I

Sir Richard Sorabji has written a magisterial and salutary book. Magisterial, because – well, because it's by Sorabji. And salutary, because modern moral psychologists and ethical theorists really ought to be talking more about the conscience, and Sorabji here gently but insistently nudges their attention back towards that topic; he also generously bestows on them a rich treasury of resources to deploy in pursuing it.

And (we might ask at the outset) why don't moral philosophers talk more about conscience? One possible reason is because they think that to be committed to a theory of the conscience is to be committed to a theory of infallible intuition. This of course is nonsense, as Sorabji points out in his first chapter (32), citing St Paul's clear and repeated rejection of the view in 1 Corinthians (e.g. 4.4). The conscience may involve something like what people often mean when they talk about intuition, viz. direct awareness of (what are taken to be) moral truths. But that's not to say it's infallible.

Or again, perhaps moral theorists prefer inferred moral truths to perceived or intuited ones because it's tidier. Their general picture of moral knowledge is inferential, so they want as little direct perception in it as possible. For them the ideal state of things is that in Sidgwick (but, as Jonathan Dancy has shown, not Ross or those even closer to particularism), that only the first principles should be directly perceived, and everything else inferred from them. After all, if we allow direct perception to happen in general, we become vulnerable to its possibly picking up all sorts of things. And theorists are as afraid of vulnerability as they are of untidiness. The utilitarian majors on the untidiness point; she would like to position her theory at a level of abstraction well above that of any historically particular cultural formation, in order that it may the better speak to any culture – as she thinks, though as plenty of critics have pointed out, the abstracting move more usually has the opposite effect. The virtue ethicist, by contrast, often seems more concerned to avoid vulnerability. The ideally-flourishing person of virtue, with her bright eyes and glossy coat, is serenely untroubled by anything like conscience-pangs, an uneasy awareness of ‘things ill done, and done to others’ harm’, or the kind of inner division and admission of fault that a bad conscience involves.

Another reason that might be given is that conscience is a Christian notion; and that moral philosophers are pretty allergic to Christianity. Actually, as Sorabji shows, the first of these claims is mistaken: the notion of conscience is found much more widely than just in Christian cultures, though certainly Christian culture has focused it more clearly than any other. But the second, the point about allergy to Christianity, is more plausible. Nowadays many people regard it as an unforgivable indecency to use the word ‘Jesus’ in any grammatical case except the vocative; and it is not only students who know about Christianity, if at all, only via Nietzsche's critique of it, which is a bit like knowing about the Roman Empire only via the table-talk of Attila the Hun. To such people, who are not necessarily a minority, it may come as a surprise to hear that (as Sorabji shows at length) the New Testament has a moral psychology, since they have never been exposed to the possibility that the New Testament might be, or contain or imply, any seriously interesting philosophy at all.

Many of the resources that Sorabji offers us are, then, historical. Sorabji's canvas is as wide as ever, stretching from Aeschylus’ time to Tolstoy's, Gandhi's, and ours. And it is as finely-detailed as ever, with space in it for Olympiodorus (45), for Origen (33), for Salman Rushdie (209), and for ‘Four English Baptists [who] call for freedom of conscience, 1612–1620’ (128–30; their names are John Smyth, Thomas Helwys, Leonard Busher, and John Murton).

Indeed Sorabji's concerns range wider even than his title suggests. He does not, in fact, restrict himself to talking only about that particular moral-psychological formation that is properly called the conscience, and which Sorabji himself so interestingly anatomises in his first two chapters. Chapter 11 is about conscientious objection, and well before that, as we have already seen, Chapter 8 is about freedom of conscience. Neither notion is necessarily connected with the phenomenon of conscience as I would analyse it (though by now it has emerged that my analysis differs from Sorabji's; mine is narrower). There can be ‘conscientious objection’ e.g. to war, or debate about it, wherever there is a forceful moral conviction of any kind against war. Likewise there can be ‘freedom of conscience’, or debate about that, just wherever there are forceful moral convictions. A moral theorist such as Peter Singer can, I take it, have little more than a merely instrumental interest in the conscience properly so called (‘If it makes people do the right things, then good for it’)Footnote ¹ ; that does not make it an error to call Singer-style vegetarianism a conscientious stance.

Alongside these rich and various explorations in the history of ideas, and indeed running through them as an underlying thread of argument, Sorabji is also propounding a thesis in moral psychology about the nature of our concept of conscience, which he bases upon philological considerations to do with the root meaning of syneidesis or conscientia, which he thinks is ‘sharing knowledge with oneself of a defect’ (34–5):

There is much here that I agree with. There are also claims that I think questionable. In true academic style, it is these disagreements that I want to focus on in the rest of this review.

I disagree with two things in particular. The first is Sorabji's reduction of conscience to a belief. (It is ‘cognitive’, he sometimes says, as opposed – presumably – to conative. So here a standard-issue problem looms about how conscience can motivate. Sorabji does not see it as insuperable: see pages 35–6.) In contrast I want to say, as above, that conscience is a syndrome, a moral-psychological formation, a nexus of beliefs and concepts; so it is a type of historically modulated experience or phenomenon, that gives rise to beliefs – and also to motivations – of characteristic and particular sorts.

That conscience is not simply a belief can be established by looking at the history of the notion, as I will in a minute, but also, and more briefly, just by considering the phenomenological profile of conscience as we most typically experience it; which is something like this. You do something wrong;Footnote ² then having once done it, you find you are troubled by feelings of guilt and/or remorse. These feelings of guilt can become overwhelming, tormenting, unbearable, or they can just hang around in your mind like an unwelcome background noise or perhaps a mild discomfort. Violent or mild, they exert a psychological pressure on you, and the surest way to escape that pressure is to give into it. Giving into it feels like a relief. The relief here consists in examining what you did, and either admitting that what you did was wrong, or else – but this is usually harder to do – finding no fault in it and excusing yourself. Where admission of guilt is the correct solution as well as the psychologically easier one, it is, paradigmatically, not only that you have to admit that you did wrong: what you have to admit, to get relief, is that you knew it was wrong even at the time when you did it. It may also be necessary, in order to obtain relief, to take penitential or reparatory steps – to apologise to someone, to pay back the money you stole, or whatever it may be. Such steps are typically possible, though there can be special anguish of conscience where, atypically, they are not. Once completed, they bring relief to the burdened mind. No mere belief could do all this.

My second disagreement with Sorabji is that his account of conscience as ‘sharing knowledge with oneself of a defect’ is primarily individualistic: ‘Conscience as a form of personal self-awareness is not necessarily an awareness of others… conscience [is] very much concerned with what [is] or would be wrong for the particular individual in a particular context’ (36). This conception, I think, has both right and wrong in it. Let me try to separate them out.

II

I can begin to explain and substantiate my differences from Sorabji by glancing at Homer. Consider Odysseus’ encounter with the Suitors in the climactic passages of the Odyssey. As Homer depicts them, it seems obvious that the Suitors could do nothing with the question ‘Don't you feel guilty about squatting in Odysseus’ halls, eating up his substance, menacing his wife?’ For the Suitors these questions of conscience simply do not arise, any more than the question ‘Don't you feel sorry for the Saxons you are slaughtering?’ would occur to Viking raiders sacking Lindisfarne. They do these things because they can, and that is all. Come to that, Odysseus himself could do little more with the question ‘Don't you feel any remorse for killing the Suitors and their fancy-women?’ To Odysseus the Suitors are merely an infestation of exactly the type that is to be expected in one's home after a long absence abroad, and killing them is merely a part – a dangerous and difficult part, to be sure – of a long-overdue spring-clean. (Once they are all dead, Odysseus fumigates.) Or again, consider the famous debate between Achilles and Agamemnon at the start of the Iliad. What is in question here is certainly, in a sense, a question of desert, but it is not in the least about what is fair or just; it is about what is due to the honour of a hero. And that, as the sequel shows, is a very different matter, and one essentially conditioned, in part, by what the hero can get away with; Agamemnon, as he brutally observes, can get away with more than Achilles, and that is the point. Achilles may feel indignation, but Agamemnon feels no guilt; and that is not just because Agamemnon is a bully, but also because there is nothing we could do to get Agamemnon to see what we call guilt as something that it could be reasonable for him to feel about his own depredations.

There is, of course, one way in which we might challenge Agamemnon or the Suitors that they would find intelligible, and this is the challenge with which such monsters are sometimes presented in Homer: we might accuse them of hybris, of tempting the invidia – originally the evil eye, later the envy – of the gods by overstepping the human mark. But this challenge, when heeded, does not lead to guilt; it leads to self-protecting fear, of a kind that we will naturally describe as superstitious, and that indeed plenty of cultures still exercise against the evil eye. The point about the Homeric gods is not that they are consistent guardians of the moral order – they are nothing of the sort. (Athene does not intervene to tell Achilles to act more justly in his quarrel with Agamemnon, but more cunningly.) The point of the charge of hybris is rather that there is no telling what might provoke the gods’ arbitrary displeasure – but that we know that human grandeur often does; so grand humans, such as Agamemnon, had better watch out.

How do we get from this ethical world, to the world of Socrates with its instantly recognisable focus upon the notion of to dikaion, with its insistence that we should always do what is just and never anything else, and with Socrates’ notion of the daimonion who holds him back from all unjust actions? What changed in between Homer and Socrates? My answerFootnote ³ is that the Greeks became civic. As a matter of fact, I think we can see the change actually occurring in some classical Greek texts that we have; in particular, the Oresteia. Aeschylus's Agamemnon too is a swaggering pillager of a nearly pure Homeric type: he sometimes talks the language of eusebeia, just reverence for the gods, but the only thing apart from violence that really makes him pause is the danger of hybris. Yet his son Orestes is capable of being pursued by the Furies because he is guilty of the blood of his mother; and crucially, the only way for Orestes to shake off that guilt is in the civic court of Athens. Of course we should not smooth away too many of the differences from us, but neither should we miss the basic fact that this is an epochal change; what Aeschylus is showing us is that Orestes, unlike his own father, has a conscience. (A primitive and superstitious and half-formed conscience, quite possibly; and yet, a conscience.) And what Aeschylus is also showing us is that, even if (as Sorabji rightly insists) a conscience is an individual thing, still the precondition for its arising in the individual is a particular kind of social and political background.

Another way of getting in focus this epochal change between Agamemnon and Orestes, this process, or beginning of a process, that even in classical Athens had already led to something very like our own notion of conscience, is to turn to a writer from long after that process was complete: namely Kant. Here is some of what Kant has to say about the conscience:Footnote ⁴

This motif of a moral self divided into three sub-selves or homunculi raises a question about how literally Kant means us to see these three sub-selves – whether each of them is to be is taken seriously as something like a self in its own right. The motif and the ensuing question also remind us, of course, of the tripartite soul of Plato's Republic. But in fact the immediate background to Kant's remarks here is not Plato's triad of logos, thumos, and epithumiai. The immediate background is the moral psychology of the New Testament.

This can be quickly demonstrated by close reading of the texts above. Kant himself underlines the New-Testament background of his courtroom picture of conscience by virtually quoting the New Testament within the second passage from the Tugendlehre that I cite. Kant's bracketed phrase ‘(“before which his thoughts accuse or excuse one another”)’ is a clear allusion to Romans 2.14–15:

(Comparing the two German texts quickly confirms this. In the Luther Bible, the version that Kant would have known all his life, these verses read: ‘Denn so die Heiden, die das Gesetz nicht haben, doch von Natur tun des Gesetzes Werk, sind dieselben, die weil sie das Gesetz nicht haben, sich selbst ein Gesetz, als die da beweisen, des Gesetzes Werk sei geschrieben in ihren Herzen, sintemal ihr Gewissen ihnen zeugt, dazu auch die Gedanken, die sich untereinander verklagen oder entschuldigen…’; and at Ak. VI: 437–8 the last sentence of the second passage that I quoted above in English runs: ‘Das Bewußtsein eines inneren Gerichtshofes im Menschen (“vor welchem sich seine Gedanken einander verklagen oder entschuldigen”) ist das Gewissen.’)

The most prominent triad in the moral psychology of the New Testament is the triad that Kant identifies in the third quotation above: accuser (diabolos), defence-advocate (paraklêtês), and judge (syneidesis or conscience). This is indeed, as Kant brings out, a forensic triad, and it is one in which God and the devil of traditional orthodoxy have interesting roles to play. Diabolos, accuser, and ‘devil’ (or ‘Teufel’) are the same word, and paraklêtês, defence-advocate, is a familiar title of the Holy Spirit (see John 14.16). The picture is then that ‘the devil’ is the one who condemns us with as many accusations as he can find – the guilt aspect of the experience of conscience; whereas the paraklêtês pleads within us, not for condemnation, but for mercy and/ or innocence. And in this inner court it is the conscience that has to choose between the defence and the prosecution: the conscience is the judge.

In the New Testament the interplay of these three forces within us is simply one battlefield in a universal cosmic conflict between God and the rebel angels. The accuser (Satan), the advocate (the Holy Spirit), and the judge (Christ in glory) are all essentially external beings. They (can) appear as psychological forces within us only because they are already spiritual realities outside us, ‘in the heavenly places’ (ἐν τοῖς ἐπουϱανίοις, Ephesians 6.12). If and when the self is divided into three in the New Testament, it is because these spiritual forces are in conflict within the individual's psychology. This conflict can only be ended by confession and forgiveness. (Notice here that the NT Greek for ‘confess’ is homologeisthai, ‘agree with’, i.e. agree with the accuser, accept the justice of his charge: see e.g. Matthew 3.6, and cp. syneidenai as, at least according to some, the usual Greek word for ‘have a feeling of conscience.’) Unless of course it can be ended instead by the successful dismissal of the accuser as, on this occasion, a lying accuser (see 1 John 3.20: ‘For even if our heart condemns us, God is greater than our heart, and knows all things’). Caught between diabolos, paraklêtês, and syneidesis, the one self is exposed to three different kinds of heavenly influences, as it were to three ‘tidal’ pulls on it, just as, in the thought of the Graeco-Roman world in the first century AD, it was natural to suppose that all selves at all times were exposed to the kind of planetary influences that were studied by astrologers. (The language of Ephesians 6.12 is strikingly close to astrological language; though of course there are differences too.) The inner divisions caused by these different forces were felt as close to unbearable by Paul, documenting them in Romans 7.24: ‘Wretched man that I am, who will deliver me from the body of this death?’ For many others the conflicts evidently became actually unbearable. Perhaps we are meant to understand the afflicted souls that the gospels describe as ‘possessed by devils’ to be psychologies that are overwhelmed by a cacophony, or a legion, of accusing inner voices of tormenting guilt.

While taking over the New Testament authors’ forensic schema of accuser, defence-advocate, and judge to describe conscience, Kant of course does not take the courtroom schema, as they clearly did, also as literal cosmology. But the point is that he does at least take it – as the New Testament authors of course also do – as anthropology; as a description of how, as a matter of phenomenology, the human conscience works.

There is a continuity here that runs all the way from Aeschylus to us. In the Greek tradition, the notion of moral assessment as a courtroom process no doubt grew up first in Athens’ and similar cities’ own legal systems, as the Greeks ceased to be Homeric and became civic – as they ceased to even aspire to be swaggering bullies like Agamemnon, and sought to learn to live together. (And here is a fact of the first importance: the very same citizens who were spectators of the Athenian dramas presented in the Theatre of Dionysus were also members of the Athenian assembly that met on the Pnyx, and – some of them at least – also jurors in the trials that took place in the Athenian agora: three different contexts of civic life within seven hundred metres of each other.) From our point of view, the development within the Greek worldFootnote ⁵ of this process of civicisation as we might call it – or equally, perhaps, of civilisation – first comes to prominence in the closing scene of the Oresteia. The great change signalled by the courtroom scene in the Eumenides is the dawning of the possibility that that way of doing justice might not only become the dominant one for a culture, but might also become internalised: that we might learn to think about right and wrong within ourselves along the lines suggested by Aeschylus’ schema of accuser, defender, and judge. My own judgement, as Kant says, becomes something like a courtroom within me. That is a key part of how I come to have a conscience; it is also how I come to be a legislator, to myself and anyone else, of ‘universal law’. We might say that, in the dramatic process that gets us from the Agamemnon to the Eumenides, what happens is that the Furies become an internal voice: we move from the voice of the force of vengeance for blood-guilt, to the voice of guilt for blood-guilt.

The key moment in the emergence of conscience as a psychological phenomenon is the moment at which we move from the civic to the internal. That is also, I suggest (following Kant), a key moment in the process whereby autonomous moral agents appear: the moment when it becomes possible for individuals, without the aid of any civic forensic apparatus, to ask themselves whether what they are doing is justifiable, or open to any criticism that a court might make of it, or, to put it simply, morally right. (In another inflection of this development, it explains too why dikaiosyne, Plato and Aristotle's word for ‘justice’, is by the time of the New Testament rightly understood not as the name of one particular cardinal virtue out of four, but as a word that the first English translators of the Bible can reasonably render by ‘righteousness’. The sources of this development too, and the pictures of morality that go with this understanding of dikaiosyne, are essentially forensic.)

This is the sense in which I reject or at least want to heavily qualify Sorabji's view of conscience as being primarily individual. To the contrary, I want to suggest, conscience can only become an individual and psychological phenomenon, as of course it is in us, by first having been an external and juridical phenomenon, as it was for the ancient Greeks up to about the time of Aeschylus; it was only as humans living in civic society learned to internalise that form of assessment and criticism that it became individual and psychological.

If this is right it substantiates a sense in which Nietzsche, in the famous critique of Christian morality that he brings most clearly into focus in the Genealogy, has things almost exactly upside down about conscience. Certainly conscience leads to divisions within the self, as we learn to replicate within our own thought-processes the divisions that there are within a courtroom. But the appearance of these divisions in our psychology is not a sign of sickness or decadence, but of civicness, of (dawning) civilisation: they are a condition of a kind of moral maturity. Before they appear we are primitives. Blond beasts perhaps, but barely human. It is of course a limitation of our nature in one respect – we cease to be simple quasi-mechanical forces in an undifferentiatedly uncivilised universe, we become members of a political community who have to respect the forces that other members of that community represent. But that is not a loss of individuality as Nietzsche claims; rather, as Rousseau claims, it is a gain in individuality. For so we become agents rather than mere animals; not just Hobbesian momentums, but self-monitoring and self-regulating and self-legislating members of a political community. Learning to see the courtroom drama within ourselves is learning to make our own moral judgements, to be moral agents. And that is a fulfilment of our nature, if man is a political animal.

III

However, there is another aspect to the experience of bad conscience that is not captured by the courtroom metaphor, either in its Kantian or in its New Testament version, and which might indeed lay philosophers of the conscience, whether Christian or post-Christian, open to something like a Nietzschean critique; Sorabji touches on this matter in his discussions of the ‘terrorisation of conscience’ and of Nietzsche. This is an inner division, not between accuser, defence-advocate, and judge, but between the me who performs my wrong action and the me who repents of it.

The main point here is not that my view of that action's permissibility or rightness changes between two points in time. Certainly there is the experience of doing something and only later changing your values in a way that makes you realise now that what you did then was wrong. But the focal case of bad conscience is not this retrospective case. In the focal case, as I put it in my initial characterisation of the experience of conscience in Section I, it is ‘not only that you have to admit that you did wrong: what you have to admit, to get relief, is that you knew it was wrong even at the time when you did it’. The pain of conscience is the pain of admitting that when I acted wrongly I already knew it was wrong. When I experience the guilt of conscience in this full form, the question that tortures me is the question how it was possible for me-then to see the same moral data as me-now sees, and yet disregard them in what I actually do. This seems to have been a significant part of Augustine's self-torturing over the stolen pears in Confessions II; despite its absence from the NT's courtroom metaphor, it is clearly there when St Paul writes that ‘I do not do the good that I want to do; the evil that I do not want to do, that is what I do’ (Romans 7.19). For, as St Paul thinks, everybody does know right from wrong – in their hearts: consider here the implications of the famous phrase of Romans 2.14, ‘the writing of the law in their [the gentiles’] hearts’, a phrase which in its turn is clearly alluding to the Old-Testament promise of Jeremiah 31.31–34, that ‘in the latter days’ the God of Israel would write a new Torah, not on tablets of stone, but on his people's hearts.

This question how I can see the same moral data as two different times, evaluate them the same way, yet in what I do at one time respect and at the other disregard them, has a self-disintegrating momentum. In contemplating it I become, as Augustine and Paul did, a riddle to myself, an alien being, an incomprehensible monstrosity. One way to bring out what is puzzling here is a parity argument. There is nothing in the moral data that me-now can see, that was not equally available for me-then to see; how is it then that I have, at those two times, drawn (or at any rate acted on) two such different practical conclusions?

Alongside the puzzling moral epistemology of the clear-eyed wrongdoing that is central to the paradigm cases of bad conscience, there is the vital practical question of how I am to respond to myself when I find that I am guilty in the way described in such riddling cases. If I am to do justice to what my conscience is telling me, it can easily seem to become necessary for me-now simply to disown me-then. But this choice of simple rejection of the me, or the part of myself, that chose the bad comes at a high price. It seems to mean that I must reject the possibility of ever achieving full self-integration; it seems to mean cutting off part of myself from part of myself. Perhaps the lesson of Jesus’ famous saying ‘If thy right hand offend thee…’ (Mt 5.30) is that this kind of radical extirpation can be called for? In Romans 7.23 St Paul at any rate does not hesitate to draw the self-disowning conclusion: there he tells us that there are two whole systems of law in conflict in him, the Law of God and the Law of sin. This seems close to Manichaean, because in tension with the orthodox Christian teaching that there is only one divine lawgiver; and it seems very close to dualistic: in conflict with the orthodox Christian teaching that body and soul are an indivisible unity and that both of them are essentially good.

These theological flaws matter, because they mean that this solution to our riddle is not philosophically stable either. The question that sets our riddle is: how me-then could with open eyes and fully responsibly choose the bad that me-now rejects – and that me-then also understood was bad. And if we now say, apparently with St Paul, that me-then, in choosing them, was actually overborne by the forces of ‘flesh’ and ‘sin’, then it is no longer clear that the me that chose them was either fully me or fully responsible after all. Apart from its other difficulties the extirpatory alternative, the alternative of demonising me-then as a chooser who is simply wicked and as such should just be rejected, apparently makes no progress with this riddle.

One way to respond to this difficulty is to rephrase what St Paul is saying in the terms of a modern psychology informed by evolutionary considerations.Footnote ⁶ In humans, we might say, there is an overlaying of different systems of response and reaction. Going below the level of the fully conscious and rational mind, we may find ourselves drawn to respond to situations in ways that reflect our pleistocene or earlier ancestry more than they reflect the needs of our current social and ethical contexts. Our history bubbles up in us, often to destructive or even calamitous effect. That history is just as much a genuine part of us as is our best and most rational judgement of how we ought to act and live; nonetheless, it is not always a good thing for us to act on its promptings. We do not need, indeed we need not, to extirpate the caveman who still lives in us; but we do need to listen to him and understand him and tame him. Just as if we were in charge of a powerful wild animal – Plato's image of the lion within us is not far away here – there are energies in the caveman that we can use; but only if we learn to channel and direct his abundant energies, and are always aware of those energies’ dangerous potential.

To take this route would be to show how psychic reintegration might come back into focus as a real possibility on the, or a, Christian view of conscience, and how we might even keep a more or less complete recognition of the agent's responsibility for clear-eyed choices of the bad. In brief: the bad choices that me-then made really were my choices; for all that they can be choices that I do well to reject, because it is possible that me-then, though truly and genuinely me, was acting, perhaps not entirely realising it, out of the untamed Stone Age rather than out of well-thought-through civilised values. The key to the manoeuvre, and to the peacemaking that it offers to do between one version of the individual and another, is its recognition that we are not naturally integrated within, and not naturally (to use a word of Aristotle's) enkratic either. Integration is an achievement, and so is enkrateia (let alone temperance). We do not start off that way, and we have work to do to get there.Footnote ⁷