gular infliction being counteracted by the unwillingness of injured parties to prosecute, of witnesses to give evidence, and of tribunals to convict, as well as by the disposition of the government to comply with petitions for a mitigation of the sentence), the fundamental condition of the theory is not satisfied, and the dictum is inapplicable to the case. This, therefore, is not an instance of erroneous prediction-it is an erroneous application of a general theorem to a case not properly falling under it. The proposition, or prediction, is as true with reference to this case as with reference to a case in which the punishment is regularly inflicted, and the crime is in consequence repressed. In the one case, however, the data assumed by the theorem are present-in the other they are not. All such predictions in speculative politics are similar to the second law of motion, which declares that a body in motion will move for ever in a straight line, with a velocity proportioned to the force and mass. This prediction with respect to moving bodies is founded on the supposition, that the retarding influence of friction and all other physical impediment is excluded; and it would doubtless be verified, if the case supposed were to exist. The reality, however, never coincides with the hypothesis, and yet the general prediction, being not absolute but conditional, is true; and it approaches to fulfilment in any individual case, in proportion as the real approach to the hypothetical circum stances. Writers on speculative politics do not in general set out, in terms, their hypothetical data, though these ought to be distinctly implied in their reasonings, if their scientific basis is properly laid. Sometimes, indeed, they have not themselves formed a clear conception of their own hypothesis, and of the elements of their own problem. For these reasons, it often happens that the conditional nature of the proposition is overlooked, and that it is understood to be absolute and categorical, when it is, in fact, qualified by a reservation which belongs to its very essence. Many true principles of speculative politics, which predict the operation of a given cause, have been taxed with unsoundness, because, in a given case, the prediction was not fulfilled. But before the soundness of any such principle can be successfully impugned, it should be shown that the cause did not exert itself in the direction indicated, and that its tendency was not repressed by counteracting influences. Unless this point is established by proper evidence, nothing is proved against the soundness of the principle at issue. It would be as if a mechanical philosopher were to argue against the second law of motion, by saying that he had projected a body in a direction perpendicular to the earth's surface, and instead of moving continually upwards in a straight line, it had, after a time, fallen to the ground. § 16 The predictions of practical politics differ from the predictions of speculative politics in being particular, and not general. The speculative politician predicts the general tendency of all laws or institutions of a certain class the practical politician predicts the particular tendency of this or that law or institution, or measure of a government. The former, for instance, lays it down that all treaties containing certain stipulations produce certain effects—the latter predicts that a particular treaty, actually made and concluded between two nations, will produce certain effects. These two processes are quite distinct in their character, and it is only the latter which announces absolutely beforehand the occurrence of a real event, and therefore deserves the name of a prediction. Hence it may happen, that persons who are well qualified to conduct the one are not well qualified to conduct the other. A political speculator may be unfit for practical statesmanship, and a practical statesman may be unfit for political speculation. We have, in a former part of this treatise, examined the process by which the practical politician determines the probable future effects of a proposed law or measure of a government, (*) and we have also seen how he applies a general theorem, or maxim, to an actual case.(24) We have likewise found that (23) Above, ch. xiii. VOL. II. (24) Ch. xx. A A political conduct consists in a choice of alternative courses, which present themselves to the free agent. (2) Now, whenever a politician takes any practical step he makes, either avowedly or tacitly, a prediction-he forms an opinion with respect to some portion of futurity. With respect to matters simple, proximate, and dependent on the will of a few persons whom he can guide, or whose conduct can be reasonably anticipated, his prediction may be absolute: he may say, or think, that a certain event will positively happen. But with respect to matters intricate, remote, and dependent on persons many in number, and beyond his control, his prediction, like those of the speculative politician, will be qualified, though, unlike those predictions, it will be special. He does not, indeed, consider the probable operation of a principle acting in vacuo-he takes a measure clothed in circumstances, and anticipates its effects in the midst of those circumstances. But in proportion as those circumstances multiply, the problem becomes more complex. Elements of doubt manifest themselves in various directions. Discordant influences threaten from opposite quarters of the horizon, and disturb the vision of the political seer. Hence his prediction, though confined to an actual case, and founded on real data, will be conditional. He will, if his judgment be discreet, content himself with saying that, provided certain disturbing circumstances, which may perhaps be partly foreseen, should not occur, a certain event will happen. Such a prediction, if made with the requisite theoretical guides, after a careful investigation of the facts of the case, and with the other conditions requisite for skilful practice, is the nearest approach to rational prediction which the conditions of political practice admit. It approaches far closer to a probable anticipation of futurity than any abstract theory, founded on unreal hypotheses, which, as we have already explained, does not properly predict any event. It may, indeed, be less happy than a fortuitous conjecture, or divination, not founded on a careful or comprehensive view of facts-which attracted little notice when (25) Ch. xxiii. made, but when subsequently verified was rescued from oblivion. But such random shots into futurity, though they may sometimes be more lucky than those of the most skilful marksmen, may be compared with the one lucky hit of the weather-prophet, which is remembered when his thousand misses are forgotten.(26) § 17 In the application of physical knowledge to practice, the difficulty of prediction arises, in general, from the imperfection of the theory. For the imperfection of physical theories there are two main causes-viz., 1, the unapproachableness of the field of observation; 2, the complexity of the data. Of the first of these, meteorological science affords an instance. The agencies by which weather is determined are so remote and inaccessible to our senses, that the laws of meteorology cannot be determined with sufficient precision to furnish rules for calculating the temperature at a given spot for a future day. For the uncertainty of predictions respecting constructive works, the complexity of the data is the principal cause. The uncertainty of medical predictions arises from both causes combined: the mechanism of the human body is in the highest degree complex, and its internal vital operations are, to a great extent, withdrawn from the observation of our senses. The prediction of Darwin respecting the application of steam to navigation, and to locomotion by land, may be considered as a happy anticipation of the future, founded on philosophical data. His prediction respecting the application of steam to motion through the atmosphere is probably not destined to receive a similar fulfilment.(") (26) See Nov. Org. i. 46, on the fallacy of attending to the hits and overlooking the misses. (27) 'Soon shall thy arm, unconquered steam, afar Darwin's Economy of Vegetation, canto i. There is reason to believe it may in time be applied to the rowing of barges, and the moving of carriages along the road. As the specific levity of air is too great for the support of great burthens by balloons, there seems no probable method of flying conveniently but by the power of steam, or some other explosive material, which another half-century may probably discover.'-Note on v. 254. In practical politics, the difficulty of prediction arises not at all from the inaccessibility of the field of observation-the whole of which, as we have already seen, is within the compass of our senses and experience-but it proceeds partly from the imperfection of theory, and partly from the complexity of the data which it is necessary, in every practical problem, to take into account. In addition to these, however, there is a cause of uncertainty more powerful than either, viz. the peculiar nature of the human will, which is so large an element in political affairs. In judging of the probable operation of a political measure, the practical politician is, or ought to be, materially assisted by theory. For instance, let the measure of which the effects are to be predicted be one affecting the currency or trade of the country, or the numbers of its population. Upon each of these subjects, the practical man would be naturally provided with the principles which have been more or less established by scientific reasoning. Wherever the political theory is imperfect, either from incompleteness, or from admixture of error, all prediction in which such theory bears a part is likely to be defective. The complexity of the data is, again, a source of uncertainty in the predictions of practical politics. Even where all the elements of a judgment are ascertained, and lie before us, it is not easy to calculate what will be their combined result when they are set in motion, and begin to act upon one another. In this respect, a new law is analogous to a new machine. All the component parts of a new machine are given, and it is known to what purposes it will be applied, yet a practical experiment is necessary in order to reduce to certainty the preliminary conjectures formed by the mechanist with respect to its probable working. In like manner, however carefully a law may be framed for accomplishing its object, yet its authors can never be confident of its success until it has been tried in practice, so complex are the data which a judgment upon almost every law involves. There is, however, an element of uncertainty in politics which is not shared by physics, which is beyond the imperfection of |