‘The lies of the quacks collapse. There were neither miracles, nor healings. It is time for the authorities to intervene.’Footnote 1 Thus read a headline in a Greek national newspaper in 1930. The article below it referred to an untrained healer, Monk Gymnasios, who had attracted the attention not only of thousands of people who hoped to be cured but also, repeatedly, of the authorities. He dominated the news in 1930–31, when, time after time, he was arrested, imprisoned, released and removed to a different part of the country, where, invariably, he was awaited by thousands of people who wanted to consult him and be healed by him.Footnote 2 While he owed his reputation, at least initially, exclusively to word of mouth, newspaper reports such as that cited above inevitably contributed to it. Nevertheless, Gymnasios, whose therapies were a mix of herbal recipes and religious incantations, emphasised that ‘I do not pretend to be a physician. I do not steal from the people. I do not take money.’Footnote 3 He never remained imprisoned for long, partly because he received no remuneration from his patients and partly because of his enormous popularity. However, in 1932, he stopped practising because of the continual persecution he suffered and because he failed in his efforts to obtain a licence;Footnote 4 like all irregulars, he lacked training and did not possess the necessary licence to practise.
Although the distinction between those who could and could not practise medicine legally in Greece was codified as early as 1834, unlicensed practitioners continued to practise well into the twentieth century; while Gymnasios was the most famous unlicensed practitioner of inter-war Greece, he was only one of many practising at the time. But was he a typical representative or, rather, was he exceptional? Who were these unlicensed practitioners and why did they practise illegally, if not for profit? How did their patients know about them and why did they choose them – if, indeed, they did – instead of physicians? What were the mechanisms that enabled their survival beyond the first half of the twentieth century? In contrast with untrained practitioners in the West, we know very little about those in Greece beyond their certain existence. This paper will therefore explore the nature of irregular practice in Greece in the first half of the twentieth century, offering an alternative to what we have come to expect from Western accounts of quackery; but one that, nevertheless, shares many common attributes with Western accounts of cunning-folk, wise women and popular healers.
A significant range of history of medicine studies has focused on the plurality of medical and healing practices from the eighteenth century to the present.Footnote 5 Such studies reveal the co-existence of the healing practised by learned physicians with alternative medical practices such as homeopathy and hydropathy, among others. But, as Robert Jütte points out, the professionalisation of medicine during the nineteenth century critically influenced the hardening of conceptual boundaries between medicine and its alternatives.Footnote 6 However, Roger Cooter has argued that, even at the turn of the twentieth century, the division between quacks and learned medics was not always clear,Footnote 7 while, more recently, Takahiro Ueyama has noted that, beyond the official (and dominant) narrative of the medical profession, medical practice, in fact, experienced significant commercialisation in late nineteenth-century London, so blurring rather than better defining the boundary between the quack and the physician.Footnote 8 And, just as in Britain, in the USA the professionalisation of medicine neither eliminated quackery nor made it easier to identify;Footnote 9 indeed, the consensus among scholars is that, in the USA, quackery, or plurality, was never eliminated and, while better controlled in the twentieth century, is ‘still as popular as ever’.Footnote 10
Studies in this area thus demonstrate the existence of non-learned medical practitioners alongside their learned counterparts, at least from the eighteenth century to the present day. Such practitioners, however, differed enormously in terms of their methods and their relationships with the learned physicians, the state and the patients from place to place and over time. Nevertheless, there are some common elements to be found in virtually all Western histories of non-learned medical practitioners: first, that profit was the main driver for their practice; and, second, that from the eighteenth to the twentieth century the major means of communicating with their ‘patients’ was advertising, in which they constantly adapted their therapeutic promises in response to legislative and other reforms.Footnote 11 Already, in the 1970s, Matthew Ramsey had distinguished between the quacks – whose practices included the elements outlined above – and the popular healers, who were ‘part of the local village culture’.Footnote 12 Yet, only occasionally have the grassroots health care practices and the reactions and adaptations to change of the latter been studied. Francesca Moore, for instance, has investigated such issues in a working-class Lancashire community at the turn of the twentieth century,Footnote 13 presenting evidence of the ‘survival of a localised pattern of alternative care’ that was embedded in the local working-class culture and which, she argues, was partly responsible for the slow acceptance of orthodox medicine by that community.Footnote 14 Prior to Moore, Owen Davies pointed to the lack of attention shown by historians to the cunning-folk and the role they played as medical providers in nineteenth-century England, identifying as reasons for their enduring popularity not only the fact that orthodox medicine ‘continued to be ineffective into the twentieth century’ but also that witchcraft continued to be seen ‘as a source of illnesses’.Footnote 15
The perspective offered in the works of Davies and Moore is different from that of much other work on alternative medicine partly because of the sources these historians employed: namely, oral history and ethnographic accounts. These are also the main sources utilised in this paper. Oral histories constitute a significant and valuable source that throws light on how unlicensed practitioners – and physicians – were viewed and employed by ordinary people – primarily in terms of determining why, how and when they consulted unlicensed practitioners – and can assist in elucidating physicians’ views on unlicensed practitioners. Such information cannot be obtained from the writings of physicians in medical journals and thus these histories offer a significant corrective to the official and the physicians’ point of view. The interviews utilised in this paper were conducted by the author over a number of years: twenty-eight interviews on Mykonos in 1994, twenty-two on Chios in 2000, twenty-two on Syros in 2001 and nine in 2011 in Athens and on Chios.Footnote 16 The interviewees ranged in age from 71 to 101 and therefore their recorded experiences span as far back as the early decades of the twentieth century. The research aims of which the interviews formed a part varied over time and, therefore, the information provided by interviewees on healers also varied in scope and detail. Significantly, however, the last set of interviews focused primarily on physicians and their experiences. In none of the interviews was an interviewee asked directly about unlicensed practitioners or physicians; rather, such information was volunteered by interviewees without prompting.
A further source significant for examining the presence and practice of unlicensed practitioners is folklore accounts. Because popular medicine was seen not only as a significant element of Greek rural life and tradition but also as a remnant of ancient Greek medical practices, information on it was intensively collected, especially in the twentieth century. While the accounts of the first half of the century come primarily from professional folklorists, later accounts were collected by a variety of individuals,Footnote 17 and the majority of the Greek publications cited here make use of folklore material and oral accounts collected by the publications’ authors. A significant example of such work is the folklore project undertaken by undergraduate students of Athens University, who, since 1964, have collected accounts from their own places of origin and submitted their manuscripts as part of their course assessment.Footnote 18 Such accounts are of particular significance because the illiteracy of the majority of unlicensed practitioners and the illegality of their practice meant that very little written evidence has been left behind by them or by the state.Footnote 19 In his discussion of this source material, David Gentilcore has questioned both the reluctance of historians of medicine to utilise such ethnographic accounts and, contrastingly, the ways in which such material has been used.Footnote 20 In the former case, the majority of the reasons offered by historians to justify their reluctance to engage with these sources do not, in fact, apply to early twentieth-century Greece. For example, an objection often raised is the ease or otherwise of distinguishing between popular and professional medicine;Footnote 21 but in Greece their differences were very apparent by the twentieth century, as they were also in late nineteenth-century Italy. Ramsey has pointed to the bias that can affect such collections when the investigator has looked specifically for ‘folkloric’ activities,Footnote 22 a problem overcome here by the juxtaposition of folkloric evidence against other sources, such as oral histories and anthropological works. In what follows questions regarding the identities of the unlicensed practitioners, their reasons for undertaking such work and the factors that operated in their survival beyond the mid-twentieth century in Greece are addressed in advance of a discussion of how these findings compare with what we know from similar studies of unlicensed practitioners and healers in the West.
1 Naming the Untrained Healers
As elsewhere, a variety of names was used in Greece to describe untrained practitioners.Footnote 23 These terms, which might depend on the type of services provided, included: empeirikos or praktikos [empirical, someone who learns his/her art empirically]; giatrina or giatrissa [female ‘doctor’]; magos [magician]; agyrtes; and kompogiannites.Footnote 24 The last two terms refer to wandering, rather than sedentary, healers and correspond to Ramsey’s itinerants.Footnote 25 Comparatively little is known about them and their practice, which seems to have been curtailed by the end of the nineteenth century and, consequently, they will not be examined here.Footnote 26 While the meanings of some of these terms have changed over time, they have also varied according to their user: so, for example, while the term empeirikos had a positive connotation for most of the rural lay population, among physicians it was used pejoratively.Footnote 27 Because of the impossibility of determining exactly what each of the descriptive terms found in the existing sources encompasses in terms of practice, all sedentary healers, upon which this paper will focus, will be referred to here, interchangeably, as either empirics or ‘popular healers’ (although this term is not one used in Greek, these practitioners’ work was referred to in Greece as ‘Popular Medicine’ [Laike Iatrike]).Footnote 28 These two names are used only for reasons of simplicity and their use should not be taken to represent the existence of a uniform practice among popular healers in Greece, since the opposite was very much the case.Footnote 29
Despite the ambiguity in terminology noted above, untrained sedentary (that is, resident) practitioners tended to specialise in specific illnesses or conditions and, as a result, it was usual for a given geographical area to boast a number of practitioners each dealing with different ailments.Footnote 30 These might include: bone-setters;Footnote 31 giatrines, who would deal primarily with children’s and women’s ailments;Footnote 32 healers dealing specifically with jaundice or with erysipelas; and others specialising in the removal of the ‘evil eye’.Footnote 33 Most sedentary healers apparently combined religious/magical elements with practical healing – as discussed below – using each to a different degree.Footnote 34 Even on occasions where no religious or magical elements were directly employed, however, the presence of religion was emphasised.Footnote 35 In terms of practical measures, empirical practitioners relied primarily on herbal remedies,Footnote 36 and there is no evidence of alternative medical practices such as homeopathy, mesmerism and Thomsonian medicine acquiring any significance in Greece until the 1950s; therefore, this paper does not engage with such practices. Hydropathy, in contrast, existed from the creation of the Greek state, but as it was, from the start, associated with the state and trained physicians it is not considered in this paper.
Sedentary healers were both women and men, and included priests or monks:Footnote 37 since healing was embedded in religious practice, there was no contradiction between the two. Aspects of spiritual healing included charms, incantations [xorkia], blessings [euhes], magical spells and prayers,Footnote 38 while many empirics reputedly owned an iatrosophic text [iatrosophia/giatrosophia], which raised their prestige.Footnote 39 These hand-written healing manuscripts were written in vernacular Greek and listed diseases and their therapies, including incantations.Footnote 40 Christos Papadopoulos has emphasised that such texts expressed ‘a vernacular philosophy of healing’ that did not differentiate between the pagan and the Christian; or, rather, I would argue, they brought together the pagan, the Christian and the magical beliefs of the society they represented.Footnote 41 Both Papadopoulos and Patricia A. Clark, as well as a majority of Greek authors dealing with such healing manuscripts, stress the links between these manuscripts and Ancient Greek medicine.Footnote 42 Moreover, it is apposite to note that Greek populations were under Ottoman rule for centuries, in consequence of which it would seem highly likely that the popular healing tradition of the Greeks – as it survived to the twentieth century – and that of Islam were strong influences upon each other.
2 Acquiring the Art of the Empiric
Almost invariably, the skills required of an empiric were transferred through generations within families, and thus were ‘inherited’ or learned from the parent or sometimes the grandparent.Footnote 43 Therefore, the option of becoming an empiric was not, generally, open to all.Footnote 44 In some exceptional cases, however, a few individuals learned the vocation from a non-related empiric to whom they were apprenticed.Footnote 45 The strong emphasis in the oral and written accounts on the ‘inherited’ nature of the art suggests that empirics who mentored non-relatives did so either because they were childless or because they had no children wishing to learn.Footnote 46 In addition to the instruction that empirics received, it is evident that most also tried to expand their knowledge throughout their lives, mostly through observation and experimentation.Footnote 47
The learning required of an apprentice empiric was crucially dependent upon the existence or creation of a strong relationship between apprentice and empiric. Such relationships were created only in the empiric’s older years;Footnote 48 for younger practitioners, secrecy was a characteristic both of the ‘occupation’ in general and of the ownership of an iatrosophic book. Although those scholars who have discussed empirics have mentioned the issue of secrecy, none has offered a possible explanation except in cases that involved incantations and spells;Footnote 49 in such cases, passing on the knowledge was considered to render the magic ineffectual for the one who passed it on and effectual only for the receiver.Footnote 50 A similar reason was offered by one of my interviewees, who possessed the knowledge of an incantation that removed the ‘evil eye’ cast upon someone. Only the interviewee, among all her siblings, received knowledge of the charm from her mother and, choosing among all her children, she transferred that knowledge to her son. As she explained, the charm would cease to be effective if either she or her son were to divulge its secret.Footnote 51 As all manuscript medical books included such charms and incantations, this rule of secrecy applied to them too. Such an explanation accounts well for the observed secrecy of the healing arts and for the practice of passing them on only to the most favoured and possibly most ‘talented’ or spiritual child, but fails to explain the employment of the same level of secrecy for all kinds of healing. It might be that all healing practices, whether ‘magic’, religious or empirical, were viewed as truly enmeshed and not differentiated as such by either the healers or their patients. On the other hand, it is perhaps plausible that restrictions were placed upon the dissemination of healing knowledge because this was a profitable profession that ran within families. It made sense, in order to retain the benefits of such profits (see below) within the family, to exercise discretion by passing on the art to one select individual. Furthermore, such strict control of the dissemination of such knowledge, and the consequent control of the numbers in the ‘profession’, ensured that the market would not become saturated with empirics.
3 Reasons for Becoming an Empiric
In Greece, at this time, it was illegal to practise healing without a licence. This raises the question of why someone would become an empiric, given that the consequences were persecution and possible imprisonment. In fact, empirics very rarely practised healing as their prime occupation, at least from the mid-nineteenth century onwards.Footnote 52 This applied both to men, who would have a main occupation, and to women, who would have only rarely had a formal occupation in rural pre-Second World War Greece.Footnote 53
The prime reason for the practice of ‘quackery’ in the West was income generation, and it seems an obvious reason in this case as well.Footnote 54 However, most scholars who discuss Greek healers point out that they did not demand (monetary) payment, although it is noted that people would offer payment in kind: ‘[T]heir payment was always various produce…eggs, chickens and the rest.’Footnote 55 When healing is referred to as a profitable pursuit, the clarification is made that the empiric would ‘receive wheat, barley, corn, almonds, clothes etc.’ for his/her services.Footnote 56 Such rare references to monetary payments as exist come from the early nineteenth century: for example, a cash exchange is mentioned in an 1825 extract from the cross-examination of an empiric accused of his patient’s death,Footnote 57 and further references from the islands of Sifnos and Melos in 1838 and 1840, respectively, show empirics documenting the monetary settlements they expected to receive.Footnote 58 The physician G. Fasoulakes, who held, unsurprisingly, somewhat negative views regarding empirics, emphasised the income-generating part of their work, especially the selling of the ‘medicinal preparations’ that they prescribed.Footnote 59 Fiction writers also emphasised the empirics’ focus on enrichment. A popular novella published in 1903 and written from the perspective of a giatrissa emphasises the paramount significance to the practitioner – the Murderess – of the ‘payment’ she received for her services, although the payment was only partially monetary.Footnote 60 Similarly, the significance of the payment was emphasised by an interviewee whose mother was a bone-setter. The interviewee mentioned in several separate instances the ‘payment’ that his mother received: ‘my mother [was a] widow, [a] very old woman[;] she knew some pseudo medical [practices] and they would give her something [payment]…. They would give her some broad beans, some oil.’Footnote 61 Yet another individual, writing in his memoirs, noted that his grandfather offered ‘help for free’, describing an incident in which his grandfather was asked about his fee by the patient’s husband and replied that he was not practising for the money; twenty days later, however, the fully recovered patient sent the empiric a quantity of local produce in appreciation of his help.Footnote 62 In inter-war Thrace a giatrissa claimed that she practised healing for ‘her soul’ and was thus not paid in money, but was allowed to take a symbolic minimal payment that she would use to light a candle in the church.Footnote 63 Marlene Sue Arnold’s interviewee, an empiric, emphasised in the mid-1970s that he neither took payment nor accepted gifts, while pointing out the materialism of contemporary physicians.Footnote 64 Still more recently, an interviewee who sought the help of an empiric in the late 1990s explained that the empiric asked for ‘whatever she had the pleasure to give, he did not say he did not want anything, he said to me whatever you want’, and the interviewee gave a small symbolic monetary payment.Footnote 65 However, the practitioner told her that some of his clients made significant gifts: one expatriate offered the practitioner his car.
It is clear that practitioners did enjoy some kind of revenue from their practice, although, almost certainly, such revenue was monetary only occasionally. It was, rather, primarily in kind and flexible in nature, given only when healing occurred, and corresponded to the economic and social standing of the healed. As the majority of healing took place in small communities which enjoyed little anonymity, expectations of appropriate gifts were very well informed and in all probability closely fulfilled.Footnote 66 Moreover, ‘gifts’ of healing would create an ‘obligation’ which invariably would be returned in due course, as local rural economies were based on systems of balanced reciprocity.Footnote 67 Although it could be argued that such gifts occurred in such economies because these were not monetised, this is unlikely to be the only reason. The great diversity of local economies in the first half of twentieth-century Greece included many that were highly monetised, and ‘gifts’ were given by patients not only to empirics but also to physicians.Footnote 68 Significantly, the offer of healing services without payment accorded with Christian teaching, as the two patron saints of all healers, Agioi Anargyroi, refused monetary payment for their services, as their name suggests.Footnote 69 More practically, most empirics perhaps refused monetary payment because the acceptance of cash could lead to incrimination in the event of prosecution by the state. To a degree, therefore, foregoing cash payment was a way for empirics to protect themselves from prosecution in order to be able to continue practising.Footnote 70
As important to the empiric as any ‘payment’ was the acquisition of significant social standing/capital in their society – a standing that would have modified substantially the everyday life and economic opportunities available to an individual.Footnote 71 As Mikes Paidouses and others have indicated, such a standing could be equal to, and at times higher than, that of a physician.Footnote 72 While the emphasis of the existing studies on the absence of payment or the ‘symbolic’ payment given to empirics, which is seen also in discussions of African traditional medicine, is partly due to an idealised view of the rural past emanating from folklore studies,Footnote 73 it also results from an implicit juxtaposition of the ‘gifts’ given to empirics with the requirements of physicians: monetary payment. Such payments were particularly difficult to provide in the Greek rural economy, even in the early 1950s: ‘In January I went to Makris [physician’s name], may he rest in peace, he was a good friend of ours; we did not have a penny left because we did not have insurance, all [our cash] was taken by Makris, we did not have insurance.’Footnote 74 Paidouses, along with Richard and Eva Blum, emphasised the significant expense involved in seeing a physician, while interviewees indicated that some physicians financially exploited the villagers: ‘[The physician] was taking from you all you had.’Footnote 75 In contrast, empirics did not demand monetary payment and, arguably, partly because of this, received instead elevated social standing within their local society and sometimes beyond – as well as the aforementioned ‘gifts’ – while also simultaneously protecting themselves against the risk of prosecution.
4 Legality, Availability and Practitioner Relationships
There is sufficient evidence to suggest that empirics enjoying a good reputation existed and were employed in equal measure with trained physicians in the early nineteenth century.Footnote 76 In 1834, only two years after the creation of the modern Greek state, its Bavarian administration established the rules of medical practice under which the Medical Council [Iatrosynedrio] was the only authority allowed to grant an individual with a licence to practise medicine in Greece.Footnote 77 Each candidate had to both ‘prove through valid certificates’ that he had studied and pass oral, written and practical examinations set by the Council.Footnote 78 By December 1834, all those practising without a licence were ordered to cease doing so,Footnote 79 although a provision seems to have been made for those ‘already practising [surgery, pharmacology and midwifery] empirically’ to attend a School of Surgery, Pharmacology and Midwifery, which would provide them with a licence if they were successful in their studies.Footnote 80 Attendance at the school was limited, however, partly because it was based in Athens and partly because of the illiteracy of most empirics.Footnote 81 Significantly, those empirics who attended the school could, at best, practise as surgeons [heirourgoi], but not as physicians.Footnote 82 In 1837 there were 130 licensed empirics in Greece, who had presumably attended the first run of the school, and eighty-five physicians.Footnote 83 In the town of Hermoupolis, there were eleven physicians (not all necessarily practising on the island continually), five empirics, two surgeons and nine phlebotomists practising in the 1830s and 1840s. The neighbouring community of Ano Syros had two empirics.Footnote 84 The possession of a licence defined the legality of any medical practitioner until 1939, when a new law decreed that a licence to practise would be automatically given to the graduates of the Greek [Athens] Medical School.Footnote 85 After the implementation of the law, some empirics petitioned for the right to be given a licence to practise gratis, but none was given such a licence.Footnote 86 Evidence from specific cases, such as those of Gymnasios and Nikolaos Theodorakis, suggests that such licences were not easily obtained, if ever.Footnote 87
Between 1907 and 1923 the Greek state doubled its territory and population; despite this, Greece did not suffer from a lack of physicians, with the ratio of physicians to inhabitants ranging around 10:10 000 in the first half of the twentieth century. Their concentration in Athens and in other urban centres, however, was astonishing. While, in 1940, the ratio was 29:10 000 for the county that included Athens, for other counties it was below 2.5:10 000.Footnote 88 Within Crete, in 1948, the same ratio was 17:10 000 for cities but 3.6:10 000 for the non-urban population.Footnote 89 The government made more than one attempt from 1923 to improve the provision of rural areas with physicians, but all such efforts proved unsuccessful for quite some time,Footnote 90 and it was not until well after 1950 that the situation began to improve.
Those few physicians who did accept a post in rural Greece rarely remained there for long, as working conditions were extremely harsh. For example, Petros Apostolides, a physician armed with a leftist ideology, took up a post in inter-war rural Greece in which he assumed responsibility for twenty-four villages situated so far apart that, he felt, he could not properly serve them all. So, having made an enormous effort to carry out his work, he abandoned his position and left for the nearest town on sick leave.Footnote 91 Rural inhabitants with access to a physician or a hospital ‘nearby’ would have to either bring the physician to their homes on a mule, often walking for hours, or take the patient to the ‘nearest’ hospital, ‘on the mule five to six hours’; still others would have to – or chose to – travel to Athens.Footnote 92 Even in the 1960s, just as in the 1930s, a newly qualified physician serving the rural community would have been acutely aware of the limited service he could offer, which would have comprised only a ‘consoling visit, a visit of support; you could not do many things’ because there were no medicines available locally. He had no instruments and, significantly, his training had left him lacking in practical experience.Footnote 93 Moreover, as rural inhabitants had no confidence in young physicians, it seems that their presence in the countryside was largely decorative.Footnote 94 The most significant ‘help’ that such a physician could offer would be to refer the patient to the nearest hospital, where he/she would be treated for a fee;Footnote 95 in the absence of such facilities, the physicians were just as helpless as they believed empirics to be, a fact readily acknowledged by some of their number:Footnote 96
A. There were two or three empirics, at various points [of the island] …. They had the empiric attribute/ability to cure some [illnesses]…and they would advise and I confess that they did good to/[benefited] the place because there were no scientists [physicians] to face the situation. But I will tell you a secret: even if there were physicians they would not have known how to face such illnesses.
Q. Did the empirics offer something to the people?
A. The empirics always offered to the people, as long as they were not money-driven.Footnote 97
The presence of traditional empirical healers during the first half of the twentieth century cannot be quantified since their practice was illegal and because the state took action against them not collectively but individually, and only extremely rarely, but the references to their existence and practice are so numerous and come from such a variety of sources that there can be no doubt about their strong presence, especially in the countryside.Footnote 98 In Crete, in 1948, it seems that in rural areas there were more bone-setters than physicians (the ratio of bone-setters to the general population was 1:2000 and that of physicians was 1:2750). ‘Herb-doctors’ appear to have been much less common, but were clearly present (with ratios of 1:48 040 in rural areas and 1:43 730 in the cities).Footnote 99 Sevaste Haviara-Karahaliou, employing exclusively oral accounts from Chios, cites no fewer than sixty named individuals who acted as empirics, primarily in the first half of the twentieth century but also earlier; they specialised in different ailments and resided throughout the island.Footnote 100 In 1951, as the folklorist Polyd. Papahristodoulou emphasised in his speech on behalf of the Folklore Society, the presence and use of empirics in rural Greece was a given: ‘Popular medicine is a common theme in our lives. And we have [all] been cured by a giatrissa or a popular healer in the rural life [in which] we grew up.’ He added that ‘popular or empirical healing is used even today in the remote areas of Greek lands’.Footnote 101 The existence and, in many cases, the names of empirics have been widely documented in all the sources cited here that refer to popular medicine in Greece, ranging geographically from Thrace to Rhodes and from the Cycladic island of Syros to the Ionian island of Cephalonia.Footnote 102 Still, as knowledge of the identities and abilities of popular healers depended entirely on reputation and word of mouth, such references cannot reveal the full extent of their presence.Footnote 103
In the twentieth century, especially in rural areas, inhabitants would have had no difficulty in distinguishing between physicians and empirics. The former was invariably an outsider to the area, an educated urbanite who had to present his licence to the local authorities before settling there.Footnote 104 In contrast, the latter was a local person, perhaps born in the village or of long settlement there. The empiric was also largely uneducated, even if he/she could read, and was unlicensed. Moreover, the kinds of healing practised by popular healers and physicians were very different.Footnote 105
5 Protecting (Healing) Tradition
Western physicians’ invective against the illegal practitioners with whom they competed for patients has been the subject of much scholarly discussion. It is possible to trace similar attacks on empirics in Greece, made not only by physicians but also by educated individuals, but these are not abundant.Footnote 106 At the same time, many references to popular healers made explicit connections between some of the methods employed by them and methods practised in ancient Greek medicine.Footnote 107 The purpose of folklore studies in Greece, which flourished from the late nineteenth century, was exactly that, namely, to trace the links between contemporary rural practices and those of Ancient Greece, so ‘proving’ the existence of cultural continuity.Footnote 108 In 1909 Nikolaos Polites, the creator of Greek folklore studies, argued that popular medicine should be one of the field’s main themes.Footnote 109 So, in 1925, the professional medical journal Klinike published in its first year a series of research articles on Greek popular medicine. Stilpon Kyriakides, the author and director of the Folklore Archive, used the archive’s collections to shed light on popular medicine, routinely making connections between the folklore material and Ancient Greek practices.Footnote 110 He also emphasised, however, the cultural gap that existed between educated physicians and overwhelmingly uneducated patients, the two having a different understanding of the origin and nature of disease, and added that ‘it was time that the physicians turned their gaze towards the Greek people and studied the way the populace [o laos] understands diseases and the means that [the populace] uses to avert or cure them, long before they seek the advice of the physician’.Footnote 111 Kyriakides argued that it would be beneficial for physicians to learn about ‘popular beliefs concerning diseases’ in order to narrow the ‘chasm that exists between the people and the physician’, so that the latter could become more ‘able to advise and instruct [the patients] in a language comprehensible by them and if necessary, [the physician should] utilise the popular beliefs for the benefit of his therapeutic purposes’.Footnote 112 The articles written by Kyriakides and his position towards popular medicine were of enormous significance: it was the restrained but explicit support of Kyriakides and other folklorists and local historians for popular medicine that partly averted the possibility of an aggressive discourse against popular healing.Footnote 113
That is not to say that action against individuals who practised healing without a licence was never taken: the example of Monk Gymnasios (above) demonstrates that it was. Another healer, practising in an urban area and accepting payments, was arrested and taken to court in the 1950s.Footnote 114 The famous Vlachos, who practised as a bone-setter for much of the twentieth century, was repeatedly taken to court.Footnote 115 In an earlier case, a popular healer who practised in rural Leukada was turned in by the physicians of the town of Leukada but escaped prosecution because no witnesses came forward to testify against him.Footnote 116 It is notable that the community that he was serving ‘protected’ him by refusing to testify against him; indeed, evidence for such protection comes in many instances and in diverse locations. For example, in the mid-nineteenth century, Aggelos Katerelos was practising in a village while also working as a teacher in British-governed Cephalonia. When Cephalonia became Greek territory in 1864, his healing practice became illegal. At times of changing power structures Greek politicians were able to remove those individuals employed by the previous administration. However, Katerelos’ position as a teacher was spared because ‘the villagers want him because he practises as a doctor’.Footnote 117 In the cases of two other empirics who were included in Cephalonia’s medical list of 1872, a note was made next to each name to the effect that the empiric ‘does not practise [medicine] as an occupation but if he is asked, due to necessity, he will provide an opinion’.Footnote 118 In this instance the tactful articulation by the local authorities of how the empirics operated – being sought by, rather than seeking, patients – protected them from persecution.
In yet another case in the village of Pyrgi, on Chios, the practitioner K. Keras and his successor Magkaikas were both persecuted by the authorities,Footnote 119 but the former enjoyed the protection of the local society in which he practised. In 1915 – that is, as soon as the Greek state established itself on Chios – the community issued a certificate assuring the police, invariably represented by a non-local, that Keras’ only occupations were those of pottery maker and tobacco cultivator, and ‘no other’.Footnote 120 Keras was not a native of the village or of the island, and it is possible that this was why he was persecuted in the first place. In the late 1950s, in mainland Greece, an empiric accused by the local physician of practising without a licence was arrested and spent a few hours in jail, but was released by the local policeman, whose wife had received treatment from him.Footnote 121 Like Keras, this empiric was not a native of his village, although he had been resident there for decades.Footnote 122 In contrast, other empirics who operated in the same area were never turned in,Footnote 123 although when an itinerant woman appeared in the same village selling healing herbs, the pharmacist called the police immediately.Footnote 124 It is thus apparent that protection was offered to esteemed empirics particularly if they were native to the village. But, even then, empirics would be jailed when patients died.Footnote 125
In some situations, and as more physicians moved to, or near to, rural areas, physicians and empirics seem to have co-existed amicably. For example, during the majority of his career as a practitioner in a northern Chios village, Giannes G. Heilas, the third generation of renowned empirics in his family, encountered no physicians.Footnote 126 In the later years of his practice, however, when physicians were available in the nearby town of Volissos, Heilas would send serious cases to the physicians while he would treat minor injuries, as ‘the expense of visiting a physician was considerable’.Footnote 127 The physicians never attempted to have Heilas prosecuted, and he practised until his death in 1965.Footnote 128 Local communities, too, were instrumental in determining both practitioners’ and physicians’ success, or otherwise. Not only did they ‘protect’ local empirics if he/she were valued, but incoming physicians, if they were to become established in such a community, had to be, or be willing to become, familiar with local ways of thinking about healing, or at least be tolerant and respectful of the unwritten rules of the community.Footnote 129 For example, the two physicians who were established in the town studied by Blum and Blum stated that they were ‘impressed’ by the therapies offered by the empirics and that they accepted the concept of the evil eye and employed locals to have it ‘removed’. The trained midwife, though not a local, also declared her trust in empirics.Footnote 130 Yet another physician working in a Cretan village in the early 1950s assured his patient’s family that he, too, would invite the priest to perform a blessing when someone was ill in his family, just as the family planned to do after his visit.Footnote 131 Although none of these physicians expressed outright support for the empirics, their compliant attitude was necessary to give them a foothold in the locality where they practised. However, some physicians were prepared to go further: one, in trying to eliminate the cultural gap between himself and his patients, pretended to be an empiric, although this occurred in an urban environment where his true identity could be more easily kept secret. His spectacular success in attracting patients and the fact that he accepted payment meant that he was soon reported to the police and arrested, only to reveal to the court that he was a licensed physician. He explained that he had carried out the deception because when he had practised as a physician he had been unable to attract any patients.Footnote 132
The protection afforded empirics by their communities, then, was a significant factor in allowing them to continue practising throughout the nineteenth and much of the twentieth centuries in spite of their legal status.Footnote 133 Moreover, the association of popular healing with Ancient Greek healing practices, as articulated by eminent folklorists and others, including Greek physicians, constrained the medical establishment from lashing out publicly against popular healing.
6 Choosing a Practitioner
Since both empirics and physicians operated in Greece, a degree of patient choice was available, dependent largely upon location. In the first half of the twentieth century, empirics and physicians occupied largely separate ‘catchment’ areas, with the former being established in rural areas and the latter based mostly in urban areas, although some also visited nearby villages or treated rural patients who had come to urban centres for that purpose.Footnote 134 On Chios, for example, the rather inaccessible northern villages were served by local empirics, whereas the towns and the southern villages were relatively well served by physicians.Footnote 135 Not all communities had an empiric, so, for many people, a consultation with any practitioner involved travelling and inconvenience;Footnote 136 but even when both a physician and an empiric were available, the patient’s choice varied from case to case: ‘My mother, who broke her leg, did not want to go to the physician. She went to an empiric.’Footnote 137 Even in urban centres where not only physicians but also hospitals were available, as in Hermoupolis, for some people the possibility of seeking a physician’s assistance did not seem, at times, even to be considered, as one interviewee revealed when discussing the death of many of his own siblings as young children:
A. That’s what I would hear, the giatrines, the local ones. Because they existed in the past, a giatrina was then considered [to be] better than a physician is [considered to be] today.
…
Q. Wouldn’t they take [the child] easily to a physician?
A. It wasn’t so much…to take you to the physician. You would become twenty years old and you would have never been to the physician.Footnote 138
But when this interviewee’s own infant son became ill in 1949 it was the physician he called, signalling how, over time, they had come to be accepted.Footnote 139 Nevertheless, for some diseases, empirics continued to be preferred to physicians: ‘People then went more to the various empirics and believed in the mysterious herbs for venereal diseases rather than [going to] the physicians.’Footnote 140
As Philip P. Argenti and H.J. Rose explain, in Chios, during the 1940s, rural people maintained a ‘prejudice’ against physicians.Footnote 141 The cultural ‘proximity’ of empirics appears to have been of considerable importance to patients in their choice of healer, especially for women, for whom modesty was an important consideration.Footnote 142 The physician Petros Apostolides describes the reaction of a 40-year-old female patient when he recommended an injection: she suggested that he make a small cut in her dress at the injection site to avoid her having to reveal her body.Footnote 143 Richard and Eva Blum stressed that ‘[t]he folk-healers share the culture of the peasants and shepherds…. They have the same assumptions about man’s relation to man, nature, and the supernatural that guide the lives of the uneducated people of the region.’Footnote 144 It is unsurprising, therefore, that a physician who practised in rural Greece in the 1960s felt strongly the cultural distance that existed between him and the villagers:
…someone had a stroke…do you know what the first question was? Not ‘what is the matter with the patient?’, which is the logical question to ask…[but] ‘will he become as he was?’ The answer was of course ‘no’. And then came the tragic, for me, reply. ‘Then doctor, let’s not do anything’.Footnote 145
In late twentieth-century Greece, biomedicine was generally perceived as ‘modern’ and effective and empirical practice as old-fashioned; yet it is important to note that, as late as the mid-1970s, many women in places such as rural Crete chose to consult an empiric first and then, only if necessary, a physician, with the question ‘it might still work, why not try it?’Footnote 146 Since consulting an empiric did not involve a physical examination, it is not difficult to comprehend the reluctance of women to visit the ‘culturally distant’ urban physician except as a last resort.
The distance between patients and physicians was increased, too, by their different understanding and interpretation of disease. The villagers studied by Blum and Blum had an understanding of diseases and their causes that was very close to that of the empirics; it downplayed the significance of contagion while also involving disorders that ‘doctors do not know’, such as the ‘wandering navel’, the ‘waist’ and the ‘lifting of the kidneys’ – diseases that the folklorist Nikolaos Polites referred to as ‘non-existent’.Footnote 147 These types of condition were described by Enrique Perdiguero, in his discussion of popular medicine in Spain, as ‘culture-bound syndromes’.Footnote 148 Even in the case of biomedically recognised diseases, physicians and patients used different names to refer to them.Footnote 149 More generally, then, language was another separating force between patients and physicians. Physicians tended to use Katharevousa, the formal language in which they were educated, a version of Greek that would have been incomprehensible to most rural patients.Footnote 150 This point was masterfully, if unwittingly, articulated by a physician who referred to ‘the foreign language that physicians speak in front of the patient so that he and his circle cannot understand the scientific or professional communications …’.Footnote 151 In contrast, empirics and their patients spoke not only the same form of Greek but also the same local dialect.Footnote 152
7 Discussion and Conclusions
The Greek case reveals a situation in which the legality and primacy of licensed physicians was institutionalised right from the beginning of the modern Greek state in the 1830s. Unlike in England and elsewhere, where distinguishing between a physician and a quack was not easy, in Greece, the distinction between physicians and popular healers was very clear, certainly in the twentieth century.Footnote 153 However, even in England, the distinction between physicians and what Davies calls cunning-folk was easy to make in the nineteenth century.Footnote 154 The numerical dominance of physicians was well established in early twentieth-century Athens and the main urban centres, but their presence in rural Greece was limited. The converse appears to be true for popular healers, but this does not mean that they did not also exist in Athens and urban centres: rather, if they did, little evidence of their existence survives.Footnote 155 Moreover, the close ‘regulation’ of the training of new popular healers meant that their numbers remained, at best, stable. As a result, they did not pose an ever-increasing threat to physicians. In rural areas the state was not quick to prosecute empirics, partly because few other medical practitioners were available there but also because local communities protected their popular healers. Community support for healers is not apparently unique to the Greek case, although its existence is not made explicit in the relevant literature either in Greece or elsewhere. For example, in England, both Davies’ cunning-folk and Moore’s wise woman received support from community members. Davies describes how, when Ellen Hayward was taken to court in 1906 for practising magic, community members wrote to the court, the press and the local magistrates supporting her presence and practice of herbalism in the community and arguing that a stop should be put to her persecution in the ‘interest of [the] community’.Footnote 156 Similarly, Moore repeatedly refers to the ‘trust’ conferred on the wise woman Nell Racker and the community’s loyalty to her.Footnote 157
Although, as in Greece, non-trained practitioners continued to exist in the West, the reasons offered for their survival differ drastically from the ones offered here for Greece. Firstly, the literature implicitly and explicitly accepts that profit was the reason for the existence not only of quackery but also of most of the practitioners of alternative medicines in the West. But, in distinguishing between the illegal practitioners, different pictures emerge. So, while the nineteenth- and twentieth-century quacks operated for profit and advertised their wares and skills, cunning-folk, wise women and popular healers did neither of these things explicitly. In France, we are told, popular healers were not ‘entreprenant’.Footnote 158 The cunning-woman Ellen Hayward, when taken to court in 1906, argued that she ‘never demanded payment’, although her clients invariably paid her. Related is, in all probability, the fact that physicians in England or in France did not undertake a ‘concerted attack’ on cunning-folk but instead persecuted quacks.Footnote 159 As in Greece, the low ‘payment requirements’ of the popular healers, their cultural proximity to their patients and the protection of the communities in which they practised ensured the continuation of their existence into the mid-twentieth century and beyond.
The Greek case offers an alternative to the discussions of unlicensed practitioners found in the Western literature. But, at the same time, it echoes the findings of Moore and Davies, who established not only the survival of localised patterns of non-orthodox medicine in the late nineteenth and early twentieth centuries but also the survival of cunning-folk and the role they played as healing providers in nineteenth- and early twentieth-century England. Although the Greek empirics seem only very rarely to have practised magic exclusively – as some cunning-folk did – it was nevertheless an integral part of their practice. This paper has demonstrated, for Greece, not only the empirics’ survival but also how and why such a survival was sustained. It is apparent, moreover, that Greece was not unique in the possession of such popular healers, but rather very much in line with England, Spain (where healers were known as curanderos) and perhaps also Italy.Footnote 160 While a wide variation in terminology and practices, as well as such healers’ elusive presence, accounts for their relative neglect hitherto by historians of medicine, the findings of this paper have highlighted the urgent need for further work on popular healing in the twentieth century.
Monk Gymnasios, the healer this paper began with, seems, in fact, to have been an exception to the way in which Greek empirics operated. First, his status as an (in)famous public figure attracted thousands of ‘patients’, caused physicians to feel threatened by his presence and led the state to get involved.Footnote 161 Second, he lacked a fixed base which would have provided him with the unspoken but valuable protection that the community bestowed on its own trusted empirics. Telling is his decision, concurrent with that to stop practising in 1932, to publish his remedies, going against the unwritten rule of secrecy that governed the practice of empirics in Greece.Footnote 162