ChineseEthics

e T Blackwell Companions to Philosophy AtCompanion to Ethics; Edited by PETER SINGER BLACKWELL REFERENCE

Copyright ¢ Basi) Bl Copyri ackwe Editorial Organigatio , ek 1 n (- Peter Singer I991 First published 5 991 First published in USA 1991 Basil Blackwelt Ltd 108 Cowley Road, Oxford, OX 4 1jF. UK Basil Blackwell, Ine, 8 (ambridge Center ambridge, Massachusetts 02 142, USA » i passages for th, retrieves Ve ® - ; e ur etrieval system, or transmitted, in PublltatIOH oy e e ced. Si’orgg?lisaor o Or permission of the i Except in the United States of America o of trade or otherwi British Library ¢, sh Library Cataloguing in Publication Datg A companion to ethicg 1. Ethics. - (Black e Ethics, well ¢ anions i L Simger. peac el companions to philosophy). 170 ISBN 0--6 3I-16211~y Librar ngress Catalo, hg Hlublllll n Data ¢ /Of(k) Q4ress ( ILgll 1 lie 0. t A Companion to cthics / edited b [ pl dCflL\(BlaCkW nctudes bibliograph ISBN 0«631-16;?1 o y Peter Singer, ell comnpanions to phi e s to philoso ical references and index ohy) -9 1. Ethics 2. Socia ethi mnge ial nger. Peter I. Series o | hics. 1.8 ger, II. Se I70—dc20 ? CIp Typeset in 10! on 1! Frome and lflnd()n ’ ' anner Ltdv means, electronic, mechanical rm gty e AR e — Introduction The title A Companion to Ethics may suggest a volume with short entries, in alphabetical order, providing summary information about leading theories, ideas and people in the academic discipline of ethics. As a glance at the outline of the volume (following this introduction) will show, this book is very different. It consists of 47 original essays. These essays deal with the origins of ethics, with the great ethical traditions, with theories about how we ought to live, with arguments about specific ethical issues, and with the nature of ethics itself. (In accordance with current usage, in this book ‘ethics’ will usually be used not only for the study of morality (that is, as a synonym for ‘moral philosophy’ but also to refer to the subject matter of that study, in other words as meaning ‘morality”.) I have chosen to organize the book in this way because it is vital that ethics not be treated as something remote, to be studied only by scholars locked away in universities. Ethics deals with values, with good and bad, with right and wrong. We cannot avoid involverhent in ethics, for what we do — and what we don't do - is always a possible subject of ethical evaluation. Anyone who thinks about what he or she ought to do is, consciously or unconsciously, involved in ethics. When we begin to think more seriously about these questions, we may begin by exploring our own underlying values but we will also be travelling over roads that have been trodden by many other thinkers, in different cultures, for well over twc thousand years. For such a journey it is helpful to have a guide with informatior about the path we shall tread, how it came to be laid out, the major forks where people have taken alternative routes, and who has been there before us. Mort valuable still, however, is the kind of companion who will stimulate our though about the route we are taking and warn us of the traps and culs-de-sac that have stopped others making progress. So the best way to use this book is to go first to the Qutline of Contents, a king of map of the book, with explanatory notes at points where the map migh otherwise be unclear to those who do not already know the territory. Then depending on your interests, you may wish to start at the beginning and worl your way through, or you may prefer to read specific essays on subjects tha interest you. To find any subject not mentioned in the Outline of Contents, consul the Index. It is designed to make it easy to find not only specific concepts o theories, for example justice, or utilitarianism, but also particular aspects of topict Thus under ‘killing’ you will find not a single heading, but also sub-headings tha will lead you to discussions of ethical aspects of killing in Buddhism, Hinduist

PART 1T - THE GREAT ETHICAL TRADITIONS conduct — the li (‘ritual’) inherited from ancient sage-kings. The Book of Li makes up the Confucian dao. Confucius taught his disciples the dao in all the classic texts. He did not see himself as a philosopher but as a historical scholar. His task was not justifying or systematizing the code but learning and relaying it. Confucius’ disciples collected dialogues and exchanges that they remembered having with Confucius. These make up the aphorisms in the book known as The Analects. His disciples disagreed about Confucius’ dao. Some focused on Ii (‘ritual’} and others on ren (‘humanity'). Ren occurred often and somewhat mysteriously in The Analects. Confucius treated the code of ritual as consisting of names and role descriptions. His pronouncements included no explicitly prescriptive ought-statements. He did not segment the li into sentence-like rules which generate duties. For Confucius, the basic role of all language is to guide our acts. We hear, study and learn the way of the sage-kings via transmitted texts, inherited traditions and conventional strings of words. They used language to regulate behaviour. Their guidance is still available to us now because they recorded and relayed their words. We detect the li in explicit written form. Li epitomizes a positive, cultural way conveyed in literature. Translations of Ii include ‘ritual’, ‘etiquette’, ‘manners’, ‘ceremonies’ and ‘propriety’. The most general term we might use for Ii is ‘con- ventions'. Li guide, for example, forms of address, funeral wear, even how to sit at meals. Confucius’ attachment to li is uncritical. He never raises the Socratic ethical question, ‘Why follow just these conventional norms?’ Yet he does seem aware that conventional norms vary in different areas and at different times. Classical Chinese philosophers expressed their disputes as accounts of i (‘mora- lity"). Translations normally translate i puritanically as ‘righteousness’ or deon- tologically as ‘duty’. T avoid ‘duty’ since Chinese thinkers do not segment their systems of guidance into sentential rules. They do not individualize obligations and duties as we do. Think of i (‘morality’) as the ideally correct social guidance in language. Confucius thus has a conventional morality (Ii i) in contrast to Mozi's. utilitarian morality (li i — a different li character) and Mencius' intuitive morality (ren i). Copfucius argued for neither the content of the Book of Ritual nor for its guiding authority. He fixed instead on an intriguing practical concern. How do we correctly ext'ract guidance from the language in the text? How do we use the book in building the human virtues that would result in our following the sages’ intended way? Confucius addressed the intellectual problem of interpretation of the guiding texts. His was a dao of education. The key to communicating a literary dao, he argued, was to rectify names. (Analects 13:3.) The first step is study — ingest the cox.ltr.ent. Confucius had his disciples study the classics. They contained both the guiding codes and the accepted description of historical models of appropriate virtue. . We learn to embrace these cultural roles and play them by studying models in life apd literature. We learn to play music or act out our parts from models. The society must provide us with example of rulers, ministers, fathers, and sons, 70 Sy 6 - CLASSICAL CHINESE ETHICS and correctly identify them. Only then can we learn from their interpretations of their parts. The desire to internalize roles flows from our nature as social beings. We make ourselves fully human by expanding our repertoire of roles. We show our excellence by the quality of our interpretation of our parts. Interpretation of ritual, thus, shares the sense it has when we speak of musical interpretation. Confucius often pairs yue (‘music’) with li (‘ritual’). Given any piece of music used as a guide, we may perform it well or badly. We call our performance an interpretation of the piece. Building good character requires that we interpret the Ii in playing our roles. The mysterious concept ren (‘bumanity’) guarantees a correct performance of the guiding dao. The interpretive ability must be separate from the code which we interpret and yet closely tied to it. Humanity is an interpretive intuition. We cannot teach ren by explicit instructions, since we have to interpret those instructions. This makes it look as if humanity must be inborn. Yet we do get better at interpretation as we learn and practice. Without this interpretive skill, the ritual code cannot function correctly. Thus humanity must be the ability to rectify names. We rectify a name when we make the right discrimination, apply it to the right setting. Think of the code as an internalized programme. To execute the programme containing the term X, we have to discriminate whether this is a case of X or non-X. We must rectify any term that occurs in the code book. To rectify a name is to see that people use it of the right objects. Otherwise the code book’s instructions will misdirect people. Since he regards the tendency to convey and receive a dao as natural, the main political concern for Confucius is rectifying names. ‘If names are not rectified ... affairs cannot be accomplished . ..*and the people will not know how to move hand or foot.' The ruler rectifies names in appointing people to status positions. Most of the code’s content directs proper action toward people filling social roles - father, brother, ruler, minister etc. The names are primarily social, hierarchical roles. The ruler dubs someone minister. We then all conform in using that title to guide our behaviour toward the person filling the role. Similarly, Confucius argues, the political system should identify model fathers and sons. This gives us our link to the code. It also gives us our models of how to act out the role patterns described in the ritual code. Without an accepted, shared social pattern of name use, the code could not guide us. Plainly, rectifying names requires something beyond the code book itself, since we must rectify names to use the code book. Ren (‘humanity’) is the intuitive ability to interpret the li correctly. We may apply ren either in our own action or in guiding others. Rectifying the use of language requires rern. Without people to model roles and rulers to recognize and title a fine performance, we cannot relay the role-based way intended by the sage-kings. Some passages suggest Confucius pointedly would not teach about this nur- tured interpretive intuition he called ren. He used the term often, but was notori- ously elusive about it. In response to his students’ frustration Confucius twice hinted without explanation that all the details of his guiding dao had some unifying 71

PART Il - THE GREAT ETHICAL TRADITIONS core. ‘Confucius said, "My dao has a single thread’’. Zengzi replied, ‘I hear you'". Confucius left. The other asked “What did he refer to?"" Zengzi answered, “Our master’s dao is loyalty and reciprocity”.’ (Analects 4:15.) Most Confucians accept chgz?'s guess. Confucius himself bequeathed little of his theory of either loyalty Ot reciprocity. Confucius does formulate a negative version of the Golden Rule, ‘What you do not desire, do not effect on others.” This may count as a gloss on ‘reciprocity’. In its simplicity, it conflicts with the elaborate code of li which Confucius usually stresses. It is implausible that Confucius meant the Golden Rule to replace his role- based morality — the way Christ repealed ‘the law’ with his Golden Rule. Still, such passages suggest that while the code is conventional, interpretation of the code appeals to more universal, moral considerations. Confucius’ refusal to elaborate on ren invites speculation that it is a universal moral sense intended to guide interpretation. The conventional translations — ‘humanity’ or ‘bene- volence’ — suggest a universal utilitarian standard for interpretation. Confucians would find this result uncomfortable. The orthodoxy treats Confucius as an anti- utilitarian because his most vociferous critic advocated utility. Apart from pointing to a developed intuition for interpreting, Confucius did not theorize about abstract axioms of conduct. He bases his explicit normative system on roles. He does not assign a normative value to persons apart from their social relationships. All your duties are duties of your station towards other socially described persons or things. These roles are natural and the family roles are the core examples. This leads Confucianism to characterize itself as a system of ‘partial’ or ‘graded’ love. We deal with people qua ‘mother’, ‘neighbour’, ‘mentor’, ‘daughter’. Contrast this with the Kantian respect for individuals as bare persons or agents. The basis for this special Kantian status of moral respect is the rationality of moral agents (persons). (See Article 14, KANTIAN ETHICS.) Confucius also exhibits little sense of desert in moral theory. Nor does he exhibit familiar deontological attitudes such as categorical requirements to tell the truth or keep promises, to be just or to respect an agent's autonomy. He strgngly opposes the rule of law. In part he objects to the tendency of punishment to induce egoism. Confucians also oppose the role-corroding egalitarianism of a legal code. In place of law, Confucius would have social education — modelling of name use and role performance, together with a traditional socio-cultural dao set in the classics. The basis for normative relations among people is social role relations. not some bare rational agency. 2 Mozi and the utilitarian Dao Mozi, the first rival philosopher, adopts much of the structure of Confucius’ normative scheme. He too discusses how names in codes guide behaviour rather thgm discussing ought sentences. We use a name by making a shi (‘this:right’) or a fei (‘not this:wrong') assignment. To know the name is to know to shi what should have the name and fei what should not. This ability to divide things in response to language triggers a tendency to treat each discriminant in the proper way. Still Mozi raises familiar philosophical doubts about the explicit linguistic ~1 tw [ 6 - CLASSICAL CHINESE ETHICS content of the guidance. He asserts that we need an argument for the traditional content. Why should we regard our specific traditions or customs as i (‘morality’)? Why regard ren (‘humanity’) as merely an intuitive interpretive ability applied to a customary dao (‘way')? Any such ability should also guide us to create new moral codes — to revise social morality. Thus he directly questions the authority of the ancestral guiding daos. Customs can be very wrong. (Mozi recites or invents a story of a tribe who customarily eat their first-born sons. That should shock good Confucians!) So, he argues, we must have a gauge for selecting among different dao contents. He proposed utility as the measure. He treated this norm as the standard of shi (‘this:right’) and fei (‘not this:wrong). Thus it became a mode! both for ordering and rectifying the terms in the dao. Mozi argued that his criterion came from a natural or heavenly will — the natural preference for benefit over harm. This naturally guiding name pair (benefit- harm) becomes the basis for using all other guiding name pairs. If we do not start from that distinction, Mozi says, we can never be clear on shi-fei. So a correct, positive dao should contain whatever passages will increase benefit when appro- priately engaged in guiding our behaviour. Mozi is a language utilitarian. We should use a single criterion to choose both the code we follow and the dis- criminations we make with the terms in that code. We should choose both in ways that constantly or reliably advance Ii (‘benefit’) and diminish hai (‘harm’). Mozi thinks this proposal entails that we should use the phrase ‘universal love’ in public discourse rather than the Confucian ‘partial love’. A dao that includes frequent use of the phrase partial love will not be constant. It will guide someone who adopts it to prefer that others have a more universal attitude. (Mozi, Section 16.) The question for Chinese ethics is what dao should we instil in people to guide their de. Partial love doctrines will entail that we ought to instil a universal love dao. Since it is self-condemning in this way, the Confucian dao is not a constant dao. Moz assumes that we should make both our name assignments and the patterns of shi-fei judgements uniform throughout society. Thus we should curtail a host of wasteful orthodox Confucian ritual practices such as elaborate funerals, expensive concerts and especially aggressive warfare. They waste resources that could be better used to benefit the people. He condemns Confucians for being able to separate morality from immorality in small matters. while in large matters, such as a state’s going to war, they praise the ruler and call him ‘moral’. He likens it to calling a little bit of white ‘white’ and a lot of white ‘black’. Confucians set a bad example of moral term use. Mozi's way is utilitarian. He does not link his utility to subijective states such as pleasure, happiness or desire satisfaction. Utility is a matter of objective, material well-being. In other ways as well, Mozi's viewpoint is less individualistic than a typical Western moral theory. His version of the Socratic philosophical question is social rather than individual. Socrates asks whether he should obey socially - accepted codes. Mozi asks if society should accept or change its public code. 73

'ART 11 - THE GREAT ETHICAL TRADITIONS dvice. As Mozi showed, etiquette Confucianism cannot be a constant dao. Without ome interpretive instinct, its guidance is indeterminate and its partial love con- emns itself as a collective guide. And Mozi may be no better off. His appeal to niversal utility may be inconstant too. Mencius argued that a dao based on utility 1ight not be able to justify appealing publicly to utility. (Mencius 1A:1.) Promoting alk of benefit may not benefit society. Further, even if we accept a certain linguistic ontent, as Confucius noted, the interpretive problem still faces us. We need an xtra-linguistic guide to use a linguistic one. We need a moral intuition to interpret he terms of a guiding code. The code itself can never guarantee constant patterns { conduct. The Daode-jing gives an explicit linguistic rationale for its scepticism that any ac can be constant. No dao can be constant because no ming (‘'names’) can be onstant. We can interpret any dao in various ways because the application of any .:ame contained in a dao requires both an interpretive distinction and preference 1duced response. Mozi merely assumed the li-hai (‘benefit-harm’) distinction rovides guidance. It may also be interpreted, and the interpretation would likely ary in different utilitarian theories. He seemed to ignore something. If social iscourse can modify our preference for our own family, it can modify our reference for benefit. Furthermore, social discourse can instil different ways to alculate benefit. Each will generate different courses of action — different daos. ne could even have an anti-benefit dao. Thus benefit is not a constant dao. Names supply all (linguistic) moral ways with their capacity to guide in the eal world. A name guides discrimination, desires, and action. Learning a name uides because. when we add it to our vocabulary, we become disposed to make socially approved distinction. Qur social superiors, our teachers and models, pprove when we discriminate as they would with a given term. This training is n integral part of our socialization. Our social models train us to choose ways of cting toward the object named. In a particular context, they teach us to treat Or:;; thing as shi and other things as fei. To learn any name is to learn to shi-fei vith it. These learned inclinations to select and reject things are conventional or cquired desires. An internalized linguistic dao translates into a body of inclinations o classify things. We use the classification in executing our internalized guiding rogramme. We select or avoid those things in conduct. Thus language guides aur wel (‘deeming:actions’). Actions based on deeming things to be this or not- his are unnatural, conventional actions. Laozi's famous slogan, wu-wei (‘avoid leeming action’) enjoins us to vacate this social, linguistic, conditioning. The ontrasting ideal is spontaneous, natural action. Whenever conventional categ- ries generate the actions, the actions are unnatural. Natural actions, by contrast, lo not require any artefact terms. No one needs to teach us to eat, sleep, or rocreate, ' To wei ~ engage in discriminating behaviours guided by names — interferes vith our natural spontaneity. Laozi's theory explains Mencius’ criticism of lan- uage-based ways. He also adopts a more realistic view of the range of natural sehaviour. Mencius had assumed that our intuition was potentially rich enough 76 [reve— 6 + CLASSICAL CHINESE ETHICS to make us sage-kings of a unified moral empire. Laozi's version of pre-linguistic inclinations remains strongly optimistic from a Western individualist perspective but less idealistic than Mencius. He assumes that without language and cultural accumulation, we would be social enough only to form small farming villages. We would live in peace because, without language, we would lack curiosity to interact with the other villages. That natural primitive behaviour is the only chang (‘constant’) dao. We follow that natural path when we refuse to act by deeming — wu-wei. As we saw with Mozi and Confucius, conflicting ways use the same terms, e.g. good and bad, beautiful and ugly, high and low. However, they disagree about how to draw the distinctions in guiding our behaviour. Laozi's Daode-jing invites us to contemplate an anti-conventional way. That way reverses all the con- ventional pairs of opposing guiding terms. The dao of reversal reverses valuation. We normally value dominance, the male, the active, having, benevolence, wisdom, and clarity. Laozi presents considerations for valuing submissiveness, the female, the passive, lacking, non-benevolence, and dullness. For each pair of opposite guiding terms, the text tries to motivate reversing the conventional choice. This is enough to show that these names do not give constant guidance. Once we see the inconstancy of social guidance by names or language, what follows? Here the Daode-jing becomes enigmatic. Confucian interpretations treat Laozi as recommending his negative way as the constant way. Legalism also treats the advice as a serious guide, expecially in its political reversal sections. Legalist writers draw quotations from the text justifying Machiavellian ‘scheming methods’ of government — keeping people ignorant, finding benevolence unnatural. I recommend a different view of Laozi. He urges, as Mencius does, that we abandon all language guides to behaviour. The issue between Mencius and Laozi lies in their view of the content of our genetic mechanisms. On Laozi's account, our natural bent, unembellished by culture and language, would only support society at the level of the agrarian village. Call this primitive Daoism. We may also take the point of reversing values associated with names to express pure scepticism. This pairs him with Zhuangzi. Finally, Buddhists read the text as entailing mystical monism and, like the metaphysics of the Buddha-nature, justifying stoic resignation. There may be other possibilities. The Daode-jing alleges simply that no statement of a way can be constantly sound. It could not coherently tell us what constant, practical conclusion to draw. iii The schools of names: formal meta-ethics The obvious importance of names and language in Chinese ethical doctrines led to a period of intense analysis of names. Three schools of thought emerged. One conclusion was that we should reform language along ideal theoretical lines. Ethical guidance, using names, should be unambiguous to be chang (‘constant’). This proposal is a formal version of Confucius' doctrine of rectifying names. It adopts the slogan of one-name-one-thing. Another school, the neo-Mohists, noted that natural language does not and 77

ART It © THE GREAT ETHICAL TRADITIONS ¢ed not conform to the one-name-one-thing. Our ordinary ways of speaking do ot follow any consistent principle. We ordinarily do think a white horse is a orse. We also agree that riding a white horse is riding a horse. But sometimes e do not think doing something to an object under one description is the same s doing it under another description. We think a thief is a person, but we treat tecutions (killing thieves) as different from murders (killing people). This school thought that we could perfectly well make sense of our language » long us we based it on what we knew of reality. Given common sense and the <ternal similarities and differences, we can rest the patterns of name use on an iternal, constant reality. This school, an outgrowth of Mozi’s thought, thus :veloped a linked theory of linguistic and moral realism. The world provides the -ounds of assertability of our shis and feis. The third school challenged even this qualitied realism. Similarities and differ- wes among things do guide our naming, but we can count and group similarities - unlimited ways. Reality privileges no constant classification scheme. Reality mnot settle our disputes about how to draw distinctions. Interpretation of any 0 (‘'way') must capriciously adopt one of a limitless range of perspectives. Zhuangzi: Daoist relativism 1at third position underwrote the Daoism of Zhuangzi. He could no longer follow ¢ anti-language views of Mencius and Laozi. The realists had shown that any atement of an anti-language position was incoherent. To say ‘all language storts the dao’ was to distort the dao. They had further argued that in any sagreement about shi-fei one party must be correct. Zhuangzi reversed Laozi's Daoism. Do not reject language for natural, spon- neous action. Instead he dropped the assumption that ‘heavenly’, ‘natural’, or 2ality’ provides a coherent standpoint for constructing daos. Zhuangzi likened I the warring schools to ‘pipes of heaven’. Each claims to be expressing the iural or heavenly point of view. Trivially, each does. They are, as actual points ‘view, natural. In their naturalness, however, none is superior to the others. All e actual points of view about shi-fei are equal from the point of view of heaven ‘reality. Of course, once we take this lofty cosmic view, we must grant the same status the poinis of view of animals. The cosmos assigns no special significance to the e or death of the human species. The goal of deriving guidance out of natural make-up fails, Zhuangzi shows. always assumes some prior interpretive shi (‘this:right’). Consider, for example, »w Mencius tried to get guidance out of the natural qualities of the heart-mind. ¢ presumes that the heart-mind should rule the other natural inclinations. He 1agines that sage-like taste is this:right and the fool's taste in cultivating his :art-mind is not this;wrong. The whole idea that the heart-mind can be a storted standard - the distinction between a natural and deficient development "the heart-mind - presupposes some standard for developing the heart. It must » other than appeal to the natural, organic heart. No mature heart-mind yields 3 6 - CLASSICAL CHINESE ETHICS a shi-fei judgement without having accumulated and assumed a prior arbitrary shi-fei standard of judgement. Thus Zhuangzi arrives at a sceptical view of guidance by language. We can show that natural or evaluative distinctions among types are correct only when we presuppose an evaluative point of view. All assignments of shi are indexical. What is shi (from one vantage point or intellectual lineage) is fei (from another). Anyone trying to resolve a difference takes a third point of view. Knowledge neither reaches all the way down nor all the way out. Our lives do end. To pursue what has no limit (perfect knowledge) with what has a limit (our lives) is foolish. Even if we had knowledge, we would not know it. We would not know if we had found the constant knowledge-ignorance distinction. So what conclusion would Zhuangzi have us adopt from this non-cognitivist analysis? (For an account of non-cognitivism in Western ethics, see Article 38, SUBJECTIVISM.) Zhuangzi seems, with some hesitation, to draw three practical conclusions. First, he extols flexibility and tolerance — openness to other perspec- tives. He seems aware that even this advice also presupposes a point of view ~ a point of view on points of view. Once we take Zhuangzi's perspective, we lose the motivation to condemn all alternative ways of guiding discourse and behaviour. Some new way of assigning categories and guiding action, e.g. science, might turn out to give us amazing powers — like the ability to fly. Being open to new conceptual schemes is characteristic of youth and flexibility. Being closed and rigid is characteristic of old age and impending death. Zhuangzi, of course, worries that our preferring life to death might arise from ignorance. Second, we can go with ‘the usual since it provides the basis for useful co- operation and interchange with others. We could practically, intelligibly require little more of a point of view or a way. Finally, given any guiding way we ingest — even being a butcher — we can hone it to artistic perfection. We can develop any skill to the point of second nature. We lose ourselves in our practice. When we have that trained intuition guiding our actions, our inner view is that an external force evokes and guides the skill. We can make any learned activity into skilled artistry and create satisfying beauty in practice. Of course to cultivate any such skill is to ignore others. Perfected at one thing, we are tragically condemned to be flawed at another. v Xunzi: pragmatic Confucianism Xunzi also learns from the analytic theories of language and sees in their analysis a way to resurrect Confucianism. As Hui Shi and Zhuangzi have argued, we have no natural basis for shi-fei distinctions. The only legitimate grounds of correct and incorrect language are the very conventions from which Mencius and Laozi fled. Only a fixed social pattern of shi-fei can make the use of names correct. The world cannot do it alone. Humans are linguistically social animals and the standards of acceptability are social. Our impulses impel us to adopt, preserve, and transmit such guiding conventions. Rather than undermine them, any responsible person 79