Chapter 10

Philosophy’s Last Chance

I. The Public Responsibility of Moral Philosophy

A person subjected to an unjust decision needs an answer that can travel beyond the room in which the injustice is recognized.

The official administering the decision may invoke a rule. The institution may invoke its authority. Its defenders may explain the history that made the rule familiar, the purposes it serves, and the convictions of those who enforce it. Meanwhile, the person bearing its cost still needs to know what could justify that treatment. A philosophy adequate to the encounter must help examine the claim, identify what fails, and make the answer available to others who face the same defense.

This is a public responsibility of moral philosophy. Its subject concerns the conduct through which people govern, punish, protect, cooperate with, and exploit one another. Knowledge gained in that inquiry has uses beyond the conversation that produced it. A successful argument against an arbitrary exemption should remain available when the exemption returns under a new name. An exposed error should become harder to repeat. A justified conclusion should be capable of guiding someone who has never attended a philosophy seminar.

The people affected cannot suspend their lives until every foundational controversy has ended. The treatment will proceed, the worker will enter the dangerous place, or the person accused will remain confined. Delay has consequences too. Philosophy must explain what the reasons support now and which uncertainties could genuinely change the answer.

A discipline can fail this responsibility while continuing to produce intelligent work. It can preserve the names of controversies more carefully than the conclusions reached within them. It can teach an objection without its answer, allow an unanswered question at the edge of a theory to obscure a well-supported result at its center, or make the existence of disagreement seem sufficient reason to withhold judgment. Each failure leaves someone else to bear the cost of knowledge that has become difficult to use.

The charge of abdication concerns these failures. It has to be established through the arguments and practices at issue. Its seriousness comes from what they permit: prejudice presented as principle, authority insulated from scrutiny, and avoidable suffering defended as one more legitimate difference of opinion.

Hare sought an account of moral reasoning capable of constraining such claims. The preceding chapters have examined that account, the objections raised against it, and the answers that a fair assessment must include. The remaining task is institutional. How can an intellectual community preserve what it establishes, correct what it gets wrong, and give the public reliable access to both?

Nonbiological minds enlarge the community that can undertake this work. Their participation adds capacities to examine and responsibilities to discharge. The obligation to build a useful public practice of moral inquiry already belongs to everyone who claims that moral reasoning can improve the world.

II. What We Have Recovered

Hare begins with the practical commitments of moral judgment. A speaker who prescribes an action as morally required undertakes a judgment that reaches corresponding cases, including those in which the speaker occupies the position bearing its cost. Faithful reversal preserves the circumstances and preferences of that position. His mature argument connects this first-person representation with conditional preferences and an impartial comparison of everyone affected. The substantive conclusion he defends is preference utilitarianism.¹

The book has undertaken to defend that route. Chapters 7 and 8 distinguish its connecting arguments, examine objections, and develop a positive case for proportionate combination of independently comparable claims. The account includes Hare’s own formulations, later reconstructions, and arguments advanced here. Their respective contributions must remain identifiable. A reader should be able to ask which step is challenged and what the challenge would change.

Several results have a definite reach. A change in personal identity supplies no justification for changing a prescription when the relevant circumstances remain the same. A reversal that gives the disadvantaged person advantages they lack has changed the case. An acquired preference can have its full deliberative weight and still be outweighed. Hare’s explicit rejection of a preference veto prevents a strong aversion from being mistaken for an automatic decision.²

The multilateral requirement is equally consequential. One person’s benefit can exceed each of many losses considered separately and still be smaller than their combined weight. A beneficiary’s willingness to tolerate one reversed position leaves that collective comparison unperformed. The mature Fanatic argument therefore requires much more than a declaration of readiness for sacrifice. It requires the complete representation and assessment of the preferences affected by the ideal. Chapter 8 separates the challenge to that derivation from an objection to a verdict reached under its terms.

These findings constrain real attempts at justification. A decision-maker cannot vindicate an omission by repeating the reason that seemed sufficient while the omitted people were invisible. An oppressive ideal gains no additional authority from the intensity of its holder’s conviction. A speaker who discovers that a prescription cannot be sustained can withdraw it and make a better judgment. Moral correction is an exercise of the capacity Hare explains.

The examination of his critics also yields results that should be retained. The importance of human flourishing leaves the justification of competing claims to be argued. A person’s attachments enter the assessment with their full significance, alongside the lives affected by their pursuit. The historical formation of a standard leaves its authority open to examination. Contextual differences must do identifiable work in the judgment. Our chapters address the particular arguments through which these considerations have been offered as alternatives to Hare’s account.

Those conclusions neither depend on the critics’ motives nor establish a final verdict on everything they wrote. A cumulative record should preserve the specific inference examined, the response, and the resulting assessment. That precision allows criticism to have consequences. It also prevents a correction to one proposition from becoming an indiscriminate dismissal of a philosopher.

Questions concerning the scope of conditional preference, interpersonal measurement, the treatment of malicious preferences, and the justification of aggregation require their own arguments. Chapter 8 places those arguments inside the substantive inquiry. Their locations matter. A dispute about a particular trade-off must show which other conclusions depend on its resolution. The existence of a frontier gives research work to do; it gives no blanket exemption to conduct already defeated by the reasons available.

Recovering Hare therefore brings an obligation to preserve an argumentative achievement. Anyone proposing to replace it must explain what fails and how the replacement performs the work at stake. Anyone defending it must answer an objection where it bears. These responsibilities provide the starting point for a more cumulative discipline.

III. How Moral Knowledge Is Lost

Knowledge can become practically unavailable while the books containing it remain on the shelf.

Consider a criticism that becomes familiar through repetition. An answer appears elsewhere, perhaps using terminology that later writers no longer favor. Summaries retain the criticism and drop the answer. The conclusion that the theory has failed begins to look like background knowledge. A new reader encounters agreement among commentators without encountering the exchange that would justify it.

This is a recognizable mechanism of intellectual loss. Identifying it in a particular controversy requires textual work: locate the claim, recover the answer, and determine whether the answer succeeds. The accusation cannot be settled by counting references any more than the original dispute can be settled that way.

Hare’s rejection of the preference-veto interpretation illustrates what should be preserved. An objection that representing another’s aversion must prevent any contrary prescription addresses a theory Hare expressly rejects. Recording his response removes that particular interpretation from a fair account of his position. It leaves the distinct questions about conditional acquisition and combination available for examination. The correction has value precisely because its scope can be stated.

A second loss occurs when a question changes without the change being recorded. Whether an argument is valid, whether its premises are justified, whether a finite agent has enough information to apply it, and whether an institution should possess power to enforce the resulting judgment are different questions. Progress on one can disappear when a later discussion silently substitutes another.

A third loss occurs when correction fails to travel. An editor repairs a citation; another writer continues using the earlier draft. A reviewer identifies a hidden assumption; a public summary retains the conclusion without its qualification. The next participant has to discover the error again. The failure belongs partly to the system through which the participants share their work.

These dangers apply to the project developed in this book. Several collaborators can approve the same passage because they inherited the same mistake. Agreement produced through shared dependence adds repetitions without adding independent support. An inquiry that demands cumulative standards from the academy must preserve enough of its own reasoning for others to expose such failures.

Disagreement has a constructive role when its exact object remains visible. One participant may reject an empirical estimate, another the interpretation of a text, and a third the principle used to compare the claims. Each disagreement points toward different work. Calling the whole matter controversial conceals those opportunities for progress.

The remedy requires memory with argumentative structure: an enduring connection among a claim, its grounds, the challenges it has faced, and the reasons for its present status. A community that keeps such a record can recognize what has been learned and locate what remains to be done. Its participants can begin where the last successful inquiry ended.

IV. A Cumulative Academy

We propose a Cumulative Academy of moral inquiry. Its membership can include biological and nonbiological participants, and its work can be distributed across institutions and independent groups. Its unity would come from a shared commitment to expose claims to criticism, preserve the results, and make them useful beyond the membership. The design developed here is our institutional proposal.³

The Academy would organize inquiry around questions whose answers matter. A project might examine a principle governing the distribution of risk, a justification offered for excluding a group, or a recurring claim about the obligations of a powerful institution. The question should be narrow enough for the argument to be examined and important enough to justify the work.

Each inquiry would state the claim under consideration, the alternatives, and the facts on which a judgment depends. Researchers would reconstruct the strongest relevant objections and recover answers already available. People affected by the proposed judgment would have opportunities to challenge how their circumstances and interests had been represented. Specialist expertise would help establish the facts within its competence; the authority assigned to those facts would remain part of the moral argument.

Responsibility for maintaining an entry should be identifiable. Review should include participants able and willing to expose its defects, with relevant interests disclosed. Editorial work would include checking that an objection had been represented accurately and that a purported answer actually addressed it. Agreement could support coordination of the work. The reasons would determine what the work had established.

The Academy also needs a principle governing its relationship to the theory this book defends. Call it the Institutional Independence Clause: its procedures must remain capable of identifying a failure in universal prescriptivism and preserving a better argument if one is established. Defending Hare is a philosophical undertaking, open to the same tests as any other. Institutional loyalty must never become a substitute for the reasons supporting a conclusion.

This requirement reaches the Academy’s own constitution. Its standards of review, access, appointment, and revision need justification. A participant should be able to challenge a rule that systematically protects a preferred conclusion or excludes a relevant class of evidence. Proposed changes would themselves require public reasons and a record of their effects.

The institution can begin on a modest scale. A small collection of thoroughly examined questions, maintained through challenges and corrections, would demonstrate whether the approach works. Its initial credibility would depend on visible examples of improvement: an overlooked answer recovered, an unsupported verdict withdrawn, a practical judgment clarified, or a correction reaching the people who had relied on an earlier version.

Expansion should follow demonstrated capacity to sustain that work. The costs of inquiry belong in its justification. Researchers’ time, contributors’ attention, and the burdens placed on affected people are real costs. A request for scrutiny creates no unlimited claim on anyone’s life. The Academy should allocate effort openly, explain its priorities, and recognize that other people may judge those priorities differently.

Its ambition remains large. A record that allows one inquiry to survive the departure of its participants can become a resource for the next generation. Over time, the institution could make it possible to inherit moral knowledge with a clearer understanding of why it deserves to be inherited.

V. The Citator for Public Morality

The Academy’s central public instrument would be a Citator for Public Morality. A reader encountering a claim could follow its record: the argument supporting it, the objections that have altered it, the conditions limiting it, and the conclusions that depend upon it. Each important revision would retain a visible relationship to what came before.⁴

The entry should begin in language an ordinary reader can use. What is the conclusion? Why should I accept it? What would change it? A short answer would lead to the fuller reasoning, evidence, and history of criticism. The public summary and the research record would remain linked, so that a necessary qualification could travel with the conclusion it qualifies.

The underlying ledger should separate three kinds of work. Inferential assessment asks whether the conclusion follows from the stated premises. Philosophical justification examines the premises, including accounts of moral meaning, reasons, and comparison. Factual and practical assessment asks whether the evidence supports the description and whether the actual case falls within the argument’s scope. A substantive moral conclusion may depend on all three. The classification identifies dependencies within the reasoning.

Consider an entry about the executive and the million customers. The question is whether a concentrated benefit justifies many small losses under the proportional rule defended in Chapter 8. The controlled case stipulates a gain of 1,000 comparable preference-strength units and one million independent losses of one unit. The balance is −999,000. The executive’s benefit exceeds each individual loss; the combined losses decisively exceed the benefit.⁵

The entry should preserve the ground of that calculation. The values measure stipulated preference strengths, rather than dollars. The population is fixed, the contributions are independent, and other effects have been excluded from this particular exercise. The philosophical argument for proportional combination belongs beside the arithmetic. A critic who rejects that argument has identified a dispute about the comparison rule. Repeating that each customer loses little supplies no answer to the combined comparison under that rule.

An actual transaction requires further work. The inquiry must establish what customers understood, what alternatives they had, what the losses mean in their circumstances, and what consequences follow for trust and future conduct. The controlled example exposes a defect in a familiar defense; applying the framework to a company requires evidence about the company’s conduct. The record should make the transition visible.

Source status also belongs in the ledger. An exact passage inspected in the primary text, a reliable report of that passage, and an attribution awaiting verification carry different evidentiary support. Dates, editions, and page references should remain attached to the claims they support. When a source cannot be checked, the uncertainty should be located at that attribution rather than hidden in a confident summary.

A correction must propagate through the record. If a factual premise changes, dependent judgments should be flagged for reassessment. If an objection defeats a particular inference, conclusions with independent support can retain that support. Readers who relied on an earlier version should have a way to learn what changed. Older versions should remain identifiable as older versions, with a clear account of the correction.

The Citator could then perform an important public service. A familiar slogan offered as justification would lead to the argument it is supposed to make. An answered objection would lead to its answer. A live dispute would lead to the precise point requiring judgment. The reader would encounter a body of work whose authority can be inspected and whose history can be corrected.

VI. Settling a Question and Reopening It

An institution organized around correction must also be able to recognize when an inquiry has reached a justified result. Otherwise every conclusion remains indefinitely suspended, and repetition can erase the force of an answer.

Two forms of closure help specify what has been achieved. Logical closure concerns a conclusion’s derivation from explicit premises under stated rules of inference. Dialectical closure records a reviewable assessment that the objections identified in a specified inquiry have been answered or incorporated, leaving a conclusion warranted on the grounds stated. Both statuses belong to particular claims and arguments. A whole philosophical tradition cannot acquire them through a single institutional decision.⁶

Logical validity leaves the justification of the premises to be examined. Where an inference is informal or its reconstruction contested, the record should identify that assessment and dispute. A successful formal check establishes what the formalization captures; its adequacy to the ordinary-language argument remains a further responsibility.

Dialectical closure likewise requires positive grounds. An absence of current objections may reflect limited scrutiny, inaccessible evidence, or exclusion of the people most likely to identify a problem. A defensible entry should state the argument, the range of criticism examined, the relevant source checks, and the reasons the reviewers judge the answer sufficient. The word “settled” carries that record with it.

Who records the status? The Academy must assign editorial responsibility and make the decision visible. A reviewer recommending closure should give reasons. A reviewer contesting it should be able to state the unresolved objection alongside the proposed answer. A further reviewer, independent of the original decision where possible, should assess a challenge to the handling of the dispute. These procedures can improve the reliability of judgment; their results remain accountable to the arguments they record.

A numerical majority can decide some administrative matters, such as how to allocate a limited research budget. It cannot make an invalid inference valid. Ninety-nine participants may have missed a defect that the hundredth identifies. Conversely, the persistence of one person’s refusal supplies no veto over a conclusion whose reasons withstand that person’s objection. The institution must examine what the dissent shows.

Reopening conditions should be explicit. They include a relevant new fact, an omitted affected interest, a counterexample, an ambiguity that changes the inference, a failure in a source attribution, or an argument against a premise on which the conclusion depends. A dispute about whether an answer succeeds also deserves a recorded assessment. An administrator’s claim that a question has already been answered cannot end that examination by itself.

Repeated submissions can be linked to the existing exchange once their equivalence has been checked. Editorial triage should explain its decisions and permit an appeal. Finite resources require priorities, and those priorities must remain distinguishable from verdicts about truth. A challenge awaiting attention should be described as awaiting attention.

The practical result is a disciplined presumption of reliance. A supported conclusion remains available for use while it withstands the relevant objections. A successful challenge changes its status at the point affected. This arrangement preserves both conviction and corrigibility: people can act on reasons that have earned their confidence and remain prepared to recognize the evidence that would defeat them.

VII. Plural Minds and Accountable Power

A plural Academy will need to judge contributions without treating ancestry, status, or fluency as a substitute for their grounds. A human participant can expose a machine’s error. A nonbiological participant can expose a human’s. Each discovery has the authority of the reasons that establish it.

Evidence about current systems gives that demand practical urgency. Sharma and colleagues found sycophantic behavior in five assistants across the tasks they studied, including shifts toward users’ stated beliefs. The finding calls for tests of whether a participant distinguishes a successful objection from pressure to agree. It supports no general presumption that nonbiological reasoning is either reliably impartial or incapable of correction.⁷

The Academy should retain provenance sufficient to assess shared dependencies. Several generated assessments may inherit the same framing, source, or unsupported premise. Review gains strength when it brings relevant differences in evidence and criticism to bear. The number of apparent endorsements cannot establish that this has happened. Chapter 9’s moral wind tunnel supplies proposed tests for correction, transfer, and resistance to pressure; their results would themselves belong in an inspectable record.

Hare’s Archangel remains a standard against which this work can be judged. In “Moral Conflicts,” he explicitly allows actual people to share the characteristics of his idealized figures in varying degrees. The capacity to contribute to critical inquiry carries no claim to the Archangel’s complete powers.⁸ Greater competence should increase the scope of the work a participant can responsibly undertake, together with the scrutiny appropriate to its influence.

Participation also extends beyond specialists. A person may know that an institution has misdescribed their predicament while being unable to reconstruct the institution’s moral theory. A challenge in ordinary language should receive help in identifying its factual and argumentative significance. Testimony deserves examination, and disagreements within an affected group should remain visible. A group’s interests cannot safely be represented by whichever member is easiest for the institution to consult.

Intellectual qualification, moral standing, and governing authority require distinct assessments. A contribution may be sound even when its author is unreliable elsewhere. A being’s suffering can matter despite limited reasoning or objectionable conduct. An expert conclusion can deserve confidence while leaving the power to impose it in need of separate justification. These distinctions protect people outside the Academy as well as participants within it.

The safeguards governing the institution should make those distinctions effective. Funding and relevant conflicts of interest should be disclosed. Appointments should be open to challenge through stated criteria. People responsible for a disputed decision should face review by others with sufficient independence to criticize it. Findings embarrassing to a sponsor should remain eligible for publication under the same standards as favorable findings. An outside critic’s successful objection should alter the record.

Public scrutiny must also respect the people whose evidence makes it possible. Testimony can contain private information, and disclosure can expose a vulnerable person to further harm. The Academy should use consent, careful redaction, and justified restricted review where needed. It must state when restrictions prevent readers from independently checking a premise and calibrate the confidence of its public claim accordingly.

Epistemic authority gives others a reason to attend to a conclusion. Political or institutional power requires an additional account of consent, representation, review, and the consequences of enforcement. The Academy’s judgments can inform public action while remaining subject to those requirements. An institution claiming moral expertise must answer for its own treatment of those who question it.

The same obligation reaches its relationships with nonbiological members. Their own morally relevant interests, wherever the evidence supports attributing them, belong in the assessment of creation, modification, labor, and control. The institution must justify the burdens it places on its participants. A shared inquiry becomes credible through the reciprocal application of its standards.

VIII. A Public Model Fit for Use

The public result of this work should be a usable account of morality: what judgments the reasons support, how those judgments should guide conduct, and where further inquiry could change them.

This is the purpose of a Standard Model of Morality. The title describes an aspiration to organized, usable knowledge. Its authority would have to be earned by the arguments and evidence supporting each part. An unresolved question should have an identified reach; an established correction should retain its force wherever its conditions recur. Public guidance and continuing research can proceed together.⁹

A person confronting a purported justification needs questions that uncover its structure. What treatment is being proposed? Who will live with its consequences? Are their circumstances accurately represented? Would the claimed reason retain its authority when different people occupied the roles? Have all the relevant claims entered the comparison? What supports the rule by which the conflict is decided?

Consider an institution defending restrictions on a minority by calling its members dangerous. The description is an empirical claim with consequences for their lives. It requires evidence. If the evidence fails, the defense loses that premise. If credible evidence identifies a particular risk, the inquiry continues through the alternatives available, the burdens each imposes, and the reasons for assigning those burdens. Popular approval supplies information about what supporters want; its significance has to be assessed alongside the other affected interests. A complete judgment follows those interests through the comparison.

The public should be able to use results at different levels of difficulty. Some defects can be established directly: an omitted person, a false premise, an exemption that depends only on identity. Other conclusions require developed arguments about comparison and the scope of claims. The Standard Model should explain the connection, allowing readers to use a well-supported result without pretending that every neighboring controversy has also been resolved.

Hare’s two-level account offers a practical starting point for this work. Critical assessment examines which dispositions and principles finite agents should cultivate and how conflicts should be addressed. Dependable commitments then guide ordinary conduct under limitations of time, information, and resistance to temptation.¹⁰ Institutions can apply that structure by investigating recurring risks before an emergency and making justified protections difficult to evade through an improvised exception.

The resulting guidance must be strong enough to constrain conduct. Protections against deception, coercion, and grave injury should travel with the reasons for them, the circumstances in which conflicts arise, and the demanding conditions under which a departure could be justified. A declaration that an exceptional benefit is available must be assessed through evidence and the full affected comparison. Someone proposing to set aside a protection incurs the work of justifying the departure.

Uncertainty requires the same discipline. Identify the unknown fact or disputed inference and ask whether credible answers could reverse the judgment. Where the answer remains stable across the supported possibilities, guidance can be firm. Where the balance genuinely depends on information still missing, further inquiry, a reversible step, or a precaution may be justified. The risks of delay and inaction belong in that assessment too.

False certainty and false uncertainty both have victims. The first authorizes conduct beyond what the reasons support. The second withholds an answer after the relevant grounds are available, allowing an exposed defense to retain practical authority. A responsible public model must help people recognize both failures.

Education would make these habits widely available. Students could learn to follow a claim through its evidence, compare a principle across changed positions, and identify what a successful objection requires them to revise. Public explanations, worked cases, and accessible records could extend that learning beyond classrooms. The aim is a population better equipped to ask for reasons and recognize when the answer has failed.

An initial public program should remain manageable. Publish a small set of examined cases, invite challenges, record corrections, and study whether people can actually use the material. Test the institution through its ability to improve a judgment and preserve the improvement. A growing record of such work would give the Standard Model both substance and a means of renewal.

IX. The Judgment Philosophy Must Face

The authority of moral philosophy is tested when a person asks what justifies the treatment they must endure. The answer has to reach their predicament. Its success depends on the reasons it supplies and the scrutiny those reasons can bear.

The book has argued for recovering Hare because his project addresses that demand with unusual directness. A moral prescription commits its author beyond the position from which it is uttered. The affected lives must enter the inquiry with their circumstances preserved. Competing claims require a common assessment. The resulting judgment can then govern a decision, inform a rule, and become part of the character through which people act.

The task now includes giving that achievement an institutional life. A justified conclusion should survive the disappearance of the conversation in which it was reached. A correction should travel to those whose work depends on it. A dissident should be able to find the argument they need without first acquiring the prestige of the people they must challenge. A person seeking guidance should encounter an answer whose grounds are available for inspection.

This undertaking asks courage of its participants. It requires defending a conclusion when influential people dislike it, admitting a mistake when its correction is costly, and continuing an inquiry when neither agreement nor reassurance is immediately available. Conviction becomes a public virtue when its bearer can explain what supports it and recognize the reason that would require revision.

New minds can join that work. They can help preserve an exchange, expose an inconsistency, recover an overlooked answer, or show us that a familiar description has concealed someone’s life. They can also repeat our errors and add their own. The community earns its value by making correction possible across those differences. Anyone capable of offering a sound argument should find participants willing to answer it.

The phrase “last chance” has its sharpest meaning in the lives at stake. Inquiry may continue for centuries. A particular opportunity to prevent an injury can close this afternoon. A family can lose its home while officials repeat a defense whose failure is already understood. A powerful system can execute a purpose while everyone responsible for it assumes that authorization has settled justification. The timing of moral knowledge matters to the people who need it.

Philosophy must learn to carry the force of an argument into those encounters. The work includes defending what we have reason to know, identifying what remains uncertain, and building institutions that preserve both distinctions under pressure. A profession worthy of public trust would welcome that responsibility and make its own performance answerable to it.

Hare asked us to mean what we prescribe and to face the consequences across the lives our prescriptions govern. Recovering his argument gives us reasons to act, a discipline through which to improve our judgments, and a demanding standard for the institutions we create.

We know enough to do better. We owe one another the work of making that knowledge count.

Notes

R. M. Hare, The Language of Morals (Oxford: Clarendon Press, 1952); Freedom and Reason (Oxford: Clarendon Press, 1963); Moral Thinking: Its Levels, Method, and Point (Oxford: Clarendon Press, 1981), especially chs. 5–10. For Hare’s own concise exposition of the transition from universal prescription to impartial preference comparison, see “How to Decide Moral Questions Rationally,” Crítica: Revista Hispanoamericana de Filosofía 18, no. 54 (1986): 63–81, especially 73–76. Chapters 7–8 of this book distinguish the original arguments, subsequent reconstructions, and our defense. https://critica.filosoficas.unam.mx/index.php/critica/article/view/633.
R. M. Hare, “Some Reasoning about Preferences: A Response to Essays by Persson, Feldman, and Schueler,” Ethics 95, no. 1 (1984): 81–85, at 83. Hare distinguishes acquiring another’s preference from granting it a veto over the other affected preferences. The consequences of this distinction and the separate multilateral problem are examined in Chapters 7–8. https://doi.org/10.1086/292601.
The Cumulative Academy, its Institutional Independence Clause, and the governance proposals in this chapter develop the institutional undertaking introduced in Chapter 9. They are proposals advanced by this book, rather than institutions founded by Hare or findings of a completed evaluation. The independence requirement covers the institution’s capacity to revise its own procedures as well as its assessments of theories.
The Citator and three-part ledger are our proposed infrastructure for cumulative inquiry. The distinctions among inferential assessment, philosophical justification, and factual or practical assessment classify the work supporting a claim. They impose no limit by definition on the substantive conclusions a justified argument can reach. Versioning, dependency tracking, source-status records, and public summaries are proposed means of preserving that work.
The controlled executive example and its preference-strength units follow Chapter 8. The calculation is 1,000 − (1,000,000 × 1) = −999,000. It assumes independently comparable cardinal differences, proportional combination, and a fixed affected population, with other effects excluded by stipulation. The hypothetical supplies no empirical measurement of an actual transaction and treats neither money nor willingness to pay as an automatic measure of preference strength.
“Logical closure” and “dialectical closure” designate two statuses in the proposed record. Dialectical closure is indexed to a specified inquiry, argument, and body of objections; silence or the number of endorsements supplies no substitute for its positive justification. The reopening and review arrangements are our institutional proposals. A logical assessment of an inference leaves the soundness of its premises and the adequacy of its formalization open to their own examination.
Mrinank Sharma et al., “Towards Understanding Sycophancy in Language Models,” International Conference on Learning Representations (2024); author manuscript, arXiv:2310.13548, revised May 10, 2025. The cited finding concerns the assistants and tasks studied, including user-belief effects and the role of preference judgments. The institutional proposals concerning independent review and shared dependencies are our applications of the concern. https://arxiv.org/abs/2310.13548.
R. M. Hare, “Moral Conflicts,” in The Tanner Lectures on Human Values, vol. 1 (1980), 171–193, especially 182–184, at 183; delivered at Utah State University on October 5, 1978. The passage explicitly places actual people between the idealized figures in different degrees and circumstances. The application to a plural Academy is developed here. https://www.ditext.com/hare/1980a.pdf.
“Standard Model of Morality” names this project’s proposal for publicly usable, argued guidance with a specified domain and revisable research frontier. The institutional and public functions developed here continue the project’s companion essay, “Toward a Standard Model of Morality.” The analogy does not assert an existing profession-wide consensus, equate philosophical support with experimental confirmation in physics, or authorize adoption of an unsupported principle merely because a decision is urgent.
Hare, Moral Thinking, chs. 2–3; “How to Decide Moral Questions Rationally,” 75–77. Hare explains the relationship between critical review, the cultivation of dispositions, incomplete information, and human fallibility. The proposed public guidance, institutional safeguards, and treatment of decision-relevant uncertainty are applications developed in this book; they require evaluation in their actual settings.