blishes such a general formula as that the blood circulates, the nerves are the media of sensation, and the like, he leaves out of sight the peculiarities of individual cases, and makes no reference to any particular person, or to any time or place. This latter process is analogous to the methods of positive politics: again, pathological investigations are analogous to the speculative department of politics, while the maxims of therapeutics correspond with the art.

In

§ 13 From what was said at the beginning of this chapter, it appears that all political facts are equally open to the observation of the senses, and that they are not of such a nature as to require any artificial assistance for their determination. this respect, the observation of politics is more simple and easy than that of physics. In many departments of physics, the phenomena are so distant, so minute, and so difficult to detect and to measure, that recourse is had to instruments of various kinds for their observation. In the use of these, much skill, nicety, delicacy, and precision is necessary; and they moreover introduce a set of errors peculiar to themselves, which require a whole system of refined contrivances in order to counteract them.(58) Errors of observation in physics are often corrected by the multiplication of observations, either of the same or different

(58) See Herschel, Treatise on Astronomy, art. 160; and compare Mr. De Morgan's Essay on Probabilities, ch. 7. In p. 148 he remarks: A reader unused to astronomical works, on opening a book on the practical part of the science, might imagine that no part of the subject pretended even to ordinary accuracy. Nothing appears to be done which is unaffected by serious error; and it seems as if a little more attention to the fabrication of instruments would render nine-tenths of what has been written altogether useless. This appearance is the victory of the head over the hands: the means of detecting the errors of instruments are much more powerful than those of correcting them.'

Compare likewise the remarks of Dr. Carpenter upon the errors of microscopic observation for physiological purposes :-'Investigations into the elementary arrangement of the parts which primarily compose organized structures are often attended with much difficulty and liability to error. The minuteness of the objects which are to be examined, and the changes which may be produced in them by the preparation they are necessarily made to undergo, before being submitted to microscopic inspection, not to mention the deceptions arising from imperfection in the instrument itself, or the mode of employing it, have led to much discrepancy in the statements of different observers.'-Gen. and Comp. Phys. § 20.

observers; and the opposite errors some in excess, and some in defect-compensate and cancel each other in a large number. This process must not be confounded with the elimination of chance, by observing a large number of cases; as in gambling with dice or cards, or in the estimation of the probability of life. Politics is not subject to the former class of errors, and therefore does not require the same means of correcting them; but it is peculiarly subject to the latter class of errors, and guards against them by the same process as the physical observer.

Again, some causes of error in observation are common to politics and physics. Such are, a dislike of labour, a want of steady and sustained attention; a proneness to exaggeration, and a love of the marvellous. That physical science has not, more than political history, been exempt from the operation of the latter causes, is proved by the numerous fictions which are to be found in the ancient naturalists, and afterwards became articles of popular faith: such as the existence of the phoenix, the singing of the swan before death, the poisonous qualities of the toad, the shrieks of the mandragora, the long life of the stag, the existence of the unicorn, &c.(59)

But the chief cause of erroneous observation, and the chief obstacle to correct observation, in politics, is interest, leading to deceit and bad faith with respect to facts; to the suppression and alteration of the truth, and to the fabrication of falsehood. Whatever relates immediately to human affairs-whatever may produce pecuniary gain or loss, may affect the character and good name of individuals, or may cause pain or pleasure, by acting upon their feelings, or even upon their vanity—is drawn within the sphere of influences hostile to the accurate determination of facts. To this source of error, the practical business of government is peculiarly exposed. Both the legislature and executive are perpetually liable to be misled by false facts, which

(59) Upon the fabulous animals of the ancients, see Cuvier's Theory of the Earth, § 26 (p. 74, Engl. transl.)

Aristotle, among his voluminous works, wrote a treatise vπèp Tv μvdoλoyovμévwv (wwv, Diog. Laert. v. § 25. Compare Browne's Vulgar Errors.

interested persons fabricate for the purpose of procuring an exercise of the sovereign power in their own favour. Falsehoods of this kind are analogous to the various forms of fraud and forgery-or the crimen falsi-which are practised between private persons, and which the civil and criminal law prohibit. For example, the imposture practised on a government functionary to obtain public relief, is analogous to the imposture practised on a private individual to obtain his alms. In war, stratagem and deceit of any kind is admissible, for the purpose of misleading the enemy: dolus an virtus, quis in hoste requiret ?' A similar latitude used to be allowed in the intercourse of the governments of independent states: but, whatever the practice may sometimes be, the improved morality of international communication in modern times condemns falsehood in an ambassador. Such is the extensive operation of deceit in practical politics; but even the political theorist is not wholly exempt from its influence. The historian is subject to being misled by false historical testimony, and by literary forgeries: erroneous numerical statements are registered by the statistician; and hence the speculator, in framing political theories, may be induced by their authority to rely upon facts which he believes to have been derived from correct observation, but which, in reality, were invented for an interested purpose.

The physical sciences, considered merely as expressive of general truths, are nearly exempt from this source of error. There is little inducement to deception in observations conducted for the use of a physical science. Cheats sometimes occur in medical cases; but they are rare, and easily detected. When the earth had been less explored than at present, and distant regions were imperfectly known, travellers assumed to themselves a considerable licence as to facts; they gratified their vanity, and their desire of surprising their hearers, by inventing marvellous tales of what they had seen and heard. (6) But the exten

(60) The ancient geographers fabricated fables about distant lands, says Polyb. iii. 58.

sion of geographical knowledge, and the increased facilities of communication by land and sea, have greatly narrowed the operation of this cause of deceit.

It may be added, that when doubts exist as to the correctness of any observation or experiment in physics, it can in general be repeated, either immediately, or within a short time; and hence physical observations and experiments admit of verification. The exceptions are few; such as remarkable earthquakes, and volcanic eruptions, or changes of level; diseases which have become extinct, such as the sweating-sickness, or which have changed their character. Some astronomical events likewise are so rare, that the observations of them can only be verified at long intervals of time. (61) A historical fact, however, being non-recurrent, and being never repeated under similar circumstances, is incapable of verification: the testimony to it may be re-examined, but the fact itself can never be reproduced. (*)

But when we pass from the science to the art, in physics; when the occasion arises for the application of physical knowledge to an actual case of practice; and when the advice and services of the mechanist, the architect, the engineer, the chemist, the physician, are called in for any practical decision-then we find that their judgment is liable to be influenced by interest, not less than the judgment of persons whose principles are derived exclusively from the moral sciences;(63) and hence the facts which they adduce in support of their opinions may be erroneously or imperfectly observed, or distorted, or coloured, under the influence of the bias which they have contracted.

§ 14 There is another fundamental point of difference between the physical sciences and politics, affecting materially their respective capacities of observation, which we have not hitherto noticed, inasmuch as it is not peculiar to politics, but is common to ethics and all the other sciences conversant with

(61) The last transit of Venus was in 1769. See Herschel's Astronomy, art. 411.

(62) See above, § 6.

(63) See the Report of the Gen. Board of Health on the supply of water to the metropolis, (1850,) p. 166.

human action. It is, however, too important to be overlooked in our present comparison.

men.

When we speak of a political fact, we always refer, either directly or indirectly, to something involving human action. Whether it be a historical or statistical fact, a fact as observed in political science, or a fact in practical politics, it is, in some form or other, concerned with the acts, dealings, or conduct of Now in all human action there are two parts; one is the outward material act, as exhibited in the determination of the will, leading to muscular exertion or rest; the other is the mental state, which precedes, accompanies, or follows such act. It is only the first part of this process which can be the object of observation by an external witness; the mental state, coincident with the act, is purely subjective, and is only known, as a matter of consciousness, by the agent himself. The state of mind accompanying the act of a moral being can therefore only be determined indirectly; and the signs or marks by which it is recognised are two; either the declaration of the person himself, by demeanour, gestures, or words, or the circumstances in which the act is performed. (6) In many cases, the declarations of the party with respect to the motives or reasons which determine him to a certain act, are free from suspicion, and therefore credible. In some cases they are open to suspicion, as being partially untrue, as suppressing one portion of the truth, or as perverting, colouring, and altering the remainder. In other cases they are wholly unworthy of belief. Inasmuch as the declarations of the agent with respect to the mental antecedents of his act cannot be always trusted, it is necessary to have recourse to other indications of its moral quality.(65) This is

(64) The knowledge of men's thoughts is, according to Bacon, obtained by six modes; viz.-1, by their countenance; 2, their words; 3, their acts; 4, their character; 5, their ends; 6, the reports of others.-De Augm. lib. viii. vol. ix. p. 41. In his Essay on Negotiating, he says, 'All practice is to discover, or to work. Men discover themselves in trust, in passion, at unawares; and of necessity, when they would have somewhat done, and cannot find an apt pretext.'

(65) Upon reflection, it will appear that almost all the characteristic principles of ethics are deducible from the secrecy of motives, and the impossibility of discovering them by direct means.