The Law of the Jungle

Rules can make cooperation possible while reserving its benefits for a privileged group. A pirate crew can divide its proceeds fairly and rob everyone else. An empire can restrain violence among recognised states while denying the same standing to conquered peoples. Order and domination can therefore grow together. The existence of a code reveals less than its membership conditions: who receives protection, who bears the costs, and who can challenge the people applying it.

A rule’s origin, its continued enforcement and its justification require separate explanations. A history of conquest may explain an institution without settling every later claim made through it. Conversely, a rule can be widely accepted and carefully administered while remaining unjust. Political philosophy distinguishes legitimacy understood as belief in authority from legitimacy understood as a justified right to rule. Confusing them turns successful persuasion into a moral certificate. (Political legitimacy)

The suspicion that winners write morality catches a recurring historical mechanism. It becomes misleading when it treats everyone else as a passive recipient. People have negotiated rules without a sovereign, refused the categories imposed on them, and used universal promises to demand changes their first proclaimers opposed. The question is whether the people bound by an order can hold its makers to account, and what happens when their reasons conflict with the interests that sustain it.

Is the jungle really ruleless?

In ordinary speech, the law of the jungle means that strength decides and everyone must fend for themselves. Rudyard Kipling’s literary jungle had a different arrangement. The poem collected in The Second Jungle Book in 1895 prescribed obligations within the wolf pack, including limits on quarrels and claims to food. Its jungle had a code. The familiar expression is useful for unrestrained competition, but it is poor natural history and an unreliable guide to the original text. (Kipling, “The Law of the Jungle”)

For the condition in which people can harm one another without a common authority to settle disputes, Hobbes’s state of nature and his war of all against all are more precise. In Leviathan (1651), superior strength offered no permanent security: even a weaker person could threaten someone stronger. The danger lay partly in uncertainty about other people’s intentions. Peaceful activity became precarious when nobody could rely on continued peace.Hobbes presented an argument about the conditions of security, not an archaeological discovery of humanity’s original way of life. (Hobbes, chapters 13 to 15)

Mutual vulnerability gives people a reason to make agreements. It also gives them a reason to distrust agreements without assurance. The possibility of killing someone is a physical capacity; permission to kill is a normative claim. Moving from the first to the second already assumes a rule about what capacities entitle their possessors to do. No observation of a forest settles that question.

Nor does survival always favour solitary aggression. Gerald Wilkinson’s 1984 study of vampire bats found food sharing associated independently with kinship and opportunities for reciprocation. An unsuccessful animal could receive blood from another. This is evidence that cooperation can contribute to survival without legislation. It does not require the bats to possess a human conception of obligation. (Wilkinson 1984)

Claims about animal fairness need more care. Brosnan and de Waal’s 2003 capuchin experiment found refusals when another monkey received a better reward. The celebrated interpretation was an evolutionary precursor of inequity aversion. A 2024 meta-analysis of 23 studies, covering 18 species, found no overall evidence for that interpretation in the accept-or-reject tasks it examined. Disappointment about the available reward remains a competing explanation. Cooperation in animals is well supported; an experiment involving grapes and cucumbers does not establish a universal moral law. (Brosnan and de Waal; Ritov and colleagues)

Robert Axelrod and William Hamilton’s 1981 models showed how repeated encounters could support cooperation among parties pursuing their own interests. Future interaction could make restraint worthwhile. The result depended on the interaction’s conditions; it was not a guarantee that patience defeats exploitation. A party expecting no further dealings, or insulated from the costs it imposes, faces a different problem. (Axelrod and Hamilton)

The wreck of the Batavia in 1629 makes the distinction concrete. Jeronimus Cornelisz appointed a council on 5 July; under his faction’s authority, at least 125 survivors were murdered. Other survivors organised resistance. The catastrophe involved competing forms of organisation, including a governing council directing predation. It cannot be explained simply as people losing all rules when ordinary authority disappeared. (Paterson and colleagues 2023)

The absence of a superior enforcer leaves room for cooperation as well as coercion. Mutual vulnerability can encourage bargaining, but rules also arise under severe inequality. What changes with bargaining power is often the content of the agreement and the price of refusing it.

Why do outlaws write constitutions?

A criminal organisation needs its members to cooperate. Theft from outsiders can be its purpose while theft between members undermines it. This difference explains why a violent group can prohibit conduct that it inflicts on others. The code’s apparent contradiction often disappears once its protected population is identified.

Peter Leeson’s study of seventeenth- and eighteenth-century pirates describes elected captains, quartermasters with separate powers, and written articles governing shares and compensation for injury. Divided authority addressed a practical danger: a captain capable of directing an attack might also abuse the crew. Rules helped organise collective resistance to that abuse.The historical record is uneven and includes accounts attributed to Captain Charles Johnson, whose reports need criticism like other period sources. Still, the documented arrangements are more specific than a vague appeal to honour. (Leeson 2007)

The articles attributed to Bartholomew Roberts also excluded women and boys from the company. A constitution that limited arbitrary command among participating men was not a commitment to universal equality. Its protections did not automatically extend to captives or the people being robbed. A group can reduce domination internally while increasing its capacity to dominate outsiders.

David Skarbek found a related problem in the written constitution of the prison gang Nuestra Familia. Its recruits faced a costly, potentially lifelong commitment. Leaders needed authority to coordinate members, but recruits needed some assurance against those leaders’ predation. Skarbek analysed formal constraints alongside testimony and institutional records. Constitutional language helped solve an organisational problem even within a violent enterprise. (Skarbek 2010)

Family immunity should be treated as a separate empirical claim. Stories about Mafia and yakuza honour often turn a claimed ideal into a dependable practice. Kaplan and Dubro’s history opens by examining the yakuza’s romantic image, including the literature that sustained it. Such an image is evidence about self-presentation. It is insufficient evidence that rival relatives enjoyed a uniform, enforceable protection. (Kaplan and Dubro, chapter 1)

A prohibition on attacks against relatives can nevertheless make strategic sense. If opponents value future coexistence and can impose costs on each other, restraint can preserve an arrangement that benefits both. This is a conditional explanation, not proof that any particular gang followed the rule. The calculation changes when leaders expect impunity or value a short advantage more than future cooperation.

Diego Gambetta’s account of the Sicilian Mafia adds another distinction. Protection can be a real service to a particular buyer while its provision reinforces wider insecurity and distrust. In his retrospective assessment, he explicitly rejects the inference that a useful service must therefore be socially beneficial. A protected transaction can coexist with an exploitative market in protection. (Gambetta 2011)

Who counts as innocent?

The phrase “women and children” makes innocence appear immediately recognisable. It draws attention to vulnerability, but also turns sex and age into substitutes for conduct. Adult civilian men can disappear from the protected category. Women can be treated as dependants even when they act politically or participate in fighting. The same shorthand can offer protection and misdescribe the people it classifies.

Charli Carpenter examined this problem in “Innocent Women and Children” (2006) and in her study of violence against civilian men and boys. Forced conscription and sex-selective attacks become harder to recognise when men are presumed to be combatants. The point is not to reduce protection for women or children. It is to avoid making another person’s civilian status depend on resemblance to the preferred image of a victim. (Carpenter 2006)

Historical restraint was never confined to those two categories. The Christian Synod of Charroux in 989 condemned robbery of peasants and attacks on unarmed clergy, with ecclesiastical sanctions. Classical Islamic jurists discussed protected categories that included elderly people, clergy and hired workers who were not fighting. These traditions had different authorities and exceptions. Their classifications demonstrate that the boundary of permissible violence was an object of argument, not a single instinct applied unchanged across societies. (Charroux decrees; Ahmed Al-Dawoody)

A biological explanation sometimes claims that violent groups spare women and children because they represent the species’ future. Men also contribute to reproduction and care, while reproductive capacity supplies no sufficient measure of a person’s moral standing. The claim that predators generally spare young animals to preserve future populations is false. Packer and Pusey’s research on lions documented substantial cub mortality following male takeovers. Evolution can favour behaviour disastrous for particular young animals. It supplies no simple injunction to protect them. (Packer and Pusey 1983)

Even an explicit norm can fail to govern conduct. Elinder and Erixson studied 18 maritime disasters involving more than 15,000 people. Women generally had worse survival outcomes than men, and crew members fared better than passengers. Their 2012 analysis also found that captains could influence compliance with protective norms. The familiar story of “women and children first” described neither an automatic response nor a dependable outcome across their sample. (Elinder and Erixson)

Violence following rules does not mean violence without violations. A meaningful restriction must affect decisions when compliance is costly. Its value also depends on whether the category it protects captures the relevant vulnerability.

Is a state the gang that won?

The claim that government is a successful protection racket has serious historical support. Charles Tilly’s 1985 essay traced the connections between warfare, extraction and state formation in Europe. Rulers eliminated rivals and financed armed organisations; the institutions built for those purposes accumulated administrative power. Protection could include threats that the protector had helped create. Tilly presented this as a fruitful analogy grounded in several centuries of European history, rather than a discovery that every state began through an identical sequence. (Tilly 1985)

Mancur Olson’s stationary-bandit model explained why a coercive ruler might supply useful public goods. A raider can take what is available and leave. A ruler expecting to collect revenue repeatedly has an interest in future production. Security and investment can enlarge what remains available to tax. That incentive weakens when the ruler expects a short tenure. The model explains how predation and provision can coexist; the benefits do not establish that the population authorised the arrangement. (Olson 1993)

These accounts make taxation and policing intelligible without assuming benevolent founders. They do not make institutional differences irrelevant. A court that can enforce a claim against an official changes a subject’s position. An office whose powers survive its holder differs from personal command. Procedures for changing law can permit a dispute to continue without becoming a contest over who can threaten whom most effectively. Each difference needs evidence of operation; a constitution printed on paper is insufficient.

Weber’s definition of the state helps locate the issue. In his 1919 account, the state claimed a monopoly of legitimate physical force within a territory. The word legitimate concerned the recognised authority to employ force. It did not certify that every exercise of that force was morally right. Belief in a ruler’s right to command can help stabilise an unjust order. (Weber and legitimacy)

The gang comparison also becomes too simple when it assumes crime governs only where the state disappears. Skarbek’s prison case concerns an institution under intensive official control. State custody and unofficial governance occupied the same space because official institutions did not enforce the illicit agreements prisoners made. Capacity in one domain did not amount to accepted or usable protection in every domain.

A gang can perform governing functions without acquiring everything associated with a state. Calling it a second government identifies a question about jurisdiction and enforcement; it should not erase differences in scope or accountability. The comparison is useful when it asks where protection becomes extraction, and what institutions make that distinction enforceable against the protector.

Must order have a single ruler?

The choice between a commanding sovereign and an unregulated struggle leaves out a large institutional middle. Elinor Ostrom studied people who made and enforced rules for shared resources. Her cases included arrangements with several centres of authority, some public and some organised by users. Such systems could fail, but their very existence challenged the assumption that coordinated order required one organisation to direct every decision. (Ostrom, “Beyond Markets and States”)

Research on irrigation systems in Nepal during the late twentieth century found that farmer management could outperform government management on maintenance, water distribution and agricultural productivity. Ostrom’s synthesis connected performance to arrangements for making agreements and monitoring compliance. Her recurring institutional features included participation by affected users in changing rules, accessible dispute resolution, and sanctions that increased with the seriousness or repetition of an offence. Local knowledge mattered because rules had to fit the resource and its users.

The lesson is conditional. A small irrigation association is not a model of every political problem, and local control can preserve local inequalities. It does show why the relevant question is often the allocation of authority. Who observes a violation? Who can challenge an accusation? What stops a powerful user taking more than their share? These questions remain intelligible without imagining either perfect voluntary harmony or a ruler above every decision.

They also clarify why “rules exist so that people can progress” is incomplete. Progress needs a specified beneficiary and a comparison. A rule that increases total production can give most of the gain to those already powerful. The weaker participants may benefit relative to violence while losing relative to a feasible, fairer arrangement. Mutual benefit and just distribution require different evidence.

A workable order therefore needs more than enough force to punish disobedience. It needs information about conduct and a way to settle disagreements over that information. The machinery for revising a rule can matter as much as the original bargain, especially when membership or material conditions change.

Why did European powers set so many terms?

Several European powers acquired exceptional overseas reach between the fifteenth and nineteenth centuries. Explaining why requires distinguishing the phases of expansion and the legal categories that travelled with it. “Europe” is too coarse a unit for a single explanation: early maritime conquest and later industrial predominance were different developments. An account of the first Portuguese voyages cannot start with the steam engine.

Philip Hoffman’s explanation stresses the incentives surrounding military competition. Repeated rivalry among European rulers encouraged expenditure and learning in gunpowder warfare. His argument goes beyond the observation that political fragmentation existed: competition had to reward particular military investments and make resources available for them.Gunpowder itself originated outside Europe. Military advantage emerged through the development and deployment of technologies, not a permanent regional monopoly on invention. (Hoffman 2012)

That mechanism does not replace every regional explanation. Epidemics were devastating in the Americas; they cannot explain every conquest elsewhere. Local alliances and conflicts affected what comparatively small invading forces could accomplish. Hoffman’s own discussion of the conquest of Mexico treats disease as part of a larger explanation, rather than a sufficient cause. A theory must explain where an advantage worked and where it failed to produce the same outcome. (Hoffman, Caltech interview)

Kenneth Pomeranz examined a different problem: the nineteenth-century economic divergence between advanced regions of Europe and East Asia. He emphasised accessible coal and the resources obtained through the Americas. Those resources eased land constraints and supported paths of industrial growth that would otherwise have been harder to sustain. The causal direction could run from imperial access to economic advantage, as well as from economic advantage to conquest. Historians dispute the relative weights and timing, but an account based on timeless European superiority explains neither. (Pomeranz 2000)

One account of imperial privilege holds that conquerors first ignored rules, then wrote them to secure their gains. Legal history reveals a further mechanism: rules helped organise the conquest itself. Antony Anghie’s history places the colonial encounter within the formation of international law, including debates about relations between Spaniards and Indigenous peoples in Francisco de Vitoria’s sixteenth-century work. In the nineteenth century, the question of which societies qualified for sovereign recognition became inseparable from imperial power. The supposedly common order could assign different legal standing to the parties encountering it. (Anghie 2005, chapter 1; chapter 2)

Article 22 of the League of Nations Covenant made a later hierarchy explicit. Territories removed from defeated powers were placed under mandates. Their inhabitants were described as not yet capable of standing independently under modern conditions; more advanced nations were assigned tutelage. Welfare obligations accompanied the hierarchy. Administration was framed as responsibility while the power to judge readiness remained unequally distributed. (League Covenant, article 22)

The historical pattern is therefore more precise than conquerors acting without rules and inventing them afterwards. Legal argument accompanied expansion and helped distinguish the people entitled to decide from those whose status others decided. Power shaped admission to the rule-making community.

Can winners close the route behind them?

A rule can preserve the outcome of an earlier struggle without mentioning that struggle. If possession becomes the starting point for future legal protection, the history through which possession was obtained may disappear from ordinary adjudication. Stability has value, but the choice of a starting point distributes that value. A prohibition on further taking leaves an earlier taker differently placed from someone who has already lost everything.

Ha-Joon Chang’s Kicking Away the Ladder (2002) applies a related suspicion to development policy. He contrasts the interventionist policies used during the industrial development of several wealthy economies with the freer-market prescriptions later presented to poorer economies. The historical discrepancy challenges a simple account in which successful countries always followed the policies they eventually recommended. It raises a question about who retains room to experiment. (Chang)

The stronger causal claim needs separate testing. Economic historian Douglas Irwin criticised Chang for attributing growth to protection merely because they occurred together. A country can grow while pursuing a policy that reduces its potential growth. Establishing that tariffs caused industrial success requires evidence about mechanisms and alternatives. A historical double standard may be demonstrable even when the wisdom of reproducing the earlier policy remains disputed. (Irwin’s review)

Copyright supplies a narrower, well-documented example of changing legal boundaries. The United States’ 1891 legislation extended federal copyright eligibility to foreign authors under specified conditions. Before that extension, federal protection excluded works by foreign authors who were not resident in the country. American publishers broke no American law by reprinting foreign books; domestic law defined which authors received protection. The distinction matters because a legally permitted advantage can later be closed without its earlier beneficiaries ever having broken their own law. (U.S. Copyright Office, 1961 historical account)

This mechanism does not require a single plan maintained across generations. Industries develop interests in stronger protection as their position changes. A rule may solve a coordination problem and confer an incumbent advantage at the same time. The proper historical test compares earlier freedoms with later restrictions, identifies the beneficiaries, and asks whether the stated reason would survive a reversal of positions.

The policy question then concerns the available alternatives. Would a less restrictive arrangement preserve the shared benefit? Who absorbs the transition cost? Can newcomers influence revision? A rule that protects everyone from a recurring harm has a different justification from one whose defence depends entirely on treating yesterday’s advantage as beyond dispute. Often a single settlement contains both elements, which is why tracing its distributional effects is more useful than assigning it one motive.

How does a rule become common sense?

A rule can become difficult to question because it organises the activities through which people learn to judge. A language can be both useful for communication and a credential for entry into respected institutions. Once success repeatedly depends on mastering it, the practical advantage can be mistaken for evidence that its speakers are inherently more intelligent. The standard’s utility and the ranking attached to it are different claims.

Antonio Gramsci’s prison writings examined how a ruling group secured consent through intellectual and moral leadership. Schools and ordinary cultural practices helped organise people’s understanding of their position. His idea of common sense was fragmented and contradictory, not a perfectly installed doctrine. Existing beliefs could contain resources for criticism as well as resignation. Hegemony described a political achievement that required maintenance and could be contested. (Gramsci)

Ngũgĩ wa Thiong’o’s Decolonising the Mind (1986) treated language as a carrier of culture and colonial education as a struggle over whose experience supplied the centre of literary life. The question concerned more than which vocabulary a student used. A curriculum could make one population’s history the measure through which everyone else’s was interpreted. The book’s language argument identifies an institutional route through which power can shape aspiration. (Ngũgĩ)

Colonial domination, cultural influence and subalternity should still be distinguished. Colonialism concerns a relation of domination between peoples; foreign influence alone is insufficient to identify that relation. People borrow forms for many reasons, including their usefulness. Describing every borrowed preference as colonisation makes the claim impossible to test and overlooks what people actively make of an inheritance. (Colonialism)

Gayatri Chakravorty Spivak’s “Can the Subaltern Speak?” (1988) addresses a harder problem than literal silence. In criticising Foucault and Deleuze, she disputes the intellectual’s apparent transparency when claiming simply to transmit oppressed people’s interests. Political representation and the production of an account of someone are related but different operations. A person can speak while institutions translate that speech into categories that prevent it from counting as the intended claim. (Spivak, in Marxism and the Interpretation of Culture, University of Illinois Press, 1988.)

This makes “restore their voice” an insufficient programme on its own. The terms of recognition also need examination. Who can appear as a credible witness? Which interests are intelligible as interests? Calling every person from a formerly colonised society subaltern would obscure precisely the differences in institutional access the concept is meant to expose.

Hart’s internal point of view identifies a separate feature of rule-following. Someone can use a rule to evaluate conduct and criticise breaches, rather than merely predicting punishment. That attitude need not involve moral approval. It helps explain the social existence of a rule; it does not show that the rule was accepted freely or deserves allegiance. Gramsci’s account of political consent and Hart’s account of legal practice therefore answer different questions. (Legal positivism)

How can a default be examined from inside?

Criticising an inherited standard does not require a mind untouched by history. A more limited task is possible: make the standard explicit and compare its reasons with alternatives. The object of criticism is a proposition or institution that can be examined, rather than an entire self declared contaminated.

Take the hypothetical belief that proficiency in a prestigious language indicates intelligence. Its practical component may be accurate in a particular setting: knowing the language can be necessary to use an archive or perform a job. The general ranking requires additional evidence. A test of intelligence that primarily rewards familiarity with the language may simply reproduce the entry condition. Separating the specific requirement from the larger judgment prevents a useful skill from becoming a verdict on human worth.

Next comes institutional history. Who required the standard, and what happened to people who used another one? Evidence of coercion matters because it can explain why alternatives disappeared. Evidence of adoption for convenience matters too. The relevant comparison is between actual possibilities, including the costs of changing an established practice. Treating every surviving convention as freely chosen would hide dependency; treating every useful inheritance as an imposed delusion would hide agency.

Positions can then be reversed. Would the rule’s defenders accept its burden if they lacked the resources required to comply? Can those affected explain their objections in terms the decision-making institution recognises? Does the rule offer a way to challenge an error? These are proposed tests of justification, not a psychological instrument for measuring how colonised an individual is.

A final distinction separates a rule’s origin from the reasons for keeping it. A standard may have arrived through domination and still solve a shared problem. Its defence must then identify that benefit and address the distribution of its costs. Conversely, an old local tradition can be coercive. An ancestral origin offers no automatic exemption from criticism, just as a European origin supplies no automatic disproof.

The search for an ultimate cultural standard often promises relief from this work. It would settle in advance which traditions deserve authority. The historical evidence supports no such shortcut. Comparison becomes more informative when it asks who can participate in correcting a standard, especially people for whom compliance is expensive. Revisability provides a practical test that admiration for origins cannot supply.

Can rules turn against their authors?

Universal language creates a difficulty for people who want limited application. A declaration of equal rights can be read as a claim extending beyond the population its drafters expected to benefit. Whether that reading gains force depends on political action. Excluded people have repeatedly supplied the action and changed the claim in doing so.

The Haitian Revolution began in Saint-Domingue in 1791. Emancipation proclaimed there in 1793 was followed by the French Convention’s abolition of slavery in the colonies on 4 February 1794. These changes occurred amid insurgency and war. Enslaved people and their allies helped force the question of whether revolutionary citizenship could coexist with slavery. They were participants in determining the scope of rights. (Jeremy Popkin’s historical account)

Laurent Dubois’s work on the French Caribbean argues that struggles over slavery transformed the meaning of universal rights. This account changes the direction of intellectual transmission: the colonies were sites where political ideas were made and revised. The language of rights travelled through conflict between people with incompatible understandings of its application. (Dubois, A Colony of Citizens, University of North Carolina Press, 2004.)

The result was reversible. Slavery was restored in Guadeloupe in 1802; resistance in Saint-Domingue instead culminated in Haitian independence in 1804. A proclamation did not supply its own enforcement. Nor did formal emancipation settle every dispute about labour or authority. That unfinished history is a reason to examine institutions after victory, not to dismiss the expansion of freedom as mere rhetoric.

Decolonisation involved a similar struggle over the meaning of self-determination. On 14 December 1960, the United Nations General Assembly adopted Resolution 1514 by 89 votes to none, with nine abstentions. The adopted draft had been sponsored by Asian and African states. It treated colonial subjection as a denial of fundamental rights and asserted peoples’ right to determine their political status. The membership and voting power of the institution had changed, allowing a different group to state what its principles required. (UN legal history of Resolution 1514)

Adom Getachew’s Worldmaking after Empire shows why this history exceeds the replacement of foreign flags with national ones. African and Caribbean anticolonial thinkers pursued projects to alter the international conditions of dependence, including regional federation and economic reconstruction. Formal sovereignty could leave serious constraints intact. Their ambitions concerned the terms on which new states entered a world already organised by imperial relationships. (Getachew 2019)

The Universal Declaration of Human Rights also resists an account of norms travelling exclusively from European rulers to passive recipients. Hansa Mehta pressed for Article 1 to use “all human beings” in place of “all men” during the drafting that led to the 1948 declaration. The amendment made the scope of equality harder to evade through a supposedly generic term. Non-European participants helped write the language, not merely accept it afterwards. (UN account of Mehta’s contribution)

A norm’s later users are therefore part of its history of authorship. They can preserve its wording while altering who can invoke it, or change the wording because exclusion survives within it. Universality has often been a claim made against an institution’s existing limits.

What makes a rule worth obeying?

The claim that morality serves power has at least two classical versions. In Plato’s Republic, Thrasymachus presents justice as the stronger party’s advantage: rulers establish laws useful to themselves. In the Gorgias, Callicles argues that the weak majority establishes conventions to restrain exceptional individuals. The targets differ. A rule can secure domination, or organise a coalition capable of limiting someone stronger. Neither origin by itself settles whether the rule is just. (Callicles and Thrasymachus)

The premise that morality began entirely as imposed power is also too strong. Curry, Mullins and Whitehouse examined ethnographic accounts from 60 societies for seven forms of cooperation, including reciprocity and the division of resources. They found broad positive moral evaluation of these behaviours. Their study supports a substantial relationship between morality and recurrent cooperative problems. It does not establish that every moral judgment is cooperative, or that cooperation within a group benefits outsiders. (Curry and colleagues 2019)

Even a complete causal explanation would leave a justificatory question. Hume identified the unexplained passage from statements about what is the case to statements about what ought to be done. The success of a practice, its evolutionary history, or a ruler’s ability to enforce it does not by itself supply the missing moral premise. Explaining an obligation’s social origin is different from showing why someone should accept it. (Hume’s moral philosophy)

What, then, separates a rule from a threat? Hart’s criticism of law understood as a gunman’s command points to features threats alone leave unexplained. Legal systems contain standards for identifying valid rules and procedures for changing them. Some rules confer powers, such as the power to make an agreement, instead of ordering conduct under punishment. Officials use shared standards to assess their own actions. These features explain an institutional form. An unjust legal system can possess them. (Hart’s account)

The distinction between legality and morality became especially difficult after the destruction of the Nazi regime. Gustav Radbruch’s 1946 argument gave legal certainty considerable weight, but held that intolerable injustice could defeat a statute’s claim to legal authority. His position contested the separation of legal validity from moral merit. The disagreement was not whether injustice mattered, but whether extreme injustice should alter what counted as law. (Radbruch’s argument and its limits)

The later Berlin Wall prosecutions exposed the difficulty of applying that distinction. In its 2001 judgment on former East German leaders, the European Court of Human Rights examined the country’s earlier written law and international obligations, including protections its border practices violated. It rejected the leaders’ claim that subsequent convictions were unforeseeable. The court asked whether the defendants could use practices they had imposed to escape standards already applicable to them. (Streletz, Kessler and Krenz, paragraphs 67 to 89)

History cannot decide whether moral truths exist independently of every human judgment. It can show that enforcement and justification come apart. An unenforced claim can identify a harm or contradiction that a powerful institution refuses to recognise. The absence of a strong defender explains its vulnerability; treating that absence as proof that the claim is wrong would simply assume the rule of might under examination.

A defensible rule must offer reasons to the people subjected to it, including those who did not write it. Shared benefit is one reason, but burdens and alternatives matter. Protection is more credible when the people supplying it can be challenged through procedures they do not personally control. A claim to universality becomes stronger when outsiders can invoke it against its established beneficiaries. These are substantive standards of judgment, open to argument themselves. Their justification must remain available even when the person asking for it lacks the power to compel an answer.

Appendix

Method

Sources were checked in September 2026 against the papers, publisher pages, university copies, legal texts and reference works linked in the text. Some claims were left out because they could not be verified: that Mafia or yakuza codes reliably protected rivals’ families in practice, several quotations commonly attributed to Mafia figures, and detailed accounts of the Tongan castaways on ‘Ata. Some books were checked through publisher extracts or abstracts rather than full editions.

Related

• Functional Anatomy of Systems shares governance, government
• AI, From the Atoms Up shares governance
• Red Trees shares governance