Repetition, Variation, and Control Historical Perspectives (Imitation)

“Imitation”     Is viper meat an antidote to viper venom? Charas showed that an animal bitten in the ear by a viper had been cured by eating the head of another viper. Redi repeated the same procedure and obtained the same (!) results. “A doubt arose” in Redi. He repeated the experiment without feeding the dogs viper meat. Both dogs survived, having been bitten in the ear. “Imitation” Redi proceeded to feed viper heads to a cockerel, a capon, two dogs, eight more cockerels, two kittens, two hares, and six doves. All animals had been bitten in the thigh or the tongue, not in the ear. All died.  Redi stressed that he had done more than Charas: In some trials, the place of the bite had also been smeared with viper blood as an additional antidote.  “Imitation” Charas, in turn, pointed out that Redi’s first results had agreed with the evidence obtained by Charas.  The other experiments were not proper replications. Charas had used medium-sized dogs and Redi large ones, on which “no certain judgment could be made”.  Charas presented another kind of evidence, a case history, in his favor. A young man was bitten by a viper. The young man had so much faith in the truth contained in Charas’s books that he immediately decided to eat the head and neck of the perpetrator. Charas served the broiled head, neck, heart, and liver of the viper.  Again, he did more, so that the results could be brought out more clearly. He made the young man swallow some volatile salt of vipers, too.  “Imitation” Charas’s group imitated Redi’s experiments with the yellow liquor – and failed. Like Redi, Charas gave yellow fluid to pigeons and two kittens. Again, he did more. He made a point of “engaging a person whom I knew most favourable to Signor Redi, to make himself the incision, and intromit the yellow liquor as he should think best.  Charas noted that this gentleman cut the wounds so deep that he, “did more than Signor Redi himself said in his Letters he had done.” Yet none of the animals died, even though Charas had proceeded by “thrusting the teeth into them as deep as I could.”  “Imitation”    Is explicitly discussed when replication attempts are unsuccessful The researcher who does the unsuccessful replication typically claims that his experiment yields the correct result; the other researcher’s experiment has a flawed design or some unknown factor interfered. To make their point more forcefully, the critics stress that they are not doing the exact same thing but they are “doing more” (using more venom, making a bigger wound, etc.) Fontana on variations  The “yellow liquor” from the viper’s teeth affects the blood  Volatile alkali, a commonly used antidote, is not a remedy for viper bites “Principal methods of good experimentation” “The first is, to multiply the experiments exceedingly. ..  “The second is, to vary them in a thousand ways, changing the circumstances as the nature and species of them may require, and giving them all the precision and simplicity they are capable of...  “The third method is […] to attain to a discovery of the source of the errours that others have fallen into” (Fontana, 1787).  Fontana  “I have made more than 6000 experiments; I have had more than 4000 animals bit; I have employed upwards of 3000 vipers and may have been deceived; some essential circumstance may have escaped me: I may have neglected some other, not thinking it necessary; my consequences may have been too general, my experiments too few in number. In a word, I may very easily have been mistaken…” (Treatise on the Venom of the Viper, 1781) “Whether the Volatile Alkali is a certain Remedy against the Bite of the Viper”  “I deemed it necessary to examine this first question in the most circumstantial way, and therefore multiplied the experiments extremely, and diversified them very much. This is the only method that could lead to demonstration, and I flatter myself that my readers will be freed from all doubt” (Fontana 1787) Experiments on the effects of volatile alkali “I had a pigeon bit in the leg by a viper, and instantly treated the part. At the end of a minute it fell forward, and could no longer support itself. In twenty seconds more it died. I had another pigeon like the first bit in the same way, but did not treat it. At the end of two minutes it fell forward, and in two minutes more it died.  I had two other pigeons bit in the leg; one was treated, and the other not. The first fell at the end of three minutes, and died at the end of the twentieth. The other fell at the end of a single minute, and died likewise after the twentieth.  Of two other pigeons bit in the leg, I treated only one. The one treated died at the end of forty hours, the other at the end of an hour….”  American herpetologist Laurence Klauber, 1962: Fontana is “the first of the great experimentalists to use adequate controls” Experimentalism, repetitions, variations, and circumstances The investigation of the circumstances attending the experiment was part of an extended effort to give the experimental design the precision and simplicity that the principal methods for successful experimentation required.  Experimenters must multiply their experiments extremely and diversify them very much. Through these variations, the circumstances attending the experiment could be identified and managed.  “Discovering the sources of error that others have fallen into”  Richard Mead had observed sharp crystals in viper venom. Fontana could find no such thing, even though he tried carefully.  Mead was “not well acquainted from habit with the shape of the different salts that are found in liquors” (in other words, he did not have the observational skills)  Fontana varied the instrument, using an English lens as well as a solar microscope.  Fontana made additional experiments to make himself “still more certain” and “to remove all doubt and suspicion on a matter so important and so generally adopted” – observing how a drop of venom dried and cracked. Redi, Charas, & Fontana  Described comparative  Described trials (comparisons with comparative trials “controls”).  (comparisons with  “controls”).  Assumed that reliable conclusions can be  Emphasized drawn from successful  repetitions. comparative trials  Main concern: the  practical challenges of  Emphasized that “many” experiments dealing with accidents must be done. and contingencies.  Assumed that reliable conclusions can be drawn from successful comparative trials  Accidents cannot be treated in a systematic way. Charas Described comparative trials (comparisons with “controls”). Assumed that reliable conclusions can be drawn from successful comparative trials Emphasized variations and circumstances. Optimistic that through multiple variations of experimental settings, the circumstances impinging on experiments could be identified. Fontana “Imitation” in the 17th & 18th c.    Not much explicit discussion Not many instances If it was done, there were different approaches  Re-doing the same procedure. If unsuccessful, the original experiment is deemed flawed or it was the wrong experiment for the purpose  Doing “more” to amplify the process ○ Using more venom ○ Cutting deeper wounds  Providing other kinds of evidence (“replicating an effect”)  For scientists, it becomes increasingly important to be able to give and receive criticism and to articulate standards of good practice. To be effective peer critics and collaborators in interdisciplinary contexts, scientists need to be evaluate other people’s works according to these standards. To do so, they need to have the necessary conceptual tools to express, explain, critique, and justify their methodological standards and their values to different audiences.  Moreover, the number of critical articles about scientific methods in the popular media is growing (including the social media), which means that scientists are increasingly challenged to articulate and defend standards of good practice vis-à-vis broader publics.  Historical and philosophical study can contribute to this task by analyzing scientists’ methods discourse— how working scientists themselves understand, conceptualize, apply, and communicate norms and standards for good research practice. The puzzle of control Comparative experimental trials performed by Fontana, Redi, and others - instances of control experiments avant la lettre?  John Start Mill’s method of difference from the System of Logic as the first philosophical conceptualization of controlled experiments?  The very term “control” came to be used only in late 19th-century experimental science. 