| paper | d4-journals-mss-2004-0011 |
| authors | — |
| venue | ? |
| filed under | unclassified |
| judged by | gpt-6-luna / xhigh (triple__luna__xhigh__c6r1) |
| judge confidence | high |
| authors would recognise it | no |
Proposition 4 is computational, but its result concerns deciding whether arbitrary recursive coalitions win by checking membership of finitely many named carrier players. The proposed bloc table replaces those individual membership bits with collective role ballots, while the population masses do not affect the answer. That is a different voting model, not a continuous-population version of the result the authors would recognize.
fails bit b-none — no population with multiplicity to relax
A distribution over type masses cannot encode whether particular zero-mass carrier individuals belong to a coalition; replacing them with collective bloc roles changes the game, while adding their membership bits separately leaves the continuous population irrelevant.
fatal: True
The attempted scope is Proposition 4 and its dictatorial special case; it leaves the computability obstructions and the anonymity and insensitivity results without a mirror.
The strongest honest mirror is the paper’s finite-carrier result, and it is a narrow one. Proposition 4, proved in this paper, says: if a simple game \(\omega\subseteq \mathrm{REC}\) has a finite carrier \(S\)—so a coalition \(A\) wins exactly when \(A\cap S\) wins—then \(\omega\) is \(\delta\)-computable. Here \(\delta\)-computability means that the game’s indicator on characteristic indices for recursive coalitions has a partial recursive extension. The proof builds a finite winning/losing table for the subsets of \(S\); given a coalition’s characteristic index, it checks membership at the finitely many carrier players and consults that table. Dictatorial games are a special case.
A plausible high-multiplicity setting is a large electorate organized into voting blocs or formal roles: a few legally decisive blocs and many routine members whose votes the rule ignores. Many people may belong to each bloc, but members of a bloc share its role and ballot; the number of people \(n\) can be far larger than the number of types \(\tau\). A type records the role and preference, and its mass \(\mu_t\) is the fraction of society in that type. This is credible where blocs vote cohesively. The carrier roles preserve the distinctions the rule actually uses; the ordinary electorate can be grouped into background types.
Call the resulting decision problem Continuous Finite-Carrier Winner Determination. An instance gives a finite role set \(R\), a designated carrier \(C\subseteq R\), rational masses \(\mu_t\) over the supported role-and-preference types with \(\sum_t\mu_t=1\), and the current ballot of each carrier role. The carrier’s rule is given as an explicit Boolean table \(g:\{0,1\}^{|C|}\to\{0,1\}\), where \(g(z)=1\) means the coalition pattern \(z\) is winning. The question is whether the current coalition wins; a solution is the correct yes/no bit, obtained by evaluating \(g\) on the carrier roles’ ballots. The coalition’s population share is also specified by the masses. As in the paper, however, the rule depends on which carrier roles support the alternative, not on their total mass. This is a winner-determination problem, not an optimization problem, so there is no objective to minimize.
I would expect this explicit-table version to be tractable: the algorithm reads the masses and carrier ballots, then looks up one entry of \(g\), in time polynomial in the input length. That is a complexity claim about the proposed mirror, not a polynomial-time bound proved by Proposition 4. The paper proves Turing computability, with no runtime bound.
The scope is Proposition 4’s finite-carrier result, including its dictatorial special case. It does not mirror the paper’s broader claims about the limits of computability, anonymity, or insensitivity. Further questions are whether there is a principled extension when a type’s mass can split between ballots, and what happens when the carrier or its rule is given succinctly rather than by an explicit table.
The authors could recognize this as a high-multiplicity version of their finite-carrier winner problem: the same rule is evaluated from the same decisive pattern, while large repeated blocs are represented by mass. The weakest point is also clear: the masses largely describe the society without affecting who wins, and treating a cohort as a bloc requires genuine within-cohort agreement. If the rule distinguishes ordinary voters individually, or if a bloc can split, this mirror stops preserving the paper’s problem. It is therefore a defensible case for one structural result, not a case that the paper’s central nonanonymity results naturally continuize.
Proposition 4 is the right anchor for the case for a mirror: it is a numbered algorithmic result, and it gives a positive computability claim. The obstacle is that its computation is about named players, not population size. Given a characteristic index for a recursive coalition, the proof checks whether each of finitely many specified players is in that coalition, then consults a finite table. The paper gives no running-time bound, but more fundamentally, the inputs and the information the algorithm uses do not match ChoCo’s society \(\mu\), a distribution over voter types.
The bloc scenario is the strongest repair. Large, cohesive cohorts can be a sensible setting, and a rule over a few decisive blocs could be useful. But there is a choice to make. If a bloc stands for a role shared by many voters, its mass and shared ballot describe a new bloc game: the original proposition makes particular player identities decisive, while the bloc game makes a whole role decisive. If those particular players remain decisive, their identities must be recorded separately. In a continuum population they have zero mass, so \(\mu\) alone cannot say whether they belong to a coalition. Giving each one a positive-mass type changes the model again; including their membership bits as extra input restores the paper’s finite identity checks, with the continuous population beside them.
That leaves the proponent’s explicit-table problem as a coherent way to evaluate a bloc rule, but it does not mirror Proposition 4’s \(\delta\)-computability claim. Its input is a ballot pattern for finitely many roles, not an index describing an arbitrary recursive coalition. The masses report the size of the blocs but supply none of the information used by the computation. A richer mass-transfer problem might make those masses consequential, but it would need new costs and a new winning rule; it would be a separately motivated problem, not a continuous version of the result in this paper.
So the negative case is strongest against this anchor, rather than against every possible application of bloc voting. The residual weakness is that a bloc model might still be valuable in its own right. What the paper establishes does not, by itself, make that model a worthwhile continuous mirror.
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.