Jump to ContentJump to Main Navigation
Legal Reasoning and Legal Theory$

Neil MacCormick

Print publication date: 1994

Print ISBN-13: 9780198763840

Published to British Academy Scholarship Online: March 2012

DOI: 10.1093/acprof:oso/9780198763840.001.0001

Show Summary Details

(p.275) Appendix: On the ‘Internal Aspect’ of Norms

(p.275) Appendix: On the ‘Internal Aspect’ of Norms

Legal Reasoning and Legal Theory
Oxford University Press

Readers of Swift's Gulliver's Travels will recall how, sometime after Gulliver had been washed ashore on the coast of Lilliput after his shipwreck, and had been captured by the tiny inhabitants, people no larger than his thumb, the King of Lilliput determined that his person should be searched. To that end he sent two commissioners, Clefren Frelock and Marsi Frelock, to make an inventory of all his personal belongings. In a document full of interest, one passage perhaps stands out more than others:

Out of the right Fob hung a great Silver Chain with a wonderful kind of engine at the Bottom. We directed him to draw out whatever was at the End of the Chain; which appeared to be a Globe, half Silver and half of some transparent Metal; for on the transparent Side we saw certain strange Figures drawn and thought we could touch them, until we found our Fingers stopped with that lucid Substance. He put this Engine to our Ears which made an incessant Noise like that of a Water Mill. And we conjecture that it is either some unknown Animal, or the God that he worships; But we are more inclined to the Latter Opinion, because he assured us (if we understood him right, for he expressed himself very imperfectly) that he seldom did anything without consulting it. He called it his Oracle, and said it pointed out the time for every Action of his life.

To the old riddle, ‘What is it that God never sees, the King rarely sees, and ordinary people see every day?’ The authorized version of the answer was ‘An Equal’, but the wit's answer was ‘A joke’. The academic vice is not so much of not seeing, but rather of being under a compulsion to say what is involved in seeing, a joke; at any rate for present purposes, I must ask what is involved in seeing the joke in the passage quoted from Gulliver's Travels.

The joke turns on the point that in relation to the ‘wonderful kind of engine’ described, the Lilliputian commissioners have given so accurate an external description of the thing in terms of its visual appearance and auditory manifestations that we can know what it is; yet they have done so without themselves knowing what it is. Up to a point, but (p.276) only up to a point, ‘for he expressed himself very imperfectly’, they know how the Man Mountain used this wonderful kind of engine—indeed Swift uses their limited understanding to turn the joke back on western man by putting in their mouths the conjecture about its being the ‘God that he worships’; but that part of the joke I shall for the moment pass over.

Appearance is not the whole of human reality. Not the most meticulous physical description of every cog and spoke and their interconnections together, not even with superadded the most meticulous description of every act and utterance of the possessor of the ‘wonderful engine’ in relation to it, would amount to an understanding on the describer's part of the crucial fact which we to whom it is described know, and know even on the footing of a relatively imprecise external description: that the engine in question is a watch; that Gulliver ‘consults’ it in order to tell the time of day.

We know what the commissioners don't know, so for us the passage is a joke; we know it because we know a watch when we see one (or hear it described), and know how to use it, and know what to use it for. To use terms of H.L.A. Hart's which I find highly illuminating, we look upon these wonderful engines ‘from the internal point of view’, from the point of view of those who understand and work with that complex set of normative conventions in terms of which we can use these artefacts to measure the passage of time and thus to synchronize and co-ordinate activities with each other to a remarkably high degree of accuracy.

By contrast, the Lilliputians are outsiders. They are unaware of those conventional norms which for us make sense of our use of watches. Accordingly they can (for the moment at any rate) only appreciate Gulliver's watch and his use of it at the level of observable, or, rather, sensible phenomena. They can only describe it in such terms, together with vague conjectures as to the possible meaning of the object to Gulliver. Theirs is, by contrast with ours, an ‘external point of view’. It is ‘external’ in respect of the set of conventional norms in question.

This contrast between ‘internal’ and ‘external’ points of (p.277) view is brought out particularly vividly by the example taken from Swift's Gulliver. Swift makes us realize what it would be like to see a watch and not know it as a watch; and thus makes us able to contrast the position of somebody like that with our own actual position. But it is the contrast itself which is of present importance. To have drawn such a contrast and explained its importance for an understanding of the law is in my opinion a major element in Hart's contribution to the philosophy of law. The purpose of this essay is to consider Hart's distinction further, and to make some suggestions in the way of constructive criticism aimed at improvement on Hart's version of the doctrine.

To recall the context and content of the doctrine: in Concept of Law 1 Hart develops his exposition of the notion of ‘rule’, which is central to his theory of law, by contrasting it with the concept of ‘habit’, which he interprets in a purely behavioural sense; in his terms it is a necessary and sufficient condition of the existence of a habit that a particular pattern of behaviour of an agent or group of agents be regularly repeated over a period of time. (Whether this is plausible as an account of the notion of ‘habit’, or fair as an account of Austin, who is criticized for his use of the notion of ‘habit of obedience’ in accounting for sovereignty and therefore law, is not of present concern.)

The explanation of ‘social rules’ is developed by expounding ‘three salient differences’ between them and that behavioural conception of habit. First, where a group has a rule about doing something or another, deviation from the normal pattern is treated as a fault or lapse open to criticism, whereas in the case of a mere habit, deviation from the normal pattern ‘need not be a matter for any form of criticism’. Secondly, such criticism is regarded as justified or legitimate in the sense that someone's deviating from the normal pattern is regarded as a ‘good reason’ in itself for criticizing the lapse or fault involved.

(p.278) Thirdly, there is the point about the ‘internal aspect of rules’, with which we are presently concerned. To constitute a habit, it is not necessary that anybody think about the habitual behaviour or be conscious of its generality in a group; ‘still less need they strive to teach nor intend to maintain it’. But ‘by contrast, if a social rule is to exist, some at least must look on the behaviour in question as a general standard to be followed by the group as a whole. A social rule has an ‘internal’ aspect in addition to the external aspect which it shares with a social habit and which consists in the regular uniform behaviour which an observer could record.’ An example which Hart proceeds to give of this is drawn from the game of chess. It may be an observable fact that chess players move the Queen always in characteristically similar ways, but there is more to it than that, says Hart. They have a ‘critical reflective attitude to their behaviour; they regard it as a standard for all who play the game’. That attitude is manifested in the way in which they criticize lapses, demand conformity with standard patterns, and acknowledge that some such demands and criticisms are justified.

The internal aspect of rules is often misrepresented as a mere matter of ‘feelings’ in contrast to externally observable physical behaviour…[But] such feelings are neither necessary nor sufficient for the existence of ‘binding’ rules. There is no contradiction in saying that people accept certain rules but experience no feelings of compulsion. What is necessary is that there should be a critical reflective attitude to certain patterns of behaviour as a common standard, and that this should display itself in criticism (including self criticism), demands for conformity, and in acknowledgments that such criticism and demands are justified, all of which find their characteristic expression in the normative terminology of ‘ought’, ‘must’, and ‘should’, ‘right’ and ‘wrong’.

Surely it must be beyond doubt that Hart is correct in saying that an understanding of rules is possible only given attention to and analysis of the ‘internal aspect’ to which in this passage he draws attention (whether or not in all respects his account here and elsewhere is entirely satisfactory). Indeed, such a view would nowadays be almost a commonplace theory, and Hart himself in advancing his account with special reference to the law was following and developing accounts previously given by, among others, (p.279) Winch2, Wittgenstein3, and Weber4 (by all of whom, especially the first mentioned, he was obviously much influenced).5 The force of such an example as that with which this section commenced—Gulliver and his watch—is that it dramatizes the difference between seeing activities only as they manifest themselves externally to the senses, and seeing them with understanding, understanding in terms of the categories used by the agents themselves—for example, what is the difference between marking a cross on a paper, and voting in an election, and how would we explain the differences?6

I trust that it is obvious that the use of clocks and watches is comprehensible only in terms of the existence of shared social conventions of a kind, indeed, which could without any strain be called ‘rules’. Why, for example, is a watch to be considered defective if its hands do not move in such a way that the larger one completes twenty-four revolutions, and the smaller two revolutions between one mid-day (as determined by the perceived position of the sun, or whatever) and the next? Why is it wrong to say ‘It is 11 o'clock’ when the longer and the shorter hands of a clock are both aligned with the numeral ‘12’? Why is it right to mark out the face of a clock with sixty equidistant points around its perimeter, every fifth of them marked out with one of the cardinal numbers from 1 to 12, in order?

The only possible answer to such questions is to explain that we have conventions for reckoning the passage of time in terms of which we envisage the passage of time from one ‘mid-day’ to the next as being divided into twenty-four (p.280) equal ‘hours’ and each of these into sixty equal ‘minutes’ and so on. Clocks and watches are artefacts which people make in order to reckon the passage of time in accordance with these units of measurement. The use of these artefacts in accordance with the measuring conventions further enables us to fix on names for different ‘times of the day’, and that in turn enables us to co-ordinate and synchronize our activities quite accurately, even when complex interrelations of many people's activities are involved, as for example in running a train service or a factory. But it is all a matter of convention. The units of measurement fixed on could have been quite different (indeed, those used in Republican Rome were quite different) the design of the artefacts could have been quite different and is indeed to some extent subject to changes of technology and of fashion over the years—I saw a television interview the other day involving a person who manufactured, inter alia, digital watches.

Certainly, to take one of Hart's points of distinction as between ‘habits’ and ‘social rules’, there is no doubt that the patterns of behaviour involved in telling the time is one which most contemporary parents in our society ‘strive to teach’ to their children, and ‘intend to maintain’. Telling the time is an important skill for western man—the Lilliputians came nearer to the mark than perhaps they knew in supposing that Gulliver's watch might have been ‘the God that he worships’. And it is a skill which must be learned, and learning it involves acquiring the capacity to distinguish correct from incorrect statements of the time for any given configuration of the hands on the face of the clock, how to correct the setting of the hands on any particular clock so as to make it conform to some standard time, e.g. Greenwich Mean Time, how to keep a watch properly wound up and so forth. The very fact that we are able to describe the skill, indeed only able to describe it, by reference to correct and incorrect performances in given contexts, draws attention to that particular feature of rule governed conduct which Hart places at the centre of his picture thereof: the manifestation of critical attitudes, the appraisal of actual performance against a conceived standard performance, which is manifested by the use of terms such as ‘correct’ and ‘incorrect’.

(p.281) This learning of correct and incorrect performance is, in a distinctive sense, the learning of a (conventional) skill, and this is a fact from which interesting consequences follow. To say ‘it is 10 o'clock’ when both hands of a watch are aligned with the 12 is to make a mistake, to tell the time incorrectly, to be wrong in what one says; but it is not a case of ‘wrongdoing’, not a case of acting badly, or being bad. There is a difference in the criticism proper to deviations from or lapses of skill, from that which is proper to breaches of duty. The explanation of the difference is interesting: a conventional skill such as telling the time is constituted by mastery of a complex set of interrelated norms none of which makes complete sense apart from the whole set. To know and understand the norms is to have the skill, in the sense of being able to use it whenever one has reason to. What is more, it is only when an agent has reason to, and does decide to, ‘tell the time’ that the norms are relevantly applicable. The only actions to which the norms of the skill are relevant are those which the agent (for some reason) intends as instances of it. Somebody who said ‘It is 11 o'clock’ may either mean his utterance as a statement of what the time is, or (for example) as a statement in the children's game ‘What's the time Mr. Wolf?’ and it is his intention which is decisive of the question which one of these (or what else) it is. A statement can only be correct or incorrect as a statement of the time of day, if that is the kind of statement which it is intended to be, or to be understood as being. If it is a move in a children's game, or something else, some other canons of criticism would be relevant, but not the time telling norms. Hence, since what makes the norms of the skill or practice applicable is the agent's own intentions in the matter, we can see why the particular mode of criticism applicable to breaches is criticism in terms of ‘mistake’ or ‘incorrectness’ or ‘inaccuracy’ rather than ‘wrongdoing’. To get it wrong is to miss one's own intentions, by misapplying norms the applicability of which has been determined by one's own will. In terms of a distinction which I have drawn elsewhere the norms involved are ‘procedural’ rather than (p.282) ‘substantive’.7 Of course, it does not follow, and has not been said, that the existence of the norms is dependent upon one's will. The social practice exists, and is available for one's use whenever one has reason to choose to involve oneself in it.

To that extent, the norms of this skill parallel those norms of a legal system which regulate procedures whose invocation is at the option of the individual. Conveyancing, for example, is a difficult and complex skill to the learning of which much effort has to be devoted. The law specifies what steps must be taken in order that ownership of Blackacre shall pass from A to B. But it is up to individuals to decide whether to convey or not to convey their own property, and if they do, to ensure that they (or their lawyers) conform to the required patterns. That the norms of conveyancing (considered as a practice) or of time telling (seen likewise) leave it entirely at one's option whether to invoke them on any given occasion or not in no sense implies that it cannot be on other grounds obligatory to invoke either set of norms. For example A may by contract bind himself to transfer Blackacre to B on 1st July, thus obligating himself on that date to execute a valid conveyance of Blackacre, in accordance with the relevant institutive rules. And so too, whoever accepts a job in a factory which works under a shift system accepts an obligation to tell the time at any rate to the extent required for keeping time properly. Indeed for us all time keeping is so very important a skill, just because the whole round of our business and social obligations, attending meetings, keeping appointments, going to the theatre, or whatever, is dependent at every point upon the clock; that is, on our accurate use of properly set and maintained clocks.

In criticizing an act as wrongdoing or breach of obligation, we are not concerned at all with the question whether the agent under criticism acted, or failed to act, with reference to the norms whose breach is at stake. Whether or not ‘the (p.283) subjective meaning’ of his act was as a fulfilment or breach of the norm is irrelevant to the question whether it was such, and whether it can be relevantly criticized as such. (That is not to say that the agent's intentions are irrelevant—the criminal law in relation to murder or assault is certainly concerned with the question whether the accused intended to injure or kill his victim. What is not necessary at all to the application of the rule against assault or the rule against murder is that the act under consideration should have been performed with the intention of its being recgnized as an instance of ‘assault’ or ‘murder’. This is a sharp contrast to what is necessary for an act to constitute a conveyance, or for that matter an act of telling the time.) So there is at least this characteristic difference between the norms of a skill and the norms of obligation: the former are only relevant to those acts which are, or are presumed to be, performed with reference to the norms themselves; but the applicability of the latter is not conditional on an agent's subjectively intending to invoke them. These are distinctions which it is only possible to make from the perspective of an ‘internal point of view’.

What has been said so far has shown how an activity which anyone who reads this essay is presumed to take for granted, the activity of using clocks and watches and telling the time by them, despite its superficially simple and taken-for-granted quality, is one which can in truth be understood only by reference to a set of conventional norms which is really very complex (so much so that most of us would find it difficult if not impossible to give any very clear or coherent explicit account of the whole set which we implicitly use with such ease and assurance). They are essentially social norms, both in the sense that all of us learned them from other people rather than inventing them for ourselves, and in the sense that their whole point is dependent on their being shared norms constitutive of a practice of telling the time which we can all understand in more or less the same way as each other. That it is the norms which make sense of the activity can be demonstrated by showing what one would make of the activity if one didn't understand the norms; a demonstration which Swift's literary skill makes it the easier to give. (p.284) By seeing that there really could be an external point of view, we make ourselves aware of the internal point of view and of its importance.

It is important to investigate further what is involved in this awareness. There is and must be here an appeal to an awareness of the content of our conscious thoughts and processes of thinking. A behaviouristic account of the use of clocks and watches would necessarily, however elaborately it were constructed, rest at the level of the Lilliputian commissioners' description, restricted to the perceptible phenomena, without reference to the categories which are essential to the conscious use of clocks and watches as instruments for measuring the passage of time. I do not say that there could not be interesting behaviourist accounts, e.g. of how children are taught to and learn to tell the time. But the product of the teaching, the skill learnt, could not itself be explained in such terms.

And so to generalize that: any account of rules and norms and standards of conduct must be in terms of this ‘internal point of view’. That is to say, it must take account of the consciousness of those who use and operate with whatever standards of conduct may be in question. To anyone who would object that this is an unscientific approach, one could only retort by reiterating the wise saying that it cannot be wrong to be anthropomorphic about people. Let whoever doubts that, contemplate how he would seek to explain to the Lilliputians the true nature and uses of the wonderful kind of engine discovered in Gulliver's right fob.

It is not clear to me how far Hart himself would go in accepting this wholeheartedly mentalistic view of the nature of what he identifies as ‘the internal point of view’. Let us recall that he treats as the necessary element of the internal point of view the existence of ‘a critical reflective attitude to certain patterns of behaviour as a common standard’, of which he at once goes on to say that ‘this should display itself in criticism…demands for conformity, and in acknowledgements that such criticism and demands are justified, all of which find their characteristic expression in the normative terminology of “ought”, “must”, and “should”, “right” and “wrong”’.8

(p.285) It isn't entirely clear whether such expressions of demands and criticisms are envisaged as being constitutive of, or merely evidentiary of, the critical reflective attitude envisaged. The use of the terms ‘Characteristic expression’ has an empiricist flavour about it which may increase doubt. In fact, it is not so much that criticism is characteristically expressed in normative terminology, as that it is appropriately so expressed; and as that more candid phrasing indicates it is the conventional norms of the language which determine how we should express the criticisms, demands, and such like which we may wish to make. But to say that the attitude in question is appropriately expressed by such terminology is frankly to concede that it could be otherwise expressed, and indeed (most important of all) that its existence is logically independent of how it is expressed. What is more, it would not seem very satisfactory to offer in explanation of the nature of norms an account of the appropriate use of normative language. For among other things, what is at stake is to understand how we can have standards of appropriateness of this kind.

On the other hand, if we strip off these problems about how the necessary attitude displays itself, we are left with the unadorned idea of ‘a critical reflective attitude to certain patterns of behaviour as a common standard’, which is illuminating, but only incompletely so, since it seems almost to be circular. What, after all, is ‘a standard’? Do we not have to account for it in much the same way as ‘a rule’, or in exactly the same way? The road to a further explanation in terms of the characteristic expression in which this is displayed being cut off, for reasons already given, there is a problem about the way in which to advance towards an improved explanation.

Let us consider what must essentially be involved in our having a critical attitude to, or making a critical appraisal of, the actual conduct of oneself or other people. In this there must be some element of appraising the actual against some other possible configuration of action in the given circumstances which confront us. We must indeed have in view some ‘pattern of behaviour’; that we use such a conceived ‘pattern of behaviour’ as that against which to appraise actual behaviour (p.286) is indeed what constitutes that pattern as a standard for us. But only given a certain sense of the term ‘appraise’. When we speak here of appraising we mean more than merely indulging in some quite neutral comparison as between the actual and the conceived state of affairs, followed by a recording of any divergence between the two. The conceived pattern must be a pattern of preferred or approved as against disapproved conduct. What makes ‘not wilfully attacking other people’ or ‘saying “It is 12 o'clock” only when both hands of the clock are aligned with the 12’ standards of conduct? Not just that these are possible patterns against which we could conceivably appraise actual conduct, but that these are patterns of conduct actually willed, desired, preferred, approved as patterns for conduct and its appraisal in our own society just now. The relevant appraisal is appraisal for conformity or not with some pattern which is envisaged as a pattern of preferred conduct.

For a person to be sincerely critical of any given action of himself or another, there must be some alternative mode of acting in the circumstances in question which he would in some degree prefer over that actually undertaken. Common social standards must depend on shared preferences for certain modes of possible action over others. It can only be a will for conformity of actual actions to conceived patterns of action which is constitutive of standards of conduct. To make sense of the idea of ‘critical appraisal’ of actual conduct against some conceived pattern it is necessary that we consider the ‘conceived pattern’ as being willed by somebody. Such will, preference or approval need not be envisaged as arbitrary; for example, there may be eminently good reasons for preferring that people refrain from assaulting each other, or that they follow certain practices of measuring and telling the time. Such reasons for preference may themselves be further norms of a higher order, though outside of theological ethics, they cannot all be.

There is at least one obvious line of objection to the view here advanced. It is surely not the case that every judgment of action by reference to a given standard, every appraisal of any state of affairs by reference to some norm is necessarily an expression of the will or preference of the person who (p.287) passes judgment or appraisal. Equally, and conversely, not every expression of will or of preference is an expression of a social (or any other kind of) norm, nor presupposes any such. How then could it be possible to assert any necessary connection between wills or preferences and standards of conduct? Is it not indeed a common and persistent source of error to set up a simple correlation of norm with will, as in all variants of ‘the command theory’ of law?9

The points made, though true in themselves, are not decisive as objections to the view here put. Plainly, for any given social grouping, a social norm can exist in relation to some behaviour even if not everybody wills conformity to the envisaged pattern. Some members may be indifferent, some even in some degree actively unwilling to conform to the pattern, or hostile to its being maintained. It would, one may suppose be highly unusual in any group of any size to find exactly similar attitudes among all members of the group, though perhaps there are easy cases like those of assault and murder.

What is more, one can understand, and work with, social norms, even in the case of being indifferent or hostile to them. Two years ago, there were many Scots lawyers who would greatly have liked to see the divorce law radically reformed, but who could nevertheless understand and work along with the law as it stood. But always the position of those who, while understanding a norm, and able to frame judgments in terms of it, remain indifferent or hostile to it, is a position which is parasitic on that of those who do for whatever reason will the pattern of behaviour in question as a standard for all in their group. This detached view of social rules makes sense only if those who hold it suppose (and it may be a false supposition) that there are some who do care about maintenance of the pattern of conduct in question. That there can be common patterns of criticism of conduct or states of affairs depends upon our conceiving that some patterns are willed as common patterns for all people in given circumstances. We can conceive of that independently of our own will in the matter, but not independently of our beliefs (p.288) about the will of other members of our social group—or, if we are anthropologists or tourists, of the groups we are studying or passing through and seeking not to offend.

Why should anybody ever conform to a pattern of conduct which he does not himself regard as a desirable or preferable standard of social conduct? The answer is obvious: other people's willing it as a standard entails their disapproving of deviant conduct, and few people are very keen to attract the avoidable disapproval of their fellows. What is more, part of what is expected of us may be ostensible commitment to the supposedly shared rules of our society, so that one may seek to avoid disapproval by oneself expressing criticism of deviant conduct, against one's own private sympathy. Hypocrisy, as the saying goes, is what makes the world go round; it certainly seems to be an essential ingredient in the cement of an ordered society.

In his own account of the ‘internal aspect’ Hart, as we saw, is anxious to reject the view that it is a matter of ‘feelings’ about conduct. Specifically, he argues that feelings of restriction or compulsion—‘feeling bound’—are not essential to the existence of rules though some individuals may in fact experience such feelings ‘where rules are generally accepted by a social group and generally supported by social criticism and pressure for conformity’. That is true, but it should not lead us, as perhaps it has led Hart, to ignore the important affective elements in the ‘internal aspect’ or the ‘internal point of view’. There is an important, indeed essential, volitional as distinct from cognitive element in the internal aspect of rules, understanding of which is essential to an understanding of rules. When Hart himself speaks of the ‘acceptance’ of rules in a social group, he seems to have in view precisely such a volitional element as that which has been discussed here, though one of the weaknesses of his account taken as a whole is that he fails to give anywhere a single specific explanation of the relationship between the various intertwined conceptions which throughout the book are central to his theory—namely the ‘internal aspect’, the ‘internal point of view’, ‘internal statements’, and ‘acceptance (p.289) of rules’.10

What seems beyond doubt is that the volitional element of the ‘internal point of view’ must be recognized as central thereto. It is the fact of people's will for conformity to a conceived pattern of action, of their preference for some rather than other possible configurations of such action in given circumstances, that is for them the primary ground of criticism and reflective appraisal of actual conduct in society. Any conceived pattern of action such as:

‘People taking reasonable care to avoid harming other people’

‘People marking out the perimeter of clock faces into sixty equal divisions’

could be the content of a norm; could be envisaged as a pattern for the critical appraisal of actions. But only such as are actually willed by some people as common patterns for a social group can be thought of as actual social norms.

But for the existence of a norm in a social group, not all its members need have this volitional commitment to it. Some indeed must, but others may simply ‘play along’ not out of conviction but from a kind of salutary hypocrisy. Different again is the ‘delinquent’ position, that is, the position of those who accept and prefer the common patterns, subject to exceptions for themselves, so far as they can get away with it. Thieves who are capitalists (as most are said to be) are a case in point. Again, there is the position of the ‘rebels’ who know and understand but actively reject the social norms willed by the more dominant groups in their society.

The ‘playing along’, ‘delinquent’, and ‘rebel’ positions (and any variants or intermediate types) are all comprehensible only in apposition or opposition to the volitionally committed position; it is necessarily presupposed by them while they are not presupposed by it. Of course, in the life of any complex social grouping it may be difficult or impossible to identify with confidence the members of the ‘committed’ group or (p.290) fully to understand all the cross-currents of attitude which may be in play. But understanding the social norms and rules of a group involves an assumption of some people's will as underpinning and sustaining the patterns which thus underpinned and sustained are the norms of and for the group. Such understanding does not necessarily entail sharing in the relevant attitude.

The assumption or presupposition of that sustaining and underpinning will which is involved in understanding or a fortiori using and applying social rules norms and standards is of course not stated in any given judgment passed by reference to the norm; nor does the judgment necessarily express any such will. Understanding, using, passing judgment in terms of given social norms, are not to be thought of as acts which must themselves be accounted for in terms of an expression of the subjective will of the person who understands, passes judgment, etc. At one end of the spectrum, critical reflective attitudes are much more markedly reflective than actively critical; though at the back of them all must rest some sense of a genuinely felt preference on somebody's part against which, ultimately, the critical reflection makes sense.

There are all kinds of statements which may be made which presuppose an understanding of given norms and all that that understanding in turn presupposes. Even to say ‘It is 11 o'clock’, meaning it as a statement of the time, is to make a statement which presupposes an understanding of the given conventions, whose existence as a set of operative social norms in turn depends on a widely disseminated will to maintain them as such. Yet the statement itself does not express such a will. And so too for statements of, or in, the law: ‘As a seller of goods by description, you must deliver goods conform with the description’, ‘As occupier of these premises, you must take reasonable care for your lawful visitors’, and so on. Such statements presuppose an understanding of a set of legal norms, whose existence is explicable only in terms of some volitional commitments by some members of a society, though they do not in themselves express a commitment on the speaker's part.

Such statements belong to the category of what Hart has (p.291) called ‘internal statements’,11 on the ground that they can only be made ‘internally’ to the norm systems which they presuppose by way of truth conditions. And at least at some points in his narrative, he assumes that to make such statements is to evince the ‘internal point of view’. But it will be noted that what determines the ‘internality’ of a statement is the understanding, not the will of the speaker. So far as that goes, indeed, an entire outsider to a given society could form an understanding of its norms, to the extent of being able to make ‘internal’ statements, ‘internal’ to the social norms of that society. I take it that that is precisely what social anthropologists try to do.

To observe that is to observe a crucial ambiguity in the internal/external distinction as drawn by Hart: is it a distinction between levels of understanding, or a distinction between degrees of volitional commitment? There are, it is submitted, two very important distinctions there; yet it seems to be the case that Hart has to some extent at least conflated them. We started this appendix by considering the ‘externality’ of the Lilliputian commissioner's report, which was external in the sense that it revealed a failure to understand Gulliver's conduct except in its overt behavioural manifestations. In a similar way, the study of human beings from certain scientific perspectives might involve ignoring the norms which people as agents would regard as guiding their conduct; to do so would be to study human behaviour only in its external aspect.

Whoever seeks to go beyond that level of understanding, and to appreciate conduct in terms of the categories which for the agent are crucial, is taking a radically different view of it and one which deserves to be described for some purposes at least as ‘internal’. But this is ‘internality’ only at the level of understanding, for the observer in this case may remain entirely detached and uncommitted as regards the norms understanding of which is vital to his enterprise.

(p.292) But of course that sort of detached understanding, like the not dissimilar uncommitted attitude of the person who plays along, or the delinquent, is parasitic upon and presupposes the position of volitional commitment which we have seen as essential to the existence of norms, and central to the internal point of view. For that reason it is important to draw a further line of distinction here in terms of differing dispositions of will.

To summarize the point: There is a genuine distinction as drawn by Hart between ‘external’ and ‘internal’ points of view with reference to human activity. But the ‘internal’ point of view as characterized by Hart contains essentially distinguishable components, which ought to be distinguished. There is ‘cognitively internal’ point of view, from which conduct is appreciated and understood in terms of the standards which are being used by the agent as guiding standards: that is sufficient for an understanding of norms and the normative. But it is parasitic on—because it presupposes—the ‘volitionally internal’ point of view: the point of view of an agent, who in some degree and for reasons which seem good to him has a volitional commitment to observance of a given pattern of conduct as a standard for himself or for other people or for both: his attitude includes, but is not included by, the ‘cognitively internal’ attitude.

I do not think that in Concept of Law Hart has observed the need to distinguish between differences in levels of understanding and differences in degrees of commitment, and to that extent I find his account ambiguous. But if there are defects in his account, that is merely a challenge to others to improve it. That is what I have here tried to do.


(1) H.L.A. Hart, The Concept of Law (Oxford, 1961). The passage with which I am primarily concerned here, and from which I quote extensively in the present and succeeding two paragraphs, is to be found at pp. 54–6; the themes first developed there recur frequently throughout the book, notably at pp. 86–8, 96, 99–100, 101, 105, 112–14, 197.

(2) See Peter Winch, The Idea of a Social Science and its Relation to Philosophy (London, 1958), esp. ch. 2, passim.

(3) See Ludwig Wittgenstein, Philosophical Investigations, ed. G.E.M. Anscombe and R. Rhees, 2nd edn. (Oxford, 1958), e.g. at 1.197–241.

(4) See, e.g., M. Rheinstein (ed.), Max Weber on Law in Economy and Society (Cambridge, Mass., 1954), pp. 11–12; also the passage and arguments from Weber cited in Winch, op.cit., pp. 45–51.

(5) See Concept of Law, p. 242, note to p. 54, citing Winch, op.cit., with reference to passages in which Winch's main argument involves close discussion of themes from Wittgenstein and Weber.

(6) For a discussion of this example, see, e.g., M. Lessnoff's The Structure of Social Science (London, 1974), p. 30; Lessnoff's book, esp. in chapter II contains a review of themes similar to those mentioned in this paragraph. Cf. also B. Wilson (ed.), Rationality (Oxford, 1973).

(7) See D.N. MacCormick, ‘Legal Obligation and the Imperative Fallacy’, in Oxford Essays in Jurisprudence Second Series, ed. A.W.B. Simpson (Oxford, 1973), esp. at pp. 116–29 for my attempt to elaborate and justify that distinction. The same essay contained a discussion of the idea of ‘internal intention’ (pp. 104f.) reference to which would be necessary fully to justify my assertion made above about intentions in relation to ‘telling the time’.

(8) Op.cit., p. 56.

(9) Cf. MacCormick, op.cit. supra, p. 282 n. 7, esp. at pp. 100–16.

(10) Fur uses of these expressions see the passages in Concept of Law referred to above, p. 277 n. 1. Raz in Concept of a Legal System (Oxford, 1970), p. 148 n. 3, draws attention to apparent confusion as between these various phrases. But in my opinion he throws the baby out with the bath-water.

(11) Op.cit., pp. 99–100. See also p. vii, ‘neither law nor any other form of social structure can be understood without an appreciation of certain crucial distinctions between two different kinds of statement, which I have called “internal” and “external” and which can both be made whenever social rules are observed.’