It is evident that the Socratic search for the essence by an analysis of instances - an induction ending in a definition - has a strong resemblance to the Baconian inductive method.

[83] N. O. i. 105.

[84] That is to say, differing in nothing save the absence of the nature under investigation.

[85] Distrib. Op. (Works, iv. 28); Parasceve (ibid. 251, 252, 255-256); Descrip. Glob. Intel. ch. 3.

[86] Works, ii. 16; cf. N. O. i. 130.

[87] A Barnabite monk, professor of mathematics and philosophy at Annecy.

[88] Letters and Life, vii. 377.

[89] For a full discussion of Bacon's relation to his predecessors and contemporaries, see Fowler's N. O. introd. § 13.

[90] Cf. what Bacon says, N. O. i. 130.

[91] Brewster, Life of Newton (1855) (see particularly vol. ii. 403, 405); Lasson, über Bacon von Verulam's wissenschaftliche Principien (1860); Liebig, über Francis Bacon von Verulam, etc. (1863). Although Liebig points out how little science proceeds according to Bacon's rules, yet his other criticisms seem of extremely little value. In a very offensive and quite unjustifiable tone, which is severely commented on by Sigwart and Fischer, he attacks the Baconian methods and its results. These results he claims to find in the Sylva Sylvarum, entirely ignoring what Bacon himself has said of the nature of that work (N. O. i. 117; cf. Rawley's Pref. to the S. S.), and thus putting a false interpretation on the experiments there noted. It is not surprising that he should detect many flaws, but he never fails to exaggerate an error, and seems sometimes completely to miss the point of what Bacon says. (See particularly his remarks on S. S. 33, 336.) The method he explains in such a way as to show he has not a glimpse of its true nature. He brings against Bacon, of all men, the accusations of making induction start from the undetermined perceptions of the senses, of using imagination, and of putting a quite arbitrary interpretation on phenomena.

He crowns his criticism by expounding what he considers to be the true scientific method, which, as has been pointed put by Fischer, is simply that Baconian doctrine against which his attack ought to have been directed. (See his account of the method, über Bacon, 47-49; K. Fischer, Bacon, pp. 499-502.)

[92] Mill, Logic, ii. pp. 115, 116, 329, 330.

[93] Whewell, Phil. of Ind. Sc. ii. 399, 402-403; Ellis, Int. to Bacon's Works, i. 39, 61; Brewster, Newton, ii. 404; Jevons, Princ. of Science ii. 220. A severe judgment on Bacon's method is given in Dühring's able but one-sided Kritische Gesch. d. Phil., in which the merits of Roger Bacon are brought prominently forward.

[94] Although it must be admitted that the Baconian method is fairly open to the above-mentioned objections, it is curious and significant that Bacon was not thoroughly ignorant of them, but with deliberate consciousness preferred his own method. We do not think, indeed, that the notiones of which he speaks in any way correspond to what Whewell and Ellis would call "conceptions or ideas furnished by the mind of the thinker"; nor do we imagine that Bacon would have admitted these as necessary elements in the inductive process. But he was certainly not ignorant of what may be called a deductive method, and of a kind of hypothesis. This is clear from the use he makes of the Vindemiatio, from certain hints as to the testing of axioms, from his admission of the syllogism into physical reasoning, and from what he calls Experientia Literata. The function of the Vindemiatio has been already pointed out; with regard to axioms, he says (N. O. i. 106), "In establishing axioms by this kind of induction, we must also examine and try whether the axiom so established be framed to the measure of these particulars, from which it is derived, or whether it be larger or wider.

And if it be larger and wider, we must observe whether, by indicating to us new particulars, it confirm that wideness and largeness as by a collateral security, that we may not either stick fast in things already known, or loosely grasp at shadows and abstract forms, not at things solid and realized in matter." (Cf. also the passage from Valerius Terminus, quoted in Ellis's note on the above aphorism.) Of the syllogism he says, "I do not propose to give up the syllogism altogether. S. is incompetent for the principal things rather than useless for the generality. In the mathematics there is no reason why it should not be employed. It is the flux of matter and the inconstancy of the physical body which requires induction, that thereby it may be fixed as it were, and allow the formation of notions well defined. In physics you wisely note, and therein I agree with you, that after the notions of the first class and the axioms concerning them have been by induction well made out and defined, syllogism may be applied safely; only it must be restrained from leaping at once to the most general notions, and progress must be made through a fit succession of steps." - ("Letter to Baranzano," Letters and Life, vii. 377). And with this may be compared what he says of mathematics (Nov. Org. ii. 8; Parasceve, vii.). In his account of Experientia Literata (De Aug. v. 2) he comes very near to the modern mode of experimental research.

It is, he says, the procedure from one experiment to another, and it is not a science but an art or learned sagacity (resembling in this Aristotle's ἀγχίνοια), which may, however, be enlightened by the precepts of the Interpretatio. Eight varieties of such experiments are enumerated, and a comparison is drawn between this and the inductive method; "though the rational method of inquiry by the Organon promises far greater things in the end, yet this sagacity, proceeding by learned experience, will in the meantime present mankind with a number of inventions which lie near at hand." (Cf. N. O. i. 103.)

[95] See the vigorous passage in Herschel, Discourse on the Study of Natural Philosophy, § 105; cf. § 96 of the same work.

[96] Bacon himself seems to anticipate that the progress of science would of itself render his method antiquated (Nov. Org. i. 130).

[97] Nov. Org. i. 127.