Chapter 2 - Phrenitis in Classical (Fifth–Fourth Centuries bce) and Hellenistic (Third–First Centuries bce) Medicine

Summary

Chapter 2 begins with classical medicine, exploring the sources of so-called Hippocratic medicine, nosological and clinical, as well as other lesser-known authors from the fourth century BCE such as Diocles and Praxagoras. The limited material preserved on our topic from Hellenistic medicine (Herophilus and Erasistratus) is also surveyed. The richest information is preserved by the writings of the Hippocratic Corpus, where phrenitis first appears, and where it is richly described, both in nosological profiles and with reference to specific patients. Its core traits are by this point established: fever, localization in the chest , and an association with winter also showed by the co-morbidity with, and analogy to, pleuritis and pneumonia. Interestingly, the phren/phrenes are seldom mentioned in discussions of phrenitis, and when they are, not in their traditional, ‘Homeric’ psychological function, or directly as locus affectus, thus signalling a desire to distance the pathological narrative from traditional poetic models.

As noted earlier, the first appearance of the noun phrenitis preserved for us is in the earlier, classical nucleus of texts within the Hippocratic Corpus.¹ This does not necessarily imply that the disease concept was a Hippocratic creation – indeed, it is reasonable to think that a disease by this name and with comparable characteristics must have been recognized before the written testimonies by medical communities, given the level of elaboration and codification associated with it in fifth–fourth-century bce medicine.² It is impossible to reach back to this previous stage or to know anything precise about these earlier medical communities. Nonetheless, a comparative exploration reveals the existence of clusters of symptoms and circumstances that may constitute a precedent to the disease, specifically high fever, heat, derangement and a range of associated symptoms. If one were to adopt greater flexibility than programmatically allowed for in this study, where the focus is on the recognized nosological entity phrenitis, and while doing so expand the object of enquiry to other areas of ancient Mediterranean culture, numerous comparable syndromes and cases, if not perfect parallels, could be traced in Babylonian and Egyptian medicine,³ as well as in other Hippocratic cases of fevers with derangement not labelled ‘phrenitis’.

The Hippocratics

The medicine of the fifth and fourth centuries bce displays a strong awareness of phrenitis as a well-demarcated illness with clear nomenclature and fixed characteristics. The key Hippocratic evidence is found in a number of texts, which offer elaborate, paragraph-length accounts.⁴ But mentions of the disease are much more numerous and suggest its pathological and doctrinal importance for these physicians and their patients.

As a general impression, in classical medicine phrenitis emerges as an acute, severe disease, often deadly, which belongs to a group of affections characterized by a high fever that seem to concentrate in the chest and respiratory tract and occur in winter.⁵ Aff. 6 (14.7–11 = 6.214 L.) says expressly that ‘diseases of the cavity … pleurisy (pleuritis), pneumonia (peripleumoniē), ardent fevers and phrenitis … occur most frequently and violently in winter’;⁶ at Aff. 10 (21 Potter = 6.218 L.) phrenitis is said to ‘sometimes change into pneumonia (peripleumoniē)’, underlining the association with the chest. At Epid. 1, 18 (25.8–10 Jouanna = 2.650 L.), in the third constitution, phrenitis occurs ‘around the equinox up to the settings of the Pleiades, and during winter’, while at Nat. Hom. 5 (212.1–4 Jouanna = 6.78 L.) we read that emetics and clysters – both purging methods – are to be used in the periods of the year that engender more phlegm, such as winter, when ‘diseases that attack the head and this region above the diaphragm (to chorion touto to hyper tōn phrenōn) occur’; phrenitis is not mentioned explicitly, but the details offered seem to point in that direction. At Epid. 7, 53 (84.21 Jouanna = 5.422 L.) the phrenitic (phrenitikē) sister of Hippis falls ill cheimōnos, ‘in winter’, while another patient, the man from Halicarnassus at Epid. 7, 112 (15–20 Jouanna = 5.460 L.), develops phrenitis after having fallen ill with earache and headache, again in winter (en cheimōni). As a winter ailment affecting the chest, our disease is similar to and often discussed in association with peripleumoniē, pleuritis and ardent fevers.

Symptomatology

In Hippocratic nosology, as much as in the clinical cases, a vast repertoire of manifestations of the disease appears, and a comprehensive reading reveals a clear and consistent picture. What is missing is a comprehensive, consistent aetiology, localization and course of therapeutic action. From the perspective of contemporary diagnostics, it is important that any account of a disease isolate the factors specific to it – for example, those that are not extensible to mental disorder generally – and that as such are necessary and sufficient to identify the disease.⁷ These must be distinguished from symptoms which might be present but are insufficient or weak indicators and common to other conditions. Such precise ranking in cogency among signs was not recognized, at least explicitly, in fifth- and fourth-century medicine. The signs mentioned below thus form a constellation of manifestations, a composite ‘story’ rather than a reliable ‘symptom checker’.

Fever is from the beginning the indicator that qualifies phrenitis as a disease, as well as singling it out among mental disorders. The inclusion of the disease in the category of fevers is obvious, due to signs linked with overheating (shivers, chills, thirst, dryness and sweating). Phrenitis is deemed acute and fatal from the start:⁸ ‘acute fevers, such as peripleumoniē or phrenitis’, specifies the author of Progn. 4 (13.3–5 Jouanna = 2.122 L.).⁹ Phrenitis thus appears, in a sense, to be a possible outcome or development of an ardent fever, a kausos, and is treated as exemplary among clear-cut cases of pyretoi, fevers. Phrenitic fever can perhaps be milder and more gradual than other kausoi. At Coac. 223 (158.1–2 Potter = 5.632 L.), in a class of individuals suffering from eye symptoms, moderate heat is a phrenitic sign: ‘Patients who are not burning hot to the touch develop phrenitis (φρενιτικοὶ γίνονται)’, while at Morb. 1.30 (86.19–88.13 Wittern = 6.200 L.) the entire description of the disease is constructed from the heating of the patient’s blood, and of his body as a consequence, producing a formidable fever. In the haematocentric perspective of this particular treatise, the affection of the blood and the ensuing fever are responsible for mental derangement.

Most succinctly, the patient at Epid. 3, 17, case 4 (98.1–11 Jouanna = 116–18 L.) is declared a phrenitic from the start. The description, a short case ending with death, is quite expressive:

The patient suffering from phrenitis on the first day that he took to bed produced copious thin vomit the colour of verdigris; much fever with shivering; continuous sweating all over; painful heaviness of head and neck; urine thin, with small, scattered substances floating in it, which did not settle. Copious excreta at a single evacuation; delirium; no sleep. Second day. In the early morning speechless; acute fever; sweating; no intermission; throbbing all over the body; convulsions at night. Third day. General exacerbation. Fourth Day. Death.¹⁰

Fever has a wide range of specific manifestations in phrenitic cases. At Prorrh. I, 27 (78.2–3 Polack = 5.516 L.) restlessness (dysphoriai) during a chill in a patient with fever who is perspiring in the upper half of his body is a phrenitic sign (phrenitikai): we have here fever with a sense of unrest, as well as a pathological focus on the torso. Likewise, at Coac. 69 (120.13–14 Potter = 5.598 L.) phrenitis (and death) are foreshadowed by ‘restlessness together with a general cooling that does not end with the fever, in a person who is perspiring over the upper part of the body’. Frequent changes in the signs that typify fever are unfavourable indicators: at Prorrh. I, 12 and 13 (76.7–10 Polack = 5.514 L.) we learn that ‘in the early stages of phrenitic cases, signs that are mild, but change frequently, are a bad sign; salivation (ptyelismos) is also bad’, and below that ‘in patients with phrenitis, a white evacuation (leukē diachōrēsis) is bad, as it was for Archecrates. Does torpor follow in these? Chills too are very bad signs in these patients.’ At Prorrh. I, 15 (76.13–14 Polack = 5.514 L.) we find a description of the onset of the disease in which pyretic signs are critical: ‘Persons out of their wits (hoi ekstantes) who are suddenly attacked by an acute fever and sweating, are phrenitic.’¹¹

The voice of the feverish patient is also mentioned: Epid. 3, 3 offers a description of spring illnesses with ‘many malignant cases of erysipelas’. Here various items are listed as signs (80.15–16 Jouanna = 3.70 L.): voices impaired (phōnai kakoumenai), phrenitic ardent fevers (kausoi phrenitikoi), ‘muffled mouths’ (stomata aphthōdea). The inclusion of a disease, phrenitis, in a list of signs does not surprise, but expresses the magmatic stage in the thematization of ‘disease entity’ in the sense in which we understand it today. The muffled voice returns as an indicator of phrenitis – as well as of consumption and ardent fevers – at Epid. 3, 5 (83.7–10 Jouanna = 3.76–80 L.): ‘Many had the symptom of impaired and muffled voices (phōnai katillousai), first at the beginning of cases of consumption, but also in ardent fevers and in those with phrenitis.’ At Epid. 3, 6 (85.3–5 Jouanna = 3.82 L.) ardent fevers and cases of phrenitis (hoi phrenitikoi) are described together as not being thirsty and displaying a particular kind of derangement, which involves not mad delirium but stupor: ‘None of these phrenitic patients was raving mad, as in the other cases, but they passed away overpowered by a dull oppression of stupor (tini kataphorēi kakēi, nōthrēi, bareōs apollynto).’ The deaths of these patients are described as similar to what happens with other ardent fevers, ‘varying with the individuals, usually irregular, at the crises, but in some cases after a long loss of speech, and in many with sweating’. At Epid. 3.11 (88.1–2 Jouanna = 3.90–92 L.), moreover, phrenitic patients are said to be ‘comatose for most of the time’ (like people suffering from ardent fevers, kausōdees, those suffering from ardent fever and most other diseases involving a high fever).

The topic of fever and its massive presence in Hippocratic medicine played an important role in the history and historiography of ancient medicine, especially a few decades ago, when retrospective diagnoses of malaria and other infectious diseases were proposed to make sense of these depictions.¹² Malaria was endemic in ancient Greece,¹³ but this bio-historical datum is irrelevant to the study of the constructed notion of phrenitis as a mental disease, and there is little to gain from pursuing the diagnosis.

The quality of urine, an established area of Hippocratic observation, is especially important in cases of phrenitis, and will return for centuries in accounts of the disease. The urine of these patients is whitish and thin, and may contain suspended matter. At Coac. 571 (250.17–19 Potter = 5.716 L.) ‘colourless urine with dark suspended material in it, in association with sleeplessness and trouble, indicates phrenitis (phrenitika)’. At Aph. 4.72 (426.7 Magdelaine = 4.528 L.) a particular kind of urine – white and transparent – is bad (ponera) and characteristic of phrenitic patients.¹⁴ At Prorrh. I, 4 (75.8–10 Polack = 5.510–12 L.) it is said that ‘in concomitance with disturbance (tarachos) and bad sleep, urine of a bland colour, with dark suspensions, is a sign of derangement (parakroustika); with sweating, phrenitis … in cases of disturbance and insomnia’.¹⁵

With fever come dryness and thirst/lack of thirst¹⁶ as part of the pathological portrait. At Prorrh. I, 3 (75.7–8 Polack = 5.510 L.) we learn that ‘muffled tongues’ (daseiai glōssai kai kataxēroi, i.e. those that are dry and unable to speak clearly) are phrenitikai, while at Coac. 229 (158.26 Potter = 5.634 L.) ‘rough, very dry tongues indicate phrenitis (phrenitikai)’. Patients with fever, and the phrenitics among them especially, often exhibit aphasia, speech impairment and a lack of clarity in verbal expression that can be explained in terms of mental disorder, but that is also plausibly a consequence of overheating. In fact, much of the repertoire of mental disturbance in this period, when considered in this connection, is of a feverish sort.

Along similar lines, spasms and motor disturbance are important signs of the disease in the Hippocratics, and are generally linked to mental health. At Epid. 7, 112 (112.3–10 Jouanna = 5.460 L.) Polyphantes of Abdera had a high fever, suffered from continuous headaches and pain in the throat, and ‘was mad in the manner of the phrenitics (tropon phrenitikon) and then died of intense spasms’. At Progn. 23 (76.3–7 Jouanna = 2.186–88 L.) motor disturbances are discussed and their severity placed in relation to the age of patients: older children and adults do not fall prey to these symptoms ‘unless one of the most severe and unfavourable signs appear, as is precisely the case with cases of phrenitis’. Likewise at Prorrh. I, 28 (78.4 Polack = 5.516 L.) the spasms of phrenitics are said to be not only intense but frequent (pykna metapiptonta spasmōdea); in these patients especially, violent trembling is fatal (Coac. 96, 126 Potter = 5.604 L.).

Among types of motor disturbance, one in particular is significant for these cases, as is hinted at in the case of Hippis’ sister at Epid. 7, 53 (84.21–24 Jouanna = 5.422 L.), who ‘was phrenitic: … very busy with her hands (tēisi chersi pragmateuomenē), lacerating herself … puffing into her jaws and lips like sleeping people’. This kind of compulsive hand movement is often noticed with the disease. Such behaviour, usually referred to as ‘crocydism’, ‘carphology’ or ‘flocillation’, is a known neurological datum, the compulsive and perhaps hallucinatory movement of the hands that is a consequence of fevers (e.g. in cases of typhus); it is a sign between motoric compulsion and hallucination – perhaps, from the modern point of view, an intersection of cognitive datum and neurological disturbance. At Progn. 4 (13.3–14.2 Jouanna = 2.122 L.) the discussion centres on precisely this aspect, and phrenitis is mentioned with reference to compulsive movement of the hands, together with other cases involving high fevers (peripleumoniē and kephalgiē):

About movements of the hands, this is my opinion: in those who suffer from high fevers, in cases of peripleumoniē, or phrenitis, or kephalalgiē, to bring them before one’s face and search through the empty air, and try to pick bits of wool from the cover, and peel threads from one’s clothes, and scratch dirt from the wall. All these are bad and anticipate death.

The claim that ‘derangements accompanied by tremors, unclear/confused, with carphology, are eminently indicative of phrenitis’ (tromōdees, asaphees, psēlaphōdees parakrousies, pany phrenitikai) also appears at Prorrh. I, 34 (78.15–79.1 Polack = 5.518 L.).¹⁷ Again, being necessary and/or sufficient is not considered a requirement in these pathological discussions, contradicting the expectations of modern medicine, or even of Galen: many of these signs are extended elsewhere to deranged or feverish patients generally.¹⁸ But the lack of a cogent symptom checklist should not prevent us from inferring patterns from the descriptions.

Last but not least, phrenitis always carries the markers of mental disturbance. This can take various forms. Familiar terms and expressions for derangement – delirium, talking nonsense, and other cognitive disturbances, but also unexplained silences, sensory impairment (numbness, deafness) and trouble sleeping (from insomnia and disturbed sleep to comatose states) – all belong to typical portrayals of mental affection. Full discussions of phrenitis (in nosology as much as in patient cases) necessarily mention mental suffering:¹⁹ ‘He is deranged in the mind’ (tou nou parakoptei, Aff. 10, 18.19 Potter = 6.217 L.); ‘Patients with phrenitis most closely resemble melancholics in their derangement (kata tēn paranoian), for melancholics too, when their blood is disordered by bile and phlegm, have this disease and are deranged (paranooi ginontai) – some even rage (mainontai)’. In phrenitis it is the same, only here ‘the raging and the derangement (hē maniē te kai ē paraphronēsis)’ are less to the extent that the bile in one case is weaker than that in the other (Morb. 1.30, 88.7–13 Wittern = 6.200 L.). Phrenitic patients are ‘out of their mind’, ekphrones eisi (Morb. 3.9, 76.21–22 Potter = 7.128 L.); they are ‘deranged (paraphroneousin) throughout the course of the disease’ at Morb. 1.34, 92.7–8 Wittern = 6.204 L.). In the patients at Epid. 7, 79, 80, 95–96.10 Jouanna = 5.44–46 L. (in themselves rare examples of individuals who survive the disease) external appearance and neurological manifestations (as we would define them, using our own distinctions) are emphasized at the expense of a report of cognitive alterations, in line with a general tendency in these texts: trembling, a broken although still intelligible voice, a burning fever (7, 79), along with a dreadful disorganization of physical posture, a thin, broken voice, disorientation and sleeplessness (7, 80). Within these signs, sensory disruption is an important symptom of mental suffering in this disease.

This deserves separate discussion both as hallucinatory disorder and as simple sensory alteration, since it characterizes phrenitis throughout its history and engenders important theoretical debates. In the earlier, classical sources these impairments are not opposed to or even categorically distinguished from cognitive ones, but belong to the core of the psychopathological portrayal. At Epid. 5, 52 (24.6–7 Jouanna = 5.236 L.) (= Epid. 7, 71, 92.13–14 Jouanna = 5.432 L.), for example, a patient is introduced as ‘blind due to phrenitis’ (to kōphōma ek phrenitidos); kōphōma in these texts is a less clear-cut experience than our term ‘blind’ suggests,²⁰ but a long-lasting disturbance in vision, caused by our disease, is patently in question. Vision is apparently distorted and its sensitivity intensified, as with flocillation: Prorrh. I, 5 (75.10–11 Polack = 5.512 L.) informs us that ‘their dreams are especially vivid’ in phrenitic patients (enypnia ta en phrenitidi enargē).²¹ This sign relates to sight: dreaming and seeing are notoriously contiguous in Greek vocabulary and imagination, and feature importantly in Galenic psychology.²²

The discussion of hallucination, which would be the richest in association with phrenitis in the Hippocratic texts, at Morb. 2.72 (Jouanna 211.15 = 7.110 L.), is philologically problematic. Both manuscripts M and θ, duly followed by Jouanna, offer phrontis (φροντίς), ‘anxiety’, rather than phrenitis (φρενῖτις) here. Potter, however, in the Loeb text corrects the passage in a forced manner that converts it into an account of our disease, printing φρενῖτις.²³ The opening is unique among the surviving descriptions of the disease and resembles other Hippocratic passages depicting derangement: it starts with patients feeling that ‘a thorn (akantha) seems to be in the inward parts and to prick them; loathing (asē) attacks the patient, he flees light and people, he loves the dark, and he is seized by fear … the patient is afraid, and he sees terrible things, frightful dreams, and sometimes the dead’. Nor is vision mentioned in these sources as the only area impaired in phrenitics, although it is the most important of the senses: at Coac. 95 (124.25–26 Potter = 5.602–04 L.) the portrait of the phrenitic patient is expanded to include ‘being over-sensitive to noise (psophou kathaptomenoi)’ as an indicator for the onset of ‘trembling and convulsion’.

Many of the disparate signs displayed by phrenitic patients are found clustered in Prorrheticon 1, a text that offers many aphorisms regarding mental health. These instances are important because they allow us to begin to construct a picture of the disease in which frequent signs appear, signs which might not be exclusive but recur in a meaningful way. In a discussion involving the neck and throat at Prorrh. I, 1, this telling question is posed: ‘Do patients who have been comatose at the beginning, but later lie awake with pains in the head, loins, hypochondrium and neck, develop phrenitis?’ (75.2–4 Polack = 5.510 L.). Neck and head, as well as the diaphragmatic location, come into play, apparently along with a general link to cold-like ailments: a bit later we read that ‘a running nose in these is a fatal sign, especially if it begins in the fourth day’.²⁴ At Prorrh. I, 2 (76.4–6 Polack = 5.512–14 L.) the localization of the affection in the throat is related to our disease. The whole aphorism reads: ‘Slight, suffocating pains in the throat, felt on opening the mouth by a person with an acute disease who cannot close it easily and is feeble, announce derangement (parakroustikon); those with phrenitis are doomed’.²⁵ Here patients with the disease are in an especially weak position if they suffer from ailments of the neck and throat.

At Coac. 223 (156.25–26 Potter = 6.632 L.) phrenitis is mentioned as associated with symptoms that are eminently mental: ‘Fixation of the eyes in an acute disease, or a sharp movement of the eyes together with disturbed sleep or sleeplessness, sometimes also provokes a haemorrhage from the nostrils. Such patients who are not burning hot to the touch become phrenitic, especially if a haemorrhage occurs’. These shifting clusters or patterns lack the coherence of a consistent nosological account, but details begin to coalesce around a few points that become the backbone of the disease in later nosology. It is important at this stage to note that a localization around the chest and throat, the respiratory process and its bodily parts, seems to dominate.

Aetiology and Additional ‘Co-factors’

References to patient profiles are rare in the Hippocratic sources in all cases. When phrenitis is involved, there are only a handful of such references. At Aph. 3.30 (408.11–13 Magdelaine = 4.500 L.) we read that phrenitis (like many other diseases) tends to occur after age twenty-five, and at Prorrh. I, 9 (76.2 Polack = 5.512 L.) that ‘phrenitic illnesses in the young end with spasms’. At Aph. 7.82 (475.11–13 Magdelaine = 4.606 L.) there is a distinction in the prognosis of the disease between middle-aged and younger subjects: ‘If phrenitis attacks those beyond forty years of age, they rarely recover; for the risk is less when the disease (hē nosos) belongs to one’s constitution (oikeiē tēs physios) and to age’ – which seems a reasonable observation, to be extended to other diseases as well. Otherwise, there is no reference to age and no pattern in terms of gender in cases of this disease.

The question of aetiology is also complex and marred by the pitfalls of anachronism. A precise, systematic account of causes as a fundamental chapter in pathology is not necessarily a feature of nosology at this stage in ancient science, when clinical description and collation of signs and symptoms (and, in second place, prognosis and therapeutics) occupy the most space. But there is some consistency in the limited information the Hippocratics give in this regard – or rather a number of fixed explanatory patterns: phrenitis ‘arises from bile (hypo cholēs) when, having been set in motion, it settles against the inward parts and the diaphragm (pros ta splanchna kai tas phrenas prosizēi)’, according to Aff. 10 (20.6 Potter = 6.218 L.). At Aff. 12 (22.11–16 Potter = 6.220 L.) the conversion from a simple winter fever to acute diseases such as phrenitis is better illustrated in terms of cause and consequence: ‘When, with phlegm and bile set in motion, what is beneficial is not administered to the patient’s body, the phlegm and bile collect together and fall upon some chance part (hēi an tychēi) of the body, and pleuritis or phrenitis or peripleumoniē result.’ Bile and phlegm are here the culprits, but the locus afflicted varies.²⁶

Diseases 1.30 (86.19–88.13 Wittern = 6.200 L.), by contrast, presents phrenitis in a haematocentric frame. Here as well, however, bile is primarily to blame: ‘When bile that has been set in motion enters the vessels and the blood, it stirs the blood up, heats it and turns it into serum, altering its normal consistency and motion; now the blood heats the rest of the body too’. Derangement and high fever follow. At Morb. 1.34 (92.7–18 Wittern = 6.204 L.) the decline towards death in phrenitis is explained as mostly caused by lack of nourishment, since the patients, being deranged (paraphroneontes), accept no food and waste away. At the origin of the cold, fever and pain is the fact that ‘when the blood in the vessels is cooled by the phlegm, it migrates and contracts into a mass at one time in one part, at another time in another part, and trembles. Finally, everything becomes cold and the person dies.’ Here phlegm is the pathogen.

In sum, the direct cause or pathological picture, when given, seems to be humoral: in most versions it is bile pressing against sensitive body parts, but it can also be phlegm (and bile) stirring the blood up and heating it, or phlegm alone cooling the blood excessively. When we read at Prorrh. I 31 (78.7–8 Polack = 5.518 L.) that ‘what is salivated in cases of phrenitis with chills is vomited back up dark’, dark humours seem to be implied.

Therapy and Prognosis

In these earlier sources, therapy for phrenitis is not always addressed in our sources as part of the discussion of the disease. One informative passage in this respect, however, is Aff. 10 (18.20–21 Potter = 6.216–18 L.), where phrenitis is treated as a disease of the central body cavity. The instructions are as follows: ‘For the pain, treat the patient with the same measures as in pleurisy’, that is, with ‘a medication to remove phlegm and bile’ from the painful area; ‘clean the cavity downwards by giving a medication and cooling it with an enema …; administer drink and gruel’ (14.18–16.2 Potter = 6.214 L.). For phrenitis,

give a medication for the cavity, and conduct the rest of the treatment along the same lines, except with regard to drink. As drink, give anything you wish except wine; give vinegar, honey and water, or water alone. Wine, however, does not benefit a deranged mind … It is of benefit in this disease to wash with copious hot water from the head downwards. For as the body is softened, sweating increases, the cavity discharges, urine passes, and the patient gains more control over himself. (18.20–20.5 Potter = 6.218 L.)

The connections between the therapy for phrenitis and that for related winter-chest diseases are evident at Nat. Hom. 5 (212.4 Jouanna = 6.78 L.), where we read that emetics and clysters – both purging methods – are to be used in periods when more phlegm is engendered, such as winter, when ‘diseases that attack the head and the region above the diaphragm, phrenes’, occur.

A second instructive passage, along similar lines, is Morb. 3.9 (76.24–27 Potter = 7.128 L.): ‘Warm this patient with moist fomentations and with drinks other than wine. If he can stand up, cleanse him upwards; he must bring up material by coughing and expectoration just as in peripleumoniē. If he fails to do this, prepare the lower cavity in order to evacuate it. Moisten the patient with drink, for that helps.’ Purging and cleansing the body cavities is central, as is the diaphragmatic location.²⁷

The clinical texts do not add much in the way of a clear protocol for phrenitis: the phrenitic butcher in Acanthus in Epid. 5, 52 (7, 71), 24.6–9 Jouanna = 5.236–38 L. (92.13–17 Jouanna = 5.432 L.), who developed a hump (kyphōma) after phrenitis,²⁸ is treated with what appear to be soothing measures: ‘No drugs helped, but red wine and eating bread, refraining from bathing, being massaged with restraint, and being warmed without much fomentation but gently.’ At 7, 71 there are more details about ‘being rubbed with oil, warming the back, not excessively, by means of a small, gentle fire’.

Competing Localizations

Some of the quotes above, which describe therapy to the head and chest alike, nicely illustrate the problematic juxtaposition of chest and head vis-à-vis phrenitis, despite its deep-seated connection with the cavities of the torso. This topic is central to the history of the disease and very influential in the development of Western psychiatry. Indeed, it runs through the whole history of phrenitis and will emerge as a leading theme in our reconstruction of it in this book.²⁹

The location in the chest is dominant in the Hippocratics, with an obvious association with winter diseases localized in the torso as the affected area. This is evident in the involvement of the respiratory system (as we would define it), as is also the case with pleuritis and peripleumoniē, and in the general location of the phrenes in the chest (whether we interpret them as the diaphragm or identify them more vaguely with the body cavities of the upper chest), here intended in an entirely material sense as ‘body part’. The association of chest, breathing and breathing affections with cognitive implications, which is traditional in Greek medicine from an early date, is also important here.³⁰ This localization explains the link between expectoration full of mucus and derangement found in some texts, most clearly at Prorrh. I, 6 (75.11–12 Polack = 5.512 L.): ‘Frequent expectoration, if another sign is present as well, indicates phrenitis (anachrempsis pyknē ge, ēn dē ti kai allo sēmeion prosēi, phrenitika).’³¹ It should not be forgotten that peripleumoniē too can cause derangement, as described at Morb. 3.15 (82.22–25 Potter = 7.136 L.) as part of a picture that shares a great deal with that of phrenitis: ‘There is violent fever, and the patient’s breathing is rapid and hot; he is distraught, weak, and restless (aporiē, kai adynamiē, kai riptasmos), and beneath his shoulder blade he suffers pain that radiates toward his collarbone and nipple. He has a heaviness in the chest, and he is deranged (paraphrosynē).’

Second, the hypochondriac affliliation, as is most evident in Aff. 10 (18.14–20.11 Potter = 6.216–18 L.):³² ‘In phrenitis, there are at first mild fever and pain over the hypochondrium … phrenitis arises from bile, when, having been set in motion, it settles against the inward parts and the phrenes.’ As the name obviously suggests, this area of the body (here phrenes, elsewhere hypochondrion) most often emerges as the pathological place in the course of this illness, although it is unclear what function would be impaired as a consequence. This point is explored in the next paragraph, but it is obvious that the name and various discussions of the disease point to a central role for the diaphragm and the upper cavity of the torso, where the lungs and the heart are located.

Third, the link with the torso also includes lower, gastric and hypogastric localizations. This appears to be a possibility at Coac. 405 (204.26–28 Potter = 5.676 L.): ‘If persons with pain in the side (meta pleurou algēmata) who do not have pleurisy evacuate favourably thin stools, they turn out to be phrenitic.’ The liver is also important, although it is never mentioned directly as a body part affected by phrenitis. The phrenes, in fact, are often described as intersected by the vein that leads to the liver, and the author of Int. 48 (230.18–236.20 Potter = 7.284–88 L.), who describes the mental effects of a swollen liver pressing against the diaphragm, is in line with the tradition which makes the liver the seat of the appetitive soul. This tradition runs from Plato to Galen’s re-elaboration³³ and is rooted in a vast popular tradition.³⁴ It was not only the whole chest, then, that was a mental centre in Greek culture as important as, or even more important than, the head;³⁵ various parts of the torso, from the upper, cardiac cavity to the gastric area, were also fundamental.

Fourth, the head (as a whole comprising skull, membranes, brain and face, or with particular reference to only one of these parts) or caput (by which I mean the uppermost section of the body, as its general termination in an upward direction) is clearly implicated. Headache as a symptom³⁶ is of great interest, since it underlines the anatomical dissonance that constitutes the backbone of the history of our disease: phrenitis is localized in the chest, but is also accompanied by head-and-neck symptoms.³⁷ Intriguingly, clinical material – the patient descriptions – rather than doctrinal nosology yields the most information regarding the head as affected in our disease. At Epid. 7, 112 (112.3–9 Jouanna = 5.460 L.) Polyphantes of Abdera has an illness characterized by a ‘phrenitic derangement’ which includes ‘continuous headaches’ (112.5–6 Jouanna = 5.460 L., ou pauomenou de tou algēmatos tēs kephalēs), as does the next patient, the maidservant of Eualcides, who had headaches (112.10–13 Jouanna = 5.460 L.). She ‘became phrenitic and died with powerful convulsions’. So too at Epid. 3, 17, case 4 (98.3–4 Jouanna = 3.118 L.) a phrenitic patient (ho phrenitikos) has ‘painful heaviness of the head and neck’.

Headache itself as a pathological entity is thematized in Coac. 116 (130.3–4 Potter = 5.608 L.), where headache in acute fevers is said to develop into phrenitis (es to phrenitikon periistatai) unless there is a haemorrhage through the nostrils. The connection with the head is also reinforced visually: at Coac. 210 (154.5–6 Potter = 5.630 L.) ‘contraction of the forehead (metōpou synagōgē)’ is phrenitikon, a phrenitic sign, in association with the idea that ‘a good colour of the face in association with sullenness in acute disease (prosōpou euchroia kai skythrōpotēs en oxei) is a bad sign’.

Finally, in delocalized terms blood can also be seen as a locus of the disease. This episodic doctrine is specific to the haematocentric views exposed in Diseases 1, which discusses aetiology, as we have already seen. In Morb. 1.30 (86.19–88.4 Wittern = 6.200 L.), for example,

Phrenitis is as follows: the blood in a human being contributes the greatest part to intelligence … (pleiston xymballetai meros synesios) … Therefore, when bile that has been set in motion enters the vessels and the blood, it stirs the blood up, heats it and turns it into serum … Due to the magnitude of his fever, and because his blood has become serous and abnormal in its motion, the person loses his wits and is no longer himself.

Morb. 1.34 (92.7–9 Wittern = 6.204 L.) elaborates on this, basing its explanation on the corruption of blood: ‘Inasmuch as [phrenitics’] blood is corrupted and does not move in its normal way (hate tou haimatos ephtharmenou te kai kekinēmenou ou tēn eōthuian kinēsin), they are deranged throughout the disease … ’ Can this isolated but representative circulatory account be interpreted as a more holistic option?

These should not be seen as contradictory or rival doctrines, or as uncertainties in medical explanations of the disease. As van der Eijk explains with reference to the more theoretical question of the discordant ‘theories of mind’ traceable in the Hippocratic Corpus, these different models are scarcely exclusive³⁸ and can in fact coexist in the same account. Even the simplified dialectic encephalocentrism-cardiocentrism plays a deeper role in our disease.

The Elusive Connection: phrenes, phrenitis and Mental Life

The dominant localization of phrenitis points to the region of the body to which the phrenes belong, as their name suggests (although hypochondrion is sometimes used instead): the ‘diaphragm’, the sheet of muscle which in modern anatomy separates the thoracic cavity that contains the lungs and the heart from the gastric cavity, and which plays a role in respiration by contracting, increasing the lungs’ volume and allowing them to be filled with air. This is a visually detachable part of the body,³⁹ tears or perforations of which can impact the respiratory functions, causing orthopnea (shortness of breath when lying down, and coughing).⁴⁰ As observed in Chapter 1, however, phrenes is also synonymous with mental life and mental soundness in classical Greek, and is a very common term which ties in with another fundamental of the disease phrenitis, its mental quality.

A combination of the following moves is thus key to explaining and assessing the localization of our disease: first, an association between phrenes as body part and phrenitis; second, an emphasis on the disease as mental; third, a contextual awareness of the phrenes as an organ of mental life or a metaphorical expression to indicate the mind. We shall see that these three points are never combined in early texts, which is surprising, as is the avoidance of (para)phronein and other cognates of phren- to describe the mental import of the illness. Let us consider three points one after the other, in order to explore their overlap: (1) the general notion of phrenes as body part in the Hippocratic texts; (2) reference there to a role of the phrenes in phrenitis in a mere locative sense, unconnected with the mental sphere, particularly in relation to vitality; (3) reference to a role of the phrenes in mental pathology, especially in phrenitis.

The Notion of phrenes in the Hippocratic Texts

As noted in Chapter 1, the noun phrēn/phrenes traditionally had two senses, which are often indistinguishable in archaic literature: a place in the body, with a locative-anatomical and strictly physiological meaning (the diaphragm; either the upper or the gastric cavity in the torso; the chest in general), on the one hand, and the mental faculties and/or a person’s character and ‘self’, on the other. In our texts, in exact countertendency to all other literature of the period, phrenes – the singular phrēn appears only once – is rarely employed in the second sense, whether in association with phrenitis or not.

At first glance, it might appear striking that this otherwise common noun, a stock term for the mind in archaic and classical literature, is not central to discussions of mental pathology in the Hippocratic sources. The datum appears less surprising if we consider that these texts systematically avoid association with traditional concepts and formulations.⁴¹ Indeed, the term phrenes/φρένες is mentioned and not avoided: it is the selectiveness of the usage that is significant. Not only do Hippocratic discussions of mental life and its disorders ignore the traditional ‘Homeric’ sense of phrenes as the seat of emotions and thought, as well as a bodily location, but even in descriptions of the disease phrenitis itself the phrenes appear in indirect, almost reluctant association.⁴²

Non-Mental phrenes in the Hippocratic Texts

Most Hippocratic uses of phrēn/phrenes are devoid of any association with the mental – or indeed with phrenitis – and clearly refer to an anatomical part, the diaphragm. The emphasis is on its ‘partitioning’ function, its position as ‘dia-phragm’ between the upper and lower cavities of the torso and its intermediate location with respect to the main veins flowing from the heart down through the liver. At Vet. Med. 24 (152.13–15 Jouanna = 1.634 L.) the nature of the phrenes is described as similar to that of other tissues rich in blood, such as the liver, insofar as they are exposed to pain and alterations (e.g. abscesses or tumours): ‘Violent pain, but much less severe, is also felt under the diaphragm (hypo phrenas). For the diaphragm is an extended, broad, resistant substance (diatasis … phrenōn plateiē kai antikeimenē), of a stronger and more sinewy texture, and so there is less pain. But here too pain and tumours occur.’

As might be expected, different treatises and topics reflect different interests in the diaphragm. In On Joints the anatomical part under the ribs is in question, as also (one is led to believe) at Artic. 41 (164.3 Kühlewein = 4.176 L.; 164.14 Kühlewein = 4.178 L.), where ‘above the phrenes’ (anōterō tōn phrenōn) indicates the position of a malignant lump in the spinal vertebrae.⁴³ Elsewhere, respiration is emphasized: at Progn. 5 (14.4 Jouanna = 2.122 L.) we read that ‘rapid respiration indicates pain or inflammation in the parts above the diaphragm, hyper tōn phrenōn’. Coac. 255, 164 Potter = 5.638 L. argues that ‘frequent and shallow breathing indicates an inflammation and pain in the parts above the diaphragm (en toisi hyper tōn phrenōn topoisi). If breaths are deep and come at long intervals, they indicate a disordering of the mind or convulsions (paraphrosynēn ē spasmon); if they are cold, they signal death.’

Anatomically, the gastric and lower cavities are also referenced for the sake of their position relative to the phrenes. At Progn. 12 (35.2–5 Jouanna = 2.142 L.) persistently thin, crude urine suggests an abscess in the area below the diaphragm (es ta katō tōn phrenōn chōria), while at Aph. 4.18 (413.3–4 Magdelaine = 4.506 L.) ‘pains above the phrenes, hyper tōn phrenōn, indicate the need for upward purging’. At Breaths 10 (118.8 Jouanna = 6.106 L.) ho phragmos tōn phrenōn, ‘the closure of the phrenes’, works as a barrier impeding the upward flux of fluids in the body.

The sensitivity of the phrenes and the danger represented by pain in this area⁴⁴ – the topic of sensitivity, which already emerged above – are described at Progn. 19 (54.6–55.1 Jouanna = 2.164 L): ‘Pains occurring with fever in the region of the loins and lower parts, if they leave the lower parts and attack the diaphragm (ēn tōn phrenōn haptontai), are very deadly’, if other bad symptoms supervene. ‘But if, when the disorder jumps to the diaphragm, the other symptoms that supervene are not bad, confidently expect an empyēma (a pocket of pus accumulated inside a body cavity).’ A primary affection moving to the phrenes, as opposed to a momentary reaction, is deadly.

In a discussion of barley gruel at Acut. 5 (42.14–15 Joly = 2.258 L.) it is said that gruel should not be offered to a stomach that is full: the consequence will be that ‘it dries the lung, besides causing discomfort in the hypochondria, the hypogastrium and the diaphragm (phrenes)’. At Aph. 7.54 (469.10–470.2 Magdelaine = 4.594 L.) pain is associated with the indication of a space or cavity in cases ‘where phlegm is confined between the midriff and the stomach, causing pain because it has no outlet into either cavity’.

Finally, the phrenes are often mentioned in order to identify the relative position of vessels in the torso. In the Epidemics (where, interestingly, the traditional term is otherwise avoided), at 2.4.1 (62.12 Smith = 5.120 L.) the liver vein is said to reach the heart through the diaphragm (dia phrenōn), while at 2.4.1 (64.17–18 Smith = 2.122 L.) phrenes are said to be attached to the liver and difficult to separate from it. Later, at 2.4.1 (64.20–22 Smith = 5.124 L.), phrenes are localized ‘at the vertebra below the ribs where the kidney separates from an artery’,⁴⁵ and are said to ‘bestride the artery’; many branching veins are described as running through the diaphragm (dia tōn phrenōn), and so forth, with various similar remarks about position with respect to the liver and spine and the presence of blood vessels. In Loc. Hom. 3 (42.7–11 Joly = 6.282 L.) the phrenes are also mentioned in order to pin down the position of vessels, here the vena cava (hē … koilē phlēps), which ‘passes over the surface/through (?) the phrenes and the heart, between the two halves (?) of the phrenes’ (42.8–11 Joly = 6.282 L., metaxy tōn phrenōn).⁴⁶ At Morb. Sacr. 3 (11.20–21 Jouanna = 6.366 L.) the vein running from the liver is also said to ‘stretch upwards toward the right diaphragm and lung (dia tōn phrenōn kai tou pleumonos tōn dexiōn)’.

The Vitality of the phrenes

The vital relevance of the phrenes is also noticeable. At Coac. 107 (128.1–4 Potter = 5.604 L.) it is a deadly sign in patients with a fever when ‘pains arising in the loins and the lower parts … seize the diaphragm at the same time they resolve through the lower parts (ekleipousai ta katō)’: disturbance to this part is definitely fatal. At Aph. 6.18 (452.1–2 Magdelaine = 4.566–68 L.) the phrenes are one of the body parts where wounds are deadly (along with other organs; repeated at Coac. 499, 230.3–15 Potter = 5.698 L.), while Morb. 1.3 (6.18–8.1 Wittern = 6.142–144 L.) declares that injuries ‘in the brain, spinal marrow, cavity, liver, phrenes, bladder, blood vessel or heart’ are bound to cause death.⁴⁷ Phrenitis in pregnant women is also inevitably fatal (Morb. 1.3, 8.3–7 Wittern = 6.144 L.).

Sensitivity and vitality obviously indicate the relevance of phrenes to mental life, especially in the biological frame of the materialistic concept of the soul in ancient medicine. The strongest indicator of this is, e contrario, given by the author of Sacred Disease as he forcefully and explicitly refutes the notion that the mental faculties should be located in the phrenes (Morb. Sacr. 17, 30.3–17 Jouanna = 6.392 L.):

Wherefore, I say that it is the brain which interprets the understanding. But the phrenes (= the diaphragm) have obtained their name from accident and usage (tēi tychēi … tōi nomōi), and not from reality or nature (tōi eonti … tēi physei), for I know no power which it possesses, either as to sense or understanding, except that when a man is affected with unexpected joy or sorrow, it throbs and produces palpitations, owing to its thinness, and since it has no belly to receive anything good or bad that may present itself to it, it is thrown into commotion by both of these, due to its natural weakness. It then perceives beforehand none of the things which occur in the body, but has received its name vaguely and without any proper reason, like the parts about the heart, which are called auricles, but which contribute nothing towards hearing.

This intriguing passage shows full awareness that the etymology is unreliable, and explicitly attacks the traditional psychological interpretation of the phrenes, which are here clearly identified with the membrane of the diaphragm (as the reference to their thinness and throbbing suggest). The author claims that they are affected by the individual’s emotions and are not their origin, hence their apparent reactivity in moments of distress.

In Morb. Sacr. 7 the term phrenes returns as a simple location that explains the discharge of excrement during epileptic attacks: ‘The liver and upper bowel are forced against the phrenes (pros tas phrenas), and the mouth of the stomach is intercepted’ (16.8 Jouanna = 6.374 L.).⁴⁸ So too at Morb. 3.14 (82.7 Potter = 7.134 L.), in cases of ileus it is recommended that one cleanse the upper cavity and ‘cool the region above the phrenes (ta anō tōn phrenōn)’. Morb. 3.16 (86.22–96.12 Potter = 7.142–56 L.) is devoted to forms of pleuritis, which are diseases contiguous to peripleumoniē and phrenitis, as was noted, with similar therapeutic recommendations (as explicitly at 90.9–10 Potter = 7.146 L.). The following therapy, for example, is described: drying the thorax by wrapping it in a plaster soaked in moist Eretrian earth, and then cauterizing or incising ‘as close to the phrenes as possible, but sparing the phrenes themselves’ (94.25–28 Potter = 7.154 L.).

This importance of the phrenes in regard to vitality and survival,⁴⁹ as well as the importance assigned them by medical (and non-medical) authors generally, I suggest, shows the persistent weight and silent influence of the archaic meaning of the word as the larger upper torso region (lungs, heart, and the area more generally) that serves as the seat of life and consciousness in early literary sources.⁵⁰

Mental phrenes?

Direct or exclusive references to phrenes as a mental organ are very rare in the Hippocratic texts, and Langholf is right, at least for the majority of cases, to define such occurrences as ‘conventional’.⁵¹ The non-bodily use of phrenes as ‘mind’ or ‘mental soundness’ appears in Regimen in Acute Diseases (sp.) 1 (68.11–12 Joly = 2.396 L.) in an expression indicating the patently mental symptom ‘derangements of the mind’, parallaxies phrenōn. The disease in question here is less fully specified than phrenitis, although it may include it: a burning fever (kausos). At Prorrh. II, 9 (244.11–27 Potter = 9.28 L.) the phrenes are mentioned as a function that can be impaired: in a discussion of the sacred disease, it is said that the physician should ‘take care if his patients are young and active, unless their mind has some defect (phrenes … ti kakon echousin) or the patient is paralysed’.

More ambiguously balanced between a literal and a metaphorical or abstract meaning is Acut. 14 (57.19–23 Joly = 2.332 L.), where sweet wine is said to ‘go less to the phrenes’ and to be ‘less affecting the phrenes’, hēsson phrenōn haptomenos, compared to oinōdea, the vinous type of wine. This suggests that phrenes are here the mental faculties that red wine affects more severely; Jones translates the word as ‘brain’!⁵² A similar mental meaning seems to be implicit a bit below (58.24 Joly = 2.336 L.), where a change from white wine is encouraged in cases of ‘no affection of the mind’ (mēde phrenōn hapsis). At Acut. 17 (64.12 Joly = 2.360 L.), if phrenōn hapsis is suspected, complete abstinence from wine is to be preferred. The same expression, phrenōn hapsis – here together with headache – is used to describe a gynaecological pathology at Mul. 1.63 (8.128 L.): ‘if she has pain in the head and there is affection of the mind (ei kephalēn algeoien kai phrenōn eiē hapsis)’.

In a material sense, the phrenes are implicitly the seat of reason and mental functions at Mul. 2.200 (8.384 L.): the womb is perceived (presumably by the woman) as pressing against the phrenes (hypo tas phrenas dokeosin hizesthai); she ‘immediately becomes speechless, her hypochondria hardens, she suffocates, gnashes her teeth and cannot hear when she is called’. The discussion of the type of epilepsy (epilepsiē) that affects young virgins in Girls (Virg. 2.8, 22.23–24 Lami = 8.468 L.) cites blood flooding into and out of the kardiē (located in the chest, if only later identified with the heart) and phrenes as causing numbness and derangement. This area (topos) of the body, the chest, is said to be ‘critical for madness and mania (epikairos es te paraphrosynēn kai maniēn)’. Notably, it is in these two gynaecological texts that the topos of the triggered phrenes most closely approximates the traditional, poetic representations, while in all the others the mediation of an anatomical or physiological grid is inserted. Girls as a whole is perhaps the most ‘literary’ text of the Hippocratic collection.

Pathology of the Diaphragmatic Region

If we decide, in the interest of a medical-historical rather than a merely philological study, to identify phrenes with the diaphragm more firmly in the medical material than is possible in the poetic, and explore this body part rather than the specific term, other testimonies become relevant to associating the region with mental affection. At Progn. 7 (17.10–18.3 Jouanna = 2.126 L.), for example, throbbing in the hypochondrium (en tōi hypochondriōi) signals thorybos, ‘trouble’, or paraphrosynē, ‘delirium’. Galen (Comm. Hipp. Progn. I 28, 245.24–246.1 Heeg = 18b.89 K.), commenting on this passage, says that the diaphragm is most prone to madness (diaphragma paraphrosynēn hetoimotata pherei) and adds – reversing the history of the concept – ‘for which reason the ancients called it phrenes (φρένας)’. See also Epid. 3, 17, case 16 (112.4 Jouanna = 3.146 L.), where ‘tension of the hypochondrium’ (hypochondriou entasis) is present in a case of illness involving wild derangement, or the similar case at Epid. 7, 25 (66.22–23 Jouanna = 5.394 L.), where the feverish and wildly deranged wife of Theodorus displays a ‘much swollen right hypochondrium’. In the nosological passage in Internal Affections 48 a ‘thick disease’⁵³ is described, with complex, obviously psychopathological consequences. These arise precisely when the liver swells and presses against the phrenes (anaptyssetai pros tas phrenas), causing pain to ‘immediately attack the head, especially the temples’, with mental consequences (Int. 48, 230.21 Potter = 7.284 L.). The patient’s condition deteriorates as the liver pushes further against this part (232.14–15 Potter = 7.284 L.).

Among the ancient scientists who prioritized the mental function in explanations of the etymology of phrenes and, preceding Galen, associated the noun with the verb of reasoning (phroneō, phrenoō, etc.) is Aristotle, who considered a localization in the chest fundamental in a cardiocentric frame. Discussing the diaphragm at PA 672b31, the philosopher writes: ‘For when, because of their proximity, the midriff absorbs the hot, residual moisture, straightaway it manifestly disturbs thought and perception (tarattei tēn dianoian kai tēn aisthēsin), which is also why they are called phrenes, as if they partake in some way in thinking (hōs metechousai ti tou phronein).’

Aristotle always discusses the phrēn/phrenes as diaphragm, clearly describing this as the bodily part that functions as a ‘belt’, diazōma, in the torso (HA 496b11; 506a6; 514a36; PA 672b11), a ‘partition’, phragmon (PA 672b20). To explain the functioning of this bodily part, he returns several times to the topic of laughter as specifically human: ‘They say that it happens also in the case of battle wounds damaging the area around the diaphragm (peri tas phrenas) that the person laughs because of the heat deriving from the wound’ (PA 673a11).⁵⁴ The pseudo-Aristotelian Probl. 35.6 (965a15–17) similarly observes that ‘laughter is a sort of frenzy or deceit’ and poses the question: ‘Is this why people struck in the midriff (eis tas phrenas) laugh? For it is not any chance part (ho tychōn topos) with which we laugh.’ Here, as in Sacred Disease, the reactive, ‘vibrating’ nature of the diaphragm seems to be at issue.⁵⁵ A passage in the Timaeus (69e2–70a2) discusses the ‘mortal part of the soul’ as located in the chest and thorax (en dē tois stēthesin kai tōi kaloumenōi thōraki to tēs psychēs thnēton genous enedoun) and assigns the phrenes a key role in separating (70a1–2) the upper part of the torso, the one that contains the soul, from the appetitive part located in the stomach; the phrenes themselves are merely an inert fence between the two.

In various ways, then, ancient biology and philosophy, as well as medicine, reworked the heritage of traditional physiology and psychology, variously recognizing the relevance of the phrenes to mental life in the body, whether pathologically (the Hippocratics, Aristotle) or within the anatomical schema of their representation of the embodied soul (Aristotle, Plato).

Pathology of the phrenes/Diaphragmatic Region and phrenitis

What happens when not only derangement and fever, but phrenitis explicitly is mentioned in association with phrenes?⁵⁶ As noted earlier, it is difficult to identify such a precise conceptualization of the phrenes as mental centre of phrenitic affection, and the neat account a modern reader would expect is ultimately missing. At Morb. 3.9 (76.20–29 Potter = 7.128 L.), in a section discussing possible onsets of forms of the disease, phrenitides, the mental part is treated as very important and reference is made to phrenes. This localization is entirely dissociated, however, from the idea that these might be mental organs:

Kinds of phrenitis (phrenitides) can also develop out of another disease. Patients suffer as follows: they experience such pain in the phrenes (tas phrenas algeousin) that they will not allow themselves to be touched, there is fever, they are deranged (ekphrones eisi), they stare fixedly, and for the rest they resemble patients with pneumonia who are deranged (toisin en tēisi peripleumoniēisi, hotan ekphrones eōsi).

The connections between phrenes, diaphragm, the chest generally and phren- as aural semantic sphere mean that they are coexistent and implicit, never clearly defined. Every passage places greater weight on one vertex of this polygonal figure; here, for instance, the phrenes are concretely sore to the touch, and the derangement is indicated by the cognate ekphrones. One cannot consider only one of these accounts in isolation, as Potter observes in his comment on this passage: ‘Therefore, I tend to understand phrenitis in terms of the specific organ, that is, in the literal sense of “disease of the diaphragm”. In fact, the author probably understands the term phrenitis to mean both a disease of the diaphragm and insanity.’⁵⁷ The choice between these items – ‘concrete illness of the diaphragm’ vs ‘madness’ – is not presented as such by the classical texts, and the distinction between the different components, physiological and psychological, is entirely our own.

A pain in the area of the chest where the phrenes reside is also referred to in the discussion of phrenitis in Aff. 10 (18.14–19 Potter = 6.216 L.), although another localization in the torso and another term, hypochondria, is used there:

In phrenitis, at first there are mild fever and pain over the hypochondria (the hypochondrium), more on the right towards the liver. When the fourth or fifth day arrives, the fever becomes more intense, as do the pains, the colour becomes somewhat bilious, and the patient’s mind becomes deranged

(kai tou nou parakopē).

Later on in the same passage, the phrenes are mentioned in a way that appears to match the locations just indicated: Aff. 10 (20.6–11 Potter = 6.218 L.) explains phrenitis as an overflow of bile ‘into the internal organs and the phrenes’ (pros ta splanchna kai tas phrenas). In contrast to the previous part of this section, however, no mental function is mentioned, and all that seems to be intended is a general localization in the chest. The same could be said about Morb. 3.9 (76.20–23 Potter = 7.128 L.), where a secondary phrenitis developing from another disease is described. Here too patients ‘experience such pain in the diaphragm (paschousi … tas phrenas) that they will not allow themselves to be touched, there is fever, they are deranged’.

In accounts which foreground blockage of fluids, the pathological picture resembles the makeup of epilepsy as described in Morb. Sacr. 7 (15.14–23 Jouanna = 6.372.5–6 L.). At Flat. 10 (117.11 Jouanna = 6.106 L.), for example, the phrenes are clearly said to constitute an impediment against the extravasated blood in the chest finding an outlet downwards, causing it to accumulate and putrefy. The process of putrefaction of the phlegm accumulated in the upper torso (epi tōn phrenōn) is described in detail at Morb. 1.15 (36.6–7 Wittern = 6.164 L.), although this disease has no mental implication. At Morb. 1.19 (50.16 Wittern = 6.174 L.) the pathology of the tubercles in the lung is described, and the phrenes are again the floor that stops or receives a putrid fluid (epi tas phrenas). Once again, however, there is no mental involvement.

The Coan Prenotions offer the two best approximations of an association between phrenes and the mental, and perhaps even the disease phrenitis. The first, Coac. 255 (164.5–8 Potter = 5.638 L.), is a rare example of phrenes as the locus affectus of a mental ailment: the passage opens by saying that ‘frequent and shallow breathing indicates an inflammation and pain in the parts above the diaphragm. If the breaths are deep and at long intervals, they indicate a disordering of the mind (paraphrosynē) or convulsions.’ At Coac. 571 (250.6–19 Potter = 5.716 L.) ‘apostasis and pain, especially in the region below the diaphragm (hypo phrena)’, are foretold by a number of signs, ‘with or without fever’; among these, ‘colourless urine with dark suspended material, in association with restlessness and sleeplessness, indicates phrenitis’. In this passage the cognate terms phrēn and phrenitis are far apart; the affected area of an illness with a possible phrenitic outcome, however, is clearly said to be in the region of the phrēn. It is also noteworthy that this is the only occurrence of the singular in the Hippocratic Corpus, perhaps with more explicit reference to a mental effect.⁵⁸

In sum, throughout the classical material phrenitis is characterized by a repertoire of fixed bodily signs, which are accompanied by mental derangement and, in terms of localization, repeatedly involve the chest or the hypochondriac area of the body, often (but not always) indicating this part with the word phrēn/phrenes, as well as the head. The aetiology, on the other hand, when indicated, is humoral or blood-related and involves no reference to the state of the phrenes or to any phrenocentric or cardiocentric theory of mind. Only in the case of the reference to blood, in fact, is there any direct mention of a place – albeit fluid – as a diseased centre of cognition.

Conceptualizing the Disease Entity: Co-morbidity and Affinity to Other Diseases

A quick survey shows that the ailment called phrenitis (hē phrenitis, ἡ φρενῖτις) as a substantive (as opposed to ‘phrenitic’ as an adjective describing types of patients or manifestations) is mentioned in Hippocratic medicine far more frequently than melancholia,⁵⁹ making it a prominent example of a disease label ‘qua label’ in which the mental aspect plays an important role. This testifies to a greater conceptualization of phrenitis as a disease in the modern sense of the term already in the Hippocratics, a quality also evident to Galen, who repeatedly picks this disease as an ontologically powerful example.⁶⁰ This ‘ontological’ robustness⁶¹ is also reflected in the fact that phrenitis and phrenitics early on enjoy a fixed set of relations to neighbouring diseases and are clearly placed as regards aspects of co-morbidity and classification. These are all signs of stronger conceptualization, taxonomic reliability and, as we shall see, the productivity of the concept for the development of medical ideas.

At Morb. 1.30 (88.7–11 Wittern = 6.200 L.), for example, we are told that ‘patients with phrenitis most resemble melancholics in their derangement (kata tēn paranoian), for melancholics too, when their blood is disordered by bile and phlegm, have this disease and are deranged (paranooi ginontai) – some even rage (mainontai)’. At Judic. 41 (13.11–12 Preiser = 9.290 L.) the association between melancholy and phrenitis is again discussed: ‘In persons suffering from melancholic conditions along with phrenitic ones, haemorrhoids are beneficial.’ The Coan Prenotions also preserve differential prognostic signs for phrenitis, while at 93 (124.21–22 Potter = 5.602 L.) the treatise mentions salivation (a sign connected to our disease): ‘Patients who become deranged with melancholia tremble and salivate: are they given to phrenitis (ēra phrenitikoi)?’ and so forth.

More concretely, phrenitis in the Hippocratic sources is contiguous to other diseases localized in the chest: at Acut. 5 (37.21–38.1 Joly = 2.232 L.) it is mentioned alongside pleuritis, peripleumoniē and burning fevers as examples of ‘acute’ diseases, in which fever is generally continuous;⁶² Aph. 3.30 (408.11–13 Magdelaine = 4.500 L.) lists phrenitis among the diseases of the young (next to asthma, pleuritis, ardent fever, kauson, etc.); and at Aph. 7.12 (462.3 Magdelaine = 4.580 L.) a transformation of peripleumoniē into phrenitis is said to be a bad development (kakon). At Morb. 1.3 (8.3–7 Wittern = 6.144 L.), again, high fever, pleuritis and phrenitis are mentioned close to one another, as also at Morb. 1.3 (10.5–6 Wittern = 6.144 L.) and Morb. 1.3 (10.7–8 Wittern = 6.146 L.), where possible conversions among diseases located in the chest are surveyed: pleuritis into ardent fever, and phrenitis into peripleumoniē.⁶³ Aff. 6 (14.7–13 Potter = 6.214 L.) makes the taxonomic point explicit by speaking of a group of ‘diseases of the cavity’, again including pleuritis, peripleumoniē, burning fever and phrenitis in a common group; these are said to be more dangerous in winter. Peripleumoniē, pleuritis and phrenitis are also mentioned together at Aff. 12 (22.7–20 Potter = 6.220 L.), where it is made clear that the same cause, a displacement of phlegm and bile, can engender all of them depending on ‘where [the fluid] happens to fall’ (ēn an tychēi).

A therapeutic discussion in Diseases 3 reinforces this grouping of chest and winter diseases. Interestingly, this chapter opens with a description of peripleumoniē that might also fit an account of phrenitis in terms of time-frame and material. The patient suffers from violent fever and breathes rapidly; ‘he is distraught, weak and restless, and beneath his shoulder blade he suffers pain that radiates towards his collar bone and nipple; he has a heaviness in his chest; and he is deranged (kai paraphrosynē)’ (Morb. 3.15, 82.25 Potter = 7.136 L.). The therapeutic measures too are explicitly said to be identical for phrenitis and pleuritis (Morb. 3.15, 84.26–28 Potter = 7.140 L.).

In the patient cases, finally, although diagnoses are only rarely mentioned, phrenitis is one of the few diseases that features more than once. Apart from individual cases, a collective description at Epid. 1, 18 is very instructive about the categorization of our disease. Burning fevers (kausoi) ‘at the equinox and about the setting of the Pleiades’ are discussed, and among these, cases of phrenitis are noted as frequent and especially dangerous and deadly (25.8–12 Jouanna = 2.650 L.). These kausoi have inter alia the following signs: ‘acute fever with slight rigour, sleeplessness, thirst, nausea, slight sweats about the forehead and the collarbones, much delirium, terrors, depressive states (poulla parelegon, phoboi, dysthymiai), very cold extremities, toes and hands … The cases of phrenitis had all the above symptoms, but the crises generally occurred on the eleventh day’; these emerge as extreme cases (26.11–27.2 Jouanna = 2.654 L.). In this discussion and those that follow, phrenitis is firmly categorized among the kausoi typical in winter.

Retrospective Diagnosis

When it comes to diagnosis and nosological conceptualization, it is important to mention a unique cue preserved at Epidemics 3: a group of ‘characters’ or sigla that mark the end of some of the patient cases in this text, as a kind of quick note made by a physician after reading the text and now incorporated into it. These sigla were known to Galen (who considered them spurious) and therefore must have entered the textual tradition earlier than that, and they are present in some manuscripts (most notably V).⁶⁴ The issue and significance of the sigla is not at stake here, but it is useful to look at Galen’s discussion and survey at Comm. Hipp. Epid. III (81–83 Wenkebach = 17a.610–13 K.). Here Galen interprets some sigla as meaning ‘recovery’ and ‘death’, ‘miscarriage’, ‘destruction’, ‘urine like semen’ and so forth – all shorthand markers for what made the case interesting for a reader. Apart from M for μανία (mania) (also used for μήτρα/η, mētra/ē ‘womb’), no other siglum corresponds to a clear disease label. Instead, they seem to indicate variable signs or states (‘irritation’, ‘dryness’) or concrete items (‘sputum’, ‘urine’, ‘wheat’). The Φ (F/PH) used for phrenitis is thus exceptional and can be taken as corroborating evidence of the importance of the disease from the fifth century bce to Galen’s era.

This Φ (F/PH) indicating an ancient retrospective diagnosis of phrenitis is used for three patients in Epid. 3, 17: case 14 (110.1 Jouanna = 3.142 L.), case 15 (111.9 Jouanna = 3.146 L.) and case 16 (112.14 Jouanna = 3.148 L.). The first is a woman who had a difficult twin birth. Acute fever with shivering follows, along with a painful head and neck, sleeplessness, colourless urine, thirst, wandering and derangement, and finally convulsions and death. The second is a female patient who falls ill ‘from grief’ (ek lypēs). She displays acute fever and shivering, and wraps herself up and has richly described crocydism (epsēlapha, etillen, eglyphen, etrichologei). There are also tears and inconsequential laughter, as well as wandering, silence, much talk – a complete picture of insanity – slow breathing, and finally death. The next patient, a young man from Moelibea, falls ill out of drunkenness and sexual indulgence, not uncommon triggers for derangement. There is ‘thin urine, no colour’; slow, deep respiration, with long intervals between breaths; tension; softness beneath the hypochondrium; delirium (quiet, then wild); sleeplessness; and death. In addition to these three patients, case 11 at Epid. 3, 1 (77.6 Jouanna = 3.62 L.) is marked with the label phrenitis in some manuscripts (GalL) and φρενιτιαία (‘phrenitiaia’) in others (IV), to the same effect: a female patient with fever, a heavy head, comatose state and sleeplessness, delirium, fears and despondency, and no thirst, all ending with death, is identified as phrenitic.

The qualification of all these cases as phrenitis indicates that sometime between the fifth century and Galen these portrayals were recognized as clearly pointing to our disease. The significant markers are crocydism, the quality of urine, derangement and psychological distress. To these we may add one final, much later retrospective diagnosis of phrenitis, offered by Potter in his Loeb edition of Diseases 2, where against both manuscripts he changes the paradosis φροντίς (phrontis) to φρενῖτις (phrenitis) at Morb. 2.72 (326.6 Potter = 211.15 Jouanna = 7.110 L.), discussed above.⁶⁵ The passage is a portrayal of an illness with significant psychopathological details in which ‘the phrenes swell outwards and are painful when touched’.⁶⁶ Jouanna retained the original reading phrontis but wondered: ‘Did the author feel the etymological connection between φροντίς (phrontis) and φρένες (phrenes)? Was he relating the swelling of the phrenes to mental derangement? Perhaps there is a trace here of the archaic belief that the phrenes are the source of intelligence.’⁶⁷ The question remains open for modern readers as much as it did for ancient ones: a psychological disturbance with pain in the diaphragm (‘anxiety, worry’) is diagnosed as phrenitis (Potter) or seen as a possible indicator of the mental relevance of the phrenes (Jouanna).

After Hippocrates

What happens in the period between the composition of the nucleus of the Hippocratic texts (which date from the classical era) and our next extensive source, the section on phrenitis in Cornelius Celsus’ De medicina, some four centuries later (3.18)? The evidence is scant, since the bulk of Hellenistic medical writing (later fourth century bce to the beginning of our era) is lost. The sophistication and richness of Celsus’ account makes one regret this loss all the more, since many developments in medical approaches to mental disease must have intervened. Unfortunately, given the restricted number of uses of the term phrenitis in non-medical texts, most of the evidence for this period is negative. As mentioned above, it is striking that neither Plato nor Aristotle mentions phrenitis even once in discussions of madness, confirming that the disease remained strongly linked to bodily physiology.⁶⁸ Moreover, Aristotle’s discussion of the phrenes at Parts of Animals 3, 672b28–30 as a partition between regions of the body and as ‘drawers (of heat)’ (pros tēn thermotēta tēn … hoion paraphyades) that serve to protect the nobler upper parts from the lower ones devoted to digestion, offers no account of phrenitis. An affection of the part is described by Aristotle when he writes that when the phrenes become drenched in the ‘hot, residual fluid’ from below, this ‘evidently disturbs (epidēlōs tarattei) the intellect and perception (tēn dianoian kai tēn aisthēsin)’. The disease itself, however, is not addressed, despite points of contact with a phrenitic humoral aetiology.⁶⁹

If we turn to the fragmentary evidence, considerable information regarding these intervening centuries can be extracted from later medical writers and doxographers. Three figures stand out: the Anonymus Parisinus (AP, first century ce), Galen (first–second century ce) and Caelius Aurelianus (fifth century ce). Additional information is preserved in the encyclopaedic works of Aetius of Amida and Paul of Aegina (sixth and seventh century ce, respectively). This indirect testimony – to be read, of course, with the caution that reported opinion and doxography dominate in it – is of immense assistance in filling in the gap between the Hippocratic material and the work of Celsus; I rely on it for what follows.

In addition to these authors, important information is offered by a section of a first-century ce text, the so-called Anonymus Londinensis,⁷⁰ regarded by scholars as preserving material from the so-called Menoneia, a collection of medical writings known to Galen and composed as early as the fourth century bce, which are attributed to Aristotle but were in fact written by his disciple Menon. We therefore start from this, as the earliest source for medical developments in this period. At Anon. Lond. IV.7–10, just before the beginning of the text attributed to Menon, the author discusses how diseases get their names. These can derive, he says, from the ‘attendant affection (apo parakolouthountos)’ or the ‘affected place (apo topou)’ (5.7–10 Ricciardetto⁷¹). Thus ‘fever’, pyr, is named after the affection, the symptom of fever (to pyrōdes, 5.10 Ricciardetto), and so too in the case of ‘paralysis’ (paralysis, 5.11 Ricciardetto). Next the author mentions a different case, exemplified by phrenitis: ‘phrenitis gets its name from the place affected (apo topou). For the affection establishes itself in the phrenes – this is not the diaphragm, but the rational part of the soul (ouchi to diaphragma, alla tout’ estin to logistikon meros tēs psyches, 5.13–17 Ricciardetto).’⁷² The difficulty is in attributing this powerful remark regarding localization to a precise period. The concept of a logistikon, a ‘rational part of the soul’, is found in Plato’s tripartite schema and is also Aristotelian and Stoic. What is noteworthy here is the theoretical distinction between phrenes as location (the diaphragm, rejected by the author in this connection) and the word’s abstract, non-bodily meaning ‘mind, intellect’, which is the sense he intends. Most of all, it is remarkable that, as far as we can tell, in this second use as ‘mind’, phrenes, like to logistikon, is still treated as a spot in the human body, a locus affectus, a topos. We thus have here an early voice advocating against a localization of mental life in the diaphragm (agreeing with De Morbo Sacro), but also arguing against a bodily meaning of phrenes when it comes to our disease, in contrast with the bulk of Hippocratic material discussed above. As an alternative, another bodily location or delocalized ‘embodiment’ for the rational functions is indicated, ‘to logistikon’. To identify this with the brain, as Jones does in his light-hearted translation (‘for the affection makes its seat in the phrenes (brain, not diaphragm), which is the rational part of the soul’⁷³), is to read too much into the word and mislead the reader.

Other important testimony from the same period comes from Diocles of Carystus (fourth–third centuries bce), a physician much celebrated in antiquity whose work survives only in fragments. Diocles wrote variously on dietetics and nosology, and several later nosological works (e.g. Galen, Anonymus Parisinus and Caelius Aurelianus) refer to him as an authority, suggesting that his contribution was substantial. What is known about Diocles’ opinions on phrenitis comes mostly from Galen and Caelius Aurelianus. In fr. 71 van der Eijk (preserved in Galen’s On Critical Days) Diocles is quoted as saying that ‘people do not become affected by phrenitis (phrenitikoi) immediately from the first day’, which seems to confirm co-morbidity with other diseases and the nature of phrenitis as an unfavourable development from one. Fragment 72 belongs to the doxographic section of the chapter on phrenitis in Anonymus Parisinus, which focuses on the localization of the disease:

Diocles says that phrenitis is an inflammation of the diaphragm (phlegmonēn tou diaphragmatos⁷⁴) – he gives this name to the affection on the basis of the place (apo tou topou) [affected], not the activity (apo energeias) [affected]), the heart being affected simultaneously (for he, too, seems to posit reasoning around this) and that, for this reason, too, these affections are accompanied by mental disturbances.

The concept of inflammation (phlegmonē), said in Anonymus Parisinus to be Diocles’, is a crucial step towards a thematized localization of the disease (or any disease generally), since it places the emphasis on an impaired part suffering damage or alteration.⁷⁵ The author goes on to explain the name, but adopts the opposite perspective from Anonymus Londinensis: phrenitis is called after the anatomical place, the diaphragm, because it is close to the heart, indicating the region Diocles regards as the centre of mental functioning.⁷⁶ Derangement is the implication of this involvement of the cardiac region.

Caelius Aurelianus also preserves information about Diocles’ therapy for phrenitis (fr. 73 van der Ejik), although this is of limited significance for our purposes: in On Fevers (thus Caelius, Morb. Ac. 1.11–12 = 76.25–80.88 Bendz) he appears to have said that the ‘strong and impetuous’ should be treated with baths, but the ‘young and strong’ and full-blooded, or those who drink wine habitually, with venesection. According to Caelius in his section on causes and treatments, moreover, Diocles said that blood should be taken from the vein under the tongue as well as from the arm. The late-antique pharmacological author Gargilius in his Medicinae ex holeribus et pomis (third century ce) xviii (2 Maire) says that Diocles prescribed boiled garlic for phrenitics; this appears to be in line with the use of substances with a strong aromatic scent, or even a foul smell, to stimulate such patients.

We only have two testimonies regarding phrenitis in Herophilus (fourth–third centuries bce). One comes again from Caelius Aurelianus: at Morb.Ac. 1.11–12 (76.25–80.8 Bendz), in his chapter on phrenitis, Caelius writes that ‘neither Hippocrates nor Praxagoras nor Herophilus (T 239 von Staden) handed over any treatment for the disease, unlike Diocles’. A possible reference to Herophilus’ disciple Demetrius of Apamea is also found at Morb.Ac. 1, 4–5 (24.6–9 Bendz): Demetrius Erophilum sequens is said to ‘define phrenitis as a violent attack of madness accompanied by a loss of reason and (more frequently than not) by fever (delirationem … vehementem cum alienatione atque <frequentius cum>⁷⁷ febre), and swiftly leading either to death or at times to a restoration of health’ (T 211 von Staden). Von Staden interprets sequens as indicating doctrinal agreement (‘following the view of … ’), as opposed to ‘being a follower of … ’; both are possible.⁷⁸ Fever appears to be underplayed here, as opposed to other sources which refer to it ‘more’ or ‘rather frequently’: Caelius, uniquely, challenges the claim that fever should occur ‘most of the time’.

Erasistratus (304–250 bce), the other great Alexandrian medical authority besides Herophilus, is said by Anonymus Parisinus in his doxographical section on the causes of the disease (Erasistratus fr. 176 Garofalo) to have claimed, in accord with his doctrinal convictions, that phrenitis occurs when

the activities of the [cerebral] membrane are affected (kata ti pathos tōn kata tēn meningan energeiōn); at the place where, according to him, thinking is reasoning (he noesis phronēsis), disturbance of thinking is likely to represent a disturbance of reasoning

(he paranoēsis paraphronēsis).

This passage is the first attestation of the association between phrenitis and the brain that shapes the history of the disease from Galen to the modern era. Not only the head, but a specific part of the brain, the meninges, are involved in making the localization more concrete. The doxographer here reverses the logic of Anonymus Londinensis once again: as in the case of Diocles’ quasi-cardiocentrism, because Erasistratus located thinking in the head, as a consequence he located the disturbance of thinking, phrenitis, in the head as well.

Erasistratus’ encephalocentrism is juxtaposed in Anonymus Parisinus to the cardiocentric view of Praxagoras (fourth century bce): Praxagoras ‘says that phrenitis is an inflammation of the heart (phlegmonē tēs kardias), whose natural activity he in fact believes to be reasoning (phronēsis), and that when the heart is disturbed (tarassomenēn) because of this inflammation, it becomes productive of this affection’ (1.2, 2.7–10 Garofalo = Praxagoras 61–2 Steckerl).⁷⁹

This simplified organization of material must be taken to reflect the Anonymus Parisinus’s penchant for localization and neat categories.⁸⁰ Proof of this is found in the final paragraph, devoted to Hippocrates on phrenitis (1.4, 2.16–21 Garofalo), contrary to chronological order. Here the encephalocentric suggestion does not match at all what we know from surviving Hippocratic material: ‘Hippocrates says that the mind is placed in the brain (en tōi enkephalōi tetachthai) like a sacred statue on the acropolis of the body (kathaper ti hieron agalma en akropolei tou sōmatos), and that it uses as nutriment the blood around the chorioid membrane’; corruption of this blood causes the phrenitic pathology.

Caelius also provides information about other figures from the Hellenistic period, whom he discusses especially in regard to therapeutic measures. The first of these is Heraclides of Tarentum (third–second centuries bce), a Greek physician of the Empirical school who wrote commentaries on Hippocrates, and ‘the only empiricist’ Caelius wishes to mention, as he states explicitly (114.13 Bendz). Heraclides’ recommendations in Book I of his On Treatment of Internal Diseases (in Caelius, Morb.Ac. 1, 17 = 115.13–125.16 Bendz) can be paraphrased as follows: patients should lie in a dark place, since light can excite them; clysters should be given for the bowels, at no specified time, but every day; it is recommended that one foment the head with decoctions of laurel, then shave it with a razor and steam it again; it is beneficial to apply a poultice of flour and hydromel, iris, mastic oil and sweet flag; the head and nostrils should be anointed with sulphurwort, castor, poppy juice and bitter almond oil (or with vinegar and iris oil); a decoction of poppy and thyme should be used to warm the head at night; sleep should be induced by administering oppressive drugs in the correct dosage; if the disease subsides, chicken broth or gruel made from pearl barley should be offered.

A second type of treatment is advised in cases of phrenitis arising through indigestion (cruditati, 122.2 Bendz). This concept is itself interesting, since it connects to the gastric localization in a strand of the Hippocratic account of the disease (see above⁸¹), despite the fact that Heraclides regarded the head, caput, as central. In this second case, a poultice should not be permitted until a clyster has been administered. There is also a third type of phrenitis: ‘If the whole body is not weighed down with excess food, but only parts of the head seem congested, blood should be withdrawn from the forehead vein’ (122.32 Bendz). Finally, what may be a further type is mentioned for those ‘who have fallen into the disease through decomposition (of the humours)’ (124.4–5 Bendz). For them, Heraclides recommends a clyster, water to drink, and sometimes honey with wine.

This survey of therapies offers a confirmation of the early presence of a competition between localizations, and a division of phrenitis into distinct embodiments, so to speak: some precisely localized (in the head), some ‘removed’ or shifted (to the stomach), and some, finally, holistic and delocalized (‘putrefaction of the humours’).

Conclusions

From this survey of the doctrines preserved from the Hippocratics, on the one hand, and the traces of later developments in medicine in the centuries that follow, on the other, a number of themes and aspects emerge: the localization, increasingly polarized around the chest and head; the originary nature of phrenitis as a winter chest ailment; the strong technicality, shown by the absence of any reference even in Aristotle and Plato; the fever. If this picture appears to dominate in medical quarters, we also find traces of a competing suggestion, which gives a central place to a holistic, delocalized account (the role of blood, for example, as locus affectus or causative agent, as seen above). I focus next on this branch of the tradition, one that emerges later and endures for centuries, although it remains marginalized in the history of the disease.