You reply, 'A cognitive system isn't like a piece of apparatus with an independently specified function, such as a soil processor, or a rock crusher. Whether my computer brain "works" well or badly is a matter of comparing one set of judgements with another set of judgements. But the only judgements whose validity I recognise are those that I am prepared to make myself. I correct my perceptual or information processing "errors" when certain anomalies appear. On the basis of my experience, I might discover that certain physical states, such as a crossed wire or a worn transistor, that are correlated with those anomalies. But I do not accept that any fault I discovered in my own physical workings would ever oblige me to correct my judgements, if I was not minded to do so. Even if it turned out that my insides were all messed up, after all is said and done I can still only continue to judge things the way I see them. I might even conclude that the crossed wires and worn transistors improved my powers of judgement!'

'Yet surely you must realise,' says a third philosopher, 'that the consequences of your misjudgement are not purely theoretical. Some of the information is vital to your own survival. You rely on your senses and information processing to avoid such things as falling into a pot-hole, or getting too big a dose of radiation. Independently of what you may happen to believe, you risk injury or even destruction if certain judgements that you make turn out to have been wrong.'

That final, desperate move leaves you completely unimpressed. 'Sure, all sorts of things might happen to me. But so long as I exist as a conscious subject capable of making judgements — and obviously I could not judge that I had ceased to exist as a conscious subject — then whatever may physically happen to me is still a matter for my judgement. You might say that falling into that pot hole proves beyond a shadow of a doubt that I have made a misjudgement, in moving confidently forwards as if I judged there to be firm ground ahead of me. I reply simply that I am not in the least surprised to find myself in a pot hole, and, besides, I find it warm and comfortable down here!'

What is the moral of our science fiction tale? In terms of the solipsist's metaphysical standpoint, it is impossible to get a genuine handle on the notion that I exist as a subject within a world, a being with physical properties that affect my capacity for perception and rational thought, whose judgements may be evaluated by other subjects, and whose map of the world determines my ability to survive and accomplish my aims. All of these features may, as we have seen, be translated into the language which 'alone I understand', the language of my possible experience.

Yet this seeming impossibility of refuting the solipsist theory proves, on reflection, to be the solipsist's very undoing. We asked whether, in criticising the solipsist's world as 'indistinguishable from a dream or paranoid delusion', there might not be 'certain necessary features of experience' that we had overlooked. We now have good reason to believe that there is no aspect of our experience as subjects within a world that a determined solipsist — whether in the form of a piece of recalcitrant space hardware, or a flesh and blood human being — could not account for in terms of that theory. All that remains, therefore, is a stark choice. Either to renounce solipsism, or to give up for good the idea that, with regard to the language which I use to express my thoughts to myself, one can ever 'talk about "right"'.

Few persons would require much persuading that solipsism is false. In fact, the young Wittgenstein stands virtually alone in owning up to a theory generally regarded as nothing short of scandalous. And he was able to do so only by persuading himself that 'solipsism coincides with pure realism' (Tractatus Logico-Philosophicus 5.64), an interpretation that we have effectively refuted. A realist view involves at the very least recognition that I stand with others on equal footing as one conscious subject amongst other conscious subjects, and that is the one proposition which, as we have seen, the solipsist rejects. Other subjects are for the solipsist merely 'characters in the story of my world'. What is much harder to reach agreement on is how we are to conceive of ourselves, as conscious subjects and language users, in the light of the rejection of the solipsist theory. In particular, what grounds do we have for believing that we can, while the solipsist cannot, legitimately 'talk about "right"'?

The project of delimiting the features of experience sufficient for experience of an objective world is associated with Immanuel Kant, in the first half of his Critique of Pure Reason (the 'Transcendental Analytic'). For Kant, however, there always remained an alternative standpoint to that of 'my possible experience'. According to his theory of transcendental idealism, in addition to my phenomenal existence as a subject tracing my way through the world of my experience, I also exist as an object in a noumenal world, alongside other subjects of experience and the objects that re-appear in each of our individual experiential worlds. The noumenal world is a world conceived as being outside space and time. It cannot logically be the object of any individual subject's cognition other than that of the Deity, for it is the way things ultimately are, in themselves, unconditioned by the restricted view imposed by sense perception, or indeed the possession of a point of view as such. It follows that the philosopher is in no position to give any positive information regarding the nature of the noumenal world. The most one can say with assurance is that this world, the world of my possible experience, is not all there is.

One might say (although Kant never makes this explicit) that the existence of real subjects, corresponding to the characters who appear in the story of my world is not, on Kant's theory, an empirical fact but a metaphysical fact, a fact about the noumenal world. Similarly, the fact that I have constructed my theory of a world according to rational principles is a fact about my existence as a noumenon, for, as we have seen, so far as my phenomenal world is concerned, the possibility that my judgements might be fundamentally irrational simply does not appear in the picture. I have to proceed on the assumption that I am capable of making rational judgements in order to make any judgements at all. — In these terms, what the solipsist denies is the very existence of a noumenal world. There is nothing, in ultimate reality, other than the world of my possible experience. There is no fact, beyond the way I appear to myself as a subject making judgements, that either validates or invalidates my necessary assumption that I am capable of making rational judgements.

The Kantian response to the solipsist's claims is not one that would find favour with many philosophers today. Belief in a noumenal world appears altogether too high a price to pay for rejecting the idea that my world is the one and only world. But is there a viable alternative? Any workable theory must, it seems, have two basic components. The first component is the recognition of a standpoint from which I appear on equal terms with other I's, as one subject amongst other subjects. The second component is the idea that the existence of a genuine distinction between what is right and what merely seems right in applying concepts presupposes a standpoint from which my judgements may be evaluated by other subjects.

What we are seeking is a way of conceiving of the reality that exists beyond, or outside the solipsist's 'world of my possible experience', a positive conception to put in the place of the negative idea of a realm of unknowable Kantian 'noumena'. This reality is, on the contrary, a world that each of us knows in small part, though the knowledge of some persons may be considerably more extensive than that of others. My knowledge overlaps yours, our separate experiences merge to form a shared experience. Each of us is an expert on our own individual patch of the universe, but no-one holds logically exclusive rights to knowledge concerning any aspect of the world, ultimately not even our own mental states. It follows that the features of this common world are not drawn in the solipsist's language of thought, but in a common language in which we are able to communicate with one another, and so correct or amplify one another's judgements, or contribute our individual thoughts to the fund of human knowledge.

It is important to realise that the idea that my language is our language is a radical idea, not a re-affirmation of common sense. 'Common sense' — that is to say, our implicit, unrefined philosophical view of the place of mind in relation to reality — tells us that the subjective qualities of a person's mental states can never be communicated to others, that only I really know, for example, what I mean by 'red'. Such illusions we have left behind. We do not, each of us translate the common language into our own private language. The language which we use to communicate with one another is the language of thought — indeed of our innermost thoughts — there is no other.

It follows that I cannot be right or wrong about any matter except by reference to publicly available criteria, criteria upon which it would, in principle, be possible to reach agreement with others. The autonomy of the solitary subject that Descartes insisted upon in founding his philosophy on the proposition, 'Cogito ergo sum' is thus renounced in favour of heteronomy. My judgement can never be the final world on any matter: my world is our world. — Does that mean then that the limits of our language mean the limits of our world? If the solitary 'I' as the ultimate foundation of the world is displaced, does the world thereby come to rest upon us? Who are 'we' anyway? The author and reader? All English speakers? All human beings? — We shall not be able to claim a secure grasp on what it means to reject solipsism until we have found a way to respond to these baffling questions.