Katherine D. van Schaik
Recent literature in the field of ancient medicine has focused on patients’ experiences of health and disease (Holmes 2010, Thumiger and Petridou 2016), on the characterization of many types of healer in the “medical market place” (Israelowich 2015), on patients’ choices regarding health and disease (ibid.), and on the relationship between the healer and the patient (Mattern 2013). This paper offers a different approach to studies of the physician-patient relationship by asking two related questions: First, how did physicians decide on a particular diagnosis and treatment for a patient? Second, how did physicians teach their preferred means of clinical decision making to their protégés? The position of particular sects or physicians regarding clinical decision making relied largely on their preferred nosological categories and their means of addressing questions of observability. These two aspects of a medical sect also affected its approach to the training of future practitioners. In this paper we will address these questions with reference to Methodist and Galenic approaches to clinical decision making, though other schools of medicine in antiquity answered them differently.
How diseases should be classified was a significant source of disagreement, especially between Galen and his Methodist competitors. A cornerstone of Methodist teaching on nosological categories was the concept of κοινότητες, or ‘commonalities’. These were three broad states – laxity, constriction, and the combination of the two – that affected the entire body and were the overarching categories into which different πάθη, ‘affections’ could be grouped. The system was streamlined but still provided opportunity for the integration of multiple factors that allowed for differentiation of conditions. Van der Eijk (1999) has argued that a “criterion of relevance” affected Methodist approaches to the classification of disease, and, in the context of clinical decision making, I would further specify that a criterion of utility affects Methodist approaches to nosology. The rapidity with which a trainee could acquire and apply Methodist teaching was an important part of the sect’s appeal and, potentially, its success.
Galen’s approach to nosology on the other hand was rooted in his knowledge of anatomy (Nutton, 2013), acquired through his extensive training in the dissection of animals and his wide reading of earlier medical literature about human dissection and vivisection. Unlike the Methodists, whose point of view regarding τὰ ἄδηλα (the ‘hidden’ parts inside the body, and the hidden ‘causes’ of disease) was more characteristic of Skepticism (Frede 1987), Galen emphasized the importance of theorizing about what was beyond direct observation. He expected his students to learn as he had: through anatomical investigation and study of relevant medical and philosophical literature. Galen’s willingness to engage with the unobservable aspects of the human body contributed to a nosological system in which illnesses were linked with humors and with specific anatomical structures that were not visible from the outside of the body.
These two approaches to nosological categories and broader issues of observability affected how Methodists and Galen believed physicians should be trained. While both approaches valued ἐμπειρία, classification schemes and ideas about what could be observed affected the kinds of signs and symptoms that contributed to diagnostic and therapeutic decision making. These classification schemes relied on heuristics – the cognitive processes that humans use to streamline decision making – and on what has been termed “expert intuition” (Kahneman and Klein 2009).
Using modern literature describing clinical encounters (Groopman 2007) and studies in the fields of decision science and behavioral economics, I argue that Methodist and Galenic approaches to the classification and interpretation of medical information can be understood as efforts to deal with what we today call ‘cognitive bias.’ While both schools of thought provided different solutions to the problems that inevitably result from the application of heuristics to the decision-making process, they were both ultimately contending with an issue that remains a challenge to today’s clinician: the tension between the particular (an individual patient) and the general (data generated from large-scale studies).