as a corpus vile-and vary its circumstances at our pleasure, for the sake only of ascertaining abstract truth. We cannot do in politics what the experimenter does in chemistry: we cannot try how the substance is affected by change of temperature, by burning, by dissolution in liquids, by combination with other chemical agents, and the like. We cannot take a portion of the community in our hand, as the king of Brobdignag took Gulliver, view it in different aspects, and place it in different positions, in order to solve social problems, and satisfy our speculative curiosity. $ 5 Nevertheless, it would be an error to suppose that political science would gain any addition to its stock of positive information by the adoption of the method of experiment, or that the facts upon which it is founded could be better or more fully ascertained by experimentation, than by the method of simple observation. The physical philosopher is compelled to interrogate nature by experiments, because she is mute. But man, the

subject of politics, can speak: he can declare his feelings spontaneously; or he can answer interrogations. Hence the experiments of physical science are, after all, a feeble and rude contrivance, compared with the methods of investigation in politics. Scientific experiment is an imperfect substitute for that information which a man can give respecting his experience; respecting his internal feelings and changes of consciousness, and the events which have passed within the range of his senses. The information which experiment can extract from insentient masses of matter, or from gases and fluids, is scanty and uninstructive as compared with the answers of human intelligence. The

investigation as in the physical sciences. The political philosopher cannot treat a nation as a chemist treats matter. He can observe the facts recorded by history, or those which he has himself witnessed; but he cannot make experiments himself, or repeat the experiments of others. Governments, indeed, act upon nations experimentally, but their experiments are always made in the same direction (?), and with the view of arriving at a result which is not always avowed. They do not afford to those who are not convinced of the rectitude of their processes, the facility of instituting contrary experiments (tom. i. p. 13). See likewise, A. Comte (Cours de Phil. Pos. tom. iv. p. 428), who shows that social science cannot use direct experiment.

responses of one oracle are brief and meagre, as compared with the copious and godlike accents which proceed from the other shrine.

If, on the other hand, every portion of matter was animated; if, according to the ancient pagan faith, every tree had its dryad, every stream its naiad; if the lightning, the winds, the element of fire, and all the great powers of nature, were each subject to their appropriate deity; or if, as in the European mythology of more recent times, gnomes and fairies and elves presided over external objects, we might obtain from lifeless matter information concerning its attributes and qualities. If we could appeal to the supernatural beings described by Pope, and by evocations and magic formulas compel them to reveal the mysteries of nature, experiment might be discarded as superfluous :

Some in the fields of purest ether play,

And bask and whiten in the blaze of day.
Some, guide the course of wandering orbs on high,
Or roll the planets through the boundless sky.
Some, less refined, beneath the moon's pale light,
Pursue the stars that shoot across the night;

Or suck the mists in grosser air below ;
Or dip their pinions in the painted bow;

Or brew fierce tempests on the wintry main;

Or o'er the glebe distil the kindly rain.

With such informants as these upon the laws of physical phenomena, we should despise the tardy process of experimental investigation, and, as in human affairs, should resort to the testimony of percipient witnesses.

The different advantages which are afforded to the observer by the voluntary communications of intelligence, and by experiments upon unintelligent matter, may be illustrated by a comparison of human and veterinary medicine. Buffon recommends the more careful cultivation of veterinary medicine, as tending to throw light upon human medicine, by the facilities for scientific study which it presents; among which he enumerates the unrestricted power of making experiments, and trying new remedies.(24) Now the veterinary art has been cultivated with

(24) La médecine que les anciens ont appellée médecine vétérinaire, n'est presque connue que de nom je suis persuadé que si quelque médecin

much assiduity and skill since the time of Buffon, and yet it has thrown little or no light upon human pathology and therapeutics. It has been found that the explanations which the human patient affords to the physician, respecting his state and sensations, are far more instructive than the experiments which the veterinary practitioner may make upon a dumb, irrational animal. In the treatment of infants, the physician is subject to a similar disadvantage, without the corresponding advantage of a facility of making experiments on his patient.

Wherever there is intelligence there is sensibility; and wherever there is sensibility, experiment, as such, mere philosophical manipulation for the sake of determining truth, is inapplicable. Each method excludes the other; but the information derived from an intelligent subject is more instructive than that attainable by the method of experiment, acting upon insentient

matter.

A physical philosopher making researches into the properties of matter, compared with a political philosopher inquiring into the nature of governments and laws, and the tendencies of human institutions, is like a traveller in a foreign country who can speak the language of the natives, compared with a traveller who is unable to hold converse with them.

If political society were a vast machine, or inert material body-if it was really a great Leviathan constructed of men, after the manner of the figure delineated in the frontispiece to the treatise of Hobbes-and if the personality of the individual

tournoit ses vues de ce côté-là, et faisoit de cette étude son principal objet, il en seroit bientôt dédommagé par d'amples succès; que non-seulement il s'enricheroit, mais même qu'au lieu de se dégrader il s'illustreroit beaucoup, et cette médecine ne seroit pas si conjecturale et si difficile que l'autre la nourriture, les mœurs, l'influence du sentiment, toutes les causes, en un mot, étant plus simples dans l'animal que dans l'homme, les maladies doivent aussi être moins compliquées, et par conséquent plus faciles à juger et à traiter avec succès; sans compter la liberté qu'on auroit toute entière de faire des expériences, de tenter de nouveaux remèdes, et de pouvoir arriver sans crainte et sans reproche à une grande étendue de connoissances en ce genre, dont on pourroit même par analogie tirer des inductions utiles à l'art de guérir les hommes.'-Buffon, Hist. Nat. Quad. tom. i. p. 96: ed. 1786.

1

man was merged and lost in the composite state-then it might be convenient to resort to the method of scientific experiment in politics. But as long as men, though members of a state, retain their personal identity and individual consciousness; so long as they are able to note and describe the subjects of their intuitions and sensations, so long will scientific experiment be powerless, for purposes of information, in politics. It is only in the absence of such informants that the method of scientific experiment ought to be employed. Wherever this mode of exploring natural objects is applicable, it ought to be applied. In proportion as the subjects of any physical science can be brought under the close scrutiny of all our senses, and as, consequently, the method of experiment is applicable to it, our knowledge of it will be full and satisfactory.(2) Nevertheless, as we shall show in a future chapter, (2) the only advantages of the method of experiment which are peculiar to physics, and which cannot be imitated in political reasoning, are negative. They consist, not in the ascertainment of positive facts, but in the exclusion of concurrent phenomena. Experiment presents the elements of a problem, all enclosed within a precise line of demarcation; and hence it creates a definiteness and exactitude unattainable in politics. This advantage of the experimental method, however, consists, not in what it includes, but in what it excludes; not in what it establishes, but in what it guards a gainst.

§ 6 Although the facts within the province of politics are

(25) M.Comte points out a philosophic law, on the subject of observation, which he considers very important, and which he was the first to remark. This law is, that in proportion as physical phenomena become more complicated, they are, at the same time, susceptible by their nature of more extensive and more varied means of investigation (Cours de Phil. Pos. tom. ii. p. 18, 403). Does not this law, however, imply an inversion of cause and effect? Instead of saying that our means of observation increase as the phenomena become more complicated, would it not be more correct to say that the phenomena become more complicated as our means of observation increase? The phenomena of astronomy are simple, because we can only observe them with the one sense of sight; but if we could examine the objects on the surface of the planets or the moon with our other senses, the phenomena would immediately become more complex. (26) Below, ch. ix. § 19.

all open to the observation of our senses, yet (as was remarked in the preceding chapter) (2) there is a constant danger of their being misreported for interested purposes. The whole field of practical politics is beset with this risk; attempts to deceive the legislative and executive authorities of a country by incorrect information, for an interested purpose, are made at every turn. There is constantly a new crop of false facts, designed to mislead the persons who have conduct of the government. But the method of experiment, even if it were applicable to politics, would afford no protection against false facts of this class, inasmuch as they arise, not from any natural difficulty of ascertaining the truth inherent in the subject, but from an unwillingness to report it, or rather from a desire to misreport it.

The reluctance of persons to disclose facts, when the disclosure would bring calamity on themselves or their friends, has induced governments, in cases where the knowledge of these facts was, or seemed to be, important, to have recourse to torture for the purpose of extracting the truth. Torture, in practical politics, may be considered as analogous to experiment in physics; it is an artificial process for exploring the truth. In the one case, means are used for interrogating an insentient object, which cannot speak; in the other case, means are used for interrogating an intelligent agent, who can, but will not speak. ever, the analogy stops. For the method of scientific experiment brings out true results; whereas the torture of a man, if it succeeds in compelling him to speak, only compels him to say something; whether what he says be true or false is uncertain.(*)

(27) Above, ch. v. § 13.

Here, how

(28) Concerning Philotas, who confessed certain designs against Alexander, Quintus Curtius remarks as follows:- Philotas, verone an mendacio liberare se a cruciatu voluerit, anceps conjectura est, quoniam et vera confessis, et falsa dicentibus, idem doloris finis ostenditur.'—vi. c. 11.

'Quæstiones nobis servorum ac tormenta accusator minitatur: in quibus quamquam nihil periculi suspicamur, tamen illa tormenta gubernat dolor, regit quæsitor, flectit libido, corrumpit spes, infirmat metus; ut in tot rerum angustiis nihil veritati loci relinquatur.'-Cicero, pro Sullâ, c. 28.

6

Quæstioni fidem non semper, nec tamen nunquam habendam, constitutionibus declaratur; etenim res est fragilis, et periculosa, et quæ veritatem fallat. Nam plerique patientiâ sive duritiâ tormentorum ita tormenta