is, the mean between certainty and impossibility; (f) and probability therefore includes the whole range between those extremes.

The terms CERTAINTY and PROBABILITY are however essentially different in meaning as applied to moral evidence, from what they import in a mathematical sense; inasmuch as the elements of moral probability, notwithstanding the ingenious arguments which have been urged to the contrary, appear to be incapable of numerical expression, and because it is not possible to assign all the chances for or against the occurrence of any particular event.

The expression MORAL PROBABILITY, though liable to objection on account of its deficiency in precision, is for want of one more definite and appropriate, of frequent and necessary use; nor will its application lead to mistake, if it be remembered, that it expresses only the preponderance of probability, resulting from [*7] the comparison and estimate of moral evidence, and that if it were capable of being expressed with exactness, it would lose its essential characteristic and possess the certainty of demonstration.

The preceding strictures equally apply to the expression MORAL CERTAINTY, which must be understood, not as importing deficiency in the proof, but only as descriptive of the kind of certainty which is attainable by means of moral evidence; and it is that degree of assurance which induces a man of sound mind to act without doubt upon the conclusions to which it leads.(g)

It has been justly and powerfully remarked by a noble and learned writer, that "the degree of excellence and

(ƒ) Kirwan's Logic, part iii. ch. vii. s. 1.

(9) Stewart's Elements, vol. ii. ch. ii. s. 4. Encyclopædia Brit., art. METAPHYSICS, part i.

of strength to which testimony may rise seems almost indefinite. There is hardly any cogency which it is not capable by possible supposition of attaining. The endless multiplication of witnesses-the unbounded variety of their habits of thinking, their prejudices, their interests-afford the means of conceiving the force of their testimony augmented ad infinitum, because these circumstances afford the means of diminishing indefinitely the chances, of their being all mistaken, all misled, or all combining to deceive us." (h) But if evidence leave reasonable ground for doubt, the conclusion cannot be morally certain, however great may be the preponderance of probability in its favour.

Some mathematical writers have propounded numerical fractions for expressing moral certainty; which, as might have been expected, have been of very different values. But the nature of the subject precludes the possibility of reducing to the form of arithmetical notation the subtle, shifting, and evanescent elements of moral assurance, or of bringing to quantitative comparison, things so inherently different as certainty and probability.

*Other writers have given in a more general [*8] manner, mathematical form to moral reasonings and judgments; but it is questionable if they have produced any useful result, however they may have shown the ingenuity of their authors. (i) Though it be true that some very important deductions from the doctrine of chances, are applicable to events dependent upon the duration of human life, such as the expectation and the decrement of life, the law of mortality, the value of annuities and other contingencies, and also to reason

(h) Lord Brougham's Discourse on Natural Theology, p. 251. (i) See Kirwin's Logic, part iii. ch. vii. s. 21. Whately's Logic, b. iv. ch. ii. s. 1.

ing in the abstract upon particular cases of testimonial evidence, (k) yet it is obvious, that all such conclusions depend upon circumstances, which, notwithstanding that to the superficial and unreflecting observer they appear casual, uncertain, and irreducible to principle, unlike moral facts and reasonings in general are really based upon and deducible from numerical elements. (7)

A learned writer, whose opinions, in despite of his numerous eccentricities of matter and of style, have exercised great influence in awakening the spirit of judicial reformation, and are destined to exercise still more auspicious influences, asks, (m) "Does justice require less precision than chemistry?" The truth is, that the precision attainable in the one case is of a nature of which the other does not admit. It would be absurd to require the proof of an historic event, by the same kind of evidence and reasoning as that which establishes the quality of triangles upon equal bases and between the same parallels, or that the latus rectum in an ellipse is a third proportional to the major and minor

axes.

This conscript father of legal reforms (n) has himself supplied a memorable illustration of the futility of his own inquiry. He has proposed a scale [*9] for measuring the degrees of belief, with a positive and a negative side, each divided into ten degrees, respectively affirming and denying the same fact, zero denoting the absence of belief; and the witness is to be asked what degree expresses his belief most correctly. With

(k) Whately's Logic, b. iv. ch. ii. s. 1.

(1) Lubbock on Probability.

(m) Bentham's Traité des Preuves Judiciaires, b. i. ch. xvii. Mackintosh's Discourse on the Progress of Ethical Philosophy, p. 290.

(n) Hoffman's Course of Legal Study, vol. i. p. 364.

his characteristic ardour, the venerable author gravely argues that this instrument could be employed without confusion, difficulty, or inconvenience. (o) But MAN must become wiser and better before the mass of his species can be entrusted with the use of such a moral gauge, from which the unassuming and the wise would shrink, while it would be eagerly grasped by the conceited, the interested, and the bold.

But, though a process strictly mathematical cannot be applied to estimate the effect of moral evidence, a proceeding somewhat analogous is observed in the examination of a group of facts adduced as grounds for inferring the existence of some other fact. Although an exact value cannot be assigned to the testimonial evidence for or against a matter of disputed fact, the separate testimony of each of the witnesses has nevertheless a determinate relative value, depending upon considerations which it would be foreign to the present subject to enumerate. On one side of the equation are mentally collected all the facts and circumstances which have an affirmative value; and on the other, all those which lead to an opposite inference, or tend to diminish the weight, or to show the non-relevancy, of all of the circumstances which have been put into the opposite scale. The value of each separate portion of the evidence is separately estimated, and, as in algebraic addition, the opposite quantities positive and negative, are united, and the balance of probabilities is what remains as the ground of human belief and judgment. (p)

[*10]

(0) Bentham's Rationale of Judicial Evidence, b. i. ch. vi. s. 1, and see in Kirwan's Logic, part iii. ch. vii. s. 21, a proposed scale of testimonial probability.

(p) See some remarks on this passage in a learned paper "On the

But, as has been already intimated, there is another sense in which the word probability is often used, and in which it denotes CREDIBILITY or INTERNAL PROBABILITY, and expresses our judgment of the accordance or similarity of events with which we become acquainted through the medium of testimony, with facts previously known by experience. (q)

The results of EXPERIENCE are, expressly or impliedly, assumed as the standard of credibility in all questions dependent upon moral evidence. By means of the senses and of our own consciences we become acquainted with external nature, and with the characteristics and properties of physical things and moral beings, which are then made the subjects of memory, reflection, and other intellectual operations; and ultimately, the inferences and observations to which they lead, are reduced to general principles, and become the basis and standard of comparison in similar circumstances. The ground-work of our reasoning, is our confidence in the permanence of the order of nature, and in the existence of moral causes, which operate with unvarying uniformity, not inferior to, and perhaps surpassing even, the stability of physical laws; though relatively to our feeble and limited powers of observation and comprehension, and on account of latency, subtlety, and fugitiveness of mental operations, and of the infinite diversities of individual men, there is apparently more of uncertainty and confusion in moral than in material phænomena. (r) *Experience comprehends, not merely the facts and deductions, personal ob

[*11]

Measure of the Force of Testimony in cases of Legal Evidence," by John Tozer, Esq., M. A., Camb. Phil. Trans. vol. viii.

(9) Abercrombie on the Intellectual Powers, part ii. s. 3. (r) Hampden's Lectures introductory to the study of Moral Philo