1. Ethics as Cartography

Where ought we to begin in philosophy? We ought to begin from where we are, comes the traditional answer. Where else could we possibly begin? But where, and indeed who, are we? Contemporary analytic philosophy offers us labels, if not answers, in response to these questions. Philosophy is contrasted with the ‘naïve’, the ‘folk’, the ‘pre-theoretical’, the ‘ordinary’, and above all, the ‘commonsensical’. In moral philosophy, with which the following pages are concerned, it is a well-established practice to begin reflection in, and on, an entity usually termed common-sense morality. This reflection can serve more than one end. It may be an effort at self-understanding, an attempt to gain a detached perspective on one's practices, an attempt to resolve a local puzzle or two, or the precursor to some programme of social criticism and reform. Most people rarely have occasion, and some lack the desire, to do any of these things. They are involved in the practice but do not habitually reflect on it, certainly not in the spirit of detachment that comes naturally to philosophers. In this, they are like the people Gilbert Ryle memorably compared to ‘people who know their way about their own parish, but cannot construct or read a map of it, much less a map of the region or continent in which their parish lies’.Footnote ¹

Ryle's is a metaphor of the philosopher as cartographer. Mapping, as he saw it, was to act as a prophylactic against the errors and omissions of older maps. He recognised that maps, being stylised representations of a more complex reality, could be agents of distortion. His most important work, The Concept of Mind, was devoted to a dissection of the Cartesian ‘myth’, which he claimed ‘continues to distort the continental geography of the subject’.Footnote ² Ryle's cartographical metaphor brings a number of fruitful philosophical questions into focus. Consider this passage from the first page of W.D. Ross's Foundations of Ethics (1939):

Consider also the opening lines of the seventh (and now standard) edition of Henry Sidgwick's The Methods of Ethics:

Note the dead metaphors from geography both authors use – a ‘starting-point’, ‘boundaries’. In Sidgwick as in Ross, the philosophical question is, roughly, what one ought to do, and it is taken for granted that this is the question. Sidgwick's opening sentence, quite characteristic of his prose, sets the tone for what is to follow in his great work, its multiple qualifications (‘sufficiently defined, at the outset, for the purposes of the present treatise’) look like scars of approximation, hinting at complexities left unexplored, even when there is no sense yet of what he may have omitted. They sound like the confessions of a cartographer embarrassed about how little his map looks like the landscape it is trying to capture, but pleading that (surely) it does the job alright. No doubt it does do the job, but unlike the jobbing cartographer working to a commission, Sidgwick's constraints were not imposed by someone else. The ‘purposes of the present treatise’ were his purposes, and our question must be why he took these to be his purposes. A few pages later, Sidgwick takes up the question of how much in human life an abstract philosophical treatment of ethics must, of necessity, leave out. We cannot, he says,

But is this the question, the question, to which ‘mankind generally require an answer’? Or is Sidgwick projecting onto the human race the peculiar anxieties of the high-minded Victorian? It is certainly a question, and other people than Victorians have had occasion to ask it of themselves. Philosophers before him had made similar observations. Here, for instance, are Kant's remarks in the Groundwork of the Metaphysics of Morals:

The argument here is subtler, but the language remains one of necessity, compulsion, and need, and cartographical metaphors persist (‘field’, ‘directions’). Moral philosophy is in this vision a response to a general human need for guidance and direction. We need maps in morals, as we do in life, to get around – when we do not already know what to do, or find ourselves lost while travelling. This is true, and these considerations go a long way towards justifying both maps and something like Sidgwick's conception of the central questions of ethics. The trouble for these conceptions is that we are not always lost, and we are not always at a loss for what to do. In such circumstances, the need for maps is less urgent, and one's reasons for consulting a map – when there remain any – come from a different source: not directions but (self-)understanding.

In trying to elaborate this metaphor, we might consider these elegiac lines from the British literary critic and nature writer Robert Macfarlane about the effects of maps on perceptions of landscape:

Macfarlane goes on to explain why this matters:

If one spends long enough on motorways, the rest of the landscape might come to seem, as the history of the industrial revolution has shown, not an independent source of value and interest, but an obstruction and an inconvenience we should be rather better without. As the road map ‘encourages us to imagine the land itself only as a context for motorised travel’, it is possible that the ethical treatises quoted above encourage us to imagine ethical experience in terms of questions about our obligations, our reasons for action and so forth. And as the road atlas leaves out ‘the fells, the caves, the tors, the woods, the moors, the river valleys and the marshes’, so does ethics of this sort leave out, or reduce to generic symbols, much in our experience that has little to do with duty or reasons for action: this is, speaking broadly, the stuff of moral psychology, our habits, dispositions, sentiments – what used to go under the label of ‘character’. In Sidgwick and Ross, and even in Kant where they are allowed to put up more of a fight, the stuff of moral psychology is ultimately argued or stipulated away as not part of the real questions of ethics as they conceive it.

Yet, in this tradition of ethics, as in this style of cartography, there is this to be said in the case for the defence: maps (or ethical theories like Sidgwick's) do what they set out to do; why cavil at them for failing to do something else? The motorist's map is no good for the rambler or the recreational map-reader, but there are other maps for them with all the rich detail of the landscape beyond the motorway. Indeed there are, and there is much to be said for the simplifications of the road atlas, but the perceptions encouraged by the road atlas do not always leave the landscape as it is. The stuff of moral psychology is unlikely to go away because some philosophers ignore it. Yet, there is the risk that the preoccupations of academic philosophy leave the greater part of the ethical landscape seeming somehow disreputable and dispensable. An excusable theoretical simplification turns out to conceal an insidious ethical vision.

This is by now a familiar line of criticism of the mainstream of contemporary ethics, and critics of this stripe have tended to turn to Hume and Aristotle for an alternative conception. I do not intend to repeat those criticisms, but to historicise them. The narrowing of moral philosophy's focus to questions about action and obligation may be given a historical location – or two – if only as a convenient shorthand for a vastly more complicated historical process. But this is not the only narrowing that Ross's stipulations involve. The ‘moral consciousness’ that he takes as his starting point consists in ‘a large body of beliefs and convictions’ about what we ought to do.

This is, he continues, ‘the time-honoured method of ethics. It was the method of Socrates and of Plato; … It was the method of Aristotle, and has indeed nowhere been better formulated than it is by him. … Kant's method was the same’.Footnote ¹⁰

Sidgwick seems to have shared both this conception of ethics and this view of its genealogy. In a passage of intellectual autobiography written late in his life, he writes of the moment in his youth when he found himself less than satisfied with the utilitarianism of Mill and Bentham, and therefore, less inclined simply to dismiss their ‘Intuitionist’ opponents as votaries of prejudice and anti-intellectualism:

In both these philosophers, we find what I shall call a doxastic conception of common-sense morality – roughly, the claim that what is essential to common-sense morality is the set of shared beliefs (convictions, intuitions, judgements) it contains; both trace the conception back to Socrates. Ross's ‘moral consciousness’ is a set of beliefs; Sidgwick's common-sense morality is ‘what “we” … think’, ‘current opinion’. Now compare Kant:

The language is less obviously doxastic (though ‘judgement’ in early modern philosophy is often roughly coextensive with ‘belief’ and ‘opinion’ as used in the subsequent tradition), but the appeal to – and critique of – common-sense morality is again traced back to Socrates. The view persists in an influential modern work such as John Rawls's A Theory of Justice, where Rawls declares that ‘Moral philosophy is Socratic: we may want to change our present considered judgments once their regulative principles are brought to light’Footnote ¹³ – Rawls's view of Socrates is, unsurprisingly, Kantian through and through.

The doxastic conception of common-sense morality is – as will generally be agreed even by those who use it – a useful approximation. It can be helpfully contrasted with the more detailed account of common sense it is approximated from, let us call it the dispositional conception: common sense not as a set of beliefs but a set of dispositions – entrenched habits of thought, sentiment, action and concept-use that do not easily admit to being reduced without remainder to a set of beliefs. Certainly, the English phrase common sense is often used in the sense of a skill, of a certain quotidian but indispensable kind. The anthropologist Clifford Geertz nicely captures the ordinary use of the phrase when he writes:

A description of common-sense morality might restrict itself to its doxastic components, as the map of a landscape might restrict itself to paved motorways. But we should be clear in our minds that this involves a choice, and an omission, with the power to shape perceptions of the phenomena. If a landscape has motorways in it, it is because they have been put there. Sometimes they are built on older roads built by (say) Romans from more than a thousand years ago, themselves the formalisation of informal paths formed by walkers repeatedly treading on certain patches of grass. In that sense, they will often reflect some older, more basic, feature of the landscape – a ridge, a plateau, or a slope. But it is always a question when this is the case. Similarly, the particular dispositions, of forming convictions or judgements about the rightness and wrongness of actions, are often already there in common-sense morality. But it is an open question whether these dispositions, once they have been overlaid with their doxastic interpretations, retain their original shape, and indeed, to what extent the psychological landscape is improved by the overlay. I am suggesting, in other words, that we should not without argument assume common-sense morality to have the structure attributed to it by philosophers who assume the doxastic conception, a fortiori a doxastic conception committed to the centrality of a particular set of beliefs in it (to wit, beliefs about right and wrong action).

2. Sidgwick and the Origins of the Doxastic Conception

The doxastic conception of common-sense morality has considerable attractions, not least the convenience of having to hand a bunch of propositions that might serve as the basis for further reflection. This is the use to which Ross and Sidgwick, as the remarks of theirs quoted above suggest, put it. In this respect, the situation in ethics since Sidgwick roughly, but only roughly, parallels that in recent philosophy of mind, where the phrase ‘folk psychology’ can mean, among other things, either ‘a particular set of cognitive capacities which include … the capacities to predict and explain behavior’ or ‘a psychological theory constituted by the platitudes about the mind ordinary people are inclined to endorse’.Footnote ¹⁵

It is something like the ‘platitude conception’ that gives Sidgwick's The Methods of Ethics its structure: the idea that a careful and rigorous examination of ‘the morality of common sense’ that looked beyond its superficial diversity might well find a deeper unity in basic, self-evident axioms of practical reason that could command universal rational assent. In aiming to do this, Sidgwick set the terms of virtually every important debate in twentieth-century Anglo-American moral philosophy. His influence is evident in the writings of John Rawls, Derek Parfit and Peter Singer, all of whom speak of Sidgwick with great respect;Footnote ¹⁶ the grand, synoptic form of the former philosophers’ most substantial works have much in common with Sidgwick's aspirations.

Part of Sidgwick's influence can be observed simply in the ease with which his arguments can be assimilated into contemporary philosophy; this is certainly not the case with a work such as F.H. Bradley's Ethical Studies, published not long after The Methods of Ethics, but well-armoured by its style against any attempt to conscript its arguments into some contemporary debate.Footnote ¹⁷ Modern Anglophone moral philosophy is Sidgwick's child, if it is anyone's, and this influence can be seen in the contemporary ubiquity of a set of philosophical methods and styles that Sidgwick did so much to develop and make rigorous, among them the attempt to elicit from ordinary moral thought, discourse and practice a set of beliefs, or considered judgements, or intuitions, that are then put to the test, modified in light of the demands of other such beliefs, judgements and intuitions on the one hand, and the demands of more general principles of theoretical betterness on the other. Rawls's label for this general style of moral philosophy, ‘reflective equilibrium’, is the closest thing that contemporary ethics has to a standard method, a method it shares, to some degree, with other branches of philosophy.Footnote ¹⁸

In addition to this idea of method, we find in Sidgwick a highly distinctive style of thought and writing, one characterised by a relentless aspiration to detachment from his own beliefs and sentiments, and a tendency to push (what we should now call) the principle of interpretative charity to almost fanatical lengths. Above all, we find in Sidgwick an attitude to common-sense morality just as characteristic of modern moral philosophy. This attitude stems from the idea that there is nowhere else to start in ethics but in the beliefs taken to be constitutive of common-sense morality, but that it must always be an open question whether philosophy will leave common sense where it was. As Derek Parfit briskly puts it, ‘Philosophers should not only interpret our beliefs; when they are false, they should change them’.Footnote ¹⁹

A philosophical tradition could do rather worse than to have Sidgwick for a progenitor. Sidgwick, famously self-effacing, did not make any grand claims for his own originality, and seems not to have sought influence or celebrity; indeed, his views committed him to reticence about his heterodox ethical and religious (that is to say, irreligious) convictions. He saw himself as belonging to an intellectual tradition, and wrote with unfeigned respect of his indebtedness to his predecessors in that tradition, one that – as we have seen – he supposed to go back to Socrates. If his own part in that tradition, both as contributor and historian, have long been underrated, they are in good part consequences of that reticence.Footnote ²⁰

One example of Sidgwick failing to receive his due is the phrase, ‘common-sense morality’, itself. So common are references to it in modern moral philosophy that it is possible to doubt that there was ever a point at which the phrase needed to be introduced into philosophical discourse.Footnote ²¹ There is one sense, as we have seen, in which the doubt is justified: as long as there have been philosophers, there have been non-philosophers (variously, and ambivalently, ‘the naïve’, ‘the vulgar’, ‘the folk’), and they have had their ways of carrying on, and their ways of talking about their ways of carrying on. Equally, there was much talk of a ‘philosophy of common sense’ among philosophers of the so-called Scottish Enlightenment. Writers before Sidgwick, such as his bête noire William Whewell, had employed the general idea in moral philosophy, but it is Sidgwick who gives it the pithy name we continue to use:

Note how, in the course of a single paragraph, the ‘Morality of Common Sense’ – ‘the moral generalities that we obtain by reflection on the ordinary thought of mankind’ – has lost both the inverted commas and the capital letters, assimilated thoroughly into the parlance of philosophy.Footnote ²³ Book III of The Methods of Ethics is where Sidgwick spends longest giving an account of the generalities implicit, or sometimes explicit, in common-sense morality. Time and again, he finds these beliefs wanting, but is at pains to insist that he has done his best to represent common-sense morality in its best light. If it continues to seem inadequate, he implies, it is because it is. And if Sidgwick's manner is inquisitorial, we should remember that he too stands charged with the rest of us:

Sidgwick does not think this style of philosophy an innovation of his. His own literary style, detached and largely affectless, is Aristotelian; so is his method, of giving us ‘what “we” … think, ascertained by reflection’, combining a stern impartiality with the guarded solidarity suggested by his pronoun. This is opposed to Aristotle's most immediate inspiration, those among the dialogues of Plato that are commonly taken to represent the practice of the historical Socrates. Where Aristotle ‘ascertained’ common-sense morality ‘by reflection’, in Plato, it is ‘elicited by interrogation’. This characterisation of Plato and Aristotle marks a methodological contrast between them, but not one to which Sidgwick attributes much significance. Aristotle, he says,

When Plato's Socrates wished to have the content of common-sense morality explicated, he asked people; Aristotle, by contrast, seems to have thought he could do it by himself, and Sidgwick follows him in this. Neither seems to have thought their armchair anthropologies raised any special difficulties.Footnote ²⁶ Sidgwick for one is evidently anxious to fend off accusations that his picture of common-sense morality is in any way idiosyncratic: ‘I wish it to be particularly observed, that I have in no case introduced my own views, in so far as I am conscious of their being at all peculiar to myself’.Footnote ²⁷

One might be inclined to be suspicious of Sidgwick, as of any unelected spokesman, however insistent he be of his integrity. But even the methodologically uncomplicated Socratic method – as it may well seem – raises its own difficulties, difficulties internal to the practice as well as more general problems. In brief, the problems can be captured in this question: given Socrates’ well-known scepticism about people's claims to self-knowledge, why should they be treated as authoritative spokespersons about their beliefs?

3. Socrates: Plato's and Sidgwick's

Plato's Socrates, as we well know, engaged in a series of conversations with ordinary interlocutors about – among other things – such everyday notions as those of courage, beauty and justice. He sought definitions from his interlocutors, in part because they claimed, usually with unfounded confidence, to know about such matters. When they replied to Socrates' questions, it is usually hinted that they were not giving voice to a merely idiosyncratic opinion; they became, often without knowing it, spokesmen for the conventional beliefs, attitudes and practices of their society. The trials those beliefs and attitudes underwent at Socrates' hands were more than ad hominem attacks on their possessor's cogency, though they certainly were at least that; they were instances of what we may well call critique.

It is usually claimed that the Socrates in Plato's dialogues drew the tools of his critique not from some alternative heterodox morality of his own devising but from within the morality that was the object of critique; his methods were, in the jargon, ‘peirastic’.Footnote ²⁸ Much progress could be made simply by exposing the internal inconsistencies and ambiguities of the beliefs that constituted that morality. It is a conspicuous characteristic of the dialogues of Plato's usually classified as ‘early’ and ‘Socratic’ – that is to say, written early in Plato's career and perhaps for that reason likelier to reflect an attempt at representing the practice of the historical Socrates – that they lack the methodological self-consciousness of his ‘late’ style, a self-consciousness and reflexivity quite overwhelming in such dialogues as the Sophist and Statesman whose subject matter is their own methods. The figure of Socrates in the early dialogues does seem to have a method, after a fashion, but the more assiduously modern scholarship has tried to reduce that method to a set of formal criteria, the less they have been able to capture the freewheeling, improvisatory quality of these dialogues.

In these dialogues, Socrates asks questions and exposes inconsistencies in his interlocutors' answers, but that is not all he does. He also tells jokes and stories to his interlocutors, makes edifying remarks to them, teases them, even flirts with them, sometimes outrageously. Certainly he elicits their beliefs from them by interrogation, but that is not all he elicits. One of Plato's many literary gifts was one for characterisation. His Socrates is a vivid creation, as are many of his interlocutors. Their beliefs are certainly part of their character, but in Plato's presentation, they are embedded in a wider range of dispositions of behaviour, thought and sentiment. In this sense, Plato's form is superbly well suited to the substance of his moral philosophy, his dialogues able not just to announce the results of an inquiry into moral psychology but to dramatise that inquiry itself by showing us moral psychology embodied in individual character(s).

Aristotle's works, which Sidgwick insightfully saw ‘as a kind of Socratic dialogue formalised and reduced to a monologue’, do not present a moral philosophy radically at odds with Plato's. Certainly there are differences in the underlying metaphysical picture, but Aristotle shares with Plato a moral philosophy entwined with the fundamental categories of moral psychology – the different forms of human motivation, rational and non-rational – and presented as part of a picture that contains also a theory of moral education and a vision of politics. Yet, the form of Aristotle's surviving works (his dialogues are lost), marked as Sidgwick saw by the conventions of the lecture-room, do not allow the happy congruity between form and substance we find in Plato. Aristotle was as concerned as Plato was with human character and the role that non-rational motivations play in the ethical life. But his descriptions of method, however continuous with those of Plato's Socrates, do leave something out.

An orthodox understanding of Aristotle's most distinctive method of doing philosophy finds a succinct statement of that method's animating principle in Book VII of his Nicomachean Ethics:

‘Endoxic’ inquiry, as we might call it, may be represented schematically as a three-stage procedure. First, one gathers a set of endoxa, or reputable opinions, on some subject. This includes beliefs held in explicit form by all, or nearly, all people, the implicit presuppositions of their behaviour, the commitments behind their language-use and discourse, and even the minority opinions of people with some claim to be taken seriously. Second, one looks among the endoxa for infelicities. These may include inconsistencies or tensions between two or more reputable opinions, or instances of vagueness, ambiguity, or simple implausibility. Finally, one attempts to resolve the infelicities discovered in the previous stage by restating the endoxa so that they are free of ambiguity or vagueness, and by abandoning some of them so as to leave behind a ‘maximal consistent subset’ of the most important of the original endoxa.Footnote ³⁰

What is missing in Aristotle that is present in Plato is some visible sense of what is involved in the first step in this process: the gathering of the endoxa. In Aristotle, this is always a fait accompli – and why not? Surely the task poses no special challenge to a moderately self-conscious, intelligent mind with some experience of life – or so he may have thought. But in Plato, the gathering of opinions, or something like it, is itself part of the business of philosophy. The dialogues themselves do not of course do this so much as present it being done; they are dialogues, literary artefacts showing a high degree of authorial design, and not a sociologist's questionnaires. Yet, part of the dialogues' evident authorial design involves the dramatisation of the processes of philosophising, even if Socrates' philosophising stands at a remove from Plato's own.

In moving from Aristotle to Sidgwick, one leaves out much that is important and interesting in the history of philosophy, but it is how Sidgwick in some moods constructs his own intellectual genealogy. As one footnote has it, ‘through a large part of the present work the influence of Plato and Aristotle on my treatment of this subject has been greater than that of any modern writer’.Footnote ³¹ It is not that he did not take the other figures of the western canon seriously: he discusses Hume and Butler with some respect, and the influence of Kantian notions of universalizability on his moral philosophy is obvious. Still, on matters of method in particular, it is to the two Greeks that he constantly defers, more often than to any recent predecessor or contemporary.

However, for all that he claimed for his philosophy a classical pedigree, Sidgwick's moral philosophy was in several ways deeply unclassical. In part, this was an aspect of his self-conscious modernity. His idea of a ‘method of ethics’, as he defines it in the opening sentences of his great work quoted earlier, was ‘any rational procedure by which we determine what individual human beings “ought” – or what it is “right” for them – to do, or to seek to realise by voluntary action.’Footnote ³² This was, as he would later note, a question conceived in distinctively modern, rather than ancient, terms:

His Outlines of the History of Ethics for English Readers, a short book revised from his article on the history of ethics for the Encyclopaedia Britannica, has a sophisticated account of the part played in the history of late antiquity and mediaeval Europe by ideas of morality with their roots in Judaism. The Methods of Ethics, though its classical debts are large and numerous, is not the work of a reactionary Philhellene. It is too dissatisfied with the apparent triviality of ancient answers to the question he thought it the aim of moral philosophy to address:

Even in Aristotle's more systematic presentation of his ethics, much work is left to that demanding virtue, ‘phronesis’ – prudence, wisdom, or judgement – which does not, in Aristotle's discussions of it, seem to be amenable to statement in the form of a simple rule. There is little to be said in defence of Plato and Aristotle against the charge of triviality once Sidgwick's demands of a method of ethics are accepted as legitimate.

The question of whether they are legitimate has been raised off and on in modern ethics, like a discordant note sounding every so often in an otherwise harmonious orchestration – by, among others, such figures as Iris Murdoch, Elizabeth Anscombe, Alasdair MacIntyre, and Bernard Williams. For the most part, however, Sidgwick's demands have been accepted, and indeed, taken to be constitutive of the discipline of ethics itself. It is certainly the standard way of teaching ethics to undergraduates in Anglo-American philosophy departments. Sidgwick took his central task – roughly, finding a rational procedure to determine what one ought to do – to be one that could be accomplished, and most philosophers in the Anglophone tradition have followed him in thinking this both possible and of central importance. There remains some place for moral psychology in ethics as thus construed, but the deep connection between philosophy and psychology in Plato and Aristotle has been effectively severed. Sidgwick's picture of ethics gives it an autonomy from psychology that it did not have in the ancient world.

The ‘psychological turn’ in Plato, as we may well term it, comes from the contexts in human life that his dialogues represent as motivating ethical reflection. These are, centrally, the needs of education and those of politics, in particular, of political stability; the Meno, Protagoras and Republic exemplify these preoccupations in their purest form. It is natural that an adequate answer to these demands would prompt deep engagement in facts about the human dispositions generally regarded together under the traditional label of ‘character’. This remains true in Aristotle: his works lack the dramatic contexts of Platonic dialogues that make the link apparent, but his work is too full of asides on pedagogy and politics to be regarded as an abandonment of the Platonic position on the matter. If Plato and Aristotle have – to use a deliberate anachronism – an ethical theory, then it is a theory whose structure is responsive to the structure of the psyche. In Sidgwick, however, the motivations for ethical reflection seem rooted in the need for guidance in practical deliberation and his ethical reflection delivers strictures with a correspondingly austere, and largely action-focused, structure. The most general answer to the ethicist's basic question in the post-Sidgwick tradition takes the form ‘An action is right iff it…’. When set aside Aristotle's account of virtue in Nicomachean Ethics II.6 as a disposition consisting in a mean defined by reference to the reason with reference to which the prudent person would define it, Sidgwick's ideals embody the modern hope that deference to persons, even as heuristics, indeed deference to anything other than rationality itself, can finally be done away with.

5. Concluding Remarks

In Plato, I have suggested, the structure of the ethical theory is well suited to both its subject matter and its literary expression in the complexity, richness and detail of its psychological descriptions. In Aristotle, the structure of the theory remains well suited to its subject matter though the relative dryness of the literary presentation makes it less easy to see that this is so. In Sidgwick, we find again a harmony of theoretical structure, subject matter and literary style, but this has been achieved by a reconstitution of the subject matter itself: not the open-ended question one finds in Plato's Socrates – ‘how should one live?’ – but the more specific question of what one should do and why.

None of this has shown, or tried to show, that Sidgwick's question is a bad one for ethics to focus on. It is, rather, an attempt to place Sidgwick's question into the broader geography of human concerns to which the ancient philosophers were sensitive, and to which a heterodox strain in ethics after Sidgwick has striven to return. I have tried in this essay to draw attention to the historically Janus-faced quality in Sidgwick: the way in which he looks backward, to the origins of western moral philosophy among the ancient Greeks, and the way in which his conception of ethics – Socratic practice reinterpreted to meet the needs of philosophy in modernity – points forward to the tradition of Anglophone academic ethics he did so much to create.

A recapitulation of the argument of this paper: Mainstream moral philosophy today is characterised by a critical examination of common-sense morality, interpreted as a set of beliefs elicited by reflection – e.g. on what judgements one is inclined to make about difficult (usually imagined) cases. This approach can be shown to have its clearest historical exemplar in Sidgwick's The Methods of Ethics, where he claims to be employing a basically Aristotelian/Socratic method in ethics. Sidgwick's understanding of Socratic method reconstructs it in terms of the examination of the ethical beliefs elicited in the course of dialogue. In emphasising the doxastic elements of Socratic inquiry, it draws attention away from Socrates' general concern with the basic psychological dispositions of which beliefs are one expression. In doing this, it effectively narrows the scope of philosophical ethics, for better or worse.

I do not claim here to have answered, or even to have addressed directly, the question of how we should regard Sidgwick's reconstruction of the modern ethical project in relation to common-sense morality. I do claim, however, to have raised it and given some reasons to think it important. I have tried to show that a certain methodological orthodoxy in contemporary Anglophone moral philosophy is not an orthodoxy constitutive of the discipline, because its status as orthodoxy is a contingent matter that is best understood in historical terms – in particular, as a result of Sidgwick's formative influence on the tradition in which such a method is the orthodoxy. Further, I have assembled considerations in favour of the view that there was a period in the early history of moral philosophy – the time of Socrates, Plato and Aristotle – when such a method was not the orthodoxy. Finally, I have tried to suggest that it is a consequence of our living in a philosophical landscape structured by Sidgwick's conclusions that we are relatively mindless of Sidgwick's influence on how we perceive that landscape.