Psychiatric Disorder and the Reactive Attitudes

There are some notoriously difficult questions about when, and how far, people with psychiatric disorders are responsible for their actions. One reason for these questions being so difficult is our limited understanding of these disorders. Another is that some conditions, such as compulsions and addictions, have blurred boundaries. So such questions as “was he responsible?” and “could she help doing it?” do not always admit of “yes” or “no” answers. These complications make matters difficult enough. But there is a further set of problems, to be the focus of attention here, raised by the fact that some psychiatric and neurological conditions cause long term changes in character and personality. How should the legal system respond to antisocial actions flowing from these changes? And, at least as important, how should those who know such people respond to these deep changes in them?

1. Excuses as a defence of character.

It is worth for a moment leaving aside psychiatric disorder, and looking at responsibility and excuses in general. It is widely agreed that Aristotle was right in seeing ignorance and compulsion as the two central classes of excuses. That ignorance and compulsion have this dominant position reflects the way our judgements of responsibility are guided by an interest in the person’s motives, values and character. If a man knowingly steals money, this reflects on his honesty in a way that being oblivious of a banknote among a loose pile of papers he is picking up does not. If a woman hands over the key to the safe because the robber is armed and dangerous, this reflects fear rather than any desire to help with the robbery. What matters is the motivation behind the action, and understanding the motivation contributes to our picture of the person’s values and character. Pleas of ignorance (“He did not notice the banknote”) and of compulsion (“She was forced to hand over the key”) defend a person’s character. If accepted, they defeat the claim that he is a thief or that she was complicit with the robber.

When mental disorder counts as an excuse, it is often because it supports a claim of ignorance or compulsion. Often the psychiatric versions of these two pleas raise special problems of their own. Someone may be ignorant of what he is doing because of his deluded state, but there are often difficulties for an observer, whether a psychiatrist or anyone else, in deciding how far he was in fact deluded. And the particular cognitive distortions that are characteristic of delusions often seem strange enough to raise difficult questions of interpretation. Psychiatric states may involve a degree of compulsion, but again the “inner” compulsion of addiction or of obsessive-compulsive disorder is often much harder to establish than the external and visible compulsion when the robber brandishes a gun.

There are well-known philosophical models for thinking about cases of conflict, such as unwilling addiction. One is Harry Frankfurt’s use of second-order desires. (1.) His “wanton” is someone with only first-order desires: someone who only wants to have things, do things, etc., but does not have any preference for having some desires over others or about which desires win out over others. Most of us are not wantons, but have second-order desires and preferences, being glad not to have the desires of a paedophile, or wishing our desire to exercise was stronger than our desire to sit talking over coffee. The unwilling addict is characterized by having a second-order desire to escape the dependence, which conflicts with a first-order desire to take the drug. Where the first-order desire wins out, the addict is unfree, because the real self, identified with the higher-order desire, is defeated.

One problem for this view is why a person’s real self should be identified with higher-order desires rather than with first-order desires. This has led to an alternative strategy. One version is Frankfurt’s own: sometimes a person just does “decisively identify” with one set of desires rather than another. Another version is shared by Gary Watson and Charles Taylor: what matters is not different orders of desires, but the contrast between desires and values. I am free when the desires that win out are in harmony with my deepest values, and I am unfree when dominant desires and deepest values are in conflict. There is considerable plausibility to this approach, although it leaves open difficult questions about how the distinction between desires and values is to be drawn and about how, when values are in conflict, we are to decide which are deepest. (2.)

2. “It wasn't the real me": character-distorting conditions.

In these cases of conflict, a “psychiatric” condition seems to have taken over part of a person. Even more difficult are the issues raised when a psychiatric disorder seems to take over and distort someone’s whole personality and character. How should we think about such cases?

On the one hand, accepting or rejecting an excuse is largely about how far an apparently bad action does or does not reflect the person’s character. Excuses protect character by disconnecting it from objectionable actions. So it seems strange to allow as an excuse the claim that psychiatric disorder has changed a person’s character for the worse. Does not such a plea mean that the person’s present character is open to precisely the sort of criticism that the objectionable action suggests?

On the other hand, where a medical condition brings about a radical transformation of character, we are inclined to say that “it is not him but his illness”. The easygoing, friendly man we all knew for many years becomes unrecognizably difficult, stubborn, and bad-tempered as his dementia progresses. Or a woman’s whole personality is changed for the worse when she gets a brain tumour. We may feel that blaming them for what they now do shows a lack of understanding. The behaviour is surely down to changes in the brain over which they have no control. The same psychological changes may lie in wait for any of us, if our brains start to go the same way.

Those close to people who seem greatly changed by dementia sometimes say such things as “He isn’t the man I married”, or “She is a different person now”. At one level we all understand what such comments mean. But at another level they raise philosophical problems about the boundaries of personal identity.

Those who make the comments may not be clear how literally they mean them. If they are meant literally, some philosophers will accept them and others will reject them. Those who see personal identity as a matter of physical continuity will think in terms of the same person having a changed psychology. Those who take a more psychological view of personal identity will be more willing to see discontinuities of memory or of personality as raising doubts about the person being the same. Others will say that the earlier person may survive to some degree. Here these issues will be sidestepped. I will talk in terms of the same person having a changed character and personality, but the discussion could be translated without loss into the language of the old and the new person, or into the language of earlier and later selves.

A. Aristotelian Common Sense.

How does the thought that this person has changed so radically bear on his or her responsibility for the actions shaped by the new personality?

One possible answer is that it makes no difference at all. In other contexts, people often change radically. People have religious conversions. They change partners and jobs and they emigrate. They take up drugs or enter convents. Any of these upheavals may change character and personality sharply. But the immigrant or the nun are still held responsible for what they do. The new actions flow from the new personality, and it is still appropriate to evaluate the person on the basis of them. The novice nun’s piety (or the new immigrant’s patriotism) is not less relevant to our picture of them because they are newly acquired. So why should the aggression of the person with dementia or a brain tumour be different?

A reply to this question could appeal to the way in which, as Nietzsche saw, we at least partly create ourselves. Nietzsche’s version of self-creation was a highly self-conscious one, making people like architects or landscape gardeners of their own character and personality. He was right that many of us do care about the sort of person we are, and do to some extent try to grow in one direction rather than another. But for most people self-creation is less dramatic than in Nietzsche’s picture: more a matter of trying to be the sort of person who answers letters, or who gives up a job in public relations because you don’t want to be the kind of person you see you are becoming. And less dramatic still is the often completely unconscious self-creation that comes about when repeated actions become character-shaping habits.

Aristotle took the view that our character is something we are responsible for, because it results from our own freely chosen actions. Pointing out that we punish people who, through carelessness, are ignorant of things they should know, he says that we assume that they have the power to take care. He then considers the objection that someone may be a careless kind of person. He replies that becoming such a person is their own fault: “they are themselves by their slack lives responsible for becoming men of that kind, and men make themselves responsible for being unjust or self-indulgent, in the one case by cheating and in the other by spending their time in drinking-bouts and the like; for it is activities exercised on particular objects that make the corresponding character.” He accepts that, once we have a certain character, we may be unable to change it at will, but says that it was still pour own fault that we acquired it. (3.)

This robustly “common sense” Aristotelian view is broadly presupposed by most legal systems and by the everyday morality of praise and blame. There is the presumption that actions are voluntary unless there is some special reason to think that they are not, together with the view that our character is also our responsibility because it is the residue left by past actions. This also applies to the character changes resulting from some voluntarily chosen major upheaval in life. The nun’s new piety and the immigrant’s new patriotism both stem from free choices they have made.

But the stubbornness of the person with dementia contrasts sharply with this. We do not do things that predictably bring on dementia. This disaster, despite being inside the body, comes on us “from outside” in the sense that it is not under our present, or even past, control. The response that the new personality is not the real person but a product of the illness seems justified.

This response also fits with our concern about being one sort of person rather than another. There has been a good deal of debate about living wills and euthanasia, in which those who think it right to respect someone’s advance directive not to be kept alive in certain states appeal to the interest people have in the pattern of their lives. The demented person I become in twenty years’ time may seem quite content and show no wish to die. But if I now hope to die earlier rather than include a few extra vacant years of dementia in the pattern of my life, this seems something to take into account. Leaving aside whether it is enough to justify acting on the advance directive, this interest in the kind of person I am over a lifetime gives me reason to hope that people will see the grumpy surliness as reflecting the dementia rather than reflecting me.

B. The Case of Schizophrenia.

Dementia mainly (though not always) comes on at a relatively late stage of life. This makes it easier to see the demented period as a coda: something after the main part of then person’s life rather than part of it. But this option is less readily available in some other character-distorting conditions. Schizophrenia, for instance, usually comes on relatively young and also often brings about radical changes of character and personality.

A person with schizophrenia may go through periods of crisis, where it is hard to see any coherent personality at all. There may be a torrent of words that do not make much sense, or which reflect some delusional state. Moods may oscillate between shallow, excited optimism and a sunken, withdrawn despair. Attitudes to people he or she knows may move, apparently randomly, from warmth and love to suspicious hostility expressed by a menacing stare.

But, as a result of medication or as a new phase of the condition, there can be long periods of stability. In these periods, the person may seem much as before the illness. But often there is a transformation of personality. Someone friendly, alert and with wide interests may have moved into a negative, strangely unreachable phase: slow, sullen, perhaps aggressive, usually uninterested in others, saying little and doing little beyond half-watching television. In these periods, friends and family may say, “it is like talking to a different person”.

For some people, the oscillation between periods of crisis and these negative phases may last a lifetime. Friends and family may have conflicting responses to the aggression they are sometimes shown. Should they react with exasperation or with detachment? Is it him or is it his illness?

There is a case for refusing to see the schizophrenic personality as the person’s real self. There is usually little doubt that the strangeness and the narrowing passivity are products of the illness. As a thought experiment, imagine that a treatment is developed that can restore permanently someone from this negative state to their original state before the onset of the illness. In the absence of other residual problems, this would be a cure for schizophrenia. It would then be natural to see the second personality as a product of the illness, and not as what the person is really like. This would be supported by the fact that many aspects of the second personality appear on lists of the “negative symptoms of schizophrenia”. In retrospect, difficulties in answering the question, “is it him or is it his illness?” would largely disappear. The decision to withhold the more critical reactive responses would seem to have been right. The person’s self-creative project would have been temporarily disrupted by an illness. The hostility or aggression they displayed should be put aside as not reflecting their real self.

Now imagine a scenario, more like the actual one for many people with schizophrenia, in which the original personality is not restored. Should we treat the current, schizophrenic, personality as not reflecting the person’s real self, but as merely the product of the illness? The case for this approach is that it is the one we would adopt if there were a cure. It seems puzzling that the schizophrenic personality should have a completely different status depending on whether we are making the decision before or after scientists have developed a cure.

On the other hand, where there is no cure, there is a case for accepting the schizophrenic personality as now being the person’s real self. This is not some brief coda at the end of life. His real self, if not reflected in the schizophrenic personality, is something hidden for decades. Can the real self of the forty year-old man really be that of the teenage boy before the onset of schizophrenia, a real self unlikely ever to be seen again? Would it not be more realistic to accept how he is now as his real self?

People confronted with this issue tend to be pulled both ways. The phrase “the real self” is of course vague and indeterminate, which makes it unlikely that the issue is one where the “right” answer can simply be read off from the evidence about the condition. Any answer seems at least as much a matter of decision as of discovery. And the reasons point in conflicting directions.

There is the desire not to give up on the possibility of a cure, a kind of keeping faith with the original person. This is given a degree of support by some of the cases in Oliver Sacks’s book Awakenings, in which he describes how people who had suffered from a severely Parkinsonian condition since the 1920s responded in the 1960s to L-Dopa. In some cases there was an extraordinary restoration of the original personality, which had been totally submerged for decades. It was not always possible to maintain the revived personality, but the fact that it could be revived at all supports the idea that, in these long-term conditions that distort character and personality, the original version of the person may not be totally lost. On the analogy of a television where the picture has been replace by visual chaos, there is the hope that if only could get the neurological or neurochemical tuning right the original picture might be restored.

There is also the thought that, as with dementia, the new personality is obviously caused by the disorder, and that it was not the product of the person’s own previous decisions or self-creative project. It seems unfair that their personality has been distorted by something outside their control, and refusal to see the new personality as really them is a kind of recognition of this.

On the other hand, the refusal to recognize the new personality as reflecting anything other than illness seems to leave the person as they are now in a kind of limbo, possibly for the rest of their lives. It is possible to have a character and personality shaped by a medical condition and still to identify with it. One autistic person protested against parents who wished his autism could be cured, by saying that their real wish was that he had not been born and that they had had a different child instead. The arguments pull both ways.


Schizophrenia typically comes on at a much earlier stage in life than dementia. But even schizophrenia’s onset comes on after there has been time for quite a bit of development of the person’s “original” character and personality. Because of this, we can contrast the person and the illness, and may see the latter as eroding or distorting the person’s previous character, But there are conditions with a comparable influence on character and personality that are present at a much earlier developmental stage, in some cases possibly from birth. This category includes the cluster of conditions known as “personality disorders”. Because there is no previously developed alternative personality, it is much harder to get any grip on the contrast implied by the question, “Is it her or is it her psychiatric condition?”

A. The Personality Disorders.

“Personality disorders” are typically said to be “deeply ingrained maladaptive patterns of behaviour”. The personality is “abnormal either in the balance of its components, their quality and expression or in its total aspect. As a result either the patient or society suffers or both.” There are obvious questions about such an account. The word “maladaptive”, with its evolutionary overtones, may disguise a residue of the bad old tendency to medicalize behaviour that deviates from socially accepted norms: the approach that once allowed being gay to count as a psychiatric problem. Or, if a really Darwinian interpretation of the word is intended, so that behaviour is “maladaptive” when it reduces the chances of survival, other problems arise. Did Socrates have a personality disorder? His deeply ingrained pattern of behaviour, his persistent asking of awkward questions, certainly seems to have shortened his life.

But objections to the general definition are too easy. It is worth mentioning some particular “personality disorders”. Although textbook lists vary, several appear on most lists. “Paranoid”, “hysterical” and “narcissistic” personality disorders involve having the kind of personality the names suggest. Another is “Schizoid personality disorder”. Typically it is said to be characterized by a defective capacity for close relationships, by lack of empathy, by withdrawal, solitariness and detachment. There is also “Obsessional personality disorder”. People in this category are characterized by perfectionism and rigidity. They are mean with money and obsessed with cleanliness and tidiness.

Lists of personality disorder invite an obvious party game, seeing in which categories ourselves and our friends fit. And there is an equally obvious danger, that people whose personalities are unappealing to others will find themselves diagnosed as having a psychiatric disorder. That there is some subjectivity here is clear. The person diagnosed as having obsessional personality disorder on the basis of perfectionism, rigidity, being mean with money, and being obsessed with cleanliness may turn round and say that the psychiatrist who diagnosed her is a slapdash, unprincipled spendthrift who needs to take a shower more often.

Personality disorders are thought of in terms of two different models. There is the model based on the thought that some personalities dispose people to particular psychiatric illnesses, so that “schizoid personality disorder” might be thought of as a kind of incipient version of schizophrenia. The alternative model is based on the idea that there are dimensions of personality, with personality disorder being at an extreme of a continuum. Each of these models has its difficulties. The link between personality disorders and psychiatric illnesses such as schizophrenia is not clear. And it is hard to see why a personality at the extreme of some dimension should count as disordered. This is the part of psychiatry where scepticism about the “medical model” is most powerful. It is hard to make a case for treating “personality disorder” as a kind of illness.

One alternative to the medical model is to think of personality disorders in an Aristotelian way, invoking a species-specific idea of the good life for human beings. In the same way that caging a bird is cruel because the good life for birds includes freedom to fly, solitary confinement for people is cruel because the good life for human beings includes the ability to mix with other people and to make friends. If some such account can be given, it may be possible to apply it in explaining some of the personality disorders. The defective capacity for close relationships said to characterize “schizoid personality disorder” can be seen as deprivation of the means of realizing one aspect of the good human life.

On this model, psychiatrists need not necessarily confine themselves to treating psychiatric illnesses. The techniques psychiatrists have developed, whether using psychotherapy or pharmacology, might benefit people who seek help in freeing themselves from the constraints of a life-diminishing personality. Some thoughtful psychiatrists notice a tendency in themselves to prescribe Prozac at first to treat clinical depression but then to continue prescribing it, not to cure an illness but because of its enhancing effects on some patients’ personalities. They are right to be concerned about the boundaries of psychiatric intervention, but the Aristotelian model could provide a justification for their crossing the “medical” boundary. Of course, those they treat must be presented with a proposal that makes clear the non-medical “enhancement”, as well as any risks involved, and they must give their reflective and voluntary consent to it. Under these conditions, psychiatric intervention directed at enriching lives rather than at curing illness sometimes may be justified.

There are some dangers in this Aristotelian psychiatry. A lot of work needs to be done on the basis of the account of the good human life. And there is the danger of an Aristotelianism that underwrites prejudice. It is all too easy to think of a crude version that would again encourage interventions against the sexuality of gays and lesbians. At the very least, Aristotle needs to be combined with John Stuart Mill. Awareness of the open-endedness of our conception of the good life and the value of “experiments in living” is a necessary corrective here. But, despite these dangers, the Aristotelian alternative to the medical model may rescue something of value from the debatable diagnosis of personality disorder.

B. “Antisocial Personality Disorder”.

One of the most problematic and intriguing psychiatric diagnoses is that of “Antisocial personality disorder”. Those with this diagnosis have the broad psychological tendencies of psychopaths. It includes psychopaths, together with others less extreme. Using the popular tool, the “Hare Psychopathy Checklist”, not all those diagnosed with “Antisocial personality disorder” would score high enough to count as psychopaths.

People with this diagnosis are often said to “lack a conscience”. This claim raises more questions than it answers. There are so many different things, some or all of which might be missing. They might lack empathy (the ability to imagine how others feel). Or they might sometimes harm people, out of a lack, not of empathy but of sympathy: they can imagine how their victim feels but do not care about it. Or they might lack feelings of guilt. Or they might lack certain moral concepts, such as cruelty or selfishness. Or they might lack a conception of the sort of person they are, or lack values that would shape a conception of the sort of person they want to be.

Because I am intrigued by these issues, I am taking part in a team project looking at the “lack of conscience” in two groups of people. One group, in a secure hospital, is made up of people having the diagnosis of Antisocial personality disorder. The other group is made up of neurological patients with the diagnosis of Fronto-temporal dementia, a group who at some stage in the development of their dementia may exhibit antisocial behaviour. My part of the study involves interviewing members of these two groups about their moral outlook, trying to find out what sort of conscience they do or do not have.

The study is not finished, so there are as yet no results that have any kind of scientific status. But, speaking unscientifically, in the interviews with the “Antisocial personality disorder” group, I am overwhelmingly struck by the terrible childhoods so many of them seem to have had. The stories are of rejection, denial of love, humiliation and denigration, violence or sexual abuse. Some of these horrendous childhood experiences took place in the family. Others took place in the public institutions into which, in Britain, children are (in the euphemistic phrase) “taken into care”. Of course, perhaps not all these stories are true. Those with Antisocial personality disorder have a reputation for being skilful deceivers. But the psychiatrists in the hospital think that perhaps eighty per cent of the patients had terrible childhoods.

I will not quote any of the patients I have interviewed. To discuss the issues raised, I will take the case of someone who, while never actually given a diagnosis of psychopathy or of Antisocial personality disorder, would fit one of these categories. This is a real case, and the person in question had a childhood not unlike that of many of my interviewees.

C. The Case of Mr. H.

Mr. H had an unhappy childhood. His father was strict and had a terrible temper. He was a “demon” about punctuality. He insisted on silence in the family. The children never dared speak in his presence unless spoken to, and were not allowed to call him anything less formal than “Father”. When he wanted his son, he never called him by name, but always whistled for him in the same way he called for the dog. The father often beat the dog, his wife, and each of his children. As a child, Mr. H was once given 230 strokes of the cane by his father. In later life, he said he remembered seeing his drunk father rape his mother.

Mr. H grew up with a very rigid personality. As an adult, he was obsessed with cleanliness, passionately hating any untidiness or dirt. He was also obsessed with wolves, sometimes thinking of himself as a wolf, and calling his Alsatian dog “Wolf”. He took the dog for exactly the same walk every day, throwing a stick for it at exactly the same place. Any suggestion of varying such routines made him agitated and angry. He hated being left alone at night, and hated the moon because he thought it was dead. He was obsessed with his own possible death from cancer, which his mother had died from.

Mr. H had difficulties in his love life. As a boy he had been terrified of being kissed. His first love affair came when he was 37. It was with a teenage girl, who tried to kill herself after he abruptly broke off the relationship. He then fell in love with his niece, who did not reciprocate his feelings. She killed herself with his pistol. At the age of 41, he had another affair, this time with an 18 year old girl, who made an unsuccessful suicide attempt early in their relationship. Mr. H seems to have been disgusted by normal sexual intercourse, saying he did not want it because he would become infected. His niece said that his main sexual pleasure was in getting her to urinate on his face.

Much of his emotional life seems to have been diverted to patriotism and politics. He fought in a war, with great patriotic enthusiasm. He was temporarily blinded during a gas attack, which seems to have left him with great resentment against those who did not fight in the war. He took up extreme right wing politics and was passionately anti-Semitic. He was highly successful at appealing to the public. He became leader of his country. He started a world war. He ordered the systematic murder of millions of his fellow-citizens. He killed himself when his country lost the war.

Now that Mr. H’s identity has emerged, there are questions about how, if at all, his childhood and character and personality are relevant to our reaction to him and to what he did.

Nothing in Mr. H’s story does much to suggest that he was ill. Rigidity, sexual problems, racism, anger and resentment, obsessions about dogs and the moon –they are all things people are better off without, but they are not obvious symptoms of illness. Nor is there much mileage in the idea that they do not reflect the real Mr. H, whose real but unexpressed personality was quite different. The anger and the anti-Semitism were characteristics as genuinely his as it was possible to find.

Mr. H raises in acute form the problem of personality disorder. It is not an illness. The excuses that defend someone’s character from criticism all fail. There is no conflict in which a set of higher-order desires or deeper values are defeated. There is no alternative “real” self obscured or distorted by a medical condition. We are confronted by someone whose real self, or character, is itself misshapen. If blame is a negative evaluation of character on the basis of actions, it is hard to see how Mr. H can escape it.

And yet, as with my interviewees, this seems not to be quite the whole story. When interviewing patients, I deliberately do not know what offences they have committed, although sometimes they tell me and afterwards I find out. As I hear in the interview about their childhood of pain, rejection and humiliation, it is natural to see them as themselves victims, to glimpse their life from the inside and to feel sympathy. Sometimes, going back from the hospital after hearing about the terrible things they have done, I struggle to get the two perspectives into one coherent picture. It is hard to reconcile the emotional response to what they have suffered with the emotional response to what they have done.

With some of the interviewees, as with Mr. H, part of the complex emotional response is linked to some secular version of the thought that, there but for the grace of God go I. This is not to endorse some form of environmental determinism. Many people have had childhoods comparable to those of the interviewees without committing crimes on their scale. Many people had childhoods comparable to that of Mr. H without committing genocide or starting world wars. The causal story must also involve things other than child abuse. But, because we do not know exactly what other factors –genetic, environmental or other- play a role, we cannot be confident that we would have responded to that kind of upbringing differently from the interviewees or differently from Mr. H. Of course we all hope that, given that terrible childhood, we would still not have been like Hitler. And no doubt many of us would not. But certainty here is over-confidence. And this adds a disquieting perspective to our attitudes to Hitler and to other psychopaths.

4. The reactive attitudes.

So far, blame has been discussed as a matter of an objectionable action giving grounds for making a negative judgement about the character of the person responsible for it. The standard excuses, ignorance and compulsion, sever the link between action and character and so prevent criticisms of the one carrying over to the other.

The standard excuses apply, with complications, in many psychiatric cases. But in cases of the “misshapen real self”, such as many cases of personality disorders, the objectionable actions often do flow from the person’s character and the standard excuses fail. Someone with a misshapen real self has precisely the kind of character to which the negative judgements apply.

This version of blame, the making of negative judgements about someone’s character, may be a very detached, unemotional affair, in the manner of the recording angel. But there is an alternative to the recording angel approach. The alternative locates blame among people’s relationships and attitudes, which may be much less detached.

In his classic paper on Freedom and Resentment, P.F. Strawson argued that those who think that determinism is a threat to the survival of blame overlook the way blame is embedded in a complex network of “interpersonal reactive attitudes” such as gratitude and resentment. He argued that it cannot be separated from these other attitudes: that they stand or fall together. And, he argued, determinism could not be a sufficient reason for giving up this whole network of attitudes. For one thing, it would be psychologically impossible for us to give them up. For another, even if it were possible, it would not be rational to do so. These attitudes are at the core of human relationships and to give them up would greatly impoverish us. It would replace our present emotionally involved reactions to each other by the detached, objective attitude perhaps often appropriate in a psychiatrist responding to a patient or a social worker responding to a client.

One merit of this approach is that it makes issues about blame less abstract. Instead of asking which judgements about someone’s character are justified, it shifts the focus to our reactions to that person. But it is perhaps too hasty to see the reactive attitudes as an indivisible web, such that undermining blame turns all relationships into ones where the temperature never rises above clinical or professional detachment. Some attitudes are more easily dispensed with than others.

A. Attitude Spirals as Part of the Core of Human Relationships

Relationships are shaped, and sometimes constituted, by the attitudes people have to each other. Some attitudes pervade a whole relationship: liking and disliking, love and hatred, friendship and enmity. Although these attitudes are often responses to what the other person does, they may not be. I may deserve neither the dislike of one person nor the love of another, and yet have both.

Not all attitudes are so free-floating. Some are based on evaluation. Respect, admiration or contempt at least purport to reflect someone’s good or bad qualities. Other attitudes are responses to people’s good or bad fortune. If I break a leg while vainly trying to windsurf, I let myself in for your sympathy, amusement or schadenfreude. Other attitudes are reflexive: Jealousy, envy and condescension all compare the other person’s qualities or situation with one’s own.

All the attitudes mentioned so far are, in a minimal way, “interpersonal reactive attitudes”. Each of them may be, and some of them have to be, a reaction to what other people are like. But, without the addition of a certain complication, they are limited.

At the core of relationships is the interplay between attitudes to each other. We have attitude spirals. I have attitudes to your attitude to me, and you have attitudes to my attitudes to your attitudes, and so on. Lovers or friends may have responses to each other’s attractiveness or wit, but the relationship is only a surface one unless they have reactions to each other’s love or trust. As enemies, you and I may dislike each other’s faces or voices, but the enmity hardly goes deep if I have no attitude to your visible contempt, or you are indifferent to my gloating when you lose your job.

Relationships become less shallow as we develop attitude spirals. Some attitudes are centrally response to the attitudes of others. At this level are gratitude, resentment, indignation, vengefulness, blame and forgiveness. These attitudes need not be in response to attitudes to ourselves: we may be indignant on someone else’s behalf, or blame a person for malevolence towards other people in general.

The emphasis on attitudes, rather than actions, as the objects of blame, gratitude and so on may be thought exaggerated. Perhaps I may be grateful to someone, not for their attitude, but for their help or for their gift? It is true that sometimes these responses seem to be more to actions than to attitudes. Your neighbour burns your house down. Your resentment may persist even after you find out that, instead of malice towards you, he had a schizophrenic delusion and thought he was burning all the evil out of the world. But this is hardly a central case of resentment, and perhaps not a clear case of it at all. For the most part, what in actions makes us grateful or resentful is the attitude that shines through them. These attitudes to attitudes are part of the core of human relationships, and to exclude someone from these responses is to exclude them from part of that core.

B. Personality Disorders and the Reactive Attitudes.

Schizophrenia and dementia often involve distortions of personality and character that had developed before the onset of these disorders. So there is scope for the contrast between the real person and the distorting effects of the illness. Personality disorders characteristically emerge early. Whether they result from treatment in early childhood, of from genes, or from things that happened in the womb or at birth, or from some combination of these, personality disorders emerge too early to allow the development of an alternative personality. There is no alternative “real” person, remembered by friends and family but now submerged by the disorder. The person’s real personality is the misshapen one. The lack of concern for others, or the malevolence and hostility, are the real attitudes of the person.

Does this mean that the reactive attitudes are in order as responses to people with personality disorder?

What counts against this is the thought that personality disorder is still a piece of bad luck. If it is caused by having particular genes, or by what happened in the womb, or at birth, none of this was under the person’s control. If it is caused by parental rejection, cruelty, or abuse, none of this was the person’s own fault. Such people are themselves victims. Given that treatment, any of us might have turned out the same. So blame and resentment seem unfair.

Yet, on the other hand, exclusion from the reactive attitudes is exclusion from part of the core of human relationships. Perhaps this exclusion too is unfair. And perhaps it is only by participating in relationships, including the spirals of reactive attitudes, that there is any chance of transcending the original, misshapen personality.

The reactive attitudes themselves are not totally under our control. We cannot just switch them on and off at will to bring about the best consequences. But we do have a degree of choice about how far we do or do not inhibit them. To the extent that we have such choices, a complex response (to, say, aggressive violence stemming from personality disorder) seems called for. The ideal seems to be both to retain the reactive attitudes to the person’s own horrible attitudes to his victims, and yet to remember that he is a victim too, someone whose whole nature has been shaped badly by causes largely outside his control. This dual response is not easy, as I found when interviewing such people.

But the dual response is necessary if we are to do justice to the complexity of the case. It is probably right that we cannot entirely or largely abandon the reactive attitudes, and that life would be impoverished if we could and did. But there is also a determinist thought that should make us a bit uncomfortable with either Aristotelian or Strawsonian commonsense. It is about those of us who do not have personality disorders. Although we shape our own characters by our voluntary actions, there is likely to be some causal explanation of how we come to choose some actions rather than others. And these causal explanations may well often go back ultimately to factors beyond our control: to genes, to what happened in the womb or at birth, or to what happened in our early childhood. As with those with personality disorders, our characteristics are certainly ours, yet are not entirely of our own making. In the long run, the dual response we develop to those with personality disorders may turn out to be what is appropriate for everyone else as well.