LAW AND LIBERTY IN J.S. MILL AND BERNARD BOSANQUET(1)

William Sweet

St. Francis Xavier University


What is the relation of liberty and the law? One finds two general answers to this question in 19th century Anglo-American social and political thought. The first--and, perhaps, the best known--is that of Jeremy Bentham, J.S. Mill and Herbert Spencer. On this view, given the value of individual liberty, strict limits must be placed on the law.

In the latter part of that century, however, one notes quite a different response--that proposed by the British idealist philosophers. Indeed, this approach is sometimes taken to be fundamentally incompatible, if not entirely incommensurable, with the ideas of Bentham, Mill and Spencer.(2) The most developed statement of this response is, arguably, to be found in the work of Bernard Bosanquet (1848-1923). Bosanquet claims that there is no conflict between law and liberty, and that liberty is dependent upon law. In fact, he writes that "the more compulsion, the more liberty, and vice versâ."(3)

Interestingly, both of these accounts of the relation of law and liberty claim to reflect the same principle and value--that of human autonomy. We have, on the one hand, J.S. Mill's remark that "over his own body and mind, the individual is sovereign"(4) and, on the other, Bosanquet's view that the fundamental legitimate authority over an individual is that individual himself.(5)

But is it not puzzling that two apparently incompatible accounts of the nature and role of law place a similar, if not identical, emphasis on the importance of individual autonomy? In this paper, I wish to solve--or, rather, `dissolve'--this puzzle by looking at how Bosanquet understood the conflict between what one might call Mill's `liberal individualism'(6) and Bosanquet's own `communitarian idealism'. To begin, I shall outline the views of Mill and Bosanquet concerning the nature of liberty and the law. I then turn to some of the presuppositions of Mill's account and argue that it is here that the real difference between Mill and Bosanquet is to be found. Nevertheless, I also suggest that Bosanquet's rejection of some of the aspects of `liberal individualism' is consistent with his inclusion of other elements of it within the idealist view. If Bosanquet's arguments here are plausible, this will explain why there was a shift in British social philosophy from (what might be called) an `individualist' to a `communitarian' understanding of the relation of liberty and the law. Furthermore, this would also provide some evidence for the claim that, to the extent that contemporary liberalism reflects 19th century presuppositions, it should--if consistently worked out--lead to a less `individualistic' view.

I

Liberal individualists see themselves as advocating principles that will ensure, if not maximize, the political liberty of the individual. In 19th century Britain, this tradition was particularly associated with a group of intellectuals called, by Elie Halevy, "the philosophic radicals."(7) Jeremy Bentham was one of the major figures of this movement, and J.S. Mill can be counted among its `spiritual descendants'.

The conception of `liberty' one finds in their work is much the same as that found at the beginning of the liberal tradition, in Hobbes.(8) There, liberty is described as "the absence of... external impediments of motion," and one is free when "in those things, which by his strength and wit he is able to do, [he] is not hindered to do what he has a will to do."(9) Bentham carries this conception over to the legal and political sphere. Consequently, as far as one is not hindered in the pursuit of the good by others in society, one has liberty and is `free'.

Corresponding to this account of liberty, the philosophic radicals, as Hobbes before them, viewed law `negatively'. According to Hobbes, law "determineth and bindeth" and is "inconsistent" with liberty.(10) It is, as Bentham saw, necessary to social order, and good laws were clearly necessary means to good government. But by its very nature, law is a restriction of liberty and is inherently painful to those whose liberty is restricted.(11) Given that pleasure and pain are fundamental to--indeed, provide--the standard of value for Bentham, liberty was a good(12) and its restriction was an evil.

While J.S. Mill's `liberalism' and his defence of it are distinct from those which one finds in his predecessors, he defines the concept of `liberty' in much the same way--as "doing what one desires" (L 95) or as "pursuing our own good in our own way" (L 12), understanding that individuals themselves (best) determine their own good (L 74; 107). Moreover, "[a]ll restraint, qua restraint, is an evil," according to Mill (L 94), and he shares Bentham's opinion that law, even though necessary to social order, is still a "restriction on the natural liberty of mankind."(13) (While one might object that Mill defended, in a way in which Bentham did not, the value of individual liberty, it is still plausible to hold that his account is not inconsistent with Bentham's own views.(14))

There is, however, a certain tension underlying Mill's account of liberty and the law. Allowing individuals to pursue their interests is, clearly, important. But if liberty is taken as an absolute, it may threaten political stability which, in turn, would threaten liberty. How might one avoid this result? If it is true that "one man is worth just the same as another man" or that, in calculating the greatest happiness "each person is to count for one and no one for more than one,"(15) then social institutions ought to be arranged so that the interests of any one individual do not dominate arbitrarily over any other. This led Bentham to the view that society should ensure that the interests of all be taken equally into account and that, if the interests of all cannot be satisfied, society should act so as to promote the satisfaction of those of the greatest number.

Mill was aware, however, that such a policy might permit a serious limitation of individual liberty. In On Liberty, then, one of his main objectives was to isolate and defend a "region of human liberty" (L 11) where one would be free from moral censure and from the penalties of the law--a goal that, he argued, was consistent with--and, in fact, based on--Bentham's greatest happiness principle.(16) Of course, according to Mill, law also is justified by this principle of utility(17)--specifically, by its necessity for security and for the maintenance of conditions for good life, such as one's liberty.(18) But even though law may be `necessary' and is to be preferred over anarchy, one must not forget that it is an `evil' and is inherently undesirable. Even in such writings as the Chapters on Socialism,(19) Mill did not see human happiness as being significantly enhanced through widespread use of the positive law.(20) Harkening back to the definition of `liberty' cited above, then, Mill would agree with Herbert Spencer that liberty is "to be measured... by the relative paucity of restraints that the governmental machinery imposes upon an individual."(21)

II

Bosanquet's account of the nature of liberty and the role of law can best be understood as a reaction to the `individualist' theories of Bentham and Mill(22)--although, as I shall note below, despite Bosanquet's view that the `liberal individualist' position was unacceptable,(23) his estimate of some of the basic values defended by it was quite positive.

To begin with, what is Bosanquet's view of law? By `law', Bosanquet means a set of "general rules" enforced by the state in order to "carry out a universal end in a plurality of units" (PTS 173).(24) But the law is not the mere command of the sovereign (PTS 37). For Bosanquet, the law is an act of what he calls `the general will', aims at a common good, is supported by the positive convictions or sentiments of individuals (PTS 36) and is explicitly formulated by the state.

Law is important, according to Bosanquet, for many reasons, not the least of which is that it reflects and allows for a multiplicity of social `institutions' to operate. The existence of the family requires, for example, the existence of laws concerning marriage and the transmission of inheritances. Again, one's membership in a community is, in the final analysis, rooted and sustained in the law by the state, which recognizes it, gives individuals the rights necessary to performing their various social functions, and protects them from interference with the exercise of their rights.(25) Consequently, social institutions cannot exist apart from the existence of the state and law.

Bosanquet also claims that liberty and the development of human personality both require the law. It is, for example, through the law (and, where necessary, the exercise of force) that the state hinders or punishes those who impede individuals in the exercise of their liberty. Again, certain compulsory enactments (such as those requiring and providing for the education of children) are, in fact, the means to "an enlarged liberty" (LL 369; see LL 376). Moreover, through the legal institution of punishment, the state "brings us to our senses... [and] makes us conscious of our errors and the moral decision to improve our behaviour" (PTS 208). It is, in other words, by means of the state and the law that "we find at once discipline and expansion, the transfiguration of partial impulses, and something to do and to care for, such as the nature of a human self demands" (PTS 140).

Given this account of law, it is not surprising that, for Bosanquet, `liberty' means more than `absence of restraint'. It means "becoming the best that we have it in us to be" (PTS 119), that is, `self-realization'. It is, then, not up to each individual to determine and to pursue his or her private conception of the good. Moreover, following Kant, Bosanquet notes that the `free mind' cannot exist "[e]xcept by expressing itself in relation to an ordered life".(26) He insists that "law and order" is "the condition and guarantee" of this liberty (PTS 119), and argues that there is no liberty where there is an absence of law. (In fact, he says that, as the `restrictions' imposed by law increase, so does our freedom [PTS 181-182; see LL 366-367].)

But Bosanquet also recognizes that `self-realization' is attainable only if human beings participate actively in their own development. It is for this reason that he insists on the importance of moral decision making; it is precisely through uncoerced moral activity that `the good'--the development of human personality--is possible. Thus, the law can properly be invoked only once certain criteria have been met: first, an important development in individual character must be being frustrated in some way; second, the benefits of such a development must be greater than the disadvantages involved in the restrictions to be imposed on the person; and, finally, it must be better that the action to be commanded by the law be done from any motive whatsoever than that it not take place (PTS 179-180). When the law is used to hinder the liberty--that is, the self-realization--of an individual, then, it is incompatible with itself. Clearly, law is far from being (as Mill's view would seem to have it) an evil.(27)

In short, then, law is coercive, but Bosanquet insists that the limits imposed by law are not foreign to human nature and, in fact, contribute to the development of the individual. Thus, like Mill, Bosanquet held that the function of law was to encourage intellectual and spiritual growth. Indeed, so far as the law reflects the general will,(28) its imperative end is freedom (PTS 94).

III

But given each's insistence on the fundamental value of human autonomy, how can one explain that Bosanquet and Mill should arrive at these different, if not opposed, analyses of the relation of law and liberty? The answer to this question, I think, lies not so much in their respective views of the nature of law or of liberty, but in a more basic disagreement. Bosanquet argued that Mill's work was flawed in several ways--and perhaps the most significant was that it reflected a defective view of the individual. Because of this, he claims, Mill was led to an inadequate description of law.

What is Mill's account of the nature of the human individual? On Bentham's view, individuals exhibit a natural rational self-interest--a psychological egoism(29)--and human action is described as being motivated by attempts to maximize one's own good.(30) While Mill did not obviously embrace Bentham's egoism--though there are traces of this in his work--he does, however, seem to adopt the latter's hedonism.(31) For example, in his discussion of the internal sanction of morality and of the "social feelings of mankind" (U 30, 58), Mill appeals to qualities which are, at base, those of pleasure and of pain.(32)

A second feature of this account was the apparent claim that the nature of the human individual can be adequately described without mention of social relationships.(33) The extension of the term `individual' is, in the main, no greater and no less than the biological entity. The implications one can draw from this are, first, that there is no `self' or `individual' greater than the biological individual and, second, that the individual is the basic unit in the social sphere. As Mill remarks in System of Logic, "[h]uman beings in society have no properties but those which are derived from, and which may be resolved into, the laws of the nature of individual men."(34) Thus, a person's relations with others--even if important (U 30)--are not essential and describe nothing that is, strictly speaking, necessary to its being what it is.

Consistent with this view of the individual, and (to quote George Sabine) "[a]t the core of this mode of political thought was a fundamental postulate about the nature of value, viz., that all value inheres ultimately in the satisfactions and realizations of human personality."(35) For example, as noted above, it is on the basis of pleasures and pains, which can exist only in individuals, that Mill was able to construct a calculus of value.(36) Moreover, recall Mill's view that the individual him or herself is "the final judge" of his or her "own concerns" (L 74), and that the "Greek ideal of self-development" (L 59) is an important goal to be pursued.(37) Clearly, the individual has a central value.

Admittedly, Mill does not infer from this alone--as some modern natural rights theorists have--that all individuals have, or get, a `liberty' that carries moral weight. As he writes in On Liberty, only "those capable of being improved by free and equal discussion" (L 10)--for example, mature adults in mature societies--have a claim to liberty. Nevertheless, once such a capacity is established, "[o]ver himself, over his own body and mind, the individual is sovereign" (L 9). It is the notion of the rational `atomic' individual(38) that underlies Mill's account of liberty, and it is this liberty that constitutes a limit on where and when the law can legitimately intervene in an individual's life.

In short, then, one sees here a psychological and a moral individualism, where "the individual human being is conceived as the source of values and as himself the supreme value".(39) While Mill did see social life as contributing to the development of the moral sentiments (U 30), to one's happiness and to the realization of the individual, he did not go so far as to claim that the nature of the individual was fundamentally social, or that it was dependent on (and, in some way, subordinate to) the institutions of social life (see L 56). Thus, given the physical independence of the human person, given the capacity for rationality and moral personality in human beings, and given the presumption that the individual is an essential element in the articulation of the principle of value (if not the principle itself), individuals have a status that Mill accepts as imposing limits on the law.

IV

Is this account of the human individual plausible? There are, no doubt, moments when it seems that it is. But sometimes one must look at things a second time. It is in just this way that Bosanquet approaches Mill's political thought. Bosanquet refers to the views of Mill--along with those of Bentham and Spencer--as, collectively, "theories of the first look" (PTS 73-75). By this he means that they are based upon the assumption that the nature of "the individual or society... is what it prima facie appears to be" (PTS 77). But this is precisely where the liberal individualist account fails, Bosanquet says. It reflects a view of the individual--and, consequently, of human autonomy--which, he thinks, on closer inspection proves to be inadequate.

To begin with, for there to be a "region of human liberty" outside of the control of the law, there must be some principle of `demarcation' that allows one to separate the individual from others--that is, to distinguish the sphere of individual autonomy from the sphere of social obligation. Mill, of course, claimed to have found such a principle--namely, that "acts which affect only the individual or need not affect others unless they like may not be punished by law or by public opinion" (L 11; see L 73-74). Bosanquet points out, however, that this criterion of demarcation is "perfectly arbitrary in its practical working.... For every act of mine affects both myself and others; it is a matter of mood and momentary urgency which aspect may be pronounced characteristic and essential" (PTS 60). More importantly, he believes that such a principle exhibits too narrow an understanding of the human person. It implies a view of the individual as "a sort of inner self" (PTS 178), independent of "the varied play of relations and obligations in society" (PTS 74). But this ignores, Bosanquet says, that "[a]ll individuals are continually reinforced and carried on beyond their average immediate consciousness by the knowledge, resources, and energy which surrounds them in the social order" (PTS 142). Mill's analysis of the sphere of human autonomy and the `principle of demarcation' which it reflects, Bosanquet concludes, depend on a view of the individual that is `atomistic' and which clearly misses out on essential characteristics of human personality.

But Bosanquet's criticism is not simply that Mill does not give a complete description of the nature of the individual; it is that Mill's attempt to isolate or separate individuals from their relations with others is itself highly problematic, if not inconsistent.

First, given the nature of his separation of `self' and `other', Mill suggests that the relation between the individual and society is one of `superior to inferior' or `means to end' (PTS 167, xxxiii). But, Bosanquet replies, this view ignores the interdependency between them and that the individual and society constitute "a single web of content" so that, strictly speaking, neither is the means to the other (PTS 168). Bosanquet holds that, by considering the individual and society as even potentially "alien and opposed" or "hostile" or as opposites, one cannot explain what either actually is.

Furthermore, Bosanquet would point out that Mill's `principle of demarcation' does not even do what it was intended to do--sc., ensure a `region of liberty' in which individuals are free and autonomous, and can `pursue their own good in their own way' (L 12). For example, sometimes "the maintenance of external conditions of good life" (PTS 64) lies within the power of the law but, because this could violate the putative boundary between `self' and `other', Mill's view would forbid it.(40) On the other hand, Bosanquet believes that it would certainly be wrong to suggest that, simply because our moral obligations touch on the legitimate interests of others, they can or ought to be enforced by law (PTS 63; xxxv). But Mill's view seems to allow just this.(41)

Bosanquet would argue that, in these latter cases, because Mill's `principle of demarcation' would permit (if not demand) the legal enforcement of our moral obligations, it not only might well destroy "the springs on which moral action depends" (PTS 64), but risks eliminating or thwarting individual moral development--one of the very values that Mill professes to recognize. To the extent that Mill leaves room for attempts to promote morality through direct state action--such as the threat of force--he either reduces the number of occasions in which individuals may be said to act morally, or simply replaces a moral motive to do the good with the prudential motive to obey the law.(42) Indeed, according to Bosanquet, enforcing morality is "per se a contradiction in terms" (PTS 105; xxxv). Clearly, what is important when it comes to deciding whether to appeal to the force of law is what effect the interference will have on the development of moral personality--not (as Bosanquet claims Mill suggests) what its source is.(43)

Finally, Bosanquet suggests that this principle may be turned against itself for, since little or nothing affects only the individual concerned, there is little or nothing which is exempt from "social interference" (PTS 59). And if one cannot separate `self' from `other', how can Mill insist that the law respect the distinctiveness of the self? Again, even if one could make such a separation, Bosanquet notes that Mill himself sometimes appears to violate his own principle, by forbidding contract into slavery (PTS 64; see L 101-102) and by forbidding certain self-destructive activity. (He reminds us, for example, that Mill allows one to restrain a person from engaging in such purely self-regarding activities as walking over a bridge which is sure to fall down and cause his or her death [PTS 64-65; see L 95].) To this extent, at least, Mill appears to be inconsistent.

Bosanquet notes that there is a yet further `inconsistency' in Mill--i.e., between his use of the principle of utility (which, in Utilitarianism, is claimed to be the basis of rights)(44) and the insistence on the importance of the moral rights of the individual (which one finds in On Liberty). Presumably the value of individual liberty and of moral right is determined by their utility, just as utility is also the basis for the moral authority of the state and law. But when, in On Liberty, the law is forbidden from restricting individual liberty, the reason given, Bosanquet maintains, is no longer that it is contrary to utility, but simply that it violates the `principle of demarcation'--that is, that it is a force `outside' individuals and that it cannot impose limits on what is of legitimate interest to them alone.(45)

V

Bosanquet argues, then, that given the inadequacies of Mill's `principle of demarcation' and of the description of the human person that it implies, the concepts of `law' and `liberty' which follow from them cannot be sustained. But there are a number of additional problems in Mill's account of the nature of law and its relation to liberty.

Consider, for example, Mill's view that, although law is the source of rights, it essentially involves a diminution of one's `liberty'. Bosanquet argues that the liberal individualist position here is inconsistent. The claim that law is possible and that "the citizen can acquire rights only by sacrificing a part of his liberty" implies that one has "a certain area of liberty, of which a portion is abandoned to save the rest" (PTS 54). But, Bosanquet notes, "the idea of such an antecedent liberty is just such a fiction, as Bentham himself delighted to expose" (PTS 54). There is no `original liberty' and hence there can be no basic liberty to `sacrifice' in order for the law to exist and to secure one's rights.

Or again, consider Mill's discussion, in On Liberty, of the nature and role of law. As noted above, according to Mill, each individual has a `region of liberty' where he or she is sovereign. Any interference from `outside'--particularly from society and the law--seems to be not only alien, but hostile, to his or her liberty.(46) `Law' and `liberty' are, thus, conceptually opposed.

But is this really so? This putative antithesis between law and liberty seems to depend on a view of the individual and of liberty which, Bosanquet charges, is not only unfounded but simply false. The individualist position suggests, Bosanquet notes, that "the claims of `others' in society is a... curtailment of the liberty of the `one'" (PTS 55). But how is it that these claims genuinely diminish one's room for action? Indeed, the existence of others and of the entire array of social institutions may give one the opportunity to engage in activity otherwise impossible. "We profit at every turn by institutions, rules [and] traditions" (PTS 55) and such institutions represent "the ground won and settled by our civilization, and [leave] us free to think and will such matters as have their value in and through being thought and willed rightly" (PTS 200). In fact, Bosanquet says, "we are apt to... forget how great is the effect, for the possibilities of life throughout, of the mere fact that a social order exists" (PTS 189).

Moreover, Bosanquet argues that, even apart from this, Mill is being inconsistent. It is curious, he says, that "if law and government [are] in their nature antagonistic to the self of man... [that Bentham and Mill] nevertheless admit... that a certain minimum of this antagonistic element is necessary to the development of the sentient or rational self" (PTS 52-53). Surely, then, such a recognition should oblige them, on their own terms, to deny both the claim that every limit on liberty is an evil and the conclusion that law itself is an evil. Otherwise, Bosanquet concludes, "[w]e have here a dualism which challenges examination" (PTS 53).

In short, then, Bosanquet argues that Mill's attempt to draw some `line of demarcation' between `self' and `other' not only reflects a fundamental misunderstanding of the nature of the human person but, if adopted, would risk interfering with individual moral development and threaten the very `self' that he claims to want to protect. Furthermore, since one's pursuit of the good--one's liberty--and the development of one's moral personality (and, perhaps, more) are not threatened, but are made possible, through the law and the institutions which reflect it, Bosanquet holds that there is no a priori antithesis between liberty (or an individual's private interests) and the law.(47) To avoid the putative incoherence and inadequacies of Mill's position, therefore, Bosanquet holds that we must look at the individual self in a way that acknowledges its social nature and the contribution of law. It is in this way alone that one can do justice to both the importance of the individual and the value of liberty.

I cannot sketch out Bosanquet's alternative theory of the human individual here,(48) but it is clear what general lines it must take. The individual is fundamentally a social being; its self-awareness, moral consciousness, and personal development are dependent upon social life and the influence of others. Bosanquet says that "man's moral being lies in his social being"(49) and that even the individual will is, in some sense, constituted by society.(50) In fact, for Bosanquet, the very concept of the self cannot be articulated, independently of reference to other persons. Again, one's social character is not something accidental, or explicitly assented to, or willed by, the individual. It is, rather, something implied in its very nature. (Still, the individual is the being who, through the fulfilment of its various tasks and duties, puts the `social power into operation' [LL 372].) And, further, Bosanquet's account of the human person is one that emphasizes the importance of rationality--although the concept of rationality employed is broader than one that presents it as being merely an aspect of consciousness, instrumental to attaining conscious ends. So far as it is essential to the existence of social institutions and the functioning of social life, and because it is in itself `rational', law is necessary to the nature and development of the human person. Elements of this view are, admittedly, present in Mill, but they are not--and cannot be--developed given the fundamentally individualistic character of Mill's account of human nature.

It is important to note that the disagreements between Mill and Bosanquet here focus on the `individualistic' elements of Mill's view--not, I would argue, on the liberal ones.(51) There are, indeed, some striking similarities between Mill's and Bosanquet's accounts of liberty and the law. Like Mill, Bosanquet believed that the basis of true self-government was the recognition of "the claim to obey only yourself" (PTS 134). The origin and justification of the coercive force of government must ultimately lie, Bosanquet acknowledged, in the individual. Moreover, Bosanquet praised Mill for acknowledging the importance of the development of a sense of social solidarity and of its role in social life. And again, like Mill, Bosanquet was concerned with the question of "how far and in what way the use of force and the like by the state" is justified (PTS 171). He saw, as did Mill, that there are limits on what law or the state can do. Specifically, Bosanquet insisted that it cannot--or, at least, should not--"attempt to prevent by punishment either immorality or irreligion as such" (PTS 61). Finally, both Mill and Bosanquet maintained that "some degree of restraint on what we can now easily imagine ourselves free to do, is involved in political society" (PTS 54). In short, both the liberal individualists and the idealists agreed that the state and law are necessary to life in society, but that limits on their authority must be specified. And there is some evidence that Bosanquet saw in his own work a continuity with the liberal principles of the tradition of the philosophic radicals.(52)

Both Bosanquet and Mill agreed, then, on the liberal value of autonomy--of `giving law to oneself'. But what `autonomy' means and amounts to is, of course, dependent on what it means to be a `self'--an individual--and it is here that they parted company. It is, moreover, because Bosanquet finds Mill's notion of the individual self to be too narrow, if not altogether mistaken, that he also argues that Mill's corresponding accounts of liberty and the role of law are ultimately defective.

VI

In the last few pages, we have seen the two principal objections that Bosanquet raises to Mill's account of liberty and the law. The first is simply that Mill's arguments make assumptions about the nature of the individual and the nature of the law that are not, or cannot be, justified. Bosanquet rejects Mill's view that there is an antagonism between the individual and the law because, he says, it rests on an erroneous understanding of the character of the law--that is, as not merely restrictive and coercive, but opposed to liberty. He claims that this overlooks how the law does or can contribute to the realization of individual liberty and ignores how law is a consequence of the will of the individual. According to Bosanquet, there is little to be gained by an a priori limitation of the law,(53) and any suggestion that its value is instrumental to that of the individual--or vice versa--is fundamentally mistaken.

In the second place, Bosanquet would point out that many of Mill's arguments are incompatible or inconsistent with basic liberal principles that Mill professes to recognize or with values that he professes to respect (such as the value of human self-development, the importance of self-government, the necessity of allowing room for moral action, and the role of the law in the liberty of the individual). If, for example, law is seen as necessary to the development of the individual, it seems inconsistent to describe it as even a prima facie evil. Or again, if the individual is, in some way, the source of value, it is incongruous for Mill to insist on a boundary between `self' and `other' that is so vague that it could endanger the very values he seeks to protect.

Bosanquet, then, not only challenges Mill's account, but suggests that, to be faithful to the moral principles and values of liberal individualism one must move beyond this position. Such values require the recognition of a broader notion of the individual (on which a more complete account of human autonomy can be based) and of a positive concept of the law (where, although it is an instrument of coercion and can act only in a negative way, it is also seen as constituting a necessary condition for the liberty, autonomy and rights of the individual).

VII

On the basis of the preceding discussion, one can see how Bosanquet believes he has solved the puzzle noted at the beginning of this paper--sc., how two apparently contradictory theories of the relation of liberty and the law can reflect a similar emphasis on the importance of human autonomy.

According to Bosanquet, the problem is not so much that idealists and liberal individualists disagree about the nature of liberty and its relation to law--though they obviously do. It is, rather, that the liberal individualist view is based on an account of the human person that is simply inadequate to the task (i.e., incomplete and inconsistent). It is because of this that both Bosanquet and Mill could agree on the liberal value of autonomy, and yet arrive at radically different views of liberty and the law.

One should realize, however, that Bosanquet's `solution' was not to reject all of the underlying principles of traditional liberalism (e.g., the value of human liberty, the importance of individual development, and the necessity of imposing limits on the law), but to develop them and (he would argue) make them more consistent. If the preceding description of these two theories is correct and if Bosanquet's criticisms of individualism are well-directed, it would suggest that while there is a conflict between individualism and idealism, this conflict is, in fact, dissolvable. Thus, there need be no fundamental incompatibility or incommensurability between these two traditions, and `communitarian idealism' can be seen as a natural development of the social concerns of liberal individualism in 19th century Britain. Mill's view, then, comes to be understood not only as a development of the legal and political philosophy of Bentham, but also as an exposition--albeit incomplete--of those principles that later formed idealist social and political thought.

ENDNOTES

1. Previous versions of this paper were read at the Tenth International Conference on Social Philosophy (Helsinki, Finland), the 1993 Annual Meeting of the Canadian Political Science Association (Ottawa, Canada), and at the School of Economics and Social Studies, Politics and Philosophy Sectors, University of East Anglia (Norwich, England). I am grateful to the commentators and the audience for their helpful remarks.

2. It has sometimes been argued, for example, that their views are quite foreign to Anglo-Saxon philosophy--that British idealism is largely a transplant of 19th century German philosophy--particularly of Hegel. (See Rudolf Metz, Die philosophischen Strömungen der Gegenwart in Grossbritannien, Leipzig: Felix Meiner Verlag, 1935; [Translated as A Hundred Years of British Philosophy by J.W. Harvey, T.E. Jessop and Henry Sturt; ed. J.H. Muirhead], London: Allen and Unwin, 1938 and Klaus Dockhorn, Die Staatsphilosophie des Englischen Idealismus, Köln/Bochum-Langendreer: Heinrich Pöppinghaus o. H.-G., 1937, pp. 61-116.)

3. "Liberty and Legislation," in The Civilization of Christendom and other studies, London: Sonnenschein, 1889, pp. 358-383, at pp. 306-307. All future references to this essay will be abbreviated as LL and incorporated in the text.

4. On Liberty, (ed. Elizabeth Rapaport), Indianapolis IN: Hackett Publ. Co, 1978, p. 9. All future references to this volume will be abbreviated as L and incorporated in the text.

5. The Philosophical Theory of the State, 4th ed., London: Macmillan, 1923, p. 134. All future references to this volume will be abbreviated as PTS and incorporated in the text.

6. Although now in common use, this term is arguably anachronistic. James Crimmins, citing Fred Rosen's Bentham, Byron and Greece: Constitutionalism, Nationalism, and Early Liberal Political Thought (Oxford: Clarendon Press, 1992, p. 5), notes that "the terms `liberal' and `liberalism'... gained ideological purchase only in the second quarter of the nineteenth century" (Crimmins, "Benthamic Theory and Practice," in CPSA Annual Meeting 1993, Ottawa, ON: Canadian Political Science Association, 1993 (on microfiche), p. 21, n. 15). Michael Taylor suggests that the same holds true for the term `individualism' (M.W. Taylor, Men versus the State: Herbert Spencer and Late Victorian Individualism, Oxford: Clarendon Press, 1992, pp. 2-4).

7. Elie Halevy, The Growth of Philosophic Radicalism, [La formation du radicalisme philosophique] translated by Mary Morris, London: Faber & Faber, 1928.

8. For a discussion of this, see my article "Hobbes, Jaume et les droits inaliénables," in Carrefour, Vol. 13, No. 2 (1991): 121-135.

9. Thomas Hobbes, Leviathan, (ed. Michael Oakeshott), New York: Collier Books, 1962. Ch. 21, p. 159.

10. Leviathan, Ch. 14, p. 103.

11. According to Bentham, "[t]he evil of... restraint [is]... the pain which it gives a man not to be able to do the act." See his An Introduction to the Principles of Morals and Legislation, (ed. J.H. Burns and H.L.A. Hart), London: Athlone Press, 1970, p. 163 (All future references to this volume will be abbreviated as IPML.) (See also An Introduction to the Principles of Morals and Legislation, in The Works of Jeremy Bentham, [ed. John Bowring], London, 1838-1843; reprinted New York, 1962, Ch. 13, sec. 14, Vol. I, pp. 1-154, p. 85); cf. PTS 53.

12. According to Bentham's Principles of the Civil Code, Pt. I, Ch., 2 (in his Works, Vol. I, p. 302), liberty is a good, though subordinate to security (which is one of the four main ends of the civil law--the others being subsistence, abundance and equality). (Note that Bentham has in mind here what one might call political [as distinct from `individual' or `personal'] liberty [see A General View of a Complete Code of Laws, in his Works, Vol. III, pp. 155-210, p. 185].) Political liberty is, moreover, of two kinds--that which the law explicitly allows us to do (as correlatives to existing obligations), and that which the law does not expressly forbid (i.e., `permissive rights') (Bentham, A Complete Code, p. 181).

Still, Bentham denies that political liberty is natural, for then it could not be limited (see his Anarchical Fallacies, in his Works, Vol. II, pp. 497-498). See also John MacCunn, Six Radical Thinkers, second impression, London, 1910, p. 25 and Frank Thakurdas, The English Utilitarians and the Idealists, Delhi: Vishal Publication, 1978, pp. 59 and 81. There is some considerable debate on the fundamental character of liberty in Bentham's thought. For a general account of the `authoritarian' versus the `liberal facilitative' interpretations of Bentham, see Crimmins, op. cit.

13. Utilitarianism, (ed. George Sher), Indianapolis, IN: Hackett Publ. Co, 1979. p. 43. All future references to this volume will be abbreviated as U and incorporated in the text.

14. See Metz, British Philosophy, p. 73.

15. George Sabine, A History of Political Theory, 4th ed., Hinsdale, IL: The Dryden Press, 1973, p. 621.

16. Admittedly, one of Mill's arguments for the moral justification of the law appears to reflect a de facto contract. Mill writes that "everyone who receives the protection of society owes a return for the benefit" (L 73). Thus, while the law is required for the existence of one's liberty, it also implies a sacrifice of some part of it (L 73-77), i.e., there would have to be a recognition and agreement among persons about what liberties would have to be given up to save the rest (see L 73).

17. Of course, here we are addressing the moral basis of law; this is not the same as saying what the basis of law actually is. For Bentham, political obligation depended on external "sanctions" (such as the threat of punishment) and the "habit of paying obedience" (see A Comment on the Commentaries and A Fragment on Government, [ed. J.H. Burns and H.L.A. Hart], London: The Athlone Press, 1977. Ch. 5, ftn. 3, p. 496; Ch. 1, sec. 10, p. 428); for Mill, it was based on internal sanctions. On this issue of social instinct as reinforcing the feeling of moral obligation, Plamenatz writes that "[t]he external sanctions of duty, to which both the elder utilitarians devote so much attention, are described in a few words, and it is on the internal sanction that the younger Mill lavishes his descriptive powers" (see John Plamenatz, The English Utilitarians, Oxford, 1949, p. 139).

18. On Mill's view, liberty seems in many respects "natural": it is simply freedom from restraint. Still, he argues that political liberty is neither an intrinsic nor basic good, nor do we have an abstract right to it (L 10). It is, rather, a good derived from the principle of utility. By its existence, he says, "[m]ankind are greater gainers" (L 12). There is, of course, much debate on the relation between Mill's utilitarianism and the positions taken in On Liberty. See D.P. Dwyer, "Justice, Liberty and the Principle of Utility in Mill," in New Essays on John Stuart Mill and Utilitarianism, (ed. Wesley E. Cooper, Kai Nielsen and Steven C. Patten), Guelph: Canadian Association for Publishing in Philosophy, 1979 (Canadian Journal of Philosophy, Supplementary Volume V), pp. 63-74.

19. In Essays on Economics and Society, in The Collected Edition of the Works of John Stuart Mill, (ed. J. M. Robson), Vol. V, pp. 703-753, Toronto: University of Toronto Press, 1965-91, p. 97.

20. But see his `interventionist' attitude towards law in the Principles of Political Economy (1st edition, 1848; 7th edition, 1871) in Vols. II-III, Collected Works, esp. Book V, Ch. i and Ch. xi, sec. 16, pp. 970-971, and see his favourable attitude towards some forms of socialism in Bk. II, Ch. 1, sec. 4. For a discussion of this aspect of Mill's liberalism, see Peter P. Nicholson, The Political Philosophy of the British Idealists, Study V, Section 2, "J. S. Mill and Social Expedience," Cambridge: Cambridge University Press, 1990, pp. 140-157.

21. Herbert Spencer in "The New Toryism" in The Man versus the State, Caldwell, ID.: Caxton, 1960, p. 19.

22. Bosanquet's social and political thought was, of course, influenced by that of T.H. Green (and, though to a lesser extent, F.H. Bradley). Still, it must not be reduced to either. Bosanquet gradually became convinced that Green's account of the role of the state and law was too cautious and saw that Bradley's account was fragmentary and incomplete. See Nicholson, British Idealists, and my review of Nicholson in Laval théologique et philosophique, Vol. 48, no. 3 (1992), pp. 477-480.

23. See, for example, Bosanquet's remarks on Bentham's view of the relation of the individual to society in PTS 167-168.

24. See also PTS 240-243.

25. See my "Individual Rights, Communitarianism and British Idealism," in The Bill of Rights: Bicentennial Reflections, (ed. Y. Hudson and C. Peden), Lewiston, NY: Edwin Mellen Press, 1993, pp. 261-277.

26. PTS 236. Bosanquet cites Kant that, in living in civil society, man "has totally abandoned his wild lawless freedom in order to find his entire freedom again undiminished in a lawful dependence, that is, in a condition of right or law; (undiminished), because this dependence springs from his own legislative will" (PTS 226). See Immanuel Kant, Metaphysische Anfangsgründe der Rechtslehre (1797) (ed. Benzion Kellermann) (Die Metaphysik der Sitten, 2. Teil, 1. Abschnitt) in Kant's Werke, Bd. 7, (ed. Ernst Cassirer) Berlin, 1916. S 47, p. 122.

27. See PTS 53ff.

28. For a discussion of what Bosanquet means by the notion of `the general will', see my "Bernard Bosanquet and the Development of Rousseau's Idea of the General Will," in Man and Nature / L'homme et la nature, Vol. X (1991): 179-197.

29. See Sabine, Political Theory, p. 640.

30. See Halevy's list of the three principal characteristics of utilitarianism which constitute the basis of Bentham's philosophy: the greatest happiness principle, universal egoism and the artificial identification of interests (Halevy, pp. 12-17; see also Thakurdas, p. 87). (For a discussion of the coherence among the various aspects of Bentham's views, see also Plamenatz, Utilitarians, esp. pp. 70-72.)

According to Mill, "[h]abitually, and throughout his works, the moment he [Bentham] has shown that a man's selfish interest would prompt him to a particular course of action, he lays it down without further parley that the man's interest lies that way..." ("Remarks on Bentham's Philosophy," (1833) in Essays on Ethics, Religion and Society, in Collected Works, Vol. X, pp. 3-18, p. 14). Mill also cites Bentham's The Book of Fallacies (London, Hunt, 1824, pp. 392-3) that "[i]n every human breast... self-regarding interest is predominant over social interest; each person's own individual interest over the interests of all other persons taken together" ("Bentham (1833)", p. 14). Of course, Mill sees this as a failure in Bentham's view, and he complains that Bentham ignores that one might desire "for its own sake, the conformity of his own character to his standard of excellence" ("Bentham (1838)" in Collected Works, Vol. X, pp. 75-115, p. 95).

31. Recall Bentham's famous remark, at the beginning of IPML, that "[n]ature has placed mankind under the governance of two sovereign masters, pain and pleasure. It is for them alone to point out what we ought to do, as well as to determine what we shall do. On the one hand the standard of right and wrong, on the other the chain of causes and effects, are fastened to their throne. They govern us in all we do, in all we say, in all we think: every effort we can make to throw off our subjection, will serve but to demonstrate and confirm it" (IPML (ed. Burns and Hart), Ch. 1, sec. 1, p. 11). For the relation to Mill here, see note 32 below.

32. Mill does, admittedly, suggest that these moral sentiments (such as the `social feelings of mankind') are something more than pleasure and pain and Thakurdas and others insist not only that Mill abandons `psychological hedonism' (Thakurdas, p. 198; Plamenatz, Utilitarians, p. 144; John Robson, On The Improvement of Mankind, Toronto: University of Toronto Press, 1968, pp. 119-159, esp. pp. 127-136), but that he no longer believes in `universal selfishness' (see Mill, Considerations on Representative Government, in his Collected Works, Vol. XIX, Ch. 3, p. 405). Plamenatz says, for example, that "[j]ustice is the product of two sentiments... the instinct of self preservation and the feeling of sympathy (Utilitarians, p. 143). John Skorupski, too, argues that Mill had abandoned hedonism by the time he wrote System of Logic (John Stuart Mill, London: Routledge, 1989, p. 296).

Still, there does seem to be at least an element of self interest at work in Mill's account of `sympathy' (in U, Ch. 3, p. 27 and p. 30), and the development of sympathy seems to be, at best, a matter of the perfecting of human nature. (See Mill's comments on the motives of, and the relations among, persons where there is no representative government, in Considerations on Representative Government, in his Collected Works, Vol. XIX, p. 412.) Again, according to Alan Ryan, "the point of sanctions... is precisely to make it the individual's interest to do what is in the general interest" (John Stuart Mill, New York: Random House, 1970, p. 201; cf. pp. 197-204). (See also Thakurdas, p. 133 and Halevy, p. 502.)

33. Ellen Wood, for example, argues that this can be seen in Mill's `associationist,' rather than dialectical, epistemology. See her Mind and Politics: An Approach to the Meaning of Liberal and Socialist Individualism, Berkeley: University of California Press, 1972, p. 13 (quoting R. Amschutz, The Philosophy of J.S. Mill, Oxford: Clarendon Press, 1953, pp. 179-180), p. 42, note 41, pp. 51 and 109-110. Wood also claims that Mill's "determinism" is incompatible with his individualist emphasis in On Liberty. Martin Hollis also draws attention to the relation between Mill's epistemology and his atomistic individualism, in "J.S. Mill's Political Philosophy of Mind," in Philosophy, 47 (1972): 334-347, esp. pp. 336, 340. See also Mary Peter Mack's remarks on the similarity between Bentham and Hobbes's nominalism in Jeremy Bentham: An Odyssey of Ideas 1748-1792, (London: Heinemann, 1962) (cited in Thakurdas, p. 101).

34. As cited in Graeme Duncan and John Gray, "The Left Against Mill," in New Essays on John Stuart Mill and Utilitarianism, (ed. Wesley E. Cooper, Kai Nielsen and Steven C. Patten), Guelph: Canadian Association for Publishing in Philosophy, 1979 (Canadian Journal of Philosophy, Supplementary Volume V), pp. 203-230, p. 216. One finds the same view in Bentham, where he remarks that "the community is a fictitious body;" it is but "a sum of the interests of the several members who compose it" (IPML ch. 1, 3.4, p. 12).

35. Sabine, Political Theory, p. 608.

36. Sabine, Political Theory, p. 640.

37. One finds a similar focus in Spencer. According to this latter view, life is fundamentally valuable and "if we say that life on the whole yields more pleasure than pain... then [the] actions by which life is maintained are justified, and there results a warrant for freedom to perform them." See The Man versus the State, p. 195.

38. Although Mill would certainly hope that the early education of individuals would include moral training and the development of a concern for others (see U ch. 3, pp. 30-33), this is by no means a necessary prerequisite for liberty.

39. See Duncan and Gray, p. 215. One notes that in Mill's later essays, his work appears to have a less individualist tone. Much of this interpretation, however, refers to the Chapters on Socialism and the (almost certainly incorrectly attributed) On Social Freedom (ed. Dorothy Fosdick, [New York, 1941]). (For discussion on the authorship of this latter work, see J.C. Rees in his appendix to his Mill and his Early Critics, [Leicester: University of Leicester Press, 1956], pp. 38-54, reprinted in his John Stuart Mill's `On Liberty', [ed. G.L. Williams, Oxford: Clarendon Press, 1985, pp. ix, 175-185].)

40. Bosanquet argues, for example, that publicly-financed education is necessary for the moral and intellectual development of the young--and points out that this policy has been effective (PTS 63). Mill's objection to this, recall, is not merely that this "establishes a despotism over the mind" by public authorities (L 105) and leads to conformity, but also that it works against "the importance of individuality of character" (L 104). Bosanquet suggests, however, that Mill's fear of public intervention is unwarranted, and that his alternative proposal for "universal State-enacted examination" (PTS 63; see L 103-107) is more likely to produce bland conformism and a hierarchical society, inattentive to individual differences.

Still, it might appear that there is little fundamental disagreement between Mill and Bosanquet on this point--that it is just an empirical question of which policy will best lead to the development of individual character. But one must be attentive to the difference between each's view of the role of the state. Bosanquet is concerned with what would positively contribute to moral and intellectual development. Mill's view seems to be that this goes too far; the state can intervene only when a parent harms or risks harming (e.g., through negligence in assuring the education of) his or her children. But, Bosanquet would say, the issue should not be whether someone might be harmed, but whether the state can help.

41. Bosanquet sees this as a logical implication of Mill's view that the law should regulate marriage in order to ensure that parents have the financial resources to support and educate their children (PTS 63-64; see L 104 and 106-107). Bosanquet argues that, in such cases, the law can be only ineffective, since it is impossible to foresee what one's future economic circumstances will be. Moreover, he thinks that an appeal to moral obligation or sentiments (e.g., love for one's children) is more likely to be successful in providing one's children with a moral and practical education than the threat of legal sanctions. But (to Bosanquet's mind, at least) most importantly, coercion here eliminates or reduces the chances for the development of the moral character of parents (e.g., by the parents having to cope with the situation and having to choose to act in appropriate ways). Bosanquet is not, of course, saying that moral obligations should never be enforced, but that they should be so only as a last resort and when to do so makes a significant contribution to individual development that exceeds the negative effect of the coercion.

42. According to Bosanquet, neither law nor the state can directly bring about the existence of an appropriate moral motive. Morality is tied to the will of the agent, and this is something which the state cannot control.

43. In LL, Bosanquet argues that "[i]t is not true that a laissez-faire society makes original characters or that [limited]... social pressure permits strength of mind to develop" (LL 382). Yet Bosanquet will also argue that "economic socialism" also has the same effect (see "The Antithesis between Individualism and Socialism Philosophically Considered," in The Civilization of Christendom and other studies, London: Sonnenschein, 1899, pp. 304-357).

44. The source of one's moral rights, Mill says, is "the general utility" (U 52) and, hence, they can be limited "when some recognized social expediency requires the reverse" (U 62).

45. As noted above, there is much debate on the relation between Mill's utilitarianism and the positions taken in On Liberty. According to Plamenatz, "[i]t is in the third chapter of Liberty... that Mill, without knowing it, abandons utility" by promoting the value of individual spontaneity (Plamenatz, Utilitarians, p. 129). Alan Ryan writes that, after acknowledging the principle of utility, "Mill's concern with self development and moral progress is a strand in his philosophy to which almost everything else is subordinate" (Ryan, p. 255). John Skorupski, however, argues for a compatibility in Mill's views (Skorupski, pp. 343-347).

One might argue that this tension dissolves if the value of individual liberty is taken to have the weight of a utilitarian-justified moral rule. Aside from the question of whether one can maintain a distinction between act and rule utilitarianism, there is no evidence that Bosanquet would ever have considered such a distinction.

46. See chapter 4, "Of the Limits to the Authority of Society over the Individual," in On Liberty.

47. Bosanquet would also argue that Mill's liberal individualist account of law fails to explain the nature of moral and political obligation. Even though Mill (as well as Bentham) looked to representative government as the authority that is entitled to employ coercive force on individuals, laws based on the right of the majority still involve the coercion of `others' over `the one'. Consequently, Bosanquet argues, even though "popular instinct" admits (and "common sense" insists) that there must be such a thing as law and self-government (PTS 72), Mill cannot provide a "rational ground" on which "an entire community can apply coercion to a single recalcitrant member" (PTS 71).

48. On this topic, see my "F.H. Bradley and Bernard Bosanquet on the Nature and Value of the Individual," forthcoming in Mankind at a Turning Point: Philosophical Perspectives, Proceedings of the XIXth World Congress of Philosophy (Moscow, 1993), and "L'individu et les droits de la personne selon Maritain et Bosanquet," in Etudes maritainiennes / Maritain Studies, Vol. VI (1990): 141-166.

49. See "Individualism and Socialism," p. 309.

50. See "Individualism and Socialism," p. 313.

51. For an argument supporting the liberal character of Bosanquet's political thought, see my "Liberalism, Bosanquet and the Theory of the State," in Liberalism, Oppression and Empowerment, (ed. C. Peden and Y. Hudson), Lewiston, NY: Edwin Mellen Press, 1994.

52. In a letter of January 1907 to Andrew Seth Pringle-Pattison, Bosanquet writes that there was at least "one remarkable thread of continuity between the Philos.[ophic] Radicals and the later development of the social spirit" (J.H. Muirhead, (ed.) Bernard Bosanquet and his Friends, London, 1935, p. 114.)--though it is not surprising that he would think that there would be a continuity between his views and those of his predecessors, given his view of the progressive nature of society and social change. Moreover, Bosanquet says that "the lower [standard of liberty, as `freedom from felt coercion, and power to do what you please (LL 360)] must necessarily grow upwards into the higher" (LL 361).

What is continuous in British political thought, then, is an ideal of a democratic society where "free individuals, freely discussing and investigating, may freely develop their interests and their capacities" (See Frederick Philip Harris, The Neo-Idealist Political Theory: Its Continuity with the British Tradition, New York. King's Crown Press, 1944 [Ph.D. thesis, Columbia University], p. iii, cf. p. 103).

53. But neither does Bosanquet think that the law is `absolute', nor does he believe "that it is possible to lay down any single definite direction which must be taken by the course of legislative enactment as time goes on" (LL 372; see also Bosanquet's comments about how `economic socialism' does or doesn't work, in "Individualism and Socialism," pp. 345 ff.). The key is, rather, to see whether such intervention works--i.e., whether it is "definitely necessary on specific grounds for the support of the private moral will" ("Individualism and Socialism," p. 318; see LL 373 and 383).