327

CHAPTER XXIV.

UPON PREDICTION IN POLITICS.

§ 1 IN political practice, every step which is taken concerns the future as well as the past. The decision of a court of justice upon a criminal prosecution, or upon a controversy of civil right, though founded upon antecedent facts, has a necessary bearing upon the future. In the one case, it either directs the liberation of the prisoner, or sentences him to a punishment; in the other case, it denies or grants to the plaintiff the redress which he seeks to obtain from the defendant. In either of these contingencies the decision of the court is followed by certain effects, and therefore deals with a future period of time. Administrative proceedings, in like manner, though growing out of foregone facts, determine men's conduct for the future. The steps which are taken by a revenue department, for example, for the collection of a tax, govern the subsequent conduct of the tax-payers. And, lastly, legislative measures are, for the most part, wholly prospective in their operation. They are framed with reference to a pre-existing state of things, and to our experience of the past, but their effect is necessarily future, even when it includes prior rights and contracts. (')

In every case, therefore, where a political act is done, we are concerned with futurity, and we set in motion an influence which will operate upon men's conduct after the period of the step which we take. Whether we foresee the consequences of our act or not, it will undoubtedly be attended with consequences -it will modify the conduct of others, and determine their wills

(1) According to Aristotle (Rhet. i. 3), the ecclesiast, or member of a legislative assembly, judges concerning the future, whereas the dicast, or member of a judicial assembly, judges concerning the past. This distinction, however, cannot be supported. Both come to a practical decision, founded on the past, but having reference to the future.

in some direction different from that in which they would have moved if they had not been subjected to that influence. Now the great problem which a practical politician has to solve is, the determination of the effects of a proposed measure. In proportion as he can calculate the future effects of a present act, he will exhibit one of the main qualifications of his office. We have already, in a former chapter, (2) examined the process by which the future effects of a given cause in politics are to be investigated; and we propose now to follow up that inquiry by considering, more at large, the extent to which political prediction can be carried, and the limits by which it is bounded, as compared with other departments of practice and knowledge.

§ 2 It is generally assumed that there is a broad distinction between the physical sciences, on the one hand, and the sciences relating to the moral and mental nature of man on the other, as to their respective powers of prediction. The common belief is, that the former can to a great extent anticipate the future with certainty, whereas the latter can only guess at it with various degrees of probability. For this distinction there is some foundation, but it is stated in too indiscriminating a manner, and it establishes a contrast between the physical and moral departments of theory and practice which a detailed analysis does not confirm.

Before we proceed to compare the prophetic powers of these two classes of sciences, we may previously remark that the mathematical sciences-arithmetic, algebra, geometry, and the higher calculus do not, by the form of their expression, involve any question of time, and therefore are not limited in their application to the past. The proposition that twice two is four, that two sides of a triangle are greater than a third, or that in the parabola the square of the ordinate is equal to the product of the parameter and abscissa-is as true with respect to all future, as it has been with respect to all past time. It is true that time admits of being measured by number, and that arith

(2) Above, ch. xiii.

metic is applied to the computation of time; but such arithmetical calculations are just as applicable to future as to past time.

In fact, the idea of prediction does not at all apply to the abstract sciences of magnitude and number; nor, indeed, is it applicable to science in general, in the same sense in which it is applicable to art or practice. Those sciences which lay down general laws of sequence, do not affirm that a certain cause will produce a certain effect in any individual case. They affirm a certain relation of cause and effect with respect to an entire class of phenomena. This is virtually a prediction; but it is a prediction in a different sense from the prediction of the experimenter, who affirms that, in the actual case before him, a certain event will happen—or from the prediction of the astronomer, who affirms that the moon will be eclipsed at such a moment on such a day. Prediction, properly speak

ing, refers to a single event, and affirms that, in an actual and concrete set of circumstances, a certain effect will be produced; whereas the anticipations of science are general, and merely affirm that, in a hypothetical and abstract state of things, a certain cause will produce a certain effect. In other words, it predicts generally the tendency of a cause to produce a certain effect, in so far as its operation is undisturbed by other causes, of which the pure theory takes no account.

§ 3 One important class of the physical sciences consists of those which describe natural objects, and their permanent laws of causation, without reference to a cyclical recurrence. Under this head are included mechanics, hydrostatics, optics, acoustics, chemistry, thermology, and the sciences of electricity, magnetism, and galvanism. These sciences are principally occupied with determining the sequences of the proper phenomena: e.g. mechanics, in explaining the laws of motion and equilibrium for solid bodies, and hydrostatics the same for fluids; optics, in explaining the properties of light; acoustics, of sound; chemistry, in tracing the relations of material particles, and so on. These phenomena either co-exist according to a constant relation, or succced one

another in a constant sequence, which has no reference to a fixed period of time, and hence a law of physical causation, or a fact of physical co-existence, when once determined, is as true of the future as of the past. (3) For example, the laws of motion in mechanics, or the law of equilibrium of fluids, are universal truths, not limited to the past, and will retain their truth as long as the order of nature remains unchanged. So the proposition that platinum is heavier that gold, that mercury becomes solid at a certain temperature, or that animals of the ox species ruminate, will be as true next year as it was last year, and as it is at the present moment.

When any physical fact or sequence has been completely determined, and the phenomena are of such a nature as to be within our command, we can reproduce them at will in the experimental method, and can therefore predict them with certainty in an individual case. We know, for instance, that we can repeat Lavoisier's experiment upon the absorption of oxygen by combustion; we know that white light can be analysed into its component rays, by refraction through a glass prism; we know that electricity can be generated by the friction of certain substances.(*) But if the physical sequence (though capable of determination when the elements can be isolated) is out of our power, and is liable to be acted on by disturbing elements which we cannot command, then its prediction is subject to uncertainty. For example, the pressure of water upon a bridge or dyke may be computed, and the strength of the resistance arranged accordingly; but if the pressure should be increased by unforeseen circumstances beyond the amount calculated on, the bridge or dyke may be washed away. Hence, with respect

(3) M. Comte (Cours de Phil. Pos. tom. vi. p. 815) speaks of the 'extrême imperfection, à la fois scientifique et logique, des études chimiques, où la prévision rationnelle, qui caractérise surtout la veritable science, n'est presque jamais possible aujourd'hui qu'à certains égards secondaires.' This view of chemistry is only true with respect to those substances or phenomena which have not been the subjects of experiment. Experiments in chemistry which have been made can be repeated ad libitum, and in this sense prediction is possible.

(4) Above, ch. vi. § 3.

to physical sequences which we cannot control, we can only affirm the tendency of causes to produce their appropriate effects, inasmuch as we cannot exclude the operation of disturbing or counteracting causes. Thus, we can affirm the tendency of a heavy body to roll down an inclined plane, but the extent to which that tendency may be checked, will depend upon the inclination of the plane, the smoothness of its surface, the shape of the body, its specific gravity, &c. These counteracting causes may be such as to overcome the downward tendency of the body, and to keep it at rest.

It has been remarked that time is an element in dynamical, but not in statical problems. (5) Equilibrium is conceived as existing at a single punctum temporis, whereas motion is conceived as occupying a certain period of time. This distinction, however, does not limit the prophetic powers of dynamics, as compared with statics. Although time is an element of dyna¬ mical problems, it is a hypothetical, not an actual period, and the portion of time which enters into the calculation may as well be taken with reference to the future, as with reference to the past.

So long as we describe facts or co-existences in nature- -as when we say that gold is yellow, or that lead is heavier than iron -and so long as we lay down ascertained laws of sequence with respect to physical phenomena which we can command, the proposition is independent of time, and is as applicable to the future as to the past: it is expressive of the result of our universal experience of the past, which we have every reason to believe will accord with our experience of the future. But when we come to apply our general scientific propositions in untried operations of practice-when we proceed to make new and unascertained combinations of natural substances, or new applications of mechanical powers, our prediction in each individual case is uncertain, and the result can only be determined by actual experiment. Hence, the invention of all new machines and mechanical contrivances is a tentative process: the mechanist, however

(5) See Comte, Phil. Pos. tom. i. p. 580.