![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
This paper discusses two methodological notions, the concepts Gegenprobe (countercheck) and Gegenversuch (counter-trial), which were widely applied, discussed, relied upon, and defended in German-language writings about empirical inquiry. In the decades around 1800, they were common in physiology; medicine; agriculture; chemistry; various technologies, such as printing, metallurgy, and mining; accounting; and legal and political argumentation. The ubiquity of those concepts signals a broad concern with securing empirical findings and empirical knowledge. Gegenproben and Gegenversuche – the terms as well as the practices – are evidence for the close connection among explicitly knowledge-seeking endeavors and crafts and technologies. Physiological experiments, chemical analyses, agricultural or vaccination trials were indebted to applied and practical procedures of making and assaying. What united these different fields was the fact that the problems that had to be solved were very complex yet at the same time of immediate relevance to everyday lives (like diseases or crops) or highly controversial (like vital forces or spontaneous generation). We know that late eighteenth-century and early nineteenth-century practices of inquiry were firmly embedded in applied, technical contexts; this essay shows that methodological notions were anchored in applied, technical contexts as well.
Acknowledgements
I discussed a much-abbreviated version of this paper with the participants in the virtual HPS Café Chez Paul at L.M.U. Munich and an early draft with the members of the Mellon-funded Sawyer Seminar “Rigor: Control, Analysis and Synthesis in Historical and Systematic Perspectives” at Indiana University Bloomington and thank both groups – especially Evan Arnet, Klodian Coko, Claudia Cristalli, Rebecca Jackson, William Newman, Kärin Nickelsen, and Friedrich Steinle – for their comments. I am grateful to Yves Gingras for his critical reading of the penultimate draft.
Notes
1 Of course, the presence of these notions in experimental reports has a rhetorical function, and assurances are no guarantee to the reader that these precautions were, in fact, taken. But this is true for the current scientific literature as well. Referring to controls, replications, and significance tests can bolster the argument that the text presents. We – peer, reviewer, historian – take the authors’ word for it, if sometimes with a grain of salt.
2 In the paper, I will use the German words rather than “countercheck” and “counter-trial”.
3 See, e.g. Klein, Humboldt’s Preußen; Popplow, “Ökonomische Aufklärung”; Bayerl, “Prolegomenon”.
4 Humboldt’s experiences as a mining official inspired his scientific pursuits, such as the experiments on the amalgamation of gold ores and on gases and mine air, as well as his respiration machine (see Klein, “Alexander von Humboldt”). Investigators in late eighteenth-century Germany were typically “artisan-savants”: generalists who pursued experiments, participated in the activities of learned societies and academies, worked as editors and lecturers at technical schools, while at the same time doing technical work in mining, metallurgy, agriculture, or manufacturing, solving practical problems, and serving as state bureaucrats (Klein, “Artisanal-Scientific Experts”).
5 For the broader intellectual context of Humboldt’s investigations, see Zammito, Gestation. On Galvanic experiments in around 1800, see Pera, Frog; Rothschuh, “Idee”.
6 Humboldt, Versuche, 226.
7 For an analysis of Humboldt’s arguments, see Rothschuh, “Humboldt”.
8 On Gordon’s trial, see McVaugh, “Experience-Based Medicine”. For Lind’s trial, see Tröhler, “Lind”. Lind’s trial, by the way, appears to be more indebted to Francis Bacon than to Gordon; see, e.g. Bacon’s experiments on stimulating the growth of wheat (Bacon, Sylva sylvarum, 110).
9 In nineteenth-century philosophy of science, this strategy is formally described. In John Stuart Mill’s System of Logic it becomes one of the four methods of inquiry.
10 Humboldt, Versuch, 342.
11 Ritter, “Elektrische Versuche”, 260–2.
12 Gärtner, “Beytrag”, 119.
13 Erman, “Schwimmblase”.
14 Meyer, “Auszug”.
15 Göppert, “Correspondenz”.
16 Gehlen, “Anmerkung des Uebersetzers”, 353.
17 Remer, Lehrbuch, 104.
18 Westrumb, Apothekerkunst, 177, 237. Notably, by the 1830s, the reproduction of a disease in a healthy animal was soon considered to be part of the demonstration of disease causation. Much later, the procedural steps became known as “Koch’s postulates”. (For details, see Schickore, “Parasites, Pepsin”; on Koch’s so-called postulates, see also Gradmann, “Scientific Rigour”).
19 See Berg, “Experimentieren”.
20 The early eighteenth-century encyclopedias also include a short paragraph on analysis, comprising both chemistry and anatomy.
21 Jablonski, Allgemeines Lexikon, 570.
22 Marperger, Lexicon, 1470–1, 663.
23 The section on Probe is much longer than the entry on Versuch, which is just a couple of pages long. Of course, this comparison is to be taken with a grain of salt, as the 242 volumes of Krünitz’s technical encyclopedia were published over several decades. Volume 217: Versicherungsgesellschaft – Vertheidigen came out in 1854. Still, it is noteworthy that there is no separate entry Experiment in the relevant volume, which appeared in 1785, and the end of the short entry Versuch refers the reader back to Probe.
24 Krünitz, Encyclopädie, vol. 117.
25 Hermann, Naturgeschichte, 159. He left open what needed to be done if the two results disagreed.
26 Jablonski, Lexikon, 570.
27 For a general discussion of the treatment of discrepant data from multiple measurements prior to 1800, see Buchwald, “Discrepant Measurement”. Buchwald’s paper focuses on numerical data in astronomy, in which the uniformity of the mesurand could be assumed. It would be interesting to compare the two domains of measuring and analyzing more systematically than I can do here.
28 See Zedler, Universal-Lexicon, vol 10; Krünitz, Encyclopädie, vol 16; Jacobsson, Wörterbuch.
29 Köhler, Kalender, 63.
30 Zedler, Universal-Lexicon, vol 10, 593; see also Krünitz, Encyclopädie, vol 16; Richter, Hüttenlexicon, 381.
31 “Eröfnung”, 706.
32 Gütle, Kupfer, 66; see also Krünitz, Encyclopädie, vol. 16.
33 Karl, Anleitung, 51.
34 Authenrieth, Anleitung, 146.
35 On late eighteenth-century cowpox and smallpox vaccination, see Sonntag, Pockenimpfung.
36 “Wien”, 157. In 1836, Gegenproben served as tests of the success of the immunization; almost like proofing yeast before baking bread (see, e.g. Schneider, “Bemerkungen”).
37 The translation appeared as part of Friedrich Gotthilf Friese’s 1804 translation of a French history of cowpox vaccination. Cowpox were quite rare and it was thus not easy to obtain the fluid with which to vaccinate people. Attempts to transport the vaccine from England had been unsuccessful. The fluid taken from a child – Anna – in Bassora (Basra) who had successfully been inoculated with (or had contracted?) cowpox served as the source of a vaccine which was then send to Bombay and multiplied.
38 Friese, Kuhpokkenimpfung, 30.
39 Pontet, “Verfahren”, 28.
40 Königliche Regierung des Unter-Mainkreises, “Instruction”, 1527–8.
41 Klaproth, Beiträge, 333–4.
42 For more information about the procedure, see Szabadváry, Analytical Chemistry, 122–3.
43 Becher, Carlsbad.
44 See Poirier, “Balance Sheet Method”. Lavoisier was not the first who employed the balance sheet method, but priority is not important for the present context.
45 Holmes, “Lavoisier”, 24.
46 Krünitz, Encyclopädie, vol 117. The lemma is “probieren”, not “Analysis”, so this was published only in volume 117 of Krünitz’s encyclopedia (1811).
47 Like Humboldt, Hermbstädt straddled science, industry, technology, cameralism, and pedagogy. (On Hermbstädt, see Mieck, “Hermbstädt”; Welsch, “Hermbstädt”.) The translation of Lavoisier’s work into German and various other publication projects were part of Hermbstädt’s endeavor to promote experimental agriculture, medicine, and chemistry.
48 Lavoisier, “Mémoire”, 351–7; see Hermbstädt, System, 206–14. Jan Frercks’s recent edition of Hermbstädt’s text does not contain the passages discussed here, as Frercks highlights Lavoisier’s theoretical achievements and thus includes only select passages from the third, practical part (Frercks, System).
49 The weighing procedure is repeated, reversing weights and retort, then averaged.
50 Lavoisier, by the way, used the expression “pour faire ma prevue” (Lavoisier, “Mémoire”, 359).
51 For details on Lampadius, see Walter, Lampadius.
52 Lampadius, “Versuche im Großen”, 378.
53 I discuss comparative agricultural trials in more detail in Schickore, “Comparative Trials”.
54 Lampadius, “Versuche”, 55.
55 See, e.g. Hartmann’s trials with foxtail millet (Hartmann, “Kolbenhirse”; see also Kreyßig, “Lebenskräfte”).
56 The presentation of Versuch and Gegenversuch was a rhetorical technique for presenting an extended argument, e.g. in the pros and cons regarding the legitimacy of dissolving regional abbeys, a legal-theological debate that continued until the early nineteenth century (Merz, Neueste Sammlung).
57 Fries, Neue Kritik, XXIII–XXIV.
58 I amended Meiklejohn’s translation of Kant’s Critique to bring it even closer to the German.
59 The second edition of the Critique begins with an epigraph taken from Bacon’s Instauratio Magna.
60 Kant, Kritik, BXIII.
61 Apparently, in Kant’s view, the sciences had reached a high enough degree of certainty and authority that made it plausible to rely on it in support of his philosophy. Late eighteenth-century practitioners, by contrast, often emphasized the probabilistic nature of empirical knowledge and the fact that only moral, but not absolute certainty could ever be attained.
62 Kant, Kritik, BXVI.
63 Kant, Kritik, BXIX, emphasis in the original.
64 Kant, Kritik, BXXI. I am grateful to William Newman for discussion of “reduction as synthesis”. On the long-term history of reduction, analysis, and synthesis, see Newman, Atoms & Alchemy.
65 Hacking, “Telepathy”, 427.
66 Arguably, the emergence of randomization displaced an earlier element of empirical experiments, namely the ceteris paribus condition: comparison to an untreated plot or person is only informative if everything, except the treatment, is equal across both trials. Once we use statistics and randomize, we no longer need to worry about the specific circumstances in individual trials. Prior to the late nineteenth century, however, harnessing the circumstances of an experimental intervention was a matter of significant concern, in agriculture, clinical medicine, and beyond. In fact, the ceteris paribus condition was even more fundamental than the control plot: any intervention to establish effects requires that background conditions be kept stable during the process, otherwise the intervention is not informative. I cannot pursue this here, however.
67 There are only a few systematic studies of the history of “blinding”: see Kaptchuk, “Ignorance”; Chalmers, “Comparing like with like”. The history of placebos intersects with both the history of “blinding” and the history of “untreated controls”, as a placebo in a trial blinds subjects to the fact that they are serving as controls. For a systematic discussion of placebos, see Holman, “Sugar Pills”.
68 The notion of touchstone originated in metallurgy, where an actual stone serves as a test for alloys of gold.
69 See, for instance, Robert Boyle’s essays on “unsuccessful experiments” (Boyle, “First Essay”; Boyle, “Second Essay”).
70 Why? There was no one single reason. Population growth and the ensuing needs to intensify and improve farming and food production, mining, and manufacturing, the expansion of academic training in medical and technical schools, and disputes about broader agendas (such as vital forces, electric fluids, or the existence atoms) all encouraged methodological discourse. The term “control”, by the way, was not a nineteenth-century coinage either; arguably, it was imported into science from eighteenth-century political and economic discourse, in which it was commonly used to refer to the management of people.
Additional information
Notes on contributors
Jutta Schickore
Jutta Schickore is Ruth N. Halls Professor of History and Philosophy of Science and Medicine at Indiana University. Her research interests include philosophical and scientific debates about scientific methods in the past and present, particularly about control, (non)replicability, and failure; historical and philosophical aspects of microscopy; and the relation between history and philosophy of science. Her latest book is About Method: Experimenters, Snake Venom, and the History of Writing Scientifically (University of Chicago Press, 2017).