The Mechanism of Conviction

In trying to understand the way savants came to believe in the reality of meteorites, I have found it very useful to borrow a concept from the physiology of the nervous system, the concept of 'summation' s1 See P. N. R. Usherwood, Nervous Systems (London: Edward Arnold, 1973), 88-93.. This concept has to do with the way that signals are passed along the nerves. Often a single impulse arriving from a nerve fibre at a nerve synapse will not be transmitted unless another impulse arrives from another fibre within a short time. The synapse then integrates or 'sums' the signals from the fibres and transmits its own signal accordingly. I would like to argue that in a very similar way, anomalous events are subject to 'summation effects' - that is to say, that the reaction to reports of anomalous events is a function of the quantity and quality of reports received. In this case time is not so important as is the fact of having received two or more independent reports which together may produce a sense of conviction - or at least of interest - which a single report would not. Keeping this concept in mind may help us understand the behaviour of many of the savants in the meteorite controversy for whom evidence coming from several independent reports indicated that meteorites were worth studying.

Let us consider this matter of summation effects and their importance a little further. There are many instances in ordinary life where independent evidence from two or more sources produces a sense of conviction that information from a single source does not. Both in the courts and in military intelligence work this kind of concurrent evidence has a greater probative value s2 Cf. L. J. Cohen, 'How Can One Testimony Corroborate Another?', in R. S. Cohen (ed.), Essays in Memory of Imre Lakatos (Dordrecht: Reidel, 1976), 65-78.. It has a special importance, however, where the content of the communication is one which the receiver would tend to doubt. The reception of anomalous events by scientists is, as it were, a special case of this general principle. A single report of an anomalous event will usually be viewed with scepticism, if not dismissed out of hand. But as reports of anomalous events of the same kind multiply, they may begin to receive attention which they would not have received by themselves.

The effect of multiple independent reports seems to work in two ways. The first effect is to orient the scientist toward viewing the events as being worthy of interest. This can be important if it means the expenditure of further time and attention and, even more significant, the search for more evidence. If two or more reports of an unusual event carry common features, especially those which it is difficult to imagine the reporters coming up with on their own, a scientist may feel that there is 'something to' the matter n1The lack of such a non-obvious 'common factor' is one reason R. V. Jones suggests that reports of Unidentified Flying Objects ought to be rejected. See his 'Natural Philosophy of Flying Saucers', Physics Bulletin, Vol. 19 (1968), 225-30, at 229. Part of the difficulty may be the reliance on 'conjunctive' concepts. See J. Bruner, J. J. Goodnow and G. A. Austin, A Study of Thinking (New York: John Wiley, 1956).. The common feature may point to a potentially solvable problem s3See P. B. Medawar, The Art of the Soluble (Harmondsworth, Middx.: Penguin Books, 1969)., one which would justify the expenditure of effort. The second closely related effect is the cross-validation of the worth of each of the reports. The fact that the reporters mention the same thing tends to make both (or all) the reports appear to possess some probative value. Thus the conjunction of two or more reports may move the scientist from doubt to suspended judgment or even to curiosity. There are many instances of exactly this effect in the meteorite controversy.

In 1790 there was a fall of a large number of stones in the community of La Grange de la Juliac in the south of France n2This is the Barbotan fall referred to earlier: Bertholon, op. cit. note 22.. This fall was witnessed by perhaps three hundred persons in the community. A professor at Agen, Saint-Amand, heard about the fall and, highly amused, asked for a formal deposition from the witnesses. To his surprise a legal protocol signed by the mayor and the town attorney arrived along with a sample of the stones. Saint-Amand was no more convinced than before, but sent the document to his friend the physicist Bertholon, who was the editor of the Journal des Sciences Utiles. Bertholon then poured scorn on the witnesses:

If the readers have already had occasion to deplore the error of some individuals, how much more will they be appalled today seeing a whole municipality attest to, consecrate, by a legal protocol in good form, these same popular sensations, which can only excite the pity, not only of physicists, but of all reasonable people ... What can we add here to such an affidavit? All the reflections which it suggests will present themselves to the philosophical reader in reading this authentic attestation of an obviously wrong fact, or a phenomenon physically impossible s4Ibid., 226. (Emphasis in original.).

Eleven years later Saint-Amand was reading an article written about the investigations on meteoric stones being conducted by Howard in England. He compared the stones which had been sent to him with the ones described in the article: they appeared to be identical! He commented:

If this news does not convince me, at least it appears to me very remarkable that all the stones from diverse countries to which are attributed the same origin, present exactly the same characters; and I remain convinced that no matter how absurd an alleged fact appears from the viewpoint of physics, judgment must be suspended, and one must not rush into regarding the fact as impossible s5 H. F. B. de Saint-Amand, letter in Bibliothèque Brittanique, Vol. 20 (May 1802), 85-89, at 88..

This kind of summation effect appears at several crucial points in the meteorite controversy. The chemical analysis of meteorites which was undertaken by Howard and which went so far toward convincing the scientific community that the stones were real was stimulated by the influence of Sir Joseph Banks, at that time President of the Royal Society. Banks had noticed a stone on exhibition in London which was supposed to have fallen from the clouds near Wold Cottage, Yorkshire in 1795. Banks noted the resemblance between the stone on exhibition and other stones which had been sent to him as having fallen near Siena, Italy in 1794. This resemblance led him to collect accounts of other falls, and finally to encourage Howard to investigate the resemblances chemically s6Howard and de Bournon, op. cit. note 23, at 175..

Another instance of a summation effect resulted in a short but very important paper by the Abbe A. X. Stütz, at the time an assistant curator of the Kaiser's Natural History Cabinet in Vienna s7A. X. Stütz, 'über einige vorgeblich vom Himmel gefallene Steine', Bergbaukunde (Leipzig, 1790), 398-409.. Stütz had been given a sample from a meteorite fall at Eichstedt by his friend the Baron Von Hompesch. He found a description of a similar 'mineral' supposed to have fallen from the sky in a book by the mineralogist Von Born. These two cases reminded him of yet a third one which had taken place in Yugoslavia in 1751: several witnesses had sworn to seeing a heavy stone descend from a fireball. The bishop of Agram's Consistory had investigated the fall and had sent the stone with an affidavit to the Kaiser's Cabinet where Stütz had seen it. The Agram stone was almost pure iron; it in turn reminded Stütz of another lump of pure iron, a huge mass which had been found on top of a mountain by Peter Simon Pallas, a Russian savant and explorer. It, too, had been said to fall from the sky. It is clear that Stütz was not predisposed to accept the depositions of the witnesses to the Agram fall. But perhaps, Stütz wondered, in view of the common theme running through all these accounts, there might be something in the matter after all? n350. Ibid., 406-09. The importance of similarity is explicitly mentioned by Stütz, but only in passing; it is more implicit in the nature and composition of the paper. Stütz in fact makes more of an experiment of Komus in which iron was reduced by an electric spark, and caused a flaming mass to fall. Stütz wondered if a similar effect might not take place in the atmosphere on a larger scale, which would account for meteors.

In many ways a museum is a natural point of summation for anomalous objects n451. Dr James Ritchie, Keeper of the Royal Scottish Museum at Edinburgh, in a letter to Rupert Gould, stated: Had you the opportunity of seeing the specimens, which are sent here for identification, supposed to be the remains, not of sea serpents, but of as monstrous land animals, and which turn out to be very ordinary things after all, you would place very little faith in the casual observations, often in difficult conditions, of sea-men and others, even when the observations were supported with affidavits. Quoted from R. Gould, The Case for the Sea Serpent (London: Philip Allan, 1930), 248-49., and museums could have played the major role in the meteorite controversy. The sad fact is, however, that often they allowed their precious specimens of meteorites to become lost, or even threw them away s8Chladni, op. cit. note 12, at 5.. Fortunately, the falls were frequent enough and a sufficient number of specimens survived in museums or private collections to make comparison between them possible n5 Private collections of natural history specimens were very popular in the eighteenth century. See Y. Lassius, 'Les Cabinets d' Histoire Naturelle', in R. Taton (ed.), Enseignement et Diffusion des Sciences en France au XVIIIe Siècle (Paris: Hermann, 1964), 659-70..

A summation effect can be produced by passive acquisition of data - as in the case of the Abbe Stütz - but it can also take place as the result of active search. What was probably the most important intellectual event in the controversy was the outcome of such a search for data. Chladni, whom I mentioned earlier, was an independent researcher and travelled widely. While he was visiting the physicist G. C. Lichtenberg in Göttingen in 1792, the conversation turned to meteors. Chladni pressed Lichtenberg to tell him what physicists thought about this unexplained phenomenon. Lichtenberg indicated that while meteors had generally been explained as electrical, he and others felt that this theory was far from satisfactory. What were they then, Chladni wanted to know. Well, suggested Lichtenberg, they might be bodies from beyond the earth. This intrigued Chladni so much that he decided to find out as much as he could about them. He relates:

With this intention I stayed nearly three weeks longer in Gottingen, to collect as many reports of fireballs as I could put my hands on from the town library. It early presented itself as an historical truth, that often stone- and iron-masses fell down following the appearance of an exploding firewall... s954. Chladni, op. cit. note 12, 8.

One of the pieces of information which he uncovered in this manner was the memoir by Stütz, which made a considerable impression on him. When he published his conclusions in a small book, its contents and title were strongly influenced by material considered by Stütz, in particular the Pallas iron mass. When he wrote, Chladni had never seen a meteorite, but based his conclusions entirely on reports; a library can thus serve as a summation point for reports of anomalies. If the relevant reports can be extracted from the mass of materials in the library's holdings, the researcher may have a readymade set of data. This is what Chladni found.

Another point, however, must be recognized here. It was not only the reality of the meteorite reports which Chladni discovered from his library research; it was also the connection of meteors and meteorites. The cases fit together not only as a set of similar events, but as a set of pieces needed to solve a puzzle. Only when the different events were juxtaposed did their relation appear as an 'historical truth'. Others (like Stütz) had made the connection before, but only in a speculative, uncertain way. Chladni, convinced by the variety of data which all pointed to the same conclusion, made the connection in an unequivocal manner.