Friday, 30 December 2011


"Intuition" implies a relation, like that of sense perception, between a perceiving subject and an independent reality which is the object of perception, so "moral intuition" is problematic in two closely related ways: the first is that philosophers have found it difficult to show irrefutably that there is a moral reality independent of the will of the experiencing subject; and secondly, that scientists have identified no organ in the human body which is recognized as being attuned, as it were, to the wavelength of that reality; and the two aspects of the problem are inter-dependent, because we cannot know that there is an independent moral reality unless we have some means of apprehending it, and we cannot believe in a "moral sense" unless we believe that there exists an independent moral reality for us to apprehend.
My aim in this essay is to provide a theory of moral intuition which establishes what "moral reality" is and how we have our intuitions of it. The account I shall give is very much my own, but it is inspired by my readings in the work of Iris Murdoch and later in the philosophy of language, particularly that of Robert Brandom, which enables us, I believe, to speak in an objective way about things Murdoch thought of as entirely subjective.
The aim of the argument offered here is not to make moral intuition possible or to teach anyone how to perform acts of moral intuition, because it is central to the idea of moral intuition that it is something you can do without understanding how you do it. The aim of the argument is only to show that a particular type of cognitive process does occur and to explain why it is right to see it as an intuitive grasping of moral facts.
The first step in this demonstration is to get some clarity about what we mean by "intuition", and I want to begin by proposing that sense-perception is our model for intuition, because when we speak of intuition we often refer to it as a "sixth sense", which implies that it shares certain important features with our senses of touch, taste, smell, hearing and vision. When we see something as red, feel it as hot to the touch, taste it as bitter, hear it as high-pitched or smell it as sweet, those sensory qualities, - redness, heat, bitterness, high pitch and sweetness, - are self-evident, in the sense that they are themselves the only evidence we have for the claim that they are red, hot, bitter, high-pitched or sweet. If others question our claims, we can only point them to the sources from which we received those sensations and ask them to apply the descriptive terms in the normal way.
We can consider sense-perception (or sensory intuition) in two ways: we can think about the object, i.e., the quality that is grasped, and we can think about the action, i.e., the grasping of a quality. I have said that the quality grasped in sense-perception is "self-evident", and now I want to propose that the act of grasping that quality be seen as direct; by which I mean that the act of grasping a sensory quality is not mediated by a process of reasoning, i.e., that the claims we make about sensory qualities are not inferences: they are not conclusions drawn from premises. It may be that, in calling the quality grasped "self-evident" and calling the act of grasping it "direct", I am just saying the same thing in different ways, but there is no harm in that as long as doing so helps to clarify what we mean when we speak of sense-perception as a form of "intuition".
Our senses are not, of course, an infallible source of knowledge, and there are three important ways in which they can fail us: firstly, they are directed by our attention, so that, if we are looking fixedly in another direction, we may fail to notice fairly obvious sensory clues right under our noses, and thereby deceive ourselves; secondly, our nervous systems may produce sensations internally, which lead us wrongly to imagine some external phenomenon as their source; and thirdly, we may thoughtlessly combine different sensory cues in ways which lead to an incorrect identification of their source. Infallibility, then, is not a feature of sensory intuition so we must not expect it to be a feature of moral intuition either.
To summarize, then: sensory intuition is a fallible act of grasping in which the quality grasped is self-evident and the act of grasping is direct, so that, if there is a grasping of moral qualities, it will not deserves the name of "intuition" unless it resembles sensory intuition in these respects.
We have established what kind of grasping is to be regarded as intuitive, so, if we are to demonstrate that there is such a thing as an act of "moral intuition", the next step must be to establish what it is that would have to be grasped for the act to count as an act of that kind. Clearly, we can begin by saying that it would have to involve the grasping of "moral qualities", but then we have to establish what kind of quality is moral, and it seems clear that moral qualities are the qualities of actions, by definition.
That moral qualities are the qualities of actions is clear from the generally accepted principle that, when we make judgments of moral condemnation, we condemn the act, but not the agent. We recognize that people who do bad things often also have a record of doing good things, so that, in condemning their bad deeds, we do not condemn them. If moral qualities are the qualities of actions, we need to be clear about which of the qualities of an action count as "moral". Bad actions involve many physical features that are also found in good actions; e.g., cutting people with knives is something we see in acts of therapeutic surgery as well as in acts of murder; so we need to clarify what it is about an action that is the object of moral judgment.
Actions are events, but there are many events which are not actions in the sense that is relevant to moral judgment. Actions are, by definition, events to which it is proper to assign intentions, i.e., motives, purposes, meanings, and it is only to the kind of event that is seen as directed by some kind of intention that moral judgments are applied. However many people are killed by an earthquake or a volcano, we do not apply moral judgments to those events as such, but only to the actions of individuals or institutions, e.g., property developers, local authorities, which may in some way have contributed to the size of the death-toll; unless, of course, we see the disaster as an "act of God", in which case we may have to pass moral judgment on God. So, if moral qualities are the qualities of actions, they must be qualities of what it is that defines those actions as actions, namely, the intentions by which they are directed. But what exactly are "intentions"?
The intention that characterizes an action can be stated explicitly as a proposal, and this can be done before or after the event, i.e., before or after the doing of the action. The proposal can then be justified, and no action can be justified until the proposal that characterizes it has been explicitly stated, because the justification applies to the proposal, not to the action considered simply as a material event. The action as a material event is justified only insofar as those of its material aspects that are morally questioned are explained as the unavoidable consequence of acting on a justified proposal; e.g., we regard the use of force as justified if it can be shown that it was the minimum necessary for an agent in a given situation to act effectively on the proposal "that he defend himself".
To speak of an "intention", then, is to speak of that which is both implicit in an action and able to be made explicit in the form of a proposal which defines the action; therefore, if moral qualities are qualities of actions, and actions are defined by their intentions, then moral qualities must be qualities of the intentions-expressible-as-proposals whereby events are recognized as actions. The next step in the demonstration must be to show that intentions-expressible-as-proposals exhibit differences of quality which can be apprehended by intuition, i.e., grasped directly as self-evident.
In order to deal explicitly with intentions-expressible-as-proposals we have to deal with them as proposals, but there is nothing objectionable in this, since this is what we do whenever we try to justify the intentions implicit in our actions, and what is obvious about proposals explicitly stated in sentences is that they have logical qualities. Most obviously, a proposal can be either self-contradictory or not. Furthermore, if a proposal is self-contradictory, its being so is self-evident: not in the sense that it is inescapably obvious, but in the sense that nothing is evidence for its being so other than its being so. If someone denies that it is self-contradictory, we can only appeal for evidence to the proposal itself: perhaps we shall repeat it more slowly, enunciating more carefully and emphasizing the key words, in the hope that the absurdity will then be recognized for what it is; but, if someone cannot see it, there is not a lot we can do.
To put it another way, the self-contradictoriness of a proposal is something we grasp not by way of a reasoning process. A proposal can be analyzed to make explicit the self-contradiction implicit in it, but, when it is explicit, the absurdity of proposing it is something that cannot be explained: it is something which, if we see it at all, we just see. In this respect it is exactly like what happens when we laugh at a good joke: if we get the point (i.e., the intention) instantly, we laugh, even though we may have great difficulty explaining why it is funny; but, if we don't get it instantly, no amount of explanation can make it work for us: our responses to jokes, which often depend upon implicit contradictions, are perfect examples of intuition in action.
There is, then, a self-evident qualitative distinction between those intentions/proposals which are self-contradictory and those which are not, and this distinction can only be grasped directly; that is to say, in the terms established so far in this demonstration, this distinction is something grasped intuitively. But to say that this distinction is logical is not yet to say that it is moral, i.e., that it is a distinction of the kind applied by way of moral judgment to actions, namely, a distinction between good and evil; however, it will be my aim to show that this is how moral qualities have to be understood.
However the idea of a "value distinction" has been understood, it has always been recognized that the words "good" and "evil" derive what meaning they have from our attempts to apply that kind of distinction. The word "good" has a meaning in so far as it has a use in the linguistic practices of proposing and justifying: we use the word "good" of an action only when we are engaged in proposing and justifying that action, so that to say of an action that it is "good" just is to imply that it is capable of being justified without absurdity. The word "evil" is a word that we use of an action only when we wish to deny that it could be justified.
When deliberating about what to do, we typically consider various proposals, some of which we reject as less good than others, but in regarding them as "less good" or "worse" we do not regard them as "evil". There are many types of action which are inherently so trivial that they are sometimes spoken of as "morally neutral", but this category is misleading, because, if "good" just means "self-consistently justifiable", then there is no act so essentially trivial that a situation cannot be imagined in which someone might propose doing it and therefore speak of it as "good".
To be responsible is to acknowledge the requirement not to do what cannot be justified, since responsibility is by definition the requirement to respond to demands for justification. To say that an action is "evil" is to say that no reason could justify it, so that to act upon it would be absolutely inexcusable on any grounds whatsoever.
To say that we have an intuitive grasp of the difference between good and evil is to say that we have a direct grasp of the self-evident difference between a "good" action, i.e., one which can without absurdity be proposed and therefore justified, and an "evil" action, i.e., one which cannot be proposed and therefore cannot be justified without absurdity.
To construct proposals which are explicitly self-contradictory is a simple matter, but these tend to be trivial precisely because they are explicit and the absurdity is therefore so obvious that no one would be tempted to propose, justify or act on them, e.g., the proposal "That you do and do not kiss (or kill) me". The danger lies in proposals where the self-contradiction is implicit rather than explicit so that we are able to deceive ourselves about it if we have strong motives for doing so.
The next step in the demonstration is to show that there is an intention implicit in the act of proposing any action which renders certain kinds of proposal implicitly self-contradictory. That is to say, the uttering of a proposal can be seen as an action in its own right, - a speech-act, - and therefore as implying an intention-expressible-as-a-proposal; and, if this is so, then it is possible for the explicit content of certain types of proposal to be seen as contradicting that implicit intention, thereby producing an implicit contradiction.
It strikes me as self-evident that anyone who proposes any course of action thereby implicitly intends to seek consent to it, and what makes it self-evident is that whoever proposes any course of action knows that, unless he somehow signals that his proposal is intended ironically, those who hear him will automatically assume that what he is doing is seeking consent for his proposal. He knows this just because he is a speaker, the user of a natural language: it is part of the interlocutory grammar which, like the grammar of sentences, every speaker applies "intuitively".
I now come to the most important part of my argument: If proposing a particular course or a general rule of action always implies an appeal for consent, then an implicit contradiction arises if it is proposed that someone be insulted, raped, tortured, enslaved or murdered, since these types of action, by definition, cannot occur unless the object of the action withholds his consent. Actions of this type cannot be proposed without self-evident absurdity, and that which cannot meaningfully be proposed cannot be justified by any reasoning whatsoever. Such actions absolutely cannot be called good and must therefore be called "evil".
I also insist, - and this is the most sensational implication of my argument, - that the same is true of the proposal that someone be punished. There is nothing self-contradictory in proposing that someone freely consent to be penalized (i.e., that he voluntarily accept a penalty as a penance for wrong-doing or, as players do, for infringements of the rules of a game); but to "punish" him is, by definition, to impose a penalty on him without his consent to your doing so at the time when you do it, which cannot be justified.
I am fully aware that these arguments portend a revolution not only in moral philosophy, but also in the law and I make no apologies for that. It is predictable that what I propose will be fiercely challenged, and I shall attempt to rebut the most common types of challenge in a supplementary paper, but for the moment I prefer to focus simply on the principle.
The demonstration is now more or less complete, in that I have explained in what sense we have an "intuitive" grasp of the difference between good and evil actions; but I have not yet named the organ through which moral intuition operates. It is the organ whereby we recognize each other's intentions in speaking, i.e., the organ which makes language possible.
Students of linguistics have spoken of a part of the central nervous system which enables us to acquire language: the so-called LAD or Language-Acquisition Device. As yet no one is entirely sure how it works, but that it does work can hardly be doubted: it enables us, on condition we have someone else to do it with, to play intuitively the game of implying and inferring that is what language is. So I propose that what is now provisionally called the LAD is to moral intuition what the five senses are to sensory intuition.
In Language, Truth and Logic, A.J.Ayer, one of the most influential voices opposing the case for moral intuition, claimed that terms of ethical evaluation such as "good" were "calculated" (i.e., intended) to "express feelings", to "arouse feelings" and to "stimulate action". In doing so he was exercising precisely the kind of LAD-ish intuition I have proposed; but, if my intuition is correct, his exercise of it was insufficiently precise. My claim that someone who proposes a course of action is appealing to others to consent to that proposal does not deny Ayer's claim, but amplifies it; perhaps my intuition just expresses more fully what is involved in Ayer's recognition that value-terms are "calculated" to "stimulate action". Seeking to "stimulate action" entails seeking consent, because acting entails intending which in turn entails consenting, so, if you want to stimulate action, the only way to do it is to appeal to the agents' consent by reasoning or by other forms of inducement.
An argument for moral intuition is an argument for moral realism and moral cognitivism; i.e., it is the claim that there are moral facts, that certain types of action just are, - and can be known to be, - absolutely unjustifiable and therefore evil. Other types of action which can be proposed without self-contradiction are more or less good in relation to each other, but there are many different ways of being good, not all of which are mutually compatible, so that choices have to be made: no practical proposal, therefore, can ever be absolutely good, i.e., the only possible good for everyone in all circumstances, but certain kinds of practical proposal can be absolutely evil.
What this demonstration has shown is what Iris Murdoch intuitively sensed, that the logical distinction between facts and values, the gap between "is" and "ought", is false. That distinction, first set out 250 years ago by David Hume, is the foundation of the dominant consensus in contemporary moral philosophy. The dogma is that judgments of value are not statements of fact and cannot validly be inferred from statements of fact alone, and therefore cannot be true in the sense in which facts are true.
My answer to Hume is that there are special kinds of fact, what we might call the facts of interlocutory grammar, - facts about what speakers recognize themselves and each other as implicitly committed and entitled to by what is said between them, - from which conclusions can be drawn which are recognizably both moral and factual. The idea that an utterance has meaning only in so far as it implicitly affects the normative status, i.e., the commitments and entitlements, of speakers in relation to each other is the central debt that I owe to Robert Brandom.
I call upon all speakers to confirm the following facts from direct observation of their relations with each other: (1) that proposing a course of action entails an appeal for consent, and (2) that appealing for consent to a proposal entails an undertaking to justify it, i.e., give reasons for consenting to it; and I call upon all speakers also to confirm that the following conclusions can be drawn from these facts: (3) that there are types of action which, by definition, cannot be consented to and which therefore cannot be proposed without implicit self-contradiction, (4) that what cannot be proposed without self-contradiction is absolutely unjustifiable, i.e., no reason could without self-contradiction be offered as a reason for consenting to it, and that (5) where its author implicitly undertakes to justify a proposal which is absolutely unjustifiable, he cannot fulfill that undertaking, and he therefore cannot responsibly refuse to withdraw it if asked to do so.
Copyright © Christopher Eddy 2011
(3492 words)


  1. Hi Chris,interesting post. But I'm going to have to disagree with you on this one. Here's why. You say:

    It strikes me as self-evident that anyone who proposes any course of action thereby implicitly intends to seek consent to it, and what makes it self-evident is that whoever proposes any course of action knows that, unless he somehow signals that his proposal is intended ironically, those who hear him will automatically assume that what he is doing is seeking consent for his proposal. He knows this just because he is a speaker, the user of a natural language: it is part of the interlocutory grammar which, like the grammar of sentences, every speaker applies "intuitively".

    It doesnt strike me as self evident. In fact the statement is so out there that it is going to require a lot more argument to back up.

    1. Dear Murali,

      Thank you very much for your question.

      In the concluding section of my paper I refer to Ayer's emotivistic intuitions about what prescriptive language ("good", "ought",etc) implies, i.e., that it seeks to "stimulate action",- a view which has gained almost universal acceptance as self-evidently true. I then point out that "seeking 'to stimulate action' entails seeking consent, because acting entails intending which in trun entails consenting, so, if you want to stimulate action, the only way to do it is to appeal to the agent's consent by reasoning or other forms of inducement." If that is so, then everyone must behave as if it's so, otherwise the language-game of prescribing simply wouldn't work; but that game does work, so it must be self-evident to everyone, - even though they may never have made it explicit to themselves, - that prescriptive language does involve an appeal for consent and an acknowledgement of the responsibility thereby incurred to supply reasons for consenting if asked to do so.

      Regards, Chris.