Lectures on the Principles of Political Obligation / Reprinted from Green's Philosophical Works, vol. II., with Preface by Bernard Bosanquet

by Thomas Hill Green

192. The answer is: As regards heaviness of punishment, it is not in the power of the state to regulate the amount of pain which it causes to the person whom it punishes. If it could only punish justly by making this pain proportionate in each case to the depravity implied in the crime, it could not punish justly at all. The amount of pain which any kind of punishment causes to the particular person depends on his temperament and circumstances, which neither the state nor its agent, the judge, can ascertain. But if it could be ascertained, and if (which is equally impossible) the amount of depravity implied in each particular crime could be ascertained likewise in order to make the pain of the punishment proportionate to the depravity, a different punishment would have to be inflicted in each case according to the temperament and circumstances of the criminal. There would be an end to all general rules of punishment.

193. In truth, however, the state in its capacity as the sustainer of rights (and it is in this capacity that it punishes) has nothing to do with the amount of moral depravity in the criminal, and the primary reference in punishment, as inflicted by the state, is not to the effect of the punishment on the person punished but to its effect on others. The considerations determining its amount should be prospective rather than retrospective. In the crime a right has been violated. No punishment can undo what has been done, or make good the wrong to the person who has suffered. What it can do is to make less likely the doing of a similar wrong in other cases. Its object, therefore, is not to cause pain to the criminal for the sake of causing it, nor chiefly for the sake of preventing him, individually, from committing the crime again, but to associate terror with the contemplation of the crime in the mind of others who might be tempted to commit it. And this object, unlike that of making the pain of the punishment commensurate with the guilt of the criminal, is in the main attainable. The effect of the spectacle of punishment on the onlooker is independent of any minute inquiry into the degree to which it affects the particular criminal. The attachment of equal penalties to offences that are alike in respect of the importance of the rights which they violate, and in respect of the ordinary temptations to them, will, on the whole, lead to the association of an equal amount of terror with the prospect of committing like offences in the public mind. When the circumstances, indeed, of two criminals guilty of offences alike in both the above respects are very greatly and obviously different, so different as to make the operation of the same penalty upon them very conspicuously different, then the penalty may be varied without interfering with its terrifying effect on the public mind. We will suppose e.g. that a fraud on the part of a respectable banker is equivalent, both in respect of the rights which it violates and of the terror needed to prevent the recurrence of like offences, to a burglary. It will not follow because the burglary is punished by imprisonment with hard labour that hard labour should be inflicted on the fraudulent banker likewise. The infliction of hard labour is in everyone's apprehension so different to the banker from what it is to the burglar, that its infliction is not needed in order to equalise the terror which the popular imagination associates with the punishment in the two cases.

194. On the same principle may be justified the consideration of extenuating circumstances in the infliction of punishment. In fact, whether under that name or another, they are taken account of in the administration of criminal law among all civilised nations. 'Extenuating circumstances' is not a phrase in use among our lawyers, but in fact the consideration of them does constantly, with the approval of the judge, convert what would otherwise have been conviction for murder into conviction for manslaughter, and when there has been conviction for murder, leads to the commutation of the sentence. This fact is often taken to show that the degree of moral depravity on the part of the criminal, the question of his character and motive, is and must be considered in determining the punishment due to him. In truth, however, 'extenuating circumstances' may very well make a difference in the kind of terror which needs to be associated with a crime in order to the future protection of rights, and under certain conditions the consideration of them may be sufficiently justified on this ground. Suppose a theft by a starving man, or a hare shot by an angry farmer whose corn it is devouring. These are crimes, but crimes under such extenuating circumstances that there is no need to associate very serious terror with them in order to the protection of the essential rights of property. In the latter case the right which the farmer violates is one which perhaps might be disallowed altogether without interference with any right which society is interested in maintaining. In the former case the right violated is a primary and essential one; one which, where there are many starving people, is in fact pretty sure to be protected by the most stringent penalties. And it might be argued that on the principle stated this is as it should be; that, so far from the hunger of the thief being a reason for lightening his punishment, it is a reason for increasing it, in order that the special temptation to steal when far gone in hunger may, if possible, be neutralised by a special terror associated with the commission of the crime under those conditions. But this would be a one-sided application of the principle. It is not the business of the state to protect one order of rights specially, but all rights equally. It ought not therefore to protect a certain order of rights by associating special terror with the violation of them, when the special temptation to their violation itself implies a violation of right in the persons of those who are so tempted, as is the case when a general danger to property arises from the fact that many people are on the edge of starvation. The attempt to do so is at once ineffectual and diverts attention from the true way of protecting the endangered right, which is to prevent people from falling into a state of starvation. In any tolerably organised society the condition of a man, ordinarily honest and industrious, who is driven to theft by hunger, will be so abnormal that very little terror needs to be associated with the crime as so committed in order to maintain the sanctity of property in the general imagination. Suppose again a man to be killed in a quarrel arising out of his having tampered with the fidelity of his neighbour's wife. In such a case 'extenuating circumstances' may fairly be pleaded against the infliction of the extremest penalty, because the extremest terror does not need to be associated with homicide, as committed under such conditions, in order to the general protection of human life, and because the attempt so to associate it would tend, so far as successful, to weaken the general sense of the wrong—the breach of family right—involved in the act which, in the case supposed, provokes the homicide.

195. 'After all,' it may be said, 'this is a far-fetched way of explaining the admission of extenuating circumstances as modifying the punishment of crime. Why so strenuously avoid the simpler explanation, that extenuating circumstances are taken into account because they are held to modify the moral guilt of the crime? Is not their recognition a practical proof that the punishment of a crime by the state represents the moral disapproval of the community? Does it not show that, however imperfectly the amount of punishment inflicted on a crime may in fact correspond to its moral wickedness, it is generally felt that it ought to do so?'

196. The answer is that there are two reasons for holding that the state neither can nor should attempt to adjust the amount of punishment which it inflicts on a crime to the degree of moral depravity which the crime implies. (1) That the degree of moral depravity implied in any crime is unascertainable. It depends on the motive of the crime, and on this as part of the general character of the agent; on the relation in which the habitual set of his character stands to the character habitually set on the pursuit of goodness. No one can ascertain this in regard to himself. He may know that he is always far from being what he ought to be; that one particular action of his represents on the whole, with much admixture of inferior motives; the better tendency; another, with some admixture of better motives, the worse. But any question in regard to the degree of moral goodness or badness in any action of his own or of his most intimate friend is quite unanswerable. Much less can a judge or jury answer such a question in regard to an unknown criminal. We may be sure indeed that any ordinary crime—nay, perhaps even that of the 'disinterested rebel'—implies the operation of some motive which is morally bad, for though it is not necessarily the worst men who come into conflict with established rights, it probably never can be the best; but the degree of badness implied in such a conflict in any particular case is quite beyond our ken, and it is this degree that must be ascertained if the amount of punishment which the state inflicts is to be proportionate to the moral badness implied in the crime. (2) The notion that the state should, if it could, adjust the amount of punishment which it inflicts on a crime to the moral wickedness of the crime, rests on a false view of the relation of the state to morality. It implies that it is the business of the state to punish wickedness, as such. But it has no such business. It cannot undertake to punish wickedness, as such, without vitiating the disinterestedness of the effort to escape wickedness, and thus checking the growth of a true goodness of the heart in the attempt to promote a goodness which is merely on the surface. This, however, is not to be understood as meaning that the punishment of crime serves no moral purpose. It does serve such a purpose, and has its value in doing so, but only in the sense that the protection of rights, and the association of terror with their violation, is the condition antecedent of any general advance in moral well-being.

197. The punishment of crime, then, neither is, nor can, nor should be adjusted to the degree of moral depravity, properly so called, which is implied in the crime. But it does not therefore follow that it does not represent the disapproval which the community feels for the crime. On the whole, making allowance for the fact that law and judicial custom vary more slowly than popular feeling, it does represent such disapproval. And the disapproval may fitly be called moral, so far as that merely means that it is a disapproval relating to voluntary action. But it is a disapproval founded on a sense of what is necessary for the protection of rights, not on a judgment of good and evil of that kind which we call conscience when it is applied to our own actions, and which is founded on an ideal of moral goodness with which we compare our inward conduct ('inward,' as representing motives and character). It is founded essentially on the outward aspect of a man's conduct, on the view of it as related to the security and freedom in action and acquisition of other members of society. It is true that this distinction between the outward and inward aspects of conduct is not present to the popular mind. It has not been recognised by those who have been the agents in establishing the existing law of crimes in civilised nations. As the state came to control the individual or family in revenging hurts, and to substitute its penalties for private vengeance, rules of punishment came to be enacted expressive of general disapproval, without any clear consciousness of what was the ground of the disapproval. But in fact it was by what have been just described as the outward consequences of conduct that a general disapproval of it was ordinarily excited. Its morality in the stricter or inward sense was not matter of general social consideration. Thus in the main it has been on the ground of its interference with the general security and freedom in action and acquisition, and in proportion to the apprehension excited by it in this respect, that conduct has been punished by the state. Thus the actual practice of criminal law has on the whole corresponded to its true principle. So far as this principle has been departed from, it has not been because the moral badness of conduct, in the true or inward sense, has been taken account of in its treatment as a crime, for this has not been generally contemplated at all, but because 'religious' considerations have interfered. Conduct which did not call for punishment by the state as interfering with any true rights (rights that should be rights) has been punished as 'irreligious.' This, however, did not mean that it was punished on the ground of moral badness, properly so called. It meant that its consequences were feared either as likely to weaken the belief in some divine authority on which the established system of rights was supposed to rest, or as likely to bring evil on the community through provoking the wrath of some unseen power.

198. This account of the considerations which have regulated the punishment of crimes explains the severity with which 'criminal negligence' is in some cases punished, and that severity is justified by the account given of the true principle of criminal law, the principle, viz., that crime should be punished according to the importance of the right which it violates, and to the degree of terror which in a well-organised society needs to be associated with the crime in order to the protection of the right. It cannot be held that the carelessness of an engine-driver who overlooks a signal and causes a fatal accident, implies more moral depravity than is implied in such negligence as all of us are constantly guilty of. Considered with reference to the state of mind of the agent, it is on a level with multitudes of actions and omissions which are not punished at all. Yet the engine-driver would be found guilty of manslaughter and sentenced to penal servitude. The justification is not to be found in distinctions between different kinds of negligence on the part of different agents, but in the effect of the negligence in different cases upon the rights of others. In the case supposed, the most important of all rights, the right to life, on the part of railway passengers depends for its maintenance on the vigilance of the drivers. Any preventible failure in such vigilance requires to have sufficient terror associated with it in the mind of other engine-drivers to prevent the recurrence of a like failure in vigilance. Such punishment is just, however generally virtuous the victim of it is, because it is necessary to the protection of rights of which the protection is necessary to social well-being; and the victim of it, in proportion to his sense of justice, which means his habit of practically recognising true rights, will recognise it as just.

199. On this principle crimes committed in drunkenness must be dealt with. Not only is all depravity of motive specially inapplicable to them, since the motives actuating a drunken man often seem to have little connection with his habitual character; it is not always the case that a crime committed in drunkenness is even intentional. When a man in a drunken rage kills another, he no doubt intends to kill him, or at any rate to do him 'grievous bodily harm,' and perhaps the association of great penal terror with such an offence may tend to restrain men from committing it even when drunk; but when a drunken mother lies on her child and smothers it, the hurt is not intentional but accidental. The drunkenness, however, is not accidental, but preventible by the influence of adequate motives. It is therefore proper to treat such a violation of right, though committed unknowingly, as a crime, and to associate terror with it in the popular imagination, in order to the protection of rights by making people more careful about getting drunk, about allowing or promoting drunkenness, and about looking after drunken people. It is unreasonable, however, to do this and at the same time to associate so little terror, as in practice we do, with the promotion of dangerous drunkenness. The case of a crime committed by a drunkard is plainly distinguishable from that of a crime committed by a lunatic, for the association of penal terror with the latter would tend neither to prevent a lunatic from committing a crime nor people from becoming lunatics.

200. The principle above stated, as that according to which punishment by the state should be inflicted and regulated, also justifies a distinction between crimes and civil injuries, i.e. between breaches of right for which the state inflicts punishment without redress to the person injured, and those for which it procures or seeks to procure redress to the person injured without punishment of the person causing the injury. We are not here concerned with the history of this distinction (for which see Maine, Ancient Law, chap. x, and W.E. Hearn, The Aryan Household, chap. xix), nor with the question whether many breaches of right now among us treated as civil injuries ought not to be treated as crimes, but with the justification that exists for treating certain kinds of breach of right as cases in which the state should interfere to procure redress for the person injured, but not in the way of inflicting punishment on the injurer until he wilfully resists the order to make redress. The principle of the distinction as ordinarily laid down, viz. that civil injuries 'are violations of rights when considered in reference to the injury sustained by the individual,' while crimes are 'violations of rights when considered in reference to their evil tendency as regards the community at large' (Stephen, Book V, chap, i), is misleading; for if the well-being of the community did not suffer in the hurt done to the individual, that hurt would not be a violation of a right in the true sense at all, nor would the community have any ground for insisting that the hurt shall be redressed, and for determining the mode in which it shall be redressed. A violation of right cannot in truth be considered merely in relation to injury sustained by an individual, for, thus considered, it would not be a violation of right. It may be said that the state is only concerned in procuring redress for civil injuries, because, if it left an individual to procure redress in his own way, there would be no public peace. But there are other and easier ways of preventing fighting than by procuring redress of wrong. We prevent our dogs from fighting, not by redressing wrongs which they sustain from each other (of wrongs as of rights they are in the proper sense incapable), but by beating them or tying them up. The community would not keep the peace by procuring redress for hurt or damage sustained by individuals, unless it conceived itself as having interest in the security of individuals from hurt and damage, unless it considered the hurt done to individuals as done to itself. The true justification for treating some breaches of right as cases merely for redress, others as cases for punishment, is that, in order to the general protection of rights, with some it is necessary to associate a certain terror, with others it is not.

201. What then is the general ground of distinction between those with which terror does, and those with which it does not, need to be associated? Clearly it is purposeless to associate terror with breaches of right in the case where the breaker does not know that he is violating a right, and is not responsible for not knowing it. No association of terror with such a breach of right can prevent men from similar breaches under like conditions. In any case, therefore, in which it is, to begin with, open to dispute whether a breach of right has been committed at all, e.g. when it is a question whether a contract has been really broken, owing to some doubt as to the interpretation of the contract or its application to a particular set of circumstances, or whether a commodity of which someone is in possession properly belongs to another,—in such a case, though the judge finally decides that there has been a breach of right, there is no ground for treating it as a crime or punishing it. If, in the course of judicial inquiry, it turns out that there has been fraud by one or other of the parties to the litigation, a criminal prosecution, having punishment, not redress, for its object, should properly supervene upon the civil suit, unless the consequences of the civil suit are incidentally such as to amount to a sufficient punishment of the fraudulent party. Again, it is purposeless to associate terror with a breach of obligation which the person committing it knows to be a breach, but of an obligation which he has no means of fulfilling, e.g. non-payment of an acknowledged debt by a man who, through no fault of his own, is without means of paying it. It is only in cases of one or other of the above kinds,—cases in which the breach of right, supposing it to have been committed, has presumably arisen either from inability to prevent it or from ignorance of the existence of the right,—that it can be held as an absolute rule to be no business of the state to interfere penally but only in the way of restoring, so far as possible, the broken right.