Coordination Without Capture
A plan, compiled in dialog with Claude.ai, examining why person-based political representation reproduces the dysfunction it is meant to prevent, and what voluntary, objective-based coordination might offer instead.
Other Information
Mission, Explained -- Rational Ignorance: Anthony Downs's "An Economic Theory of Democracy" (1957) argued that it is rational for individual voters to remain uninformed, because the expected payoff from acquiring costly information is vanishingly small when one vote among millions is unlikely to be decisive.
The Krasa-Polborn finding discussed in the dialogue underlying this plan -- that voters indulge expressive, tribal impulses once the probability of policy implementation is low -- is a close cousin of this mechanism: both describe citizens rationally disengaging from substantive evaluation once their individual stake in the outcome is diluted past the point of mattering.
This plan's Mission targets that mechanism directly: a coordination model in which individual affiliation visibly and traceably contributes to a specific, falsifiable result restores some of the stake that winner-take-all, jurisdiction-wide voting dilutes away.
Values, Grounded: Each Value below generalizes a thread already argued under the Goals that follow.
* Accountability Independent of Virtue draws on Lord Acton's 1887 dictum that power tends to corrupt and absolute power corrupts absolutely (Goal 2, Objective 2.3), on Robert Higgs's account of the "ratchet effect," whereby capacity built under one set of justifications persists past the moment that justified it, and on Madison's own design intent in Federalist 51, which sought workable government even with imperfect men rather than presupposing virtue (Goal 1).
* Epistemic Humility About Authority generalizes the Hayek, Sandel, La Boetie, and Michels critiques of rule by the qualified, including the further Hayek and Scott material on the limits of even well-intentioned power (Goal 2, Objectives 2.2 and 2.4).
* Falsifiability and Transparency points to Popper's falsifiability criterion as the philosophical ancestor of StratML's insistence on machine-readable, auditable objectives in place of an aggregated "will" that Arrow and Condorcet suggest may not coherently exist (Goal 3), and is the same standard that distinguishes a plan from a protest (Goal 4, Objective 4.3).
* Informed Consent generalizes the Downs rational-ignorance mechanism described under the Mission above: affiliation is only meaningful consent to the extent the people giving it have enough information to understand what they are agreeing to.
* Freedom of Association generalizes Hirschman's voice-and-exit framework and Buchanan and Tullock's public-choice critique of the state's monopoly character (Goal 4). It is also a constitutionally grounded principle in its own right: NAACP v. Alabama, 357 U.S. 449 (1958), held that the right to associate freely, including anonymously, is protected by the First and Fourteenth Amendments, establishing that people cannot be coerced into disclosing or forgoing their voluntary affiliations -- the same voluntary-extension-and-withdrawal logic that French and Raven's personal bases of power depend on (Goal 4, Objective 4.5).
* Subsidiarity generalizes Ostrom and Tiebout's polycentric and competing-jurisdictions models (Goal 4).
Submitter's Note: This plan was compiled and rendered in StratML format in dialog with Claude.ai focusing on William Howell's June 25, 2026, article entitled "Democracy’s Defenses, Democracy’s Undoing," which is available at https://democracyproject.org/posts/democracys-defenses-democracys-undoing
While my primary intention is to sharpen my own understanding of the underlying issues, I hope that others might benefit as well.
🏢No One in Particular NOIP No One in Particular (NOIP) is a deliberately vacant identity used across this corpus for hypothetical plans open to anyone who shares the stated mission and goals, without requiring a chartered organization to claim ownership of them.
Stakeholders:
👥 | Source & SubmitterSource:https://stratml.us/docs/CWOC.xml Period: to Published: |
Vision
Coordination occurring primarily through voluntary affiliation with shared, falsifiable objectives rather than through capture of jurisdiction-wide office -- so that no one in particular need hold disproportionate authority for public problems to get solved.
Mission
To reduce the need for rational ignorance of the potential for realizing better means of achieving common objectives
Values
Goal 1: The Dilemma 🔗
Diagnose the internal inconsistency in arguments that propose restoring governmental capacity as the remedy for strongman politics.
Other Information
Name the Dilemma ~ Howell's essay is the entry point for this plan. It observes that the constitutional machinery built to frustrate a would-be autocrat is the same machinery that frustrates competent, accountable governance, and it proposes restoring "capacity" as the way out. This goal interrogates that proposed exit, asking whether capacity, once built, can be trusted to remain on democracy's side once it changes hands.
Objective 1.1: Institutional Capacity 🔗
Show that institutional capacity built to help democracy govern transfers intact to whoever next captures the government, including a future autocrat.
Other Information
Capacity's Double Edge ~ In "Democracy's Defenses, Democracy's Undoing," William Howell argues that the same constitutional machinery that frustrates an autocrat also frustrates ordinary governance, and proposes restoring governing capacity as the remedy. Howell's Dilemma ~ The dialogue's central objection: he states the trade-off explicitly -- strengthen capacity and you arm abusers, weaken it and you feed disillusionment -- but then resolves it by simply choosing the strengthen-capacity horn, without showing why that horn does not recreate the danger he is diagnosing. Capacity, once built, does not stay loyal to the purposes it was built for; it is capture-agnostic and is inherited by the next occupant of the office, strongman or not. Robert Higgs's account of the "ratchet effect" sharpens the timing problem: capacity built under favorable conditions or virtuous intent rarely contracts once the moment that justified it has passed, so what is built to rescue a democracy from one era's dysfunction tends to outlive that era and await its next occupant, good or bad.
Objective 1.2: Legislation & Execution 🔗
Distinguish congressional capacity to check power from executive delivery capacity, since conflating the two obscures exactly where the dilemma bites.
Other Information
Two Capacities ~ Distinguishing Legislative Capacity from Executive Capacity: Strengthening Congress's ability to legislate and conduct oversight is anti-strongman almost by definition, since it checks unilateral executive action. Strengthening agency and executive-branch delivery capacity is capture-agnostic: whoever wins next inherits it. Howell's essay blends both under one elastic phrase -- governing institutions capable of solving public problems -- and the blending makes the dilemma look resolved when it has only been relabeled. The line is not always crisp in practice -- control over personnel, such as reclassifying career civil servants into at-will status, lets executive capacity bleed into the kind of structural power that determines who answers to whom, which is why proposals to expand executive "capacity" deserve scrutiny over which lever they actually pull.
Goal 2: The Fix 🔗
Examine the assumption that recruiting more capable people into more powerful offices is itself a structural solution rather than a restatement of an old premise.
Other Information
Question the Elite Fix ~ If capacity alone cannot resolve the dilemma, the next candidate is the people who wield it -- recruit more capable, less extreme officeholders into more authoritative offices, and democratic governance is expected to improve. This goal traces that proposal back to its eighteenth-century ancestor and tests it against independent traditions that doubt whether "better people" is ever the right lever to pull -- whether because the right people cannot reliably be identified in advance, or because even the right people may not stay right once empowered.
Objective 2.1: Elections 🔗
Recognize that attracting better people to more authoritative offices echoes Federalist 57's claim that elections extract the most fit characters, rather than offering a genuinely new paradigm.
Other Information
Madison Restated: Federalist 57 ~ Madison and Hamilton on Extracting Fit Characters: Howell's own empirical finding -- that offices with meaningful authority differentially attract qualified, moderate candidates while extremists remain indifferent -- restates rather than escapes the founding premise that elections filter for a natural aristocracy of merit. An essay that spends its length blaming Madison's machinery for dysfunction ends by offering a more faithful version of Madison's founding bet, not a less faithful one. Michels's "iron law of oligarchy" generalizes the point beyond Madison: any sufficiently organized system, even an avowedly democratic or voluntary one, tends over time to concentrate effective control in a small leadership stratum. If that law holds, swapping in more capable leaders changes who occupies the stratum without addressing why a stratum recurrently forms -- the question Goal 4 returns to.
Objective 2.2: Knowledge & Meritocracy 🔗
Apply the dispersed-knowledge and meritocracy critiques to test whether better people in more powerful offices can substitute for distributed information and self-interested selection criteria.
Other Information
Knowledge and Merit Limits: Hayek, Sandel, La Boetie ~ Three Critiques of Rule by the Qualified: Hayek's knowledge problem holds that the information needed to govern millions of dispersed lives and preferences cannot be centralized in any sufficiently qualified head, however measured. Sandel's "tyranny of merit" observes that meritocratic selection criteria tend to flatter the selectors -- cognitive ability and managerial experience are exactly the traits an academic selecting on those traits would value in himself. La Boetie's older and blunter claim, from his treatise on voluntary servitude, is that the subjects' belief in a ruler's fitness is the mechanism that sustains rule, not a check on it. None of the three deny that qualified people might govern competently; all three deny that competence is the bottleneck, or that authority concentrated in qualified hands is itself progress. Taken together, the three critiques attack the elite fix at three different layers -- informational, sociological, and psychological -- which is why no single counter-argument, such as pointing to a particular candidate's genuine competence, answers all three at once.
Objective 2.3: Power 🔗
Show that concentrated power tends to corrode the judgment and character of whoever holds it, regardless of how carefully or virtuously they were selected, which challenges the elite fix at a stage neither the selection nor the knowledge critiques reach.
Other Information
Acton's Dictum/Power Corrupts ~ Lord Acton on Power and Corruption: In an 1887 letter to Bishop Mandell Creighton, Acton offered the line that became the standard formulation of this problem -- power tends to corrupt, and power without limit corrupts without limit. Unlike the selection-stage critiques in Objective 2.2, Acton's claim operates after selection has already gone well: even a genuinely well-informed, well-intentioned officeholder is not guaranteed to remain that way once granted concentrated and durable authority. Hayek, cited under Objective 2.2, was directly influenced by Acton on exactly this point. David Kipnis's 1972 experiments gave the claim empirical legs: participants assigned institutional power over others used it, came to credit their own authority rather than the merits of their requests for others' compliance, and rated the people subject to that power less favorably -- a self-reinforcing pattern rather than a one-time lapse. A later replication attempt found a much weaker effect, so the finding should be read as suggestive rather than settled. A genuine counterpoint belongs here. Biographer Robert Caro, after decades studying Robert Moses and Lyndon Johnson, has argued that power less often corrupts a person than reveals them: it removes the practical constraints that had been concealing what someone was already capable of, for better or worse. Either reading damages the elite-fix proposal examined in this Goal. If power corrupts, the "better people" recruited into office will not reliably stay better. If power instead reveals, there was no way to know in advance what the office would expose. Both readings converge on the Value of Accountability Independent of Virtue, which this plan treats as a requirement for any safeguard rather than a bonus.
Objective 2.4: Intentions 🔗
Show that even officeholders whose character has not been corrupted, and who sincerely intend a good outcome, cannot fully foresee a concentrated decision's consequences for every stakeholder it touches, so good intentions cannot substitute for limits on the scope of any single decision-maker's power.
Other Information
Good Intentions Are Not Enough ~ The Limits of Benevolent Power: A problem closely related to, but distinct from, Objective 2.3: even a powerholder whose character has not been corrupted, and who sincerely intends a good outcome, cannot fully model a complex decision's consequences for every stakeholder it touches. The failure here is epistemic, not moral. Hayek made this the subject of his last book, "The Fatal Conceit" (1988): the belief that any mind, however well-intentioned, can shape a society more complex than that mind can model is itself the error, independent of the planner's motives. This generalizes the knowledge problem already cited under Objective 2.2 from a critique of selecting rulers to a critique of trusting even well-selected ones to act on society's behalf. James C. Scott's "Seeing Like a State" (1998) supplies the historical record: forced agricultural collectivization, scientific forestry, and large-scale resettlement schemes were repeatedly designed by genuinely well-intentioned planners and repeatedly produced large-scale, unanticipated harm, because none of the planners could see the local, tacit knowledge their schemes overrode. The conclusion is not that power-holders should try harder to anticipate consequences, but that the scope of any single decision-maker's power should be limited regardless of their intentions or their information -- an argument for Subsidiarity, taken up under Goal 4, as much as for the accountability mechanisms argued under Goal 1.
Goal 3: Representation 🔗
Establish, formally and practically, that no single representative can losslessly aggregate the heterogeneous preferences of a large constituency.
Other Information
Confront Representation's Limits ~ Even granting capable, well-intentioned officeholders, a deeper problem remains -- a single representative is being asked to do something that may not be formally possible: stand for the coherent, aggregated preference of a large and heterogeneous constituency. This goal examines that impossibility in the abstract, through social-choice theory, and in the concrete mechanics of how safe districts and primaries select for the kind of representative worth disputing in the first place.
Objective 3.1: Majoritarianism 🔗
Apply Condorcet's cycling problem and Arrow's impossibility theorem to show that majority will may have no coherent, representable form once there are three or more options.
Other Information
Formal Impossibility ~ The Formal Limits of Aggregating Preferences: • Condorcet showed majority preferences among three or more options can cycle, so there may be no coherent collective preference to represent at all. • Arrow's impossibility theorem generalized this: no method of converting individual rankings into a collective ranking can satisfy a small set of minimal fairness conditions simultaneously once there are three or more options and more than two voters. • Rousseau drew the sharpest conclusion from a related premise: sovereignty, the general will, cannot be represented at all, because representation collapses many wills into one voice -- a category error, not a simplification. • Pitkin's later taxonomy of representation (formal, descriptive, symbolic, substantive) shows where one elected person can succeed, namely formal authorization, and where they structurally cannot, namely substantive acting-for across a multidimensional and often internally inconsistent constituency. None of this indicts any one voting method as uniquely flawed; it shows that no method escapes the trade-offs entirely, which is why reforms such as ranked-choice or proportional voting can improve the mapping from preferences to outcomes at the margins without ever achieving lossless aggregation in principle.
Objective 3.2: Primaries, Polarization & Extremism 🔗
Show how safe congressional districts convert the primary into the operative election, handing effective selection power to a smaller and more ideologically extreme electorate.
Other Information
Safe Seats, Captured Primaries ~ The Politics Industry and Primary Capture, With a Caveat: • Gehl and Porter's industry-structure framing treats closed partisan primaries plus safe seats as a barrier to entry that rewards intra-party extremity over general-electorate appeal, since the only competition that matters happens inside the more ideologically homogeneous half of the electorate. • Fairness requires noting the countervailing research: Hall's work on primary electorates finds primary voters are sometimes less unrepresentative of their party's general-election voters than the popular story assumes, with candidates' own strategic positioning doing more causal work than turnout composition alone. The structural barrier-to-entry point stands regardless of which mechanism dominates. The 2019 Supreme Court ruling in Rucho v. Common Cause, holding that partisan-gerrymandering claims present non-justiciable political questions, removed the federal courts as a check on this dynamic, leaving redistricting reform mainly to state-level initiatives and ballot measures.
Goal 4: The Paradigm 🔗
Replace person-based representation, which asks one official to aggregate an unrepresentable totality of preferences, with voluntary coordination around discrete, falsifiable, machine-readable objectives that people may join or leave without a proxy.
Other Information
Shift the Paradigm ~ If person-based representation cannot be fixed by better capacity, better people, or better aggregation, the remaining move is to rely on it less. This goal assembles the case, from public-choice theory through StratML's own design, for voluntary coordination around discrete and falsifiable objectives as a structural alternative to asking any single official to represent the unrepresentable. Two further objectives stand guard against this goal proving too convenient: one tests whether voice without a plan accomplishes the coordinating work claimed for it, and the other tests whether anything has actually changed since the standard defense of representative democracy was last made in full confidence. A final objective asks what kind of power the paradigm itself can legitimately claim, given everything Goal 2 found wrong with the alternative.
Objective 4.1: Voice, Exit & Polycentricity 🔗
Draw on Hirschman, Ostrom, Tiebout, and Buchanan and Tullock to outline how voluntary, exit-friendly, jurisdictionally plural arrangements reduce the stakes of any single political contest.
Other Information
Hirschman, Ostrom, Tiebout, Buchanan and Tullock ~ Voice, Exit, and Polycentric Governance: • Hirschman's voice and exit distinction reframes the goal as maximizing exit options so that voice -- politics, deliberation, voting -- becomes lower-stakes, rather than trying to make voice itself more accurate. • Ostrom's polycentric governance and Tiebout's competing-jurisdictions model show that coordination can occur through overlapping, voluntary, locally accountable units rather than one coercive monopoly. • Buchanan and Tullock's public-choice critique treats the state's monopoly character, not any officeholder, as the design flaw generating rent-seeking and faction, implying the pathology is tamed by dissolving the monopoly rather than rotating who staffs it. Historical caution, drawn from cases such as the Hanseatic League and Swiss cantonal subsidiarity: voluntary federation works best where externalities are small and exit is cheap, and tends to fail where externalities are large and exit is expensive -- common defense, currency, cross-boundary pollution, and free-rider enforcement remain the hard cases for this whole tradition. Ostrom's own fieldwork across irrigation systems, fisheries, and forests is the empirical backbone of this value: communities repeatedly proved able to devise durable, self-enforcing rules for shared resources without either privatizing them or handing them to a central state, direct evidence against the assumption that property rights and state coercion are the only two solutions to collective-action problems.
Objective 4.2: Objectives 🔗
Apply the Communities of Results model as concrete infrastructure for letting people coordinate around shared, auditable objectives directly, without compressing their preferences through a single elected proxy.
Other Information
StratML, Communities of Results ~ Coordination Infrastructure as an Alternative to Proxy Representation: Where person-based representation must compress a constituency's multidimensional preferences into one vote in one chamber, an objective-based model lets people discover and voluntarily affiliate with specific, machine-readable, falsifiable goals, opting in or out without needing a proxy to get their whole political identity right. This does not resolve the hard cases that historically defeated voluntary federation -- defense, currency, large externalities -- but it shrinks the domain in which any single official's fitness to represent a heterogeneous public is even the relevant question, answering the dialogue's objection to merit-based fixes by making the office matter less rather than asking it to be filled by better people. This is the same wager underlying StratML's broader "Communities of Results" concept and its advocacy for GPRAMA Section 10 compliance: that making objectives discoverable and machine-readable lets coalitions form around shared results without first having to capture an office or agency that claims to speak for everyone within its jurisdiction. See also https://connectedcommunity.net/
Objective 4.3: Plans 🔗
Distinguish voice that takes the form of unstructured protest or symbolic dissent from voice that takes the form of a discrete, falsifiable plan, and show that only the latter does the coordinating work this Goal requires.
Other Information
Plans, Not Protests/Protest as Signal, Not Substance ~ Tufekci on Networked Protest: Zeynep Tufekci's "Twitter and Tear Gas" (2017) distinguishes a movement's capacity to mobilize large numbers quickly from its capacity to negotiate, adjust tactics, and convert momentum into a concrete outcome -- what she calls "tactical freeze" when the second capacity is missing. Networked protests, from Gezi Park to Occupy, scaled fast precisely because they skipped the years of organizational groundwork older movements needed, and several decayed for the same reason once the initial surge passed. Tufekci contrasts this with the 1963 March on Washington, by her account conceived roughly a decade earlier: the march itself was the exclamation point at the end of a long organizing sentence, not the sentence's first letter. The protest was downstream of a plan, not a substitute for one. This sharpens Hirschman's voice under Objective 4.1: voice is not self-executing. Unstructured dissent signals that something is wrong without specifying a falsifiable result, a decision-making mechanism, or who is accountable for follow-through -- exactly the gap that StratML's Communities of Results, examined under Objective 4.2, is built to close. A protest expresses what people are against; a plan is what lets them discover who agrees on what they are for, and hold the resulting coalition to it. None of this devalues protest as a signal -- Hirschman's framework still needs voice to register discontent at all. The claim is narrower: a signal is not a coordination mechanism, and the Krasa-Polborn expressive-voting pattern described under the Mission above is the same failure mode in electoral form.
Objective 4.4: Technical Alternatives 🔗
Address the standard defense of representative democracy -- that it beats every alternative ever tried -- by showing that the technological conditions for a genuinely untried alternative did not exist when that defense was first made.
Other Information
Untried Alternative/Churchill's Borrowed Aphorism ~ The Limits of an Empirical Defense: On 11 November 1947, in a House of Commons debate over the Parliament Bill, Churchill cited what he called an existing saying: democracy is the worst form of government except all the other forms that have been tried from time to time. He was quoting, not coining, the line, which had circulated anonymously at least as early as 1946; its durability rests on its candor rather than its originality. The defense is genuinely strong, and this plan does not dispute it as a claim about the historical record up to that point. But it is explicitly an empirical claim about forms already tried, not a logical proof that no untried form could do better. Churchill himself, earlier in the same speech, objected to the idea of "super-planners" remaking society on the people's behalf without their check or correction -- a caution about centralized control that this plan's own Goal 2 shares. What did not exist in 1947, or for that matter at any point before the spread of networked computing and now AI-mediated interfaces, was the technical means to make millions of voluntary, falsifiable, granular commitments machine-readable, discoverable, and auditable at population scale, without funneling them through either a centralized bureaucracy or an elected proxy. The StratML model examined under Objective 4.2, including the conversational interview that produced this very document, is one candidate for what that newly available means makes possible. This is a claim about feasibility, not a guarantee of success. The plan's own opening section already concedes that prospects for implementation may be remote. The point is narrower: the historical menu Churchill was comparing democracy against did not include this option, so its absence from that comparison is not evidence against it.
Objective 4.5: Followership 🔗
Apply French and Raven's taxonomy of social power to show that the kind of power this plan's paradigm relies on -- power voluntarily extended by the people subject to it -- is structurally different from, and less prone to higher-order illegitimacy than, the positional power critiqued under Goal 2.
Other Information
French and Raven, Reconsidered ~ Positional Power Versus Personal Power: French and Raven's classic taxonomy distinguishes six, sometimes seven, bases of social power: reward, coercive, legitimate, referent, expert, informational, and, as later commentators added, connectional. Of these, legitimate, reward, and coercive power are positional: they attach to an office or rank and would transfer to anyone who filled it. Referent, expert, and informational power are personal: they attach to an individual and travel with that person regardless of office. The higher-order legitimacy problem: positional power is self-justifying rather than earned. As one commentator on French and Raven observed, actual influence is granted by whoever is being influenced, not claimed by whoever holds the more senior title -- once a follower withdraws that grant, the position keeps its formal trappings but loses its working force, which makes leadership, on this account, a downstream effect of followership rather than its cause. A position bureaucratically equipped with legitimate, reward, and coercive power but lacking genuine referent or expert power is a hollow kind of authority, sustained mainly by inertia and the cost of defection. This reframes the elite-fix problem examined under Goal 2 from a different angle: the "better people" that fix recruits would still inherit an office's positional power -- legitimate, reward, and coercive -- largely intact, regardless of whether their personal, referent, or expert power justifies it. Acton's and Kipnis's findings under Objective 2.3 describe what concentrated positional power does to a holder over time; this objective adds why that power was disproportionate to begin with. The information age sharpens the problem and points toward this plan's own remedy. As information and expertise become more freely and widely available, the informational power that once accrued almost exclusively to hierarchical position is harder to monopolize, and personal, referent, and expert power -- the kinds that must be voluntarily extended by the people subject to them -- become comparatively more durable. A coordination model built on voluntary affiliation with falsifiable objectives, examined under Objective 4.2, is one structural expression of that shift: authority within it can only be the personal kind, since there is no office to capture and no positional power to inherit.