Sometimes, common sense morality belies what seems like reality

G.E. Moore famously replied to sceptical doubts about the existence of the external world by holding up his hands. I cannot, he said, doubt that these are mine. No matter how clever your sceptical arguments may be, they dissolve in the face of robust common sense. Some things just can't be doubted, as a matter of fact, and these common sensical observations serve as all the foundation we need for knowledge.

Common sense continues to serve us as an implicit standard when we try to understand the behaviour of others. Consider how we go about distinguishing cases of self-deception from mere ignorance. Imagine a woman whose husband is having an affair. She appears to have no suspicions regarding his conduct, despite the sudden late nights at work, the mysterious phone calls, the unexplained expenses. Is she self-deceived, or merely ignorant? Common sense provides us with our guide: we ask ourselves what it would be like to be in her shoes. Would we find the evidence overwhelming? If we would, then she must be self-deceived; she must be denying to herself what she, in some sense, knows. The clincher would come if we observed behaviour in her that could only be explained by the hypothesis that she knew what she denied: if she avoids talking about her husband's absences, for example.

Typically, we blame people for being self-deceived. It represents a failure of epistemic courage, perhaps; in any case, it is a moral failing. So our ability to distinguish reliably between self-deception and mere ignorance matters, morally. If she had been deceived by the entirely plausible stratagems of her husband, then he would take all the blame; as it is, some of it is due to her. The alcoholic who refuses to acknowledge his alcoholism, the mother who ignores her son's drug dealing, not stopping to ask where the money for the expensive presents comes from, the company director who is wilfully ignorant of the state of the books - all are blameable, other things being equal.

We sometimes use this same common sense methodology to attribute blame in criminal cases. Sometimes, we need to decide whether a defendant likely knew the consequences of an action, when we need to distinguish between negligence and recklessness, for instance. We need to determine defendants' state of mind in order to judge whether they had the mens rea for a crime. So the reliability of our common sense methodology of attributing beliefs matters. It is an essential part of our life together, as social beings, and plays a significant role in our moral practices.

It will be bad news, then, if Moore was wrong: if our common sense can let us down. It is well known that common sense judgments about the world are not always reliable: it was once common sense that witches exist, and more recently that homosexuality was wrong. But it wasn't this kind of judgment that Moore had in mind. He defended only our common sense judgments about very basic claims: that the world existed yesterday, that 2+2=4, that I have two hands (here they are). It is on the basis of these judgments that we infer the states of mind of other people. And these judgments, at least, cannot be doubted. I can say the words: I can mouth propositions such as "is this my hand?" But I cannot mean them. There are some statements that, as a matter of fact, we can't doubt.

But now consider the neurological syndrome called anosognosia. Anosognosia is the product of a brain injury, usually as a result of a stroke. Anosognosics are apparently rational people. They can talk intelligently about their lives, their families, even about politics. But they are also severely disabled as a result of their stroke. Very often, they have a paralysed left arm. What distinguishes the anosognosic from other stroke victims, however, is that they deny that they are paralysed. They usually claim they are fine. In the even more bizarre case of somataparaphrenia, they go even further, denying that the affected limb even belongs to them. Sufferers from this disorder do just what Moore claimed was impossible: they are profoundly sceptical of the claim that the hand, which they admit is attached to their arm, belongs to them.

Something very strange is going on in these cases. Sufferers from these disorders exhibit a range of symptoms. Sometimes, they confabulate responses to questions: asked whether they can move their hand, they might reply that they just did. Or they might plead tiredness, or object to the presumption of the doctor's orders. In somatoparaphrenia, the confabulations are even stranger: the patient asserts that the arm belongs to the doctor, or that another patient must have left it behind. Very often, their behaviour looks like self-deception. It is plausible to maintain they are trying to protect themselves against some very distressing knowledge: they are disabled, perhaps permanently. This is precisely the view of V.S. Ramachandran, one of the most prominent neuroscientists in the world today. For Ramachandran, anosognosia reveals the defense mechanisms of the mind at work. He hypothesises that the left hemisphere of the brain plays an important role in constructing a coherent image of ourselves, an image which is as close as possible to the idealised view of our self, while the right hemisphere has the role of detecting anomalies in our self-perception. When the anomalies get too large to be ignored, the right hemisphere ensures that they are incorporated into our self-image. But anosognosics have damaged right hemispheres, so their left hemisphere is free to confabulate unchecked. Anosognosia is therefore normal self-deception writ large.

Further evidence for Ramachandran's hypothesis comes from the striking finding that under certain conditions anosognosics admit their paralysis, and seem even to indicate that they knew of its existence all along. When an anosognosic is given an inactive injection which they are told will produce temporary paralysis in their arm, they accept that it is paralysed. Ramachandran suggests that they accept it in this case only because they are told it is temporary, and it is therefore not threatening to them. Anosognosics can also be brought temporarily to acknowledge their paralysis (oddly enough) simply by stimulating their ear canal with cold water. For a short time after so-called vestibular stimulation, not only does the patient admit that her arm is paralysed, but she also agrees that is has been paralysed all along. Once the effect of the stimulation wears off, however, she reverts not only to denying the paralysis, but also to denying that she ever admitted it.

However, the self-deception hypothesis will only account for a relatively restricted range of the paradoxical belief states of anosognosics and other patients suffering from neurological deficits. Most obviously, the hypothesis ignores the fact that anosognosia is a member of a larger class of neurological deficits, called neglect. Neglect sufferers simply ignore a part of the world. For instance, a patient suffering from visual neglect will typically ignore the left side of her world. She will eat only the food that is on the right side of her plate, put makeup only on the right side of her face, and so on. She seems even to have lost the idea of left. But, just as the anosognosic can be shown to know (in some sense) what she denies, so the neglect sufferer is sometimes aware of events in the left half of her visual field.

In one experiment, neglect sufferers were shown pairs of pictures, which seemed identical to them because the right side of the images was the same, but which in fact were very different on the left side. For instance, one pair showed a house, which in one image had flames engulfing the left-hand side of the building. Neglect sufferers claim that the two pictures are identical, yet if they are asked which house they would rather live in, they consistently choose the house that was not in flames. Though they had no conscious knowledge of the flames, it was nevertheless affecting their preferences and their behaviour.

Clearly, however, these patients have no reason to deceive themselves about a picture. Neglect shows that it is possible to be aware of something in one sense, and at the same time sincerely deny all knowledge of it. In some cases at least, this paradoxical states is entirely innocent. It is not the result of a lack of courage or any other moral failing. It is the product simply of a breakdown in communication between parts of the brain.

Do anosognosia, somatoparaphrenia and other puzzling neurological deficits have lessons for us? I suggest that they do. Our attributions of mental states and knowledge to other people relies upon a common sense picture of what is and isn't possible. Of course, we are prepared to admit that there are insane people who might hold inconsistent beliefs. We also agree that someone might know something unconsciously, without conscious awareness of it. But we don't think that it is possible that someone might have beliefs that are blatantly contradictory, and yet remain otherwise rational. If we come across such a case, we think, we are in the presence of a liar: they are lying either to us, or to themselves, and we ought to treat them accordingly. But now we see that contradictory beliefs that are, prima facie, innocent are possible. We know it can happen in the case of brain injury. But perhaps the trauma merely reveals and makes more dramatic the mechanism that is often or usually at work in run-of-the-mill self-deception.

If this is the case, if self-deception or apparently wilful ignorance, is the product of brain mechanisms over which people cannot be expected to exercise control, then a good deal of our everyday moral judgments, and at least some of our legal judgments, rest on a shaky foundation. It may be that we need to revise our legal and moral practices, in the light of a philosophically informed neurology.