An Introduction to the Principles of Morals and Legislation/Chapter XV

I. It has been shown what the rules are, which ought to be observed in adjusting the proportion between the punishments and the offense. The properties to be given to a lot of punishment, in every instance, will of course be such as it stands in need of, in order to be capable of being applied, in conformity to those rules: the quality will be regulated by the quantity.

II. The first of those rules, we may remember, was, that the quantity of punishment must not be less, in any case, than what is sufficient to outweigh the profit of the offence: since, as often as it is less, the whole lot (unless by accident the deficiency should be supplied from some of the other sanctions) is thrown away: it is inefficacious. The fifth was, that the punishment ought in no case to be more than what is required by the several other rules: since, if it be, all that is above that quantity is needless. The fourth was, that the punishment should be adjusted in such manner to each individual offence, that every part of the mischief of that offence may have a penalty (that is, a tutelary motive) to encounter it: otherwise, with respect to so much of the offense as has not a penalty to correspond to it, it is as if there were no punishment in the case. Now to none of those rules can a lot of punishment be conformable, unless, for every variation in point of quantity, in the mischief of the species of offense to which it is annexed, such lot of punishment admits of a correspondent variation. To prove this, let the profit of the offence admit of a multitude of degrees. Suppose it, then, at any one of these degrees: if the punishment be less than what is suitable to that degree, it will be inefficacious; it will be so much thrown away: if it be more, as far as the difference extends, it will be needless; it will therefore be thrown away also in that case. The first property, therefore, that ought to be given to a lot of punishment, is that of being variable in point of quantity, in conformity to every variation which can take place in either the profit or mischief of the offense. This property might, perhaps, be termed, in a single word, variability.

III. A second property, intimately connected with the former, may be styled equability. It will avail but little, that a mode of punishment (proper in all other respects) has been established by the legislator; and that capable of being screwed up or let down to any degree that can be required; if, after all, whatever degree of it be pitched upon, that same degree shall be liable, according to circumstances, to produce a very heavy degree of pain, or a very slight one, or even none at all. In this case, as in the former, if circumstances happen one way, there will be a great deal of pain produced which will be needless: if the other way, there will be no pain at all applied, or none that will be efficacious. A punishment, when liable to this irregularity, may be styled an unequable one: when free from it, an equable one. The quantity of pain produced by the punishment will, it is true, depend in a considerable degree upon circumstances distinct from the nature of the punishment itself: upon the condition which the offender is in, with respect to the circumstances by which a man's sensibility is liable to be influenced. But the influence of these very circumstances will in many cases be reciprocally influenced by the nature of the punishment: in other words, the pain which is produced by any mode of punishment, will be the joint effect of the punishment which is applied to him, and the circumstances in which he is exposed to it. Now there are some punishments, of which the effect may be liable to undergo a greater alteration by the influence of such foreign circumstances, than the effect of other punishments is liable to undergo. So far, then, as this is the case, equability or unequability may be regarded as properties belonging to the punishment itself.

IV. An example of a mode of punishment which is apt to be unequable, is that of banishment, when the locus a quo (or place the party is banished from) is some determinate place appointed by the law, which perhaps the offender cares not whether he ever see or no. This is also the case with pecuniary, or quasi-pecuniary punishment, when it respects some particular species of property, which the offender may have been possessed of, or not, as it may happen. All these punishments may be split down into parcels, and measured out with the utmost nicety: being divisible by time, at least, if by nothing else. They are not, therefore, any of them defective in point of variability: and yet, in many cases, this defect in point of equability may make them as unfit for use as if they were.

V. The third rule of proportion was, that where two offenses come in competition, the punishment for the greater offenses must be sufficient to induce a man to prefer the less. Now, to be sufficient for this purpose, it must be evidently and uniformly greater: greater, not in the eyes of some men only, but of all men who are liable to be in a situation to take their choice between the two offenses; that is, in effect, of all mankind. In other words, the two punishments must be perfectly commensurable. Hence arises a third property, which may be termed commensurability: to wit, with reference to other punishments.

VI. But punishments of different kinds are in very few instances uniformly greater one than another; especially when the lowest degrees of that which is ordinarily the greater, are compared with the highest degrees of that which is ordinarily the less: in other words, punishments of different kinds are in few instances uniformly commensurable. The only certain and universal means of making two lots of punishment perfectly commensurable, is by making the lesser an ingredient in the composition of the greater. This may be done in either of two ways.
 * 1) By adding to the lesser punishment another quantity of punishment of the same kind.
 * 2) By adding to it another quantity of a different kind. The latter mode is not less certain than the former: for though one cannot always be absolutely sure, that to the same person a given punishment will appear greater than another given punishment; yet one may be always absolutely sure, that any given punishment, so as it does but come into contemplation, will appear greater than none at all.

VII. Again: Punishment cannot act any farther than in as far as the idea of it, and of its connection with the offense, is present in the mind. The idea of it, if not present, cannot act at all; and then the punishment itself must be inefficacious. Now, to be present, it must be remembered, and to be remembered it must have been learnt. But of all punishments that can be imagined, there are none of which the connection with the offense is either so easily learnt, or so efficaciously remembered, as those of which the idea is already in part associated with some part of the idea of the offense: which is the case when the one and the other have some circumstance that belongs to them in common. When this is the case with a punishment and an offense, the punishment is said to bear an analogy to, or to be characteristic of, the offence. Characteristicalness is, therefore, a fourth property, which on this account ought to be given, whenever it can conveniently be given, to a lot of punishment. VIII. It is obvious, that the effect of this contrivance will be the greater, as the analogy is the closer. The analogy will be the closer, the more material that circumstance is, which is in common. Now the most material circumstance that can belong to an offense and a punishment in common, is the hurt or damage which they produce. The closest analogy, therefore, that can subsist between an offense and the punishment annexed to it, is that which subsists between them when the hurt or damage they produce is of the same nature: in other words, that which is constituted by the circumstance of identity in point of damage. Accordingly, the mode of punishment, which of all others bears the closest analogy to the offense, is that which in the proper and exact sense of the word is termed retaliation. Retaliation, therefore, in the few cases in which it is practicable, and not too expensive, will have one great advantage over every other mode of punishment.

IX. Again: It is the idea only of the punishment (or, in other words, the apparent punishment) that really acts upon the mind; the punishment itself (the real punishment) acts not any farther than as giving rise to that idea. It is the apparent punishment, therefore, that does all the service, I mean in the way of example, which is the principal object. It is the real punishment that does all the mischief. Now the ordinary and obvious way of increasing the magnitude of the apparent punishment, is by increasing the magnitude of the real. The apparent magnitude, however, may to a certain degree be increased by other less expensive means: whenever, therefore, at the same time that these less expensive means would have answered that purpose, an additional real punishment is employed, this additional real punishment is needless. As to these less expensive means, they consist,
 * 1) In the choice of a particular mode of punishment, a punishment of a particular quality, independent of the quantity.
 * 2) In a particular set of solemnities distinct from the punishment itself, and accompanying the execution of it.

X. A mode of punishment, according as the appearance of it bears a greater proportion to the reality, may be said to be the more exemplary. Now as to what concerns the choice of the punishment itself, there is not any means by which a given quantity of punishment can be rendered more exemplary, than by choosing it of such a sort as shall bear an analogy to the offense. Hence another reason for rendering the punishment analogous to, or in other words characteristic of, the offense.

XI. Punishment, it is still to be remembered, is in itself an expense: it is in itself an evil. Accordingly the fifth rule of proportion is, not to produce more of it than what is demanded by the other rules. But this is the case as often as any particle of pain is produced, which contributes nothing to the effect proposed. Now if any mode of punishment is more apt than another to produce any such superfluous and needless pain, it may be styled unfrugal; if less, it may be styled frugal. Frugality, therefore, is a sixth property to be wished for in a mode of punishment.

XII. The perfection of frugality, in a mode of punishment, is where not only no superfluous pain is produced on the part of the person punished, but even that same operation, by which he is subjected to pain, is made to answer the purpose of producing pleasure on the part of some other person. Understand a profit or stock of pleasure of the self-regarding kind: for a pleasure of the dissocial kind is produced almost of course, on the part of all persons in whose breasts the offence has excited the sentiment of ill-will. Now this is the case with pecuniary punishment, as also with such punishments of the quasi-pecuniary kind as consist in the subtraction of such a species of possession as is transferable from one party to another. The pleasure, indeed, produced by such an operation, is not in general equal to the pain: it may, however, be so in particular circumstances, as where he, from whom the thing is taken, is very rich, and he, to whom it is given, very poor: and, be it what it will, it is always so much more than can be produced by any other mode of punishment.

XIII. The properties of exemplarity and frugality seem to pursue the same immediate end, though by different courses. Both are occupied in diminishing the ratio of the real suffering to the apparent: but exemplarity tends to increase the apparent; frugality to reduce the real.

XIV. Thus much concerning the properties to be given to punishments in general, to whatsoever offenses they are to be applied. Those which follow are of less importance, either as referring only to certain offenses in particular, or depending upon the influence of transitory and local circumstances.

In the first place, the four distinct ends into which the main and general end of punishment is divisible, may give rise to so many distinct properties, according as any particular mode of punishment appear to be more particularly adapted to the compassing of one or of another of those ends. To that of example, as being the principal one, a particular property has already been adapted. There remains the three inferior ones of reformation, disablement, and compensation.

XV. A seventh property, therefore, to be wished for in a mode of punishment, is that of subserviency to reformation, or reforming tendency. Now any punishment is subservient to reformation in proportion to its quantity: since the greater the punishment a man has experienced, the stronger is the tendency it has to create in him an aversion towards the offense which was the cause of it: and that with respect to all offenses alike. But there are certain punishments which, with regard to certain offenses, have a particular tendency to produce that effect by reason of their quality: and where this is the case, the punishments in question, as applied to the offenses in question, will pro tanto have the advantage over all others. This influence will depend upon the nature of the motive which is the cause of the offence: the punishment most subservient to reformation will be the sort of punishment that is best calculated to invalidate the force of that motive.

XVI. Thus, in offenses originating from the motive of ill-will, that punishment has the strongest reforming tendency, which is best calculatedto weaken the force of the irascible affections. And more particularly, in that sort of offense which consists in an obstinate refusal, on the part of the offender, to do something which is lawfully required of him, and in which the obstinacy is in great measure kept up by his resentment against those who have an interest in forcing him to compliance, the most efficacious punishment seems to be that of confinement to spare diet.

XVII. Thus, also, in offenses which owe their birth to the joint influence of indolence and pecuniary interest, that punishment seems to possess the strongest reforming tendency, which is best calculated to weaken the force of the former of those dispositions. And more particularly, in the cases of theft, embezzlement, and every species of defraudment, the mode of punishment best adapted to this purpose seems, in most cases, to be that of penal labour.

XVIII. An eighth property to be given to a lot of punishment in certain cases, is that of efficacy with respect to disablement, or, as it might be styled more briefly, disabling efficacy. This is a property which may be given in perfection to a lot of punishment; and that with much greater certainty than the property of subserviency to reformation. The inconvenience is, that this property is apt, in general, to run counter to that of frugality: there being, in most cases, no certain way of disabling a man from doing mischief, without, at the same time, disabling him, in a great measure, from doing good, either to himself or others. The mischief therefore of the offense must be so great as to demand a very considerable lot of punishment, for the purpose of example, before it can warrant the application of a punishment equal to that which is necessary for the purpose of disablement.

XIX. The punishment, of which the efficacy in this way is the greatest, is evidently that of death. In this case the efficacy of it is certain. This accordingly is the punishment peculiarly adapted to those cases in which the name of the offender, so long as he lives, may be sufficient to keep a whole nation in a flame. This will now and then be the case with competitors for the sovereignty, and leaders of the factions in civil wars: though, when applied to offenses of so questionable a nature, in which the question concerning criminality turns more upon success than any thing else; an infliction of this sort may seem more to savour of hostility than punishment. At the same time this punishment, it is evident, is in an eminent degree unfrugal; which forms one among the many objections there are against the use of it, in any but very extraordinary cases.

XX. In ordinary cases the purpose maybe sufficiently answered by one or other of the various kinds of confinement and banishment: of which, imprisonment is the most strict and efficacious. For when an offense is so circumstanced that it cannot be committed but in a certain place, as is the case, for the most part, with offenses against the person, all the law has to do, in order to disable the offender from committing it, is to prevent his being in that place. In any of the offenses which consist in the breach or the abuse of any kind of trust, the purpose may be compassed at a still cheaper rate, merely by forfeiture of the trust: and in general, in any of those offenses which can only be committed under favour of some relation in which the offender stands with reference to any person, or sets of persons, merely by forfeiture of that relation: that is, of the right of continuing to reap the advantages belonging to it. This is the case, for instance, with any of those offences which consist in an abuse of the privileges of marriage, or of the liberty of carrying on any lucrative or other occupation.

XXI. The ninth property is that of subserviency to compensation. This property of punishment, if it be vindictive compensation that is in view, will, with little variation, be in proportion to the quantity: if lucrative, it is the peculiar and characteristic property of pecuniary punishment.

XXII. In the rear of all these properties may be introduced that of popularity; a very fleeting and indeterminate kind of property, which may belong to a lot of punishment one moment, and be lost by it the next. By popularity is meant the property of being acceptable, or rather not unacceptable, to the bulk of the people, among whom it is proposed to be established. In strictness of speech, it should rather be called absence of unpopularity: for it cannot be expected, in regard to such a matter as punishment, that any species or lot of it should be positively acceptable and grateful to the people: it is sufficient, for the most part, if they have no decided aversion to the thoughts of it. Now the property of characteristicalness, above noticed, seems to go as far towards conciliating the approbation of the people to a mode of punishment, as any; insomuch that popularity may be regarded as a kind of secondary quality, depending upon that of characteristicalness. The use of inserting this property in the catalogue, is chiefly to make it serve by way of memento to the legislator not to introduce, without a cogent necessity, any mode or lot of punishment, towards which he happens to perceive any violent aversion entertained by the body of the people.

XXIII. The effects of unpopularity in a mode of punishment are analogous to those of unfrugality. The unnecessary pain which denominates a punishment unfrugal, is most apt to be that which is produced on the part of the offender. A portion of superfluous pain is in like manner produced when the punishment is unpopular: but in this case it is produced on the part of persons altogether innocent, the people at large. This is already one mischief; and another is, the weakness which it is apt to introduce into the law. When the people are satisfied with the law, they voluntarily lend their assistance in the execution: when they are dissatisfied, they will naturally withhold that assistance; it is well if they do not take a positive part in raising impediments. This contributes greatly to the uncertainty of the punishment; by which, in the first instance, the frequency of the offense receives an increase. In process of time that deficiency, as usual, is apt to draw on an increase in magnitude: an addition of a certain quantity which otherwise would be needless.

XXIV. This property, it is to be observed, necessarily supposes, on the part of the people, some prejudice or other, which it is the business of the legislator to endeavour to correct. For if the aversion to the punishment in question were grounded on, the principle of utility, the punishment would be such as, on other accounts, ought not to be employed: in which case its popularity or unpopularity would never be worth drawing into question. It is properly therefore a property not so much of the punishment as of the people: a disposition to entertain an unreasonable dislike against an object which merits their approbation. It is the sign also of another property, to wit. indolence or weakness, on the part of the legislator: in suffering the people for the want of some instruction, which ought to be and might be given them, to quarrel with their own interest. Be this as it may, so long as any such dissatisfaction subsists, it behoves the legislator to have an eye to it, as much as if it were ever so well grounded. Every nation is liable to have its prejudices and its caprices which it is the business of the legislator to look out for, to study, and to cure.

XXV. The eleventh and last of all the properties that seem to be requisite in a lot of punishment, is that of remissibility. The general presumption is, that when punishment is applied, punishment is needful: that it ought to be applied, and therefore cannot want to be remitted. But in very particular, and those always very deplorable cases, it may by accident happen otherwise. It may happen that punishment shall have been inflicted, where, according to the intention of the law itself, it ought not to have been inflicted: that is, where the sufferer is innocent of the offense. At the time of the sentence passed he appeared guilty: but since then, accident has brought his innocence to light. This being the case, so much of the destined punishment as he has suffered already, there is no help for. The business is then to free him from as much as is yet to come. But is there any yet to come? There is very little chance of there being any, unless it be so much as consists of chronical punishment: such as imprisonment, banishment, penal labour, and the like. So much as consists of acute punishment, to wit where the penal process itself is over presently, however permanent the punishment may be in its effects, may be considered as irremissible. This is the case, for example, with whipping, branding, mutilation, and capital punishment. The most perfectly irremissible of any is capital punishment. For though other punishments cannot, when they are over, be remitted, they may be compensated for; and although the unfortunate victim cannot be put into the same condition, yet possibly means may be found of putting him into as good a condition, as he would have been in if he had never suffered. This may in general be done very effectually where the punishment has been no other than pecuniary.

There is another case in which the property of remissibility may appear to be of use: this is, where, although the offender has been justly punished, yet on account of some good behaviour of his, displayed at a time subsequent to that of the commencement of the punishment, it may seem expedient to remit a part of it. But this it can scarcely be, if the proportion of the punishment is, in other respects, what it ought to be. The purpose of example is the more important object, in comparison of that of reformation. It is not very likely, that less punishment should be required for the former purpose than for the latter. For it must be rather an extraordinary case, if a punishment, which is sufficient to deter a man who has only thought of it for a few moments, should not be sufficient to deter a man who has been feeling it all the time. Whatever, then, is required for the purpose of example, must abide at all events: it is not any reformation on the part of the offender, that can warrant the remitting of any part of it: if it could, a man would have nothing to do but to reform immediately, and so free himself from the greatest part of that punishment which was deemed necessary. In order, then, to warrant the remitting of any part of a punishment upon this ground, it must first be supposed that the punishment at first appointed was more than was necessary for the purpose of example, and consequently that a part of it was needless upon the whole. This, indeed, is apt enough to be the case, under the imperfect systems that are as yet on foot: and therefore, during the continuance of those systems, the property of remissibility may, on this second ground likewise, as well as on the former, be deemed a useful one. But this would not be the case in any new-constructed system, in which the rules of proportion above laid down should be observed. In such a system, therefore, the utility of this property would rest solely on the former ground.

XXVI. Upon taking a survey of the various possible modes of punishment, it will appear evidently, that there is not any one of them that possesses all the above properties in perfection. To do the best that can be done in the way of punishment, it will therefore be necessary, upon most occasions, to compound them, and make them into complex lots, each consisting of a number of different modes of punishment put together: the nature and proportions of the constituent parts of each lot being different, according to the nature of the offence which it is designed to combat.

XXVII. It may not be amiss to bring together, and exhibit in one view, the eleven properties above established. They are as follows: Two of them are concerned in establishing a proper proportion between a single offense and its punishment; viz.,
 * 1. Variability.
 * 2. Equability.

One, in establishing a proportion, between more offences than one, and more punishments than one; viz.,
 * 3. Commensurability.

A fourth contributes to place the punishment in that situation in which alone it can be efficacious; and at the same time to be bestowing on it the two farther properties of exemplarity and popularity; viz.,
 * 4. Characteristicalness.

Two others are concerned in excluding all useless punishment; the one indirectly, by heightening the efficacy of what is useful; the other in a direct way; viz.,
 * 5. Exemplarity.
 * 6. Frugality.

Three others contribute severally to the three inferior ends of punishment; viz.,
 * 7. Subserviency to reformation.
 * 8. Efficacy in disabling.
 * 9. Subserviency to compensation.

Another property tends to exclude a collateral mischief, which a particular mode of punishment is liable accidentally to produce; viz.,
 * 10. Popularity.

The remaining property tends to palliate a mischief, which all punishment, as such is liable accidentally to produce; viz.,
 * 11. Remissibility.

The properties of commensurability, characteristicalness, exemplarity, subserviency to reformation, and efficacy in disabling, are more particularly calculated to augment the profit which is to be made by punishment: frugality, subserviency to compensation, popularity, and remissibility, to diminish the expense: variability and equability are alike subservient to both those purposes.

XXVIII. We now come to take a general survey of the system of offences: that is, of such acts to which, on account of the mischievous consequences they have a natural tendency to produce, and in the view of putting a stop to those consequences, it may be proper to annex a certain artificial consequence, consisting of punishment, to be inflicted on the authors of such acts according to the principles just established.