ness; but in considering the probability of this effect being produced in any individual case, we must estimate the probability of such events as deficient harvests, difficulty of freight, insecurity arising from war or civil commotions, and the like. The reduction of a high customs duty on any article would naturally tend to increase its importation; yet a change in the public taste, or the discovery of a cheaper or preferable substitute, might prove an effectual counteraction to this tendency.

Before, therefore, we form a practical maxim out of a general theorem, or use a maxim formed from a theorem without any limitation, we must consider whether the cause is likely to operate freely, or to be impeded by counteracting influences. If the latter is likely to be the case, we must attempt to calculate their force, direction, number, duration, and combined action; and having made this calculation, we must correct our general theorem accordingly. It is in this process that the value of theory and the skill of the practician are equally perceptible. Without theory, the practical politician would be bewildered in a crowd of unconnected and unarranged particulars. He would be like a solitary wanderer in a trackless forest, without any knowledge of the ground. But with theory alone as his guide, and without any inquiry into the facts of the case, he would almost infallibly be led into error. He would apply to a concrete case a practical maxim founded upon a theorem, which supposes only an arbitrarily limited abstraction. He would therefore commit the same error as that of a military engineer who should calculate the actual course of a cannon ball by equations in which the friction of the atmosphere is neglected.

§ 5 Whenever it is affirmed that anything is 'true in theory and false in practice,' or that it is 'good in theory and bad in practice,'() the meaning intended to be conveyed is, apparently, that which has been just described. This objection is taken when it is proposed to apply a general theorem to some

(9) Compare the Essay of Kant: Ueber den gemeinspruch: Das mag in der theorie richtig sein, taugt aber nicht für die praxis.—Werke, vol. vii. part i. p. 177.

actual case, and the objector admits that the theorem is, scientifically or abstractedly, true; in other words, that if the phenomena supposed in the theory all co-exist, and co-exist alone, without anything to impede their operation in any actual case, the result predicated will also exist. But he affirms that in the actual case under consideration, either some of the phenomena supposed in the theory are absent; or (what is more common), that certain other phenomena not supposed in the theory, are present, and that the theoretical proposition therefore requires correction before it can be safely applied to the case in question. If we assume A, and A alone, to exist, it may be true that B will follow; but if, in the actual case, we have A+ a + á; then it may not be true that B will follow. In this manner, the

proposition which is true in theory may be false (that is to say, inapplicable without correction) in an actual case of practice.

The question therefore which the objector raises, is not as to the truth of the theorem, so long as it remains a mere scientific abstraction, but as to the construction from it of a practical maxim, to be applied to the case under discussion.

Now it may

indeed happen that the objection is unfounded, and that the practical maxim may be properly formed from the theory, and applied directly to the individual case. The circumstances alleged by him may be irrelevant; they may not be sufficient to differentiate the case, and to prevent the general rule from including it. The exemption may be really urged by the arguer, not because the case is abnormal, but because it is his own, and because his personal interest is at stake. On the other hand, it may happen that the general theorem is not applicable to the case in question without corrections which nullify its effect. may be apparent, upon an examination of the facts, that the natural tendency of the cause will be counteracted by so many obstacles as to prevent it from reaching its appointed goal; or that other circumstances exist which withdraw the case from the general rule, and place it among the exceptions. Thus the following are practical maxims for the management of an army in war,

It

taken from an ancient treatise, embodying the results of Roman experience. (10)

1. If you discover that your plan of operations has been betrayed to the enemy, you ought to change your arrangements. 2. Soldiers in quarters are best kept in discipline by punishment and fear; when they are in the field, hope and reward produce the best effect.

'3. It is better to press your enemy with hunger than with the sword.'

These are practical maxims founded upon general theorems which have been derived from an intelligent induction of particulars, and which represent average experience. But they are not absolute and universal; they are presumptions, not inflexible precepts; they cannot be applied blindly without a scrutiny and a consideration of the facts of the given case. Thus, it may be generally true that when an intended plan of attack has been betrayed to the enemy, a different point of attack may be selected with advantage. But this maxim is not invariable: the plan originally chosen may offer such prospects of success, and the alternative plans may be so inferior to it, that it may, on the whole, be preferable even with the drawback of being known beforehand to the enemy, who is thus able to meet it with counter-preparations. In an instance of this sort it may be said (with some laxity of language) that the rule is theoretically true but practically false, or, more correctly, that the general presumption fails, and the case is exceptional.

Again, it may be laid down generally that certainty of punishment is the most effectual mode of repressing crime.(") From this general proposition a practical maxim arises which may be safely applied to the majority of actual cases. Nevertheless, conjunctures occur in political affairs when clemency produces a more tranquillizing and healing effect, than an inflexible and

(10) Vegetius, De Re Mil. iii. 26.

(11) See Paley's Moral and Political Philosophy, b. vi. c. 9, vol. 4, p. 445.

mechanical infliction of the legal punishment; when pardons and commutations of sentences are more likely to repress crime than a strict execution of the law. It may likewise be laid down generally that punishment operates chiefly by way of example to the community at large, and that the removal of the convict himself is a matter of secondary moment. Nevertheless, cases may occur where the ostracizing principle is more important than the deterring example of the offender's punishment, and where it is the first object to get rid of the criminal, and eliminate him from the state.(1)

$ 6 The circumscription of general maxims, with reference to actual cases of practice, is thus effected by adding the circumstances of the given case, and considering the combined result. A general theorem is founded on a limited set of hypothetical data, and the more limited they are, the more abstract is the ? theorem. The intensity varies inversely with the extent of its signification. Now a theoretical proposition, when converted into a rule of conduct, may be conceived as taken in connexion with an indefinite number of sets of concomitant circumstances, which may modify its operation. If, therefore, we add a definite number of circumstances to the proposition, we exclude all uncertainty as to the possible combinations, and we in fact perform a sort of practical abscissio infiniti. We substitute a real and

definite for an ideal and indefinite compound. The addition of a limited number of terms operates as the exclusion of an unlimited number.

Thus, let it be supposed that our general theorem is as to the operation of legal punishment. Legal punishment, if left to itself, may be expected to produce abstinence from crime; but it may be accompanied, and as it were, held in solution by a vast variety of collateral circumstances which may influence its operation. Thus, it may be combined with an inefficient or unskilful

(12) The circumstances which rendered this mode of punishment expedient in a Greek state are enumerated in Mr. Grote's History of Greece, vol. iv. p. 200-17. Plato, (Leg. i. 8, p. 636) remarks, that institutions may suit some states, but not others; that no universal rule can be laid down, pyo kai λóy, both in practice and theory. Compare vol. i. p. 22, n. 29.

police, a venal, weak, or tardy administration of justice, difficulty of detection, unwillingness to prosecute or to give evidence, or a fanatical contempt of suffering. Various other circumstances might likewise be mentioned which diminish the deterring force of the fear of legal punishments on the minds of given individuals. (13) Now, all that can be said with reference to such a general theorem, so long as it remains an abstraction, is that it describes a prevailing tendency, liable to be resisted and modified by an unlimited number of counter-influences with which legal punishment may be combined. But when an actual case is laid before us, we can perceive whether any, and which of those other circumstances are present. Of such as are wanting we take no account, we note those which are discernible, and we then form a definite practical problem, in this shape: How will the denunciation of legal punishment operate, taken in connexion with a reluctance of witnesses to give evidence, or with a willingness of judges to take bribes (as the case may be)? What will be the effect of legal punishment, combined with a hope of impunity, or a disregard of pain, of some special ascertained nature?'

Again, let it be assumed that the general maxim to be applied to a particular case is, that when the population of a country becomes redundant, it ought to be relieved by the foundation of a new colony. In order, however, to determine the fitness of applying this rule to the given case, we must add the material circumstances of the actual problem: such as the distance of the proposed settlement, its situation with respect to other countries, the expense of defending it, the political relations in respect of dependence which would subsist with the mother-country, the military and naval resources of the mother-country, and the like. When all these circumstances have been enumerated and arranged, a practical problem arises in this form: Is it expedient for a country having a redundant population to found a new colony in such a position, with such political relations, with such a

(13) See in Aristot., Rhet. i. 12, an enumeration of many of the motives which induce a man to commit crime, and to brave the danger of punish

ment.