\) and about repeated conservative amendments.
The weakest point is substantive: the paper does not ask a computational question, so this proposal adds one. Also, its finite electorate gives the threshold rule a finite set of significant values, while the continuous version can lose its selected maximum. For instance, with \(\delta=1/2\), mass \(0.2\) at \(0.55\), and mass \(0.8\) at \(0.9\), upward amendments qualify for every \(x<0.8\) sufficiently close to \(0.8\), but not at \(0.8\); there is no largest qualifying \(x\). That boundary problem is part of the mirror’s challenge, not a reason to pretend the paper already solved it.
The strongest case against a ChoCo mirror is that this paper has no computational result to mirror. Its Condorcet Amendment theorem characterizes a rule from axioms and preferences; it gives no algorithmic problem, complexity classification, or computational bound. Turning that characterization into “compute the amendment from the population” is a plausible new task, but it is not a continuous mirror of a computational result in the paper.
The proposed high-multiplicity model is coherent: ideal points can be complete types, and masses can represent a large membership with repeated preferences. But with finite rational support, computing the qualifying thresholds reduces to sorting the ideal points, accumulating their masses, and comparing rational values. That is a valid algorithmic question, but the paper supplies no deeper computational structure for it. Tractability alone would not make a mirror unworthy; the issue is that this would be routine evaluation of an axiomatic rule, rather than a complexity question the paper establishes or motivates.
There is also a genuine existence gap in the direct continuous rule. With current threshold \(\delta=\tfrac12\), put mass \(0.2\) at ideal point \(0.55\) and mass \(0.8\) at \(0.9\). The upward condition \(\mu([x,1))>x\) holds for every \(x\in(\tfrac12,0.8)\), but fails at \(0.8\). Thus there is no largest qualifying amendment. Returning the unattained endpoint diagnoses the failure; choosing it as the new rule fails the stated condition, while choosing some smaller qualifying value requires an extra selection rule. Restricting thresholds or adding a boundary convention could repair the model, but would add a constitutional choice absent from the theorem.
One could instead study how much mass must shift to induce a desired amendment. That would be a more substantial population-level computation, but the paper specifies no costs, allowable changes, or manipulation objective to determine that problem. Those choices would define a new research direction rather than mirror a result here.
So the negative case is strongest on scope, not on the claim that no plausible population regime exists. A finite-support amendment computation is possible, and the population interpretation makes sense. What is missing is a computational contribution in the paper to carry across; the better extensions require adding the computational problem themselves.
The adversarial triple: the proponent anchors on up to three named results; the opponent sees that case and must defeat every anchor; the judge decides which case convinced it. These are the pipeline’s own outputs, generated by tools/triple_run.py — no human edited them. The paper’s own text is not reproduced here beyond the quoted statement above.