But Mr. Casaubon’s theory of the elements which made the seed of all tradition was not likely to bruise itself unawares against discoveries: it floated among flexible conjectures no more solid than those etymologies which seemed strong because of likeness in sound until it was shown that likeness in sound made them impossible: it was a method of interpretation which was not tested by the necessity of forming anything which had sharper collisions than an elaborate notion of Gog and Magog: it was as free from interruption as a plan for threading the stars together. (George Eliot, Middlemarch, ch. 48)Mr. Gijsbers' post got me wondering about the provenance of this idea.
Other nineteenth-century theorists stressed the importance of trying to falsify our beliefs in our quest for truth. In an article in the Monist in 1982, Susan Haack identified this theme in C. S. Peirce's work (and pointed out differences between Peirce's and Popper's views), and others take William Stanley Jevons to have emphasized the importance of attempts at falsification in our belief-forming regimen.
I doubt that Eliot was influenced by either of these theorists. Peirce's work was published too late to have influenced Eliot and it looks like Jevons and Eliot developed their thinking independently of each other. (Update [Sept. 23]: Eric Schliesser asked why I don't think that Jevons influenced Eliot on this matter. Eliot published Middlemarch in 1871-2, but Jevons' statement on the importance of seeking a theory's falsification ['objections' and 'discordance'] appeared in his Principles of Science, which was published in 1874.)
Eliot was influenced by Auguste Comte's positivism, but Comte doesn't seem to have recognized falsifiability as a hallmark of good epistemology (let alone of specifically scientific theory). Nonetheless, the Comtean background is relevant. Comte endorsed phrenology, which was ridiculed as a quack pseudo-science. It looks like some of phrenology's nineteenth-century critics were put off by the phrenologists' capacity to dispense with any and all seemingly recalcitrant bits of behavioral evidence by means of arbitrary stipulations. My source for this claim is G. N. Cantor ('The Edinburgh Phrenology Debate: 1803-1828', Annals of Science 32 : 195-218, see esp. 212 and 213).
Cantor attributes the objection to Francis Jeffrey and to P. M. Roget. Jefffrey's critique appeared in a book review in an 1826 issue of the Edinburgh Review, (v. 44 [Sept 1826]: 253-318). According to Jeffrey,
It is quite plain ... that these admissions ...reduce this whole 'science of observation,' to a series of mere evasions and gratuitous suppositions. We produce, for example, a person whose whole conduct indicates great Benevolence, but who happens to have a very small bump in the place where the organ of that propensity is said to be situated. Is not this a proof of the fallacy of the system? Oh no—by no means. The individual has had the good luck to be trained up among very benevolent people, and has had his small original stock prodigiously increased by their precepts and example, aided perhaps' by his own large endowment of the faculty of Imitation!—or, his organ of benevolence has perhaps been excited to a diseased activity by some internal inflammation,—or at all events, as he has Love of Approbation and Cautiousness very large, nothing is so probable as that his apparent benevolence is merely put on, to gain the good opinion of the world, or to secure some advantage to himself! We next produce another person with an enormous bump of Benevolence on his forehead; and, offering to prove that he is, notwithstanding, notoriously cruel, oppressive, and uncharitable, we ask, again, how this is to be reconciled with the truth of the system? O, nothing in the world so easy! First of all, he has probably had no training in the paths of benevolence, and the field, though naturally fertile, has therefore been actually barren; But besides, you have only to look, and you will most probably find the organs of Combativeness, and Destructiveness, and Acquisitiveness, still larger than that of Benevolence. These, of course, make him quarrelsome, and cruel, and avaricious: and how then can his poor benevolence find means to display itself? (Francis Jeffrey, Edinburgh Review 44 : 298-299)Jeffrey concluded that 'the phrenological theory, though absolutely incapable of any clear or satisfactory proof, abounds in those equivocations and means of retreat, by which it may often escape from. direct refutation'. (Ibid.)
As Cantor says, the response of critics like Jeffrey and Roget
resembles Popper's reaction to Adler's theory a century later. In his Encyclopaedia Britannica article P. M. Roget noted the difficulty of falsifying phrenology since its proponents employed ad hoc assumptions which covered all contingencies. (Cantor, 213)George Eliot had endorsed phrenology but later came to see its limitations. By the 1860's, she was poking fun at the theory in Felix Holt: the Radical (1866). As T. R. Wright has pointed out, Eliot's ridicule plays upon phrenology's apparent immunity to falsification. ('From Bumps to Morals: The Phrenological Background to George Eliot's Moral Framework', The Review of English Studies, New Series 33 : 34-46) In Chapter 5 of Felix Holt, the eponymous character echoes Jeffrey's above objection:
A phrenologist at Glasgow told me I had large veneration; another man there, who knew me, laughed out and said I was the most blasphemous iconoclast living. 'That,' says my phrenologist, 'is because of his large ideality, which prevents him from finding anything perfect enough to be venerated.' (Felix Holt, ch. 5)In addition to the critique of phrenology, another possible influence on Eliot's thinking was the methodological work of Claude Bernard. In 1865, Bernard published his Introduction to the Study of Experimental Medicine, the first part of which was devoted to methodological considerations. Bernard's reflections occupied Eliot and her partner, G. H. Lewes, in the 1870's while Eliot was writing Middlemarch and Lewes was working on his multi-volume Problems of Life and Mind (which was completed by Eliot after Lewes' death). I don't know much about Bernard's methodology, but at least one philosopher, Jean-François Malherbe, finds close parallels between it and Popper's views.
So, Eliot's remark about 'Mr. Casaubon's theory of the elements' places Casaubon's theory in the same pseudo-scientific wastebasket as phrenology. Scientists, for Eliot, are better advised to follow the example of Lydgate, who 'was enamored of that arduous invention which is the very eyes of research, provisionally framing its object and correcting it to more and more exact relation.' (Middlemarch, ch. 45)