CONCLUSION T O PA R T II: TRUST NO T THE PHYSICIAN For centuries many people had understood that the claims doctors made on behalf of their therapies were exaggerated: we have seen that the ﬁrst Hippocratics had to defend medicine against attack. In later centuries Christian critics would quote Mark 5: 25–7, on ‘a certain woman, who had an issue of blood twelve years. And had suﬀered many things of many physicians, and had spent all she had and was nothing bettered but rather grew worse.’ In early seventeenth- century England, plays are full of complaints about doctors. In Shakespeare’s Timon of Athens, Timon says ‘Trust not the physician; / His antidotes are poisons and he slays . . .’ In Thomas Dekker’s The Honest Whore we are told that it is far safer to ﬁght a duel than to consult a doctor. In Ben Jonson’s Volpone doctors are said to be more dangerous than the diseases they treat, for ‘they ﬂay a man / before they kill him’. Queen Elizabeth I would have agreed: she consistently refused to let the doctors treat her, even when dying. Half a century later the same complaints are to be heard in France, in Molière’s Le Médecin malgré lui (1667) and Le Malade imaginaire (1673). But the high point of anti-medical agitation was in England during the Interregnum, 1649–60. Nicholas Culpeper, publishing an unauthorized translation of the oﬃcial medical Pharmacopoeia in 1649, asked Would it not pity a man to see whole estates wasted in Physick (‘all a man hath spent upon physicians’), both body and soul consumed upon outlandish rubbish? . . . Is it handsome and well-beseeming a com- monwealth to see a doctor ride in state, in plush with a footcloth, and not a grain of wit [knowledge], but what was in print before he was born? Send for them into a Visited House [i.e. a house whose inhabitants have the plague], they will answer they dare not come. How many honest poor souls have been so cast away will be known when the Lord shall come to make Inquisition for Blood. Send for them to a poor man’s house, who is not able to give them their fee, then they will not come, and the poor creature for whom Christ died must forfeit his life for want of money. For all his complaints that doctors failed to treat the sick, Culpeper doubted the eﬃcacy of traditional medicine (he favoured a mixture of herbs, of chemical remedies, and of astrology). The goal of his publications was to destroy the medical profession’s monopoly of knowledge, so that if people wanted conventional therapies then they could employ them without recourse to doctors. For it was obvious to everyone that doctors had a ﬁnancial interest in claiming to be the only ones able to cure diseases, just as lawyers had a ﬁnancial interest in seeing disputes reach the courts. In 1651, Noah Biggs said it was impossible not to be aware of ‘the cruelties and unsuccessfulness of the medical profession’. George Starkey, in 1657, complained that doctors engaged in ‘bloody cruelty . . . tormenting the patient’. Some of the people who saw that medi- cine was not what it claimed to be were themselves doctors. Biggs was convinced that even doctors were aware of ‘their own unsuccessful- ness’. Take the example of Antonio Durazzini, who practised in the small town of Figline near Florence. From there he reported to the government of Florence in 1622 on an epidemic of a deadly fever. He had been treating those who could aﬀord it, treating them with bloodletting and other traditional remedies, but he noted in his oﬃcial report that ‘more of those who are able to seek medical advice and treatment die than of the poor’, who of course received no treatment. Here was a simple test of the eﬀectiveness of medicine, but Durazzini may not have been able (or willing) to recognize it as such. Perhaps he thought the poor were peculiarly robust. There is no such ambiguity about the views of Latanzio Magiotti, a contemporary of Durazzini’s and doctor to the court of the grand duke of Florence. Count Lorenzo Magalotti tells us that our dear friend Magiotti said quite openly [that doctors and medicine were useless] and when Grand Duke Ferdinand asked how in all honesty he could accept money from patients knowing he could not cure them he replied: ‘Most Serene Highness, I take the money not for my services as a doctor but as a guard, to prevent some young man who believes everything he reads in books from coming along and stuﬃng something down the patients which kills them.’ Similar examples could be found in every age and society, at least where the records are good enough to record indiscreet conversa- tions and private doubts. ‘Therapeutic nihilism’, the belief that most conventional medical therapies did not work, became the norm amongst sophisticated (particularly Parisian) doctors in the 1840s as a result of a new interest in statistics. In 1860, Oliver Wendell Holmes, who had received a Parisian medical education, expressed a good deal of sympathy for the view that ‘on the whole, more harm than good is done by medication’. But a scattering of therapeutic nihilists is to be found in every age and place. As long as therapies were not subject to statistical tests, however, such sceptics had limited inﬂuence. There had always been important disputes amongst doctors themselves about which therapies were most eﬀective. For example, there were disagreements about which diseases were best treated with bleeding, how much blood should be taken, where it should be taken from, and how it should be taken (by lancet, cup, or leech). These disagreements were recurrent and unresolvable within conventional medicine, and sometimes they amounted to an attack on signiﬁcant aspects of that medicine. We have already seen that Girolamo Fracastoro in 1546 insisted that many diseases were contagious. Because he no longer saw such diseases as the result of unbalanced humours, he naturally also went on to question the utility of blood- letting as a therapy when dealing with such diseases. Over time, his arguments won ground: by the 1630s a signiﬁcant minority amongst Italian doctors thought that letting the blood of plague victims did more harm than good, and they seem to have been a clear majority by the end of the century. The Dutch physician Ysbrand van Diemerbroeck, whose work was translated into English in 1666, expressed similar views that were widely inﬂuential. In 1696 a work by Dominicus La Scala entitled Phlebotomia Damnata, or bloodletting condemned, provoked a rapid response entitled Phlebotomia Liberata, or bloodletting set free.
25. This lithograph by Honoré Daumier, which appeared in 1833 shows a doctor (sitting under a bust of Hippocrates) asking himself ‘Why the devil do all my patients go oﬀ like this [i.e. in coﬃns] . . . I do my best by bleeding them, purging them, drugging them . . . I just don’t understand it!’ It marks the moment when doctors began to recognize that con- ventional Hippocratic remedies were ineﬀective. Fracastoro and Diemerbroeck did not deny that there was an appropriate use of traditional Hippocratic remedies, but sustained attacks on the whole Hippocratic tradition of treatment by contraries came from Theophratus Paracelsus (d. 1541) and Jan Baptiste van Helmont (d. 1644). They held that diseases never resulted from imbalances of humours. Rather each disease was a speciﬁc condition, it had an ‘essential thingliness’, it was a ‘lively, active thing’, and must therefore be treated with speciﬁc remedies. God had so designed the world that there were remedies for every disease if one only knew where to look for them. Their arguments were taken up by the religious radicals of mid-seventeenth-century England who, as a consequence, did not hesitate to question the eﬃcacy of traditional therapies. Why were these disputes never resolved by practical trials? Doctors relied instead on what we would now call anecdotal evidence: individual case histories of successful treatment. They continued to do so long after the unquestionable authority of the ancients had been destroyed, ﬁrst by Vesalius and then by Harvey. Real tests only began in the 1820s. Once they began, medicine was inevitably thrown into a prolonged crisis, a crisis that peaked in the 1850s when the core therapy of bloodletting was shown to be ineﬀec- tual. That crisis was only resolved by the triumph of germ theory in the years after 1865. But the triumph of germ theory was itself only the result of a major shift towards microscopy (which reassumes prominence in the 1830s) and towards laboratory research, a shift that had taken place at least thirty years earlier. Historians often puzzle over why that shift occurred, and why there was extensive investment in laboratories when at ﬁrst the results were mediocre. At least part of the answer lies in the urgent need, ﬁrst apparent in the 1820s, to ﬁnd an eﬀective form of medical therapy as traditional therapies began to be exposed as worthless. The laboratory did not bring about the demise of conventional medicine; rather the demise of conventional medicine led to investment in the laboratory. Germ theory did not supplant Hippocratic medicine and its therap- ies; rather the demise of Hippocratic medicine was a precondition for the triumph of germ theory. Medicine, until the 1820s, was rather like a religion, in that its claims were not subject to practical testing. For some 2,350 years doctors relied on a series of therapies –– bleeding, vomiting, pur- ging –– which did not work and actually did harm. These therapies, of course, looked as if they were doing good. They mobilized the placebo eﬀect, and moreover, in the case of bloodletting they resulted in what appeared to be immediate (if not lasting) beneﬁt: a slower pulse, a reduction in temperature and inﬂammation, a sound sleep. Faith in these apparent beneﬁts, however, would not have survived any comparative trial. The real puzzle with regard to the history of medicine before germ theory, as with the history of astrology, is working out why medicine once passed for knowledge. The case of medicine is, at ﬁrst sight, rather more intractable than that of astrology, for it is hard to disprove astrology: one would need to compare the lives of a group of people all born at the same moment. In the absence of such a test, it is easy for astrologers to claim their arguments ﬁt the facts. But medi- cine, it would seem, is quite diﬀerent, for it is obvious how to set about testing the eﬃcacy of a medical therapy. All that is needed is to take a group of patients with similar symptoms and treat some of them and not others. Moreover, it would seem, it is hard to tell when an astrologer is right and when wrong; but in the case of medicine there is a convenient crude measure of success to hand: the ratio of those patients who are still above ground to those who are now below ground. If it is this easy to put medicine to the test, why did traditional medicine survive untested into the nineteenth century? There are three good reasons for medicine’s unique status as a mythical technology. First, doctors were trying to achieve outcomes that the body’s natural healing processes were also working to achieve. It was easy to assume that if a patient recovered, then the therapy they had received was responsible, though in 1657 George Starkey had claimed that only a third of diseases were cured, but less than a tenth were cured by the doctors. He was trying to expose the illusory character of medical success. The ﬁrst study of the body’s capacity to heal itself, however, did not appear until 1835: the Ameri- can Jacob Bigelow’s Discourse on Self-Limited Diseases. By 1860, when Oliver Wendell Holmes set out to analyse the profession’s ‘tendency to self-delusion’, the fallacy of much traditional medical reasoning was apparent to almost everyone. Second, the placebo eﬀect meant that an ineﬀectual intervention could often result in a cure. This reinforced the illusion of success, and for a very long time the work- ing of the placebo eﬀect was entirely invisible both to doctors and to their critics. Its discovery in 1800 marks the moment at which it ﬁrst became possible to begin to assess the true eﬀectiveness of medical intervention. It is this discovery, which we will explore in the next chapter, not the new anatomy, which marks the beginning of the end for Hippocratic medicine. Third, in order to test a therapy you need to have a concept of a disease as being not a disorderly condition of a particular patient, but a typical condition of many patients, for only then can you be conﬁdent that you are comparing like with like. In order to think of diseases in these terms you needed to have either a concept of contagion, or a concept of epidemics such as was developed by Sydenham. These three obstacles made it diﬃcult to recognize that medicine did not work. In addition, there was (or appeared to be) a funda- mental ethical obstacle to the carrying out of the most simple of comparative tests. Doctors were supposed to do the best for their patients. They had no right to withhold treatment, if a treatment was available, and no right to try out an untested remedy when a reput- able therapy existed. But this ethical dilemma existed in appearance, not reality. Early modern doctors only treated those who could aﬀord to pay for them. Vast numbers of people went without treatment, so that there would never have been any shortage of people to use as a control group had anyone wanted to compare the eﬀects of treatment with no treatment. Much more constraining was the tacit obligation to give the minority who could aﬀord to pay orthodox treatment. ‘I will abstain from harming or wronging any man’, says the Hippocratic oath. From the beginning this must have been understood as meaning ‘I will do what other supposedly competent doctors would do in these circumstances.’ Underlying the apparent ethical obstacle was a pres- sure to conform that made it impossible to test new therapies against old ones. In 1663 Robert Boyle told a story that became popular of a doctor who had refused to try alternative remedies, saying of his patient ‘briskly . . . “Let him die if he will, so [long as] he die Secundum Artem”’, i.e. while receiving orthodox treatment. Such an attitude was commonplace. In 1818, Alexander MacLean, a Scottish doctor working in India, was busy recommending swallow- ing mercury as a therapy for almost all diseases. His opponents pointed out that too many of his patients died; he replied that this was because he treated only the most desperate cases. He proposed a randomized trial to compare his treatment with bloodletting. His opponents refused, saying one should not experiment with the lives of men, ‘as if’, said MacLean, ‘the practice of medicine, in its conjectural state, were anything else than a continued series of experiments upon the lives of our fellow creatures’. MacLean was quite right about that, if not about mercury. The eﬃcacy of blood- letting was completely untested; nevertheless, it was universally employed. For as the Helmontian Noah Biggs had complained in 1651, the theories of the Hippocratic doctors were quite irrelevant, ‘nothing but triﬂes and anxious disputes’. What really mattered were their basic remedies, especially purging and bloodletting. These were ‘the slender hinges’ upon which ‘the whole huge bulk of the art of healing seems nowadays to be moved’. These reasons –– the illusion of success, the placebo eﬀect, the tendency to think about patients not diseases, the pressure to conform em;go some way to explaining why medical therapies continued to be employed that were at best ineﬀectual, and more often than not posi- tively harmful. But do they go far enough? The missing element, it might be thought, is formal regulation. Once medicine became, in the thirteenth century, a profession licensed by the universities of Europe, and by governments who took for granted the value of medical degrees, Hippocratic doctors were not competing on equal terms with the various unoﬃcial suppliers of therapy, and the eﬃcacy of Hippocratic medicine was (except for brief moments, such as the English Civil War and the French Revolution) already oﬃcially pre- determined and beyond question. Oﬃcial medicine, it might be claimed, went unquestioned because doctors were legally authorized in roughly the same way that theologians were. The validity of ortho- dox medicine was established by the decrees of university, state, and church, and was thus not open to question. The test of truth within the scholastic intellectual world of the universities was intellectual coherence, not practical eﬀectiveness, and as long as medicine met that test it was subject to no other. Medicine was (in the world of the universities, if not always outside) a monopoly, and because it was a monopoly there was no need for it to prove its superiority by comparing alternative therapies. An interesting test case is provided by Gianna Pomata’s Contracting a Cure (1998). Pomata’s book is a study of a panel of doctors, the protomedicato, in early modern Bologna that adjudicated on com- plaints by patients that they had not received good value from those they had paid for treatment. She set out, she tells us, intending to write a history of how popular healers had been marginalized and outlawed by the medical profession; but she was so captivated by the voices of the patients in the judicial records that she turned aside to write a history of the sick persons’ experience of their illness and treatment, and in particular of the working of a system where patients were entitled (or at least believed they were entitled) to withhold payment or reclaim their money if they had been treated without being cured. Pomata constantly stresses that doctors and legal authorities contested the right of patients to refunds where there was no cure, but she argues that at the beginning of the seventeenth century this right was recognized by the court of the protomedicato if there was a prior agreement that payment should be by results, while by the eighteenth century it was consistently being rejected. You have to read her book with some care to discover that she does not identify a single example of a patient successfully enforcing an agreement for a cure against a licensed and qualiﬁed doctor. We are told that at the end of the sixteenth century ‘the protomedici endorsed the terms set by the agreements for a cure, including the principle of payment for results’, but this turns out to be true only in the case of claims against barber- surgeons; for qualiﬁed doctors it was already the case that ‘for the patients, a therapeutic transaction was fair if the healer respected the terms of the cure agreement; for the protomedici, it was fair if the practitioner medicated according to the oﬃcial rules’. As far as doc- tors were concerned the agreement for a cure had already ceased to exist when the protomedici were established in 1581. Doctors were prepared to reduce the fees claimed by other doctors in practising orthodox medicine when those fees were exorbitant or when patients were impoverished, but they were not prepared to rule that the failure of the therapy meant that they were not entitled to pay- ment. Not only were the sentences of the tribunal in malpractice cases ‘always favourable to the practitioners’, but in disputes over whether doctors were entitled to payment when they had failed to cure the tribunal consistently ruled in favour of the doctors. In Bologna doctors did not have a monopoly: they practised alongside, and in competition with, other licensed healers, including barber-surgeons and apothecaries. But they practised on terms that were biased in their favour, for it was they who sat on the tribunal which decided if healers were entitled to payment and if there had been malpractice. Elsewhere, though, doctors did not have even this degree of control over the marketplace for therapy. In England, eﬀect- ive regulation of the medical profession broke down in the course of the seventeenth century. In eighteenth-century America there was something close to a free market in medical training, diﬀerent therap- ies competed against each other without hindrance, and regulation only became the norm late in the nineteenth century. If we go back to ancient Greece and Rome, then competing schools of medicine had confronted each other on equal terms. Hippocratic medicine did not depend on regulation to establish itself, and it did not collapse when regulation collapsed. So it is one thing to harbour doubts, and quite another to put those doubts to a test. It is one thing for the eﬃcacy of traditional medicine to be questioned, another thing for its eﬃcacy to be tested. We need to add a further factor to the illusion of success, the placebo eﬀect, the tendency to think about patients not diseases, and the pressure to conform, and that factor is the absence of statistical think- ing. Statistical thinking does not come naturally. You can tell if a building is properly designed by seeing whether it stands or falls. You can see if a clock tells time by comparing it with a sundial. But making a comparison between two competing therapies is a quite diﬀerent enterprise. It requires a statistical comparison of two sample populations. Sophisticated statistical thinking was born along with probability thinking in the 1660s. It was soon being used to predict life expectan- cies using the London ‘bills of mortality’ (ﬁrst analysed by John Graunt in 1662) which showed the ages of all those who died in London. But for a hundred years life insurance continued to be sold, like a lottery ticket, at the same price for everyone, which meant that the young paid far too much for it and the old paid far too little. Statistical thinking was slow to develop and there was considerable resistance to it. Until that resistance had been overcome, there could be no trial of traditional medical therapies. As we shall see, that resist- ance has never been entirely overcome, and arguments from statistics are given much less prominence in histories of medicine than they ought to be. It is the combination of these ﬁve factors –– the illusion of success, the placebo eﬀect, the tendency to think of patients not dis- eases, the pressure to conform, the resistance to statistics –– which explains, if anything can, the intolerable delay in testing the eﬃcacy of orthodox medicine. At this point we need to ask what sort of historical explanation we hope to ﬁnd. Do we want to prove that there was never any possibility of testing orthodox medicine before the 1820s? We would be wrong. Do we want to prove that the obstacles to testing orthodox medicine were great, but not insuperable, before the 1820s? Then we would be right, but an argument of this sort cannot absolve all those doctors who practised an ineﬀectual form of therapy from some responsibil- ity. If we provide too strong an explanation of why traditional medicine was not put to the test until the nineteenth century, then we will inevitably lose sight of the fact that plenty of people could see that it didn’t work. It is diﬃcult to strike the right balance here, but the resilience of orthodox medicine is far more signiﬁcant than the persistent criticism it encountered. The stress needs to fall on medicine’s appearance of success, and on the ease with which it saw oﬀ competition from Paracelsians and Helmontians.