Search tips
Search criteria 


Logo of wtpaEurope PMCEurope PMC Funders GroupSubmit a Manuscript
Class Q. Author manuscript; available in PMC 2009 May 22.
Published in final edited form as:
PMCID: PMC2685474


The so-called diatritus in Graeco-Roman medicine represents a therapeutic system which was designed to provide a chronological framework for the regulation of regimen in disease. Developed in the mid-first century A.D. within the Methodist school of medicine, it appears thereafter to have been a cornerstone of this school’s therapeutics, as represented in the three major Methodist treatises to have survived. In the therapeutic prescriptions recorded in the Gynaecia, Soranus (fl. late first/early second century A.D.) mentions the diatritus specifically on four occasions,2 while Caelius Aurelianus (fifth century A.D.), drawing to some extent on Soranus’ earlier work, makes constant use of it throughout his Celeres Passiones and Tardae Passiones. Galen, too, universally hostile to the diatritus, associates it almost exclusively with Methodist physicians. There is strong evidence, however, that it was employed as a therapeutic tool also outside Methodist circles. It surfaces intermittently, but without criticism, in the treatments laid down in medical treatises with no apparent links to this school, and over a considerable period of time.3 In particular, the unknown author of the treatise On Acute and Chronic Diseases, often referred to as the Anonymus Parisinus, relies on it extensively, though not systematically, in the therapeutic sections of his work.4 Yet the diatritus has for the most part received only passing reference in scholarship, largely with regard to the doctrines of Thessalus or Methodism generally, and with little or no detailed inquiry into its exact meaning or uses.5 Given its significance not only for our understanding of the development of Methodist therapeutics, but also as an example of the broader influence which Methodism may have had over ancient therapy, a more systematic study of the diatritus is warranted. In this paper I shall examine the diatritus’ origins, its precise meaning, and what it represented in its practical therapeutic contexts, both in the hands of Methodist and non-Methodist physicians. I shall also attempt to offer some suggestions as to why the chronological framework which it describes was thought to be appropriate in the treatment of disease.


The earliest surviving attestation of the adjective διάτριτος appears in Philo of Alexandria’s work De Confusione Linguarum, composed in the early first century A.D. It appears here once, qualifying the term περίοδοι, and describing the cyclical phases which characterise the tertian fever.6 περίοδοι has a technical usage in this context, as Philo makes clear, in connection with the highly complex systems which had been developed by doctors in this period for analysing and distinguishing the various types of fever, now recognised as malarial, according to the relative frequency of their attacks.7 In this instance, διάτριτος describes that particular cycle of a fever which produces with regularity an attack every other day, the fever itself being commonly referred to as the τριταῖος (tertian).

Excepting this one instance in Philo, however, the term διάτριτος only appears in Greek as a substantive with the feminine article, and, unlike in Philo, it is found in exclusively therapeutic contexts. It is never again used in extant literature to describe the periodicity of a disease, and by the time it next appears, in the works of Soranus, it has clearly assumed a different meaning and broader connotation. It is this meaning of the term, and the therapeutic contexts in which it is used, which will form the main subject matter of this paper.

I have stated above that the development of the diatritus as a distinct therapeutic tool took place in the mid-first century A.D., but this requires some comment. Our only source for the origins of the system of the diatritus in therapeutic practice is Galen, who in several different works unequivocably attributes its invention to Thessalus of Tralles, the leading Methodist physician who was active at Rome during the reign of Nero.8 As Diller noted,9 however, the fifth century A.D. Methodist physician Caelius Aurelianus once employs the term diatritus in describing the therapeutic prescriptions for phrenitis laid down by Themison of Laodicea, generally regarded as the founder of Methodism (fl. first century B.C.), and predating Thessalus by a considerable period of time. At first sight, this might seem to suggest that Themison himself had adopted the term and its associated doctrine, and casts doubt on Galen’s attribution of its invention to Thessalus.10 But, as will be discussed below, the diatritus for Caelius can refer in a quite general sense to the third day of an illness, and in view of this usage, the fact that he employs the term in this passage need not entail that it was found also in the writings of Themison. Caelius might very easily be retrojecting the term here. Indeed, he does this again later in the Celeres Passiones with regard to the therapies of Heraclides of Tarentum, the acclaimed Empiricist physician of the first century BC,11 and a most unlikely adherent or inventor of the diatritus.12 In employing the term in these cases, Caelius seems merely to be describing the therapeutic directions which these authors set out generally for ‘the third day’ of these particular illnesses, rather than attributing to them any doctrinal dependence on or allegiance to the system of the diatritus. Furthermore, the specific therapeutic directions of Themison and Heraclides, as Caelius represents them, diverge from the therapeutic regulations which the system of the diatritus represents in the work of later Methodists, as we shall see, though it should be noted that discrepancies on this level would not automatically preclude the possibility that the diatritus itself was used in a different way by these earlier doctors. For example Themison, shortly afterwards at Cel. Pass. 1.160 (pp. 110-112 Bendz), is said to have given food every day after the first ‘third day’, contrary to what we know of the doctrine of the diatritus, by which food is given every other day (i.e. after subsequent diatriti). Heraclides, at Cel. Pass. 3.217 (p. 418 Bendz), gives food both on the first day and on the third, whereas the diatritus system demands generally that food be given only on the third.

We are left with Galen’s statements, however, which are quite clear and emphatic on Thessalus’ role in the development of the diatritus, and there is no reason to doubt his familiarity with its history. Against his testimony, the usage of the term in these passages of Caelius can be considered a simple anachronism. It may also be noted in this regard that Celsus (fl. probably reign of Tiberius), who makes no reference to Thessalus in the De Medicina, nowhere mentions or describes the diatritus. In the absence of any firm evidence to the contrary, therefore, we have no good reason to doubt that Thessalus was its inventor.

One of the leading physicians at Rome under Nero, Thessalus represents a rather flamboyant character in the history of medicine, rejecting outright the medical traditions of the ancients with claims to have revolutionised the art, and for his own part producing and refining a set of doctrines which stood at the centre of the development of Methodism.13 His assertion that the whole of medicine could be taught in a matter of six months certainly outraged Galen,14 but reflects what appears to have been one of the most important aspects of Thessalus’ project, namely to redefine what was necessary for therapy, and to discard all that was not. As will be seen, the straightforwardness of the system represented by the diatritus, its apparent emancipation from any theoretical grounding necessary for its practical employment, and its universal applicability to all diseases, would seem to be entirely compatible with the simplicity and reductionism implied by Thessalus’ 6-month medical curriculum.

Nevertheless, Themison’s doctrine, formulated perhaps a century earlier, had a strong influence in general on Thessalus, and there is evidence that the observation of the third day in some form played an important role in the former’s regulation of regimen. As with other aspects of Methodist doctrine, most notably the theory of the common conditions (κοινότητες), we might speculate here too that Thessalus owed some debt to his Methodist predecessor.15 Celsus, in the course of a passage surveying previous physicians’ doctrines on when first to give food in fever,16 summarises Themison’s position thus:

at Themison nuper, non quando coepisset febris, sed quando desisset, aut certe levata esse​<​t​>​, considerabat; et ab illo tempore expectato die tertio, si non accesserit febris, statim; si accesserat, ubi ea vel desierat, vel si adsidue inhaerebat, certe si se inclinaverat, cibum dabat.

But recently Themison took into account not when the fever began, but when it ended or certainly abated. Waiting for the third day from that time, if the fever did not attack, (he gave food) immediately. If it had attacked, he gave food either when it ended, or if it continued persistently, certainly if it abated.

(De Medicina 3.4.6)

In this doubtless simplified scheme, Celsus also presents Themison’s use of the third day for prescribing food as a departure from the practice of previous doctors. Our sources for Themison’s therapeutic doctrines are rather meagre, and Celsus’ account clearly does not correspond to the diatritus system (which for one thing was calculated from the beginning of an illness), but it is conceivable that a general observance of the third day in disease was later developed or refined by Thessalus and codified into a universal therapeutic system.17 It is worth speculating on the early development of this notion regarding the importance of the third day, but, overall, there is no compelling reason to doubt Galen’s testimony that the diatritus was an innovation by Thessalus, nor indeed that this was the same diatritus as we know from Soranus and subsequent authors.


The term ἡ διάτριτος is regularly translated as ‘the three-day period’ by scholars.18 LSJ9 s. v. διάτριτος II gives the definition thus: ‘δ. (sc. περίοδος), ἡ, period of three days’. There is, however, no basis for this sense of the term diatritus, and the notion that it is the noun ‘περίοδος’ that is suppressed has presumably been influenced by the Philo passage discussed above, which contains the sole instance in which διάτριτος is found qualifying a noun. Caelius Aurelianus, writing in Latin but retaining the Greek term, tells us clearly and specifically in several places that diatritus means ‘the third day’.19 It has been assumed that Caelius also used the term diatritus to refer to a three-day period, but this assumption is ill-founded.20 It appears to be based on his use of the preposition intra, such that when he draws a distinction between intra diatriton and in ipsa diatrito,21 it has been supposed that in the former case diatritus means ‘three-day period’, while in the latter it refers to ‘the third day’ or ‘the end of the three-day period’.22 Although intra can of course mean ‘within’ or ‘during’ a period of time, it must mean ‘before’ when a specific time limit is set, or when linked with an ordinal rather than cardinal number.23 That tertius dies and diatritus are equivalent terms for Caelius means that diatritus must act as an ordinal number in such phrases involving intra. Thus Caelius in the same way explicitly draws a distinction between intra tertium diem and in ipsa diatrito, where the former phrase does not mean ‘during the third day’, but ‘before the third day’.24 It is unnecessary, and incorrect, to posit two different meanings for the term diatritus in Caelius Aurelianus in this way, especially since he explicitly offers only one meaning himself. Furthermore Soranus, our earliest extant source for the use of the term, and an adherent, never uses it to describe a period of three days. The term διάτριτος appears four times in the Gynaecia, and in each instance (with the possible exception of the first) it cannot refer to a period of three days: τῆς πρώτης ἐνστάσης διατρίτου (3.10.5); ἔμπροσθεν τῆς διατρίτου ἢ καὶ ἐν αὐτῇ (3.11.1); ἕως διατρίτου (3.28.4); and ἄχρι διατρίτου (4.37.3). We shall also see in the next section, which will discuss the relationship between the diatritus and the paroxysm, additional evidence for this interpretation. It is thus much more likely, and indeed much more natural, that the noun which is suppressed in the expression ἡ διάτριτος should be ἡμέρα rather than περίοδος. The literal meaning of the diatritus is then ‘the day which comes every three’, or ‘the third day recurring’, that is, counting inclusively in accordance with ancient practice, the third, fifth, seventh, ninth days, etc.


The focus on the recurrent third day which is inherent to this therapeutic system was clearly not an arbitrary choice, and there is evidence of a certain rationale behind its development within Methodism. It appears that the diatritus was related, at least on a theoretical level, to the expected recurrence of a disease’s paroxysms. Many of our sources draw a direct link between the diatritus and the paroxysm, and there is good evidence to suggest that this was an important part of its theoretical background.25 In the course of a critique of Thessalus’ treatment of chronic wounds, for example, Galen describes how Thessalus’ followers, who adhere to the diatritus, think it necessary for fever patients to pass the paroxysm, which in turn is described as occurring δι[alpha] τρίτης ἡμέρας:

ὅπου καὶ τοὺς ὁπωσοῦν πυρέξαντας ὑπερβάλλειν [alpha]ξιοῦσι τὸν δι[alpha] τρίτης ἡμέρας ἤτοι γενησόμενον ἢ μὴ γενησόμενον παροξυσμόν. οὕτως [alpha]κριβ[omega with tilde]ς ἄρα τὴν περὶ κρίσεων ἐκμεμαθήκασι θεωρίαν, ἢ τῆς μελλούσης [alpha]κμῆς τοῦ νοσήματος ὑπάρχουσι προγνωστικοί.

After all, they (i.e. Thessalus’ disciples) think that those suffering from any kind of fever should pass the paroxysm on the third day, which may or may not happen. This is how accurately they have grasped the theory of crises, or make prognoses of the future culmination of the disease.

(Meth. Med. 4.4 (10.263.10-15 K))

The close association of the paroxysm and the recurring third day is emphasised here in the phrase ὁ δι[alpha] τρίτης ἡμέρας παροξυσμός, a phrase which Galen repeats elsewhere in his attacks on Thessalus or his Methodist successors.26 One of Galen’s (several) objections, too, to the diatritus is that the paroxysms of fevers can be shown not only to occur on the third day, but also every day or on the fourth day, further implying a direct relationship between the diatritus and the paroxysm on the third day.27

In the sixth or perhaps seventh century, Stephanus of Athens, in his Commentarii in Priorem Galeni Librum Therapeuticum ad Glauconem 7 (p. 245 Dietz = p. 48.4-7 Dickson), appears to elevate this notion to a doctrinal level within the Methodist school, though it must be acknowledged that Galen may be his only source for this information. The context here coincides closely with the diatritus, without mentioning it by name:

οἱ δὲ Μεθοδικοὶ οὐχ οὕτως πως, [alpha]λλ[alpha] τούτους παραλαβόντες κελεύουσιν ἐπὶ τρεῖς ἡμέρας [alpha]σιτεῖν, ϕυλάττεσθαι τὸν δι[alpha] τρίτης [alpha]ξιοῦντες παροξυσμόν. ὑπολαμβάνουσι γ[alpha]ρ ἅπαντα σχεδὸν τ[alpha] νοσήματα δι[alpha] τρίτης παροξύνειν.

The Methodists do not do this, but when they get hold of them (i.e. patients) they tell them to fast over three days, thinking it proper to wait for the paroxysm on the third day. For they believe that almost all diseases reach their paroxysm on the third day.

The Methodists as a school are here credited with the notion that a paroxysm was at least likely to occur ‘δι[alpha] τρίτης’, that is to say ‘on the third day’, counting inclusively, or cyclically ‘every other day’. Sextus Empiricus may refer to the same notion, in this case with direct reference to the diatritus.28

“κατ[alpha] τ[alpha]ς νόσους ἐν ταῖς παρακμαῖς τήν τε ποικίλην δίαιταν καὶ τὸν οἶνον δοκιμαστέον˙ ἐπὶ πάσης δὲ τυπώσεως νόσου πρὸ τῆς πρώτης διατρίτου πάντως γίνεται παρακμή˙ [alpha]ναγκαῖον ἄρα πρὸ τῆς πρώτης διατρίτου τήν τε ποικίλην δίαιταν καὶ τὸν οἶνον ὡς τὸ πολὺ παραλαμβάνειν”. ὁ μὲν διαλεκτικὸς οὐδὲν ἂν εἰπεῖν ἔχοι πρὸς διάλυσιν τοῦ λόγου, καίτοι χρησίμην οὖσαν, ὁ δὲ ἰατρὸς διαλύσεται τὸ σόϕισμα, εἰδὼς ὅτι παρακμὴ λέγεται διχ[omega with tilde]ς ἥ τε τοῦ ὅλου νοσήματος καὶ ἡ ἑκάστης ἐπιτάσεως μερικῆς [alpha]πὸ τῆς [alpha]κμῆς ῥοπὴ πρὸς τὸ κρεῖσσον, καὶ πρὸ μὲν τῆς πρώτης διατρίτου ὡς τὸ πολὺ γίνεται παρακμὴ ἡ τῆς ἐπιτάσεως τῆς μερικῆς, τὴν ποικίλην δίαιταν οὐκ ἐν ταύτῃ δοκιμάζομεν [alpha]λλ᾽ ἐν τῇ παρακμῇ τοῦ ὅλου νοσήματος.

“In disease, one must prescribe a varied diet and wine in the abatement. In every type of disease, an abatement always occurs before the first diatritus. It is necessary, then, to give a varied diet and wine in general before the first diatritus.

The dialectician will have nothing to say for the resolution of this argument, even though this would be useful, but the physician will resolve the sophism since he knows that ‘abatement’ has two meanings: the turn of the whole disease from the crisis towards strength, and that of each partial attack; and the abatement of the partial attack occurs for the most part before the first diatritus. We do not prescribe a varied diet during this, but in the abatement of the whole disease.

(PH 2.237-238)

Part of the point of the result of this sophism is that, as will be seen, food is generally never to be prescribed before the diatritus, and Sextus seems to expect his audience to have been aware of this. The most interesting point, however, is that the abatement (and of course the paroxysm which precedes it) is again conceived of as something which (generally) occurs before the diatritus.

It seems clear, however, that the adherents of the diatritus did not maintain a direct equivalence between it and the paroxysm cycle, as a passage in Galen seems to confirm:

[alpha]νάγκη δήπου κᾀμε λέγειν διάτριτον, ὅπως μὴ καὶ περὶ τοῦδε τῆς ϕλυαρίας αὐτ[omega with tilde]ν ἐνέχωμαι, ϕασκόντων ἕτερον μὲν εἶναι τὸν δι[alpha] τρίτης παροξυσμόν, ἑτέραν δὲ τὴν πρώτην διάτριτον˙ ἤκουσας γάρ ποτε καὶ σὺ μέχρι τοσούτου τῆς [alpha]δολεσχίας αὐτ[omega with tilde]ν ἐκταθείσης.

It is of course necessary that I too use the word “diatritus”, so that I am not embroiled in their nonsense on this point as well, since they say that “the paroxysm on the third day” is one thing, and “the first diatritus” another. For you have heard as well how far their frivolity extends.

(Meth. Med. 8.3 (10.559.2-6 K))

While it is significant again that Galen should regard the paroxysm on the third day and the first diatritus as broadly synomymous, it must be acknowledged that it is also to his advantage to identify the two, since he can then point to the fact that certain fevers do not reach a paroxysm on the third day, and thereby also undermine the diatritus.

Nevertheless, for those who adhered to the diatritus, there was clearly an important distinction to be drawn. In the passages of Stephanus and Sextus Empiricus quoted above, it is not asserted that the paroxysm will recur on the third day in every case, but rather only for the most part (ἅπαντα σχεδόν in Stephanus; ὡς τὸ πολύ in Sextus). It was apparently not thought that the paroxysm necessarily occurred on the third day for every affection, but the diatritus was nevertheless felt to be applicable even under circumstances in which it did not. We might also identify a tendency to distance the diatritus from the paroxysm in the fact that certain Methodists sought additional arguments to validate it. At Tard. Pass. 2.157 (p. 640.11-13 Bendz), for example, Caelius Aurelianus adduces the Methodist Antipater’s assertion of the ‘rationality’ of the number three,29 suggesting an almost arithmological justification:

ut enim Antipater ait tertio Epistolarum libro ad Gallum scribens, naturalis est ratio trini numeri, non solum in diebus, sed etiam in omnibus diatritis.

For, as Antipater says, writing in Book 3 of the Letters to Gallus, the rationality of the number three is natural, not only with regard to days, but also in all things which recur according to the number three.

This all suggests that the paroxysm cycle was by no means the only relevant factor in formulating the diatritus, but we are unfortunately left without any hints as to what other specific considerations may have contributed to its development.


None of our surviving sources is concerned to provide a precise and clear explication of what the diatritus actually represented, and much of its system is taken for granted in the more practical contexts in which it is mentioned. Nevertheless, it is possible to reconstruct some of the details behind the diatritus from a variety of rather indirect remarks made by certain authors. While the term διάτριτος must qualify ἡμέρα as described above, the diatritus in several passages appears to be employed in an implicit sense to refer to a specific, narrower period of time during the course of the third day. This implicit sense derives from the fact that the diatritus is reckoned from a specific point on the first day, i.e. from the hour at which the affection originally attacked. The diatritus in this way refers to the corresponding time on the third day. Thus Caelius Aurelianus, in a passage on the signs of epilepsy, wishes to identify as accurately as possible the time of the attack, explicitly in order to be able to calculate the time of the diatritus:

quaesitum etiam, utrum ex antecedentibus signis accessionem fore an praesentem sive factam iudicemus. sed tunc erit accessio iudicanda, quoties ceciderint in ea aegrotantes, quo discere poterimus, quo tempore diatriti tempus sit numerandum, ut recte possint curationes suis reddi temporibus.

It has also been asked whether we should judge from the antecedent signs that the attack is to come, or that it is present or over. But the attack should be judged (to occur) at the time when the patients fall over during it, from which we can discover from what point the time of the diatritus is to be calculated, so that the remedies can be correctly assigned to their own times.

(Tard. Pass. 1.105 (p. 492.3-7 Bendz))

At Tard. Pass. 2.173 (p. 650.7 Bendz), Caelius also more precisely locates the diatritus at a time during the third day: tunc tertia die accepta diatriti temporis observatione, ... (Then on the third day, when the observation of the time of the diatritus has been made, ...).

Galen in several places implies that the diatritus represents the hour on the third day corresponding to the original attack, once with reference to the notion of the suspicious hour. In recounting a case history at Meth. Med. 10.3 (10.673.14-16 K), we are informed that the patient fell ill with fever at around the eleventh hour of the first day, after which certain adherents of the diatritus got hold of him, refused to give food on the second day, and then returned on the third:

καὶ τοίνυν καὶ κατ[alpha] τὴν τρίτην ἡμέραν ἕωθεν [alpha]ϕικόμενοι τὴν διάτριτον ὑπερβάλλειν ἠξίουν. ἦν δ᾽, ὡς εἴρηται πρόσθεν, ἡ ὕποπτος ὥρα τῆς ἡμέρας ἐκείνης ἑνδεκάτη.

But then when they arrived early on the third day, they thought that he should go beyond the diatritus. The suspicious hour, as was said earlier, was the eleventh of that day.

Annia Faustina, a close relative of the emperor, criticises certain Methodist physicians accompanying her who adhere to the diatritus, with regard to a fever which had attacked at the eighth hour (on the first day):

... κατ[alpha] τὴν τρίτην ἡμέραν οὐχ ὑπερβάλλειν, ὡς ὑμεῖς [alpha]ξιοῦτε, τὴν ὀγδόην ὥραν [alpha]ναμείνας [alpha]λλ᾽ ἤδη λούσας τε καὶ θρέψας <ἐθεράπευσεν>.

(Galen) did not wait to pass the eighth hour on the third day, as you demand, but had already treated him with a bath and food.

(De Praen. 12.9 (14.664 K = p. 132.18-20 Nutton))

In these passages the diatritus certainly refers to a specific period of time on the third day, and clearly corresponds to the time of the original attack 48 hours before. At Meth. Med. 10.3 (10.678.11-14 K), in discussing his own subsequent (and successful) therapeutic prescriptions for the same case as in the former passage, Galen again makes it clear that the diatritus is a particular time during the day, and that it coincides to some degree with the paroxysm. It should be noted that in this fever, the paroxysms occur δι[alpha] τρίτης (on the third day/every other day), in accordance with the Methodist expectations for most diseases discussed above:

οὗτος ὁ ἄρρωστος ἐπαίδευσε πολλοὺς τ[omega with tilde]ν ἡμιμοχθήρων τε καὶ μὴ παντάπασιν [Greek small letter omicron with psili and oxia]νων ὡς καὶ πρὸ τ[omega with tilde]ν παροξυσμ[omega with tilde]ν ἐνίοτε χρὴ τρέϕειν, εἰ καὶ δυοῖν ὡρ[omega with tilde]ν εἴη τὸ μεταξύ, καὶ πολὺ δὴ μ[alpha]λλον ἔμπροσθεν τῆς διατρίτου.

This patient educated many of the semi-villains and not complete asses that it is sometimes necessary to give food also before the paroxysms, even if there is an interval of two hours (between giving food and the paroxysm), and much more (certainly) before the diatritus.

That the diatritus can represent a time 48 hours after the original attack of an affection can apparently also be traced to surviving testimony for Thessalus’ own doctrine. Caelius Aurelianus describes Thessalus’ prescriptions for the treatment of paralysis, where the diatritus is not mentioned specifically, but there seems little doubt that it is at work:30

etenim post phlebotomiam usque ad tertium imperat abstineri, solum potum diurna cum luce offerendum. tunc transacta die atque hora, qua sunt aegrotantes paralysi vitiati, cibum probat. For he (i.e. Thessalus), after phlebotomy, orders (patients) to abstain from food until the third day, and that drink should be offered only at dawn. Then, on the third day when the hour has passed at which the patients were attacked by paralysis, he gives food.

(Tard. Pass. 2.60 (p. 580.5-8 Bendz))

This meaning of the diatritus can also be discerned more generally with regard to the regulation of food in therapy. One of the diatritus’ most important and commonly deployed functions, as will be discussed below, was to control at what point in illness food was first to be given to the patient. Generally speaking, as attested by many of our sources, the correct time for food is during or just after the diatritus. The directions for this are variously expressed: it is stated that food should be given ‘after the diatritus’ (μετ[alpha] τὴν διάτριτον);31 that the patient should be kept without food ‘until the diatritus’ (μέχρι/ἄχρι/ἕως τῆς διατρίτου);32 or that the patient should ‘pass/go beyond the diatritus’ without food (ὑπερβάλλειν τὴν διάτριτον).33

In three places Galen tells us explicitly, or has an interlocutor tell us, that the adherents of the diatritus who give their therapeutic advice prescribe a two-day period of fasting. At De Praen. 12.8 (14.663-664 K = p. 132.12-15 Nutton), shortly before the passage cited above, Annia Faustina tells her Methodist attendants:

ἐνίους δὲ καὶ κατ[alpha] τὴν πρώτην, ἐνίους τῇ τρίτῃ τ[omega with tilde]ν ἡμερ[omega with tilde]ν [alpha]πέλυσεν ἐπὶ τ[alpha]ς συνήθεις πράξεις˙ ἐν ᾗ πάντες ὑμεῖς ταῖς πρώταις ἡμέραις δύο προασιτῆσαι κελεύσαντες, ὑπερβάλλειν τ[alpha]ς ὑπόπτους ὥρας κατακειμένους ϕυλάττεσθε.

He returned some on the first day, others on the third day, to their usual habits. In this time all of you, after ordering patients initially to fast for the first two days, make sure that they pass the suspicious hours lying down.

Galen repeats this elsewhere with regard to two further case histories which have been mentioned above:

πολλάκις γοῦν ἤκουσας ἡμ[omega with tilde]ν λεγόντων ἐνίοις τ[omega with tilde]ν τοιούτων ἰατρ[omega with tilde]ν ὡς ἔξεστιν αὐτοῖς ἐναργέστατα μαθεῖν ἡλίκον ἐργάζονται κακὸν ἐν [alpha]σιτίᾳ ϕυλάττοντες τ[alpha]ς προειρημένας ϕύσεις, ἢν ἐθελήσωσιν ὑγιαίνοντας [alpha]μέμπτως αὐτοὺς [alpha]σιτῆσαι κελεῦσαι δυοῖν ἡμέραιν.

You yourself have often heard us saying to some of these doctors (i.e. adherents of the diatritus) that it is possible to learn quite clearly what great evil they cause by keeping the natures mentioned without food, if they want to tell healthy patients with impunity to fast for two days.

(Meth. Med. 8.2 (10.542.14-18 K))

παρὼν ὁ εὐτράπελος ἐκεῖνος ἐγελωτοποίει, [alpha]ναμιμνήσκων τὸν ἐμημεκότα τρίτης ἑσπέρας ὡς ἐχρῆν αὐτὸν ἄρτι δυοῖν ἡμερ[omega with tilde]ν ἄσιτόν τε καὶ ξηρὸν καὶ ἄσης μεστὸν κατακεῖσθαι, εἰς τ[alpha]ς ὥρας [alpha]ποβλέποντα κατ[alpha] τὴν τ[omega with tilde]ν διατριταρίων πρόσταξιν.

That witty man was present and cracked a joke, reminding the one who had vomited of the third evening, that it was necessary for him to lie down for two days without food, dry and full of distress, staring at the clock according to the orders of the diatritarians.

(Meth. Med. 8.6 (10.582.8-12 K))

For the first of these two passages especially, the two-day starvation period is clearly marked in Galen’s detailed account of the doctors’ prescribed treatment according to the diatritus (Meth. Med. 8.2 (10.536-541 K)). It is thus after this two-day period of fasting (i.e. at the diatritus, or 48 hours after the original onset) that food is to be administered to patients. Throughout his surviving works, too, Caelius Aurelianus, in employing the diatritus, consistently tells us that food is to be given on the third day.

In view of this apparent agreement, it is then somewhat disconcerting to find that in two passages of the Meth. Med. we have clear and explicit testimony from Galen that adherents of the diatritus gave food on the fourth day:

οὓς ὁ σοϕώτατος Θεσσαλός ... ἐταρίχευσεν ἂν ὅλαις οἶμαι τρισὶ λιμοκτονήσας ἡμέραις, εἶθ᾽ οὕτως ἔθρεψεν ἂν οἶμαι δηλαδὴ μετρίως τεταρταίους.

These (fever patients) the most wise Thessalus ... would have worn down, starving them for three whole days, I think, and then in this way would have given them food, plainly I think, in a small amount on the fourth.

(4.4 (10.264.4-7 K))

ἔθρεψαν γ[alpha]ρ ἂν αὐτὸν ἐν τῇ τετάρτῃ τ[omega with tilde]ν ἡμερ[omega with tilde]ν, εἶτ᾽ αὖθις ἐν τῇ ἕκτῃ, κᾄπειτ᾽ ὀγδόῃ τε καὶ δεκάτῃ˙ τοιοῦτος γοῦν τις ὁ τύπος τῆς διαίτης αὐτ[omega with tilde]ν ἐστιν, ὡς ἐθεάσω πολλάκις˙ ἡ μὲν πρώτη τροϕὴ μετ[alpha] τὴν πρώτην διάτριτον, αἱ δ᾽ ἄλλαι παρ[alpha] μίαν.

For they (adherents of the diatritus) would have given him food during the fourth day, then again during the sixth, and then the eighth and the tenth. For such is the pattern of their regimen, as you have often observed: the first food after the first diatritus, the rest every other day.

(10.6 (10.692.6-10 K))

This directly contradicts what Galen tells us about the diatritus elsewhere, in the passages cited above.34 How are we to reconcile these statements? Firstly, we must take account of the contexts in which Galen states that food was given on the fourth day. With regard to the first quoted passage, Galen is attacking Thessalus’ general tendency to delay treatments needlessly, indeed dangerously. With regard to the second, he discusses how relatively innocuous fevers develop into more serious ‘hectic’ fevers, simply by forcing the patient to pass the diatritus. His objection in each case is the extreme delay in administering food involved in observing the diatritus. This would offer some grounds for the suggestion that Galen may be exaggerating the period of fasting for his own rhetorical purposes.35 Nevertheless, it would be dangerous to accuse Galen here of an outright lie. An explanation, however, appears to emerge from several passages which suggest that under certain circumstances the giving of food was in fact delayed until the fourth day, fully in accordance with the diatritus system. This involved the apparently long-standing notion that the evening was an unsuitable time for food, so that if the diatritus were to fall late in the day, it was advisable to wait until the following morning. So Caelius Aurelianus tells us explicitly in his appraisal of Asclepiades’ treatment of phrenitis:36

falsum est etiam, cum continua atque iugis fuerit passio, circa vesperum cibum dare, siquidem etiam secundum ipsum Asclepiadem releventur omnes lucis initio, vespero accessiones augeantur. ... omnino igitur ratione caret vesperum cibo tempus eligere, quo etiam secundum Asclepiadem accessiones augentur.

It is also wrong, when an affection is continuous and uninterrupted, to give food around evening, since, even according to Asclepiades himself, all attacks are relieved at dawn, but are increased in the evening. ... Therefore it is entirely irrational to select the evening as the time for food, when the paroxysms are increased, even according to Asclepiades.

(Cel. Pass. 1.141 (p. 100.19-26 Bendz))

This can also be seen in action in the case history mentioned above which Galen recounts at Meth. Med. 8.2 (10.536-541 K). When the two adherents of the diatritus return to see their patient at sunset on the third day, and are satisfied that he has passed the diatritus without food, they postpone feeding him until the morning (10.541.3-9 K):

κελεύσαντες οὖν τοῖς οἰκείοις, ἐ[alpha]ν γένωνταί τινες αὐτῷ νοτίδες, [alpha]πομάττειν ἐπιμελ[omega with tilde]ς ὅπως μὴ ψυχθείη παραγενήσεσθαί τε ϕάντες ἕωθεν, [alpha]παλλάττονται μηδὲ τότε κελεύοντες ἐπὶ τοῖς ἱδρ[omega with tilde]σι τραϕῆναι τὸν ἄνθρωπον˙ ἐδόκουν γ[alpha]ρ αὐτοῖς αἱ κατ᾽ ἐκεῖνον τὸν καιρὸν νύκτες εἶναι μικραὶ καὶ κάλλιον ἐϕαίνετο κατ[alpha] τὴν ὑστεραίαν ἕωθεν τρέϕειν.

Having told the slaves that if any moisture should appear on (the patient), they should wipe it off carefully to prevent chilling, they said they would return early, and left, telling them not to feed the man at that time, while he was sweating. For they decided that the nights were short at that season, and it seemed better to administer food early on the next day.

A brief comment found at Anon. Paris. On Acute and Chronic Diseases 14.3.6 (p. 96 Garofalo), on the treatment of ileus, must refer to the same notion:

τροϕὴν δὲ μετ[alpha] τὴν πρώτην <διάτριτον> δοτέον ..., ϕεύγοντας ἐπὶ πάντων τὴν ἑσπέραν δι[alpha] τοὺς παροξυσμούς.37

One must give food after the first diatritus ..., avoiding in all cases the evening because of the paroxysms.

This postponement of food until the morning of the fourth day appears to represent an important exception to the normal regulation of diet according to the diatritus, and it would explain the contradiction in Galen’s testimony, rendering it a selective misrepresentation rather than a deliberate falsification. Such an activity would be well in keeping with the often tendentious character of Galen’s polemics elsewhere. In any case, there is no doubt that Caelius Aurelianus’ consistent statements on the diatritus that food is to be given on the third day, coupled with Galen’s comments elsewhere (where he has no such motive to distort), should carry more weight than these two passages.

To summarise, the therapeutic system of the diatritus was an innovation by Thessalus, who may, however, have been drawing on certain features of Themison’s perhaps less codified doctrine. It describes a chronology which highlights the importance of the recurring third day in illness, i.e., counting inclusively, the third, fifth, seventh, ninth days, etc., limited by the duration of the disease. This is furthermore linked with a specific period of time on the recurring third day which directly corresponds to the hour on the first at which the disease originally presented itself. It is not always clear to what degree there is an overlap between these two senses, or precisely how it was understood by various authors, and it seems best to judge from the context of each passage. Soranus, for example, in each of the four occurrences in the Gynaecia, appears to employ the diatritus to mean simply ‘the third day’, but the hour on the third day was probably understood without the need to specify, in particular since this hour fluctuated according to the timing of the disease’s initial onset. The significance of the third day is also theoretically linked with doctrine concerning the expected recurrence of an affection’s paroxysm, but the diatritus is not dependent on the observation of paroxysms and represents a universally-applicable system designed for the regulation of therapy, the practical use of which did not necessarily require any theoretical grounding.


So far we have examined what precisely was meant by the diatritus, with reference to its origins and theoretical background, and in doing so we have come across certain aspects of its practical functions, most significantly the regulation of food in disease. We must now address the question of how it was actually applied to a practical therapeutic context. The fullest surviving account of the diatritus in action is to be found in the works of Caelius Aurelianus, who describes in detail the ways in which the diatritus was applied in treating a full range of acute and chronic diseases. This is not the place for a systematic analysis of the therapeutic procedures laid down in the Celeres and Tardes Passiones, but I wish to offer firstly some observations on the general patterns which emerge with regard to the therapeutic system which Caelius describes, before comparing this with the uses made of the diatritus preserved in other works.

The diatritus is often defined by scholars with reference solely to the regulation of food, but this constituted but one of its several functions.38 It governed the employment of most forms of therapy generally associated with regimen, most conspicuously the administering of food, baths, plasters, embrocations, clysters, as well as phlebotomy.39 An interesting exception is the use of any form of exercise.40 Caelius begins his therapeutic account for each disease with a general description of the environment suitable for the patient, addressing factors such as light, temperature, noise levels, position, etc.41 This is followed by a list in which the various kinds of dietetic therapy are arranged into a more or less fixed chronological hierarchy. Phlebotomy is generally placed first, but is also one of the least chronologically fixed forms of therapy within this hierarchy.42 It is always qualified by the injunction that the physician must pay due regard to the patient’s general strength, based also on factors such as age and sex. If the patient’s strength allows it, phlebotomy is normally to be performed before the diatritus, failing that during the diatritus, or if not then not at all, in which case it is generally replaced by the use of some other sort of purgative treatment, such as a clyster. Food itself is to be given on the third day, i.e. during the diatritus, but is very often immediately preceded by some form of embrocation or anointment. Cupping-vessels are generally prescribed after this, if necessary, and at the corresponding intervals thereafter. Plasters are often applied on the day after the diatritus, i.e. on the day when no food is to be given. Clysters, too, generally come after food, but their suitability and use are dependent on varying factors. Baths are employed towards the decline of the disease, while wine is used only as patients are recuperating.

There is a striking regularity, though it is by no means entirely inflexible, with which the patterns of treatment in Caelius are arranged into this chronological hierarchy, and this gives a clear impression of a therapeutic system of generalised rules governing the use of individual types of dietetic remedy. The chronological structure provided by the diatritus thus forms the basis of a broader dietetic scheme, and we might therefore draw a distinction between the diatritus itself and this therapeutic system in which it played a crucial structuring role. There can be little doubt, too, that this system was a peculiarly Methodist one. Caelius placed himself within a therapeutic tradition which was avowedly Methodist, and his widespread attacks on the therapeutic errors of his predecessors are primarily directed at non-Methodists. On the not infrequent occasions, however, when Methodists such as Themison and Thessalus are brought to account for their mistakes, the repeated charge is that they failed to act in accordance with Methodist principles. There are also clear parallels in Caelius’ use of the diatritus with that found in the other major work of Methodist origin to have survived, Soranus’ Gynaecia. We may also note that Galen viewed Methodist dietetics in a comparable way, recognising in it a set of generalised therapeutic rules which could be universally applied, though in making reference to this he was concerned above all to emphasise the unforgiveable rigidity and simplicity of the Methodist system. With a clear reference again to Thessalus’ 6-month curriculum, Galen imagined how the Methodists teach their students:

“ἅπαντας τοὺς πυρέττοντας ἐν [alpha]ρχῇ μὲν ϕλεβοτομήσεις, ἐ[alpha]ν ἰσχυροὶ τὴν δύναμιν ὦσι˙ καταντλήσας δ᾽ ἐλαίῳ μετ[alpha] τὴν διάτριτον, ἤτοι μελικράτῳ μετ[alpha] χόνδρου θρέψεις ἢ ῥοϕήματι. κᾄπειθ᾽ ἑξῆς τρέϕε παρ[alpha] μίαν, [alpha]ρτομέλιτι προκαταπλάττων. εἰ δ᾽ ἐπισχεθείη ποτὲ ἡ γαστήρ, ὑπάγειν αὐτὴν κλυστῆρι.” τὴν μὲν οὖν ἐκείνων διαιτητικὴν τέχνην ἤδη σύμπασαν [alpha]κήκοας ἐν τοῖσδε τοῖς ῥήμασιν, οὐκ ἐν ἓξ μησίν, [alpha]λλ᾽ ἐν ἓξ στίχοις.

“You will use phlebotomy on all fever patients at the beginning, if they are strong as to their dynamis. Having anointed them with oil after the diatritus, you will feed them either with honeyed water and wheat or with gruel. Next, feed them every other day, applying first a plaster of bread soaked in honey. If the stomach should be constricted, you must evacuate downwards with a clyster.” You have just now heard their entire dietetic art in these words, not in six months, but in six lines.

(Meth. Med. 11.15 (10.781.4-11 K))

There are clear parallels here with the patterns which Caelius presents throughout the Celeres and Tardes Passiones, and although Galen’s presentation is radically simplified, it is significant that he could characterise Methodist dietetics as a system reducible to a set of simple guidelines in this way.

We can discern in Caelius, therefore, the ways in which the diatritus, as a structuring device, could be adopted and employed in the context of a wider, specifically Methodist, therapeutic framework. But, as mentioned, the diatritus was also employed by non-Methodist physicians in therapy, and this raises questions as to how it was used and whether it could serve different purposes when divorced from its original Methodist context. Our evidence for the non-Methodist use of the diatritus consists of scattered mentions in the ps.-Galenic Introductio, Oribasius and the Anonymus Parisinus,43 as well as Galen’s account of its use by the Erasistratean doctor alongside his Methodist colleague (Meth. Med. 8.2 (10.536-541 K)),44 and I shall offer a brief comparison of its uses in these passages with the work of Caelius.45

A full evaluation of the diatritus in the Introductio is somewhat problematised by a few possible textual uncertainties in Kühn’s edition,46 but the prescription of food after the diatritus (14.731.17-18 K), the use of embrocations before the prescription of food (14.732.4-5 K), and the application of plasters and clysters on the day following the diatritus (14.732.6 K) correspond very closely to the general patterns found in Caelius. Galen’s Erasistratean physician also presents a united front with the Methodist whom he accompanies, and their recommendations are fully in keeping with the attested Methodist therapies. Nor do the excerpted passages which mention the diatritus preserved in the Eclogae Medicamentorum from the works of Oribasius display any clear divergences from what we find in Caelius’ works: there are repeated recommendations to keep the patient without food until the diatritus, generally prescribing phlebotomy beforehand if the patient is strong enough, or cupping-vessels or clysters if not (chs. 56.2 (CMG 6.2.2 p. 220.25 Raeder), 59.3 (p. 224.11-13 Raeder), 65 (p. 230.20-21 Raeder), 84.1 (p. 260.17 Raeder), 103.2 (p. 282.34 Raeder), and 149.6 (p. 307.36 Raeder)), as well as recommending embrocations around the same time as food (chs. 65 (p. 230.20-21 Raeder), and 84.1 (p. 260.17 Raeder)), and plasters after the diatritus (chs. 34.1 (CMG 6.2.2 p. 197.19 Raeder), and 84.1 (p. 260.17 Raeder)).

By contrast, however, the uses made of the diatritus in a further excerpt found in the work of Oribasius and in the treatise by the Anonymus Parisinus seem to deviate in certain significant ways from the patterns of treatment which Caelius’ work records. In his Collectiones Medicae, Oribasius preserves a passage on the therapeutic use of wine from the works of Herodotus, a Pneumatist physician of the early second century A.D., which contains a single reference to the diatritus:

καὶ τοὺς μὲν ἐπιπυρέξαντας, εἰ μὲν ἐμέσειαν, μακρ[alpha]ς μὲν γενομένης τῆς ἐπισημασίας, θρεπτέον μετ[alpha] ταύτην ὀλίγῳ σιτίῳ, μικρ[alpha]ς δέ, μετ[alpha] τὸν τῆς δευτέρας ὕπνον. εἰ δὲ μὴ ἐμέσειαν, τὴν [alpha]πὸ ταύτης ϕυλακτέον διάτριτον.

As for those in whom fever supervenes, if they vomit, and the attack is long, one should feed them after this (attack) with a little food, but if (the attack) is short, (feed them) after sleep on the second day. If they do not vomit, one should wait for the diatritus from this (attack).

(Coll. Med. 5.27.23 (CMG 6.1.1 p. 146.6-10 Raeder))

Herodotus prescribes food here at various times, immediately after the attack of fever, or on one of the next two successive days, directly contravening the Methodists’ prescription of food during the diatritus. It should be remarked, however, that this is the only mention of the diatritus in the relatively extensive therapeutic excerpts which Oribasius derives from the work of Herodotus.

The Anonymus Parisinus’ work On Acute and Chronic Diseases furnishes us with a much more substantial basis for comparing the various uses to which the diatritus could be put outside Methodism. Again, a systematic analysis of the therapies employed in it is beyond the scope of the present study, but it may suffice here to draw attention to a few examples which illustrate some of the diverse ways in which the treatise diverges from the Methodist system described above. Firstly, there is an overall lack of systematicity with which the diatritus is applied throughout the work. It is mentioned only in the therapeutic sections for 17 of the 51 diseases treated, and more often with regard to acute diseases than chronic.47 Elsewhere, as well as within some of these chapters, the third day is also simply referred to as ἡ τρίτη ἡμέρα, implying a choice on some level not to use the term diatritus. We may note too an interesting feature of the Anonymus’ therapeutics by which he switches between different chronological structuring tools, sometimes employing the diatritus, at others, apparently drawing on a more Hippocratic tradition with regard to the periodicity of disease, arranging therapies according to hebdomads, or periods of seven days. Generally these distinct tools are confined to separate sections, but occur also within the therapeutic recommendations for a single disease, as for example on paralysis (cf. esp. 21.3.1 and 3 (p. 126.5-6 and 15-18 Garofalo)). There is no such structuring device attested for Methodist therapeutics. There are also many specific points on which therapies diverge from the patterns found in Caelius. To focus again on food, certain details do indeed accord with our Methodist scheme, for example keeping the patient without food until the diatritus (e.g. 6.3.5 (p. 42.16 Garofalo); 7.3. (p. 52.13 Garofalo); 43.3.1 (p. 210.22-23 Garofalo)), but food is also regularly found to be prescribed every day, even from the beginning of a disease, and often indeed also with wine (e.g. 10.3.8 (p. 76.2-4 Garofalo); 12.3.4 (p. 84.7-10 Garofalo)), a remedy used by the Methodists only when the patient is securely in recovery. Important differences from Methodism also arise at a more theoretical level. For the Methodists, any reference to the underlying causes of a disease was felt to be irrelevant in determining therapy,48 but this is clearly not the position of the Anonymus, and not simply because the causes of the disease are listed at the beginning of each chapter. In several places, his causal framework directly affects his therapy: for example, at 14.3.2 (p. 96.9-11 Garofalo), the diverse causes of ileus demand a varied and bold form of therapy; at 19.3.1 (p. 118.13-14 Garofalo), the presence of a plethoric state in melancholy demands phlebotomy; and at 39.3.1 (p. 202.12-14 Garofalo), for ulceration of the bladder, the causes of the affection must be removed first, before turning one’s attention to the condition of the patient. Overall, the presence of the diatritus in this treatise appears in some ways to reflect an occasional employment as a useful term to denote a particular chronological structure in therapy, when this should be considered appropriate for a given pathological condition, but there are clearly many instances in which it is not felt to be appropriate, and for these different chronological referencing is used accordingly.

These sources thus strongly suggest that the concept of the diatritus, developed for specific purposes within the Methodist school, was nonetheless able to pass beyond its original context into medical practice more generally. In some cases, the appropriation of the diatritus by non-Methodist physicians appears also to bring with it the adoption of therapeutic patterns which correspond closely to those used by Methodists, suggesting that Methodism could at times have a direct and significant influence on healing practices outside this school. On the other hand, for some authors, namely Herodotus and the Anonymus Parisinus, the use of the diatritus seems to represent a thorough separation of the concept from its original Methodist therapeutic context. We find no consistent patterns emerging in the employment of the diatritus in its new situations here, and the few similarities to Methodist therapeutics are outweighed by more telling differences. We might wonder, then, why these authors would adopt the term at all in this case. The surviving evidence does not admit of any firm answers, but we might speculate that the diatritus system was able with time to exert some influence over the way in which the chronology of disease was viewed generally, principally by attaching a high degree of therapeutic significance to the recurring third day. It may be that the terminology introduced to describe this, viz. the diatritus, achieved a wider currency among physicians of various doctrinal backgrounds, and could then be used independently of its original context to signify in a neutral way the recurring third day of an illness, without necessarily implying any associated doctrine.

As a final remark on the diatritus in practice, we should also note that it could apparently be used as a tool to diagnose the seriousness of the disease’s development. Caelius Aurelianus, at Cel. Pass. 1.35 (p. 40.26 Bendz), on the signs of phrenitis, describes how mental aberration (alienatio) can signal varying degrees in the severity of a particular case of phrenitis if it occurs before or after the diatritus. He elaborates at 1.39 (p. 44.6 Bendz), specifying that those who are affected by phrenitis after the diatritus are in a worse state than those who are affected before it (peius etiam laborare dicimus eos, qui post primam diatriton fuerint hac affecti passione, quam qui ante ipsam).49


The system of the diatritus originated in the mid-first century A.D. within Thessalian Methodism, and continued to be used as a therapeutic tool throughout the Roman period into Late Antiquity until at least the fifth century, while a mention in a gynaecological excerpt in Paul of Aegina (3.64.2 (CMG 9.1 p. 280.19 Heiberg)) suggests perhaps that some may still have acknowledged its relevance in the seventh. It was a product of the Methodists’ emphasis on generality in disease and their denial of the relevance of individuality in therapy, which was reflected in a corresponding reduction in the complexities of treatment. It thus provided a simple, universally applied chronological structure around which a therapeutic system, in the form of a set of general rules, could be developed for the treatment of any disease. Part of the point of the diatritus was to create a system which could be divorced from any theoretical basis which the physician was required to understand in order to be able to employ it. As we have seen, the diatritus did have some theoretical grounding, but the practising physician needed only to be able to count the days from the initial onset of any given disease. Caelius Aurelianus’ monumental therapeutic work employs the diatritus throughout, but nowhere offers any rational justification for why it should be considered appropriate. Its adherents were thus freed from the obligation to understand the impaired faculties associated with disease or the constitutions of their individual patients, but were forced to respond to the observable changes and developments which the patient underwent.

The simplicity of the diatritus system should not be over-emphasised, however. In his six-line summation discussed above, Galen presents Methodist dietetics as a pseudo-technê characterised not only by extreme rigidity and straightforwardness, but also by the fact that those who use it need know nothing about the disease or the patient who suffers from it, or indeed about medicine generally. But this does no justice to the flexibility and complexity which is inherent to the system of the diatritus. One of its strengths and most important features was that it allowed, or rather demanded, a process of constant re-evaluation of the condition of the patient and the progress of the disease.50 The therapeutic recommendations associated with the diatritus, although they do often fall into a fairly regular pattern, are in fact always dependent on the observations of the physician when faced with a particular case. Thus the timing and form of Caelius’ therapies are repeatedly conditional upon factors such as the patient’s strength, the severity of the disease, the particular ‘commonality’ to which the disease at that moment belonged, etc. A considerable amount of experience was required to be able to identify which were the relevant factors, to read the signs, and to tailor therapy to changing circumstances. In this way, the diatritus served the physician by allowing him a necessary and sufficient period of time in which to determine the progress and development of the disease, where diseases are conceived of as constantly changing, fluid entities, affecting the patient in unpredictable and unexpected ways. An additional effect of the structured framework of the diatritus system, we might imagine, would have been to offer the doctor a certain sense of control in the face of a bewildering array of diseases and symptoms. This assurance must also have extended in part to the patient, whose treatment now required regular visitations and was refined periodically according to the distinctive development of the disease’s symptoms. These various considerations may also have been attractive to doctors from a range of theoretical backgrounds, and one of the more significant features of the diatritus system is that it enables us to trace the influence of certain aspects of Methodist therapeutics upon medical practice more broadly.


1An earlier draft of this paper benefited from comments and criticism by Prof. Vivian Nutton, and I thank him warmly for his help. I would like to express my gratitude to the anonymous reader whose comments have improved this article in a variety of ways. Much of the work was carried out during a Research Fellowship funded by the Wellcome Trust, to which I am also grateful.

2Gyn. 3.10.5 (p. 100.8 Ilberg = p. 10.123 Burguière-Gourevitch-Malinas Vol. 3); 3.11.1 (p. 100.10-11 Ilberg = p. 10.126 Burguière-Gourevitch-Malinas Vol. 3); 3.28.4 (p. 111.10 Ilberg = p. 28.80 Burguière-Gourevitch-Malinas Vol. 3); and 4.37.3 (p. 150.15 Ilberg = p. 27.108 Burguière-Gourevitch-Malinas Vol. 4).

3Notably in the ps.-Galenic Intro. (later second century A.D.), Oribasius’ Coll. Med. and Ecl. Med. (fourth century A.D.), and once in Paul of Aegina (seventh century A.D.). The diatritus is also mentioned on a papyrus dated to the end of the first or beginning of the second century A.D., preserving a text on phlebotomy, at col. ii lines 43-44, for which see I. Andorlini, ‘Trattato o catechismo? La tecnica della flebotomia in PSI inv. CNR 85/86’, in ead. (ed.), ‘Specimina’ per il Corpus dei Papiri Greci di Medicina (Florence, 1997), 153-68. In my ‘A note on the Greek medical fragment PGolenischeff’, in I. Andorlini (ed.), Greek Medical Papyri Vol. 2 (2008) no. 15, I also argue that the diatritus is referred to at col. ii line 11 of this third century A.D. papyrus.

4The text is edited in I. Garofalo, Anonymi Medici De Morbis Acutis et Chroniis (Leiden-New York-Köln, 1997), with English translation by B. Fuchs. The dating of the treatise remains problematic.

5D. Le Clerc, Histoire de la médecine (Amsterdam, 1723), 451 and 473 offers some observations, describing it as ‘le terme de trois jours’ and discussing its usage in Caelius Aurelianus. H. Diller in RE, s. v. Thessalus (6) coll. 168-82, at coll. 177-8, offers a short but helpful description with reference to Thessalus’ doctrine. M. Wellmann, ‘Herodots Werk περὶ τ[omega with tilde]ν ὀξέων καὶ χρονίων νοσημάτων’, Hermes 40 (1905), 580-604, at 596 and 599, mentions it with regard to its employment in the Anonymus Parisinus and its links with Methodism, but goes little beyond describing it as ‘die Lehre von der dreitägigen Periode’. T. Meyer-Steineg, ‘Das medizinische System der Methodiker: eine Vorstudie zu Caelius Aurelianus “De morbis acutis et chronicis”’, Jenaer medizin-historische Beiträge 7/8 (1916), 81-2, discusses the diatritus as a ‘Dreitagsfrist’. J. Pigeaud in recent years has made somewhat more mention of the diatritus, but does not go into details, referring readers to Le Clerc (1723): cf. J. Pigeaud, ‘Les origines du Méthodisme d’après Maladies aiguës et Maladies chroniques de Caelius Aurélien’, in I. Mazzini & F. Fusco (edd.), I testi di medicina latini antichi. Problemi filologichi e storichi (Macerata, 1985), 321-38, at 331-2; id., ‘L’introduction du Méthodisme à Rome’, in ANRW 2.37.1 (Berlin-New York, 1993), 565-99, at 594-5. Cf. also D. Gourevitch, ‘La pratique méthodique: définition de la maladie, indication et traitement’, in P. Mudry & J. Pigeaud (edd.), Les Écoles Médicales à Rome (Geneva, 1991), 51-81, at 74-6, where the diatritus, described as a ‘dogme méthodique des trois jours de jeûne’, is mentioned with regard to an analysis of Galen’s hostility to Methodism.

6Philo, De Confusione Linguarum 151 (ed. P. Wendland, Philonis Alexandrini opera quae supersunt Vol. 2 (Berlin, 1897), at 258): καὶ ἐπὶ τῆς πολίτιδος τὸ κατασκευαστὸν τὸ παραπλήσιον ἰδεῖν ἔστιν˙ αἵ τε γ[alpha]ρ [alpha]μϕημεριναὶ καὶ διάτριτοι καὶ τεταρταΐζουσαι παρ[alpha] παισὶν ἰατρ[omega with tilde]ν λεγόμεναι περίοδοι μεθ᾽ ἡμέραν τε καὶ νύκτωρ περὶ τ[alpha]ς αὐτ[alpha]ς ὥρας κατασκήπτουσι τὴν εἰς αὐτ[alpha] καὶ τάξιν ϕυλάττουσαι. The text is problematic on a number of points, and I offer the translation given by F. H. Colson & G. H. Whitaker in their Loeb edition of Philo, Vol. 4: ‘And the same studied regularity may be noticed in fever. For the recurrences which are called in the medical schools quotidian, or tertian, or quartan, make their visitation about the same hour of the day or night and maintain their relative order.’ On the text, see further their appendix, at 557-8. The reading of διάτριτοι at any rate is secure.

7For the complexity of the counting systems which this kind of analysis generated, sometimes involving books of tables, and the various conflicting and overlapping usages of terms such as περίοδος, τάξις and τύπος, cf. esp. Gal. Adv. eos qui de Typ. Scrips. 1 (7.475-9 K), and, much less polemical in tone, De Typ. (7.463-74 K).

8Galen’s clearest statements are found at Meth. Med. 4.4 (10.264.4-5 K): ὁ σοϕώτατος Θεσσαλός, ὁ τὴν πρώτην διάτριτον ἐξευρών (most wise Thessalus, the inventor of the first diatritus); and at De Cris. 2.3 (9.657.14-16 K): ..., [alpha]λλ[alpha] τὴν τ[omega with tilde]ν ἐριουργ[omega with tilde]ν [alpha]ναμενοῦμεν διάτριτον, ἣν νομίζων εἶναι [alpha]ναγκαῖον ἐν τῇ γυναικωνίτιδι τρεϕόμενος ὁ ληρώδης Θεσσαλὸς ὑπὸ πατρὶ μοχθηρ[omega with tilde]ς ἔρια ξαίνοντι (..., but shall we wait for the wool-workers’ diatritus, which Thessalus the ranter thought was necessary, reared in the women’s quarters by his father as he miserably carded wool). Cf. also De Cris. 2.3 (9.655-656 K), where the diatritus is again closely associated with Thessalus, and De Praen. 12.6 (14.663 K = p. 132.4-5 Nutton): π[omega with tilde]ς δὲ οὐ μέλλομεν, ἔϕη, τὴν Θεσσαλείαν ὑπερβάλλειν διάτριτον (“How”, he said, “can we not wait for the Thessalian diatritus?”). Cf. also Pigeaud (1993) (n. 5), 594 and n. 112. Furthermore, with regard to the first of these statements, from Meth. Med. 4.4, I suggest that the text may be corrupt, and that an emendation to ‘ὁ πρ[omega with tilde]τος τὴν διάτριτον ἐξευρών’ (the first to discover the diatritus) would offer much better sense than the transmitted ‘ὁ τὴν πρώτην διάτριτον ἐξευρών’. The naming of a πρ[omega with tilde]τος ἐξευρών is such a traditional motif (cf. e.g. Ar. Nu. 737; Aeschin. In Ctes. 82; D.S. 12.53.2; Plut. Glor. Ath. 346A), and it would fit well both with the context and with Galen’s sarcasm and ironic praise concerning Thessalus in the Meth. Med. generally. Galen uses this formulation elsewhere most often with reference to Hippocrates’ achievements (e.g. Meth. Med. 4.5 (10.292 K), ὡς Ἱπποκράτης ἁπάντων πρ[omega with tilde]τος ὧν ἴσμεν ἐξεῦρε τοῦτον τὸν σκοπὸν ...; De Elem. ex Hip. 1.487 K; In Hip. Epid. VI 17A.823 K), but also with reference to his own (UP 3.576 K). I have checked the fourteenth century ms. British Library Add. 6898, in which the relevant passage is at fol. 39r, and it agrees with Kühn’s text, but the proposed error is a simple one, and may have been influenced by other passages in the Meth. Med. in which ‘the first diatritus’ is certainly referred to (Meth. Med. 8.3 (10.559.5 and 6-7 K), 10.6 (10.692.9-10 K)). We could scarcely ask for a more direct statement on the origins of the diatritus than this.

9Diller (n. 5), col. 178.

10Cel. Pass. 1.155 (p. 108.15-16 Bendz): primo igitur ex prima diatrito nutriendos dicit nullis etiam cogentibus causis, quod est improprium tempori (First he (i.e. Themison) says that (patients) should be fed from the first diatritus, even when there are no compelling reasons, which is inappropriate to the time).

11The fragments are collected in A. Guardasole (ed.), Frammenti, testo critico, introduzione, traduzione e commentario: Eraclido di Taranto (Naples, 1999).

12At Cel. Pass. 3.217-219 (pp. 418-420 Bendz), on cholera, Caelius describes how Heraclides gave food on the first day and again on the third before remarking at 218-219 (p. 420.5-9 Bendz): sed huic erit respondendum, utrum in declinatione passionis dari dixerit vinum atque varium cibum an ante ipsam, quod est intemporale atque importunum. sed si in declinatione, cur declinante passione et, quantum creditur, forsitan et exclusa usque ad primam diatriton abstinendos existimat aegrotantes? (But (Heraclides) would have to answer whether he meant that wine and a varied diet should be given in the abatement of the affection or before it, which is ill-timed and unsuitable. And if (he meant that it should be given) in the abatement, why does he think that, with the affection abating and, as is conceivable, perhaps even having disappeared, they should be kept without food until the first diatritus?).

13For Thessalus generally, see esp. Plin. HN 29.9; Gal. Meth. Med. 1.1-2 (10.4-9 K). Regarding the dating of Thessalus’ floruit, M. Tecusan, The Fragments of the Methodists Vol. 1 (Leiden-Boston, 2004), 15, suggests that his letter to the Emperor Nero should perhaps be regarded as the act of an established physician, and may have occurred in an advanced stage of his career, though it must be remarked that he does not otherwise appear to have been any sort of respecter of tradition, and his shameless self-promotion is the object of both Pliny’s and Galen’s disapproval. For Thessalus and Methodist doctrine, see esp. Diller (n. 5); T. Meyer-Steineg, ‘Thessalos von Tralles’, Sudhoffs Archiv für Geschichte der Medizin 4 (1911), 89-108; Pigeaud (1993) (n. 5), 594-9.

14Cf. Meth. Med. 1.1 (10.4-5 K).

15For discussion of Thessalus’ dependence on Themison’s ideas in the development of Methodism, cf., e.g., Pigeaud (1985) (n. 5), 332-7, and id. (1993) (n. 5), where he concludes at 599, though perhaps underestimating Thessalus’ contributions, that Themison ‘en a donné les notions dynamiques, et Thessalus n’a fait qu’aménager du point de vue théoretique’.

16When one should give food to the patient first appears to have become one of the most important questions in dietetics during the Hellenistic and Roman periods. Celsus refers to it clearly here, but its formulation as a specific question seems to date back to the early Hellenistic period, and it has been suggested that it was inspired by a particularly famous discussion by Erasistratus: cf. W. D. Smith, ‘Hippocrates’ Regimen in Acute Diseases and the later history of therapy of acute diseases’, in R. Wittern & P. Pellegrin (edd.), Hippokratische Medizin und antike Philosophie (Hildesheim, 1996), 477-89, at 483-6; A. Roselli, ‘Notes on the doxai of doctors in Galen’s commentaries on Hippocrates’, in P. J. van der Eijk (ed.), Ancient Histories of Medicine (Leiden-Boston-Köln, 1999), 359-81, at 373-9. In some ways, the diatritus can function as an answer to this question, and it should be considered in the light of this tradition.

17Caelius Aurelianus also seems to suggest that Themison made use of a rather inchoate or unsophisticated notion of the third day. At Tard. Pass. 2.58 (p. 578.15-20 Bendz), he has Themison giving food on the third day to patients suffering from paralysis, as he does in cases of phrenitis in the passage referred to above, but he seems concerned here to distinguish this from the doctrine of the diatritus: tunc tertia die cibum probat atque quarta die cucurbitam imponi adiuncta scarificatione non interrogans passionis tempus, solum numerum dierum imprudenter attendens. dehinc sequentibus gestationem probat et neque in eo definiens tempus atque audaciter secundae diatriti expectationem spernens. (Then on the third day he prescribes food and on the fourth day the application of cupping-glasses with scarification. He does not inquire into the time of the affection, but only considers the number of days in an ignorant manner. Then on the following days he prescribes exercise, without setting a limit to the time there either, and boldly rejects waiting for the second diatritus). Caelius objects to the fact that Themison does not pay attention to the ‘time’ of the disease, as opposed to the number of days, which might be interpreted as a reference to the diatritus and its imposition of a chronological structure onto the disease. The force of imprudenter here may also reflect Caelius’ misgivings about a certain arbitrariness in the number of days employed by Themison, and the reference to ‘the second diatritus’ suggests that Themison did not make use of consecutive diatriti. Meyer-Steineg (n. 5), 81-2, cites this passage as evidence for Themison’s invention of the diatritus, but there is no suggestion here from Caelius that he employed the diatritus at all: Caelius is describing what Themison did not do, again from his own (diatritus-influenced) point of view. It should be noted too that this passage is prefaced at 2.57 (p. 578.7 Bendz) by one of Caelius’ recurring statements that Themison did not, at least at this stage of his career, reflect true Methodism (nondum rectam Methodon respiciens). On this passage generally, cf. Pigeaud (1985) (n. 5), 331-2.

18Cf. n. 5 above, though Diller (n. 5), col. 178 describes it more accurately: ‘man ... zählte jeweils den dritten Tag (unter Eischluß des Anfangs- und Schlußtages, d. h. also jeden ungeraden Tag des Verlaufs, die διάτριτος)’.

19Cf., e.g., Tard. Pass. 1.57 (p. 462.9 Bendz), usque ad tertium diem, quem Graeci diatriton vocaverunt; 1.81 (p. 476.21 Bendz), in tertium diem sive, ut Graeci vocant, diatriton; 2.97 (p. 602.27 Bendz), usque ad tertium diem, quem Graeci diatriton vocant; 3.24 (p. 692.16 Bendz), qua perfecta si tertia fuerit dies, quam Graeci diatriton vocant.

20Cf. Pigeaud (1993) (n. 5), 595, quoting Le Clerc (n. 5), 473, ‘il faut remarquer que Caelius donne le nom de diatritos, non seulement à l’éspace de trois jours, mais encore au troisième jour en particulier’; cf. also Pigeaud (1985) (n. 5), 332 n. 6.

21As, e.g., at Cel. Pass. 1.70 (p. 62.5-6 Bendz), si passio cogit, intra diatriton, si minus, in ipsa prima diatrito; 3.181 (p. 398.23-24 Bendz), cogentibus rebus intra diatriton, permittentibus autem in ipsa diatrito.

22Cf. also the translations of the term throughout I. E. Drabkin (ed.), Caelius Aurelianus: On Acute Diseases and On Chronic Diseases (Chicago, 1950).

23Cf. OLD s. v. intra 3 b: ‘within the period terminated by, before, by. ... apparuisse numen deorum ~ finem anni vertentis Ant. in Cic. Phil. 13. 22; ~ annum ... vicesimum feminae notitiam habuisse in turpissimis habent rebus Caes. Gal. 6. 21. 5; nisi ~ sextum mensem sciero Petr. 53. 8; ~ quadragensimum pugnae diem Tac. Hist. 2. 70’; also Lewis & Short s. v. intra II B 2: ‘intra decimum diem quam Pheras venerat, i.e. before the lapse of ten days after his arrival, Liv. 36,10,1′.

24Cf. Cel. Pass. 2.104 (p. 200.16-17 Bendz), si nullus ventris fuerit fluor, intra tertium diem, sin vero fuerit adiuncta ventris solutio, in ipsa diatrito erit phlebotomia adhibenda.

25Coincidentally, it may be relevant to note that the earliest surviving attestation of the term διάτριτος, in the Philo passage discussed above, is descriptive of the frequency of paroxysms in fever, though in a quite different sense.

26Cf., e.g., Gal. Meth. Med. 8.2 (10.540.16-541.3 K), in a passage which will be further discussed below, where two physicians who employ the diatritus congratulate a patient whom they believe to have followed their therapeutic advice: ἐπαινέσαντες οὖν αὐτὸν ἐπὶ τῷ πεισθῆναί σϕισι καὶ μηδὲν ἁμαρτεῖν πρὸ τοῦ δι[alpha] τρίτης παροξυσμοῦ˙ “τοιγάρτοι δι[alpha] τοῦτο” ἔϕασαν “ἤδη μέν σοι πέπαυται τ[alpha] τῆς ϕρίκης, ἱδρ[omega with tilde]τος δ᾽ ἐστὶν ὑπόϕασις καὶ νοτὶς πολλὴ περὶ τὸ δέρμα, καὶ ταῦτ᾽ οὐκ ἂν ἐγένετο μὴ [alpha]σιτήσαντός σου καὶ τὴν διάτριτον ὑπερβάλλοντος.” (And so they praised him for obeying them and for making no mistake before the paroxysm on the third day. “You see,” they said, “it is because of this that your (symptom) of shivering has stopped, and there is an appearance of sweat and a lot of moisture over your skin; and this would not have happened had you not fasted and passed the diatritus.”). Cf. also ibid. 4.3 (10.559.1-5 K); and Niccolò da Reggio’s translation of Galen, De Procatarcticis 2.13 (ed. R. J. Hankinson, Galen: On Antecedent Causes (Cambridge, 1998)), aliis vero medicis expectare videbatur et videre eum qui per terciam paroxismum qualis erit, where his phrase should render the Greek ‘τὸν δι[alpha] τρίτης παροξυσμόν’. The context here is certainly that of the diatritus (cf. ibid. 3.21).

27Cf. Meth. Med. 8.3 (10.561.7-11 K), μὴ τοίνυν ἔτι θαύμαζε δι[alpha] τί πολλοὶ τ[omega with tilde]ν ἰατρ[omega with tilde]ν οὐκ αἰσχύνονται ϕθεγγόμενοι τὴν διάτριτον, καίτοι μυριάκις ἡμ[alpha]ς ἑωρακότες ἐπὶ τ[omega with tilde]ν νοσούντων προγινώσκοντας εἴτ᾽ ἐϕήμερός ἐστιν ὁ πυρετὸς εἴτε καὶ κατ[alpha] τὴν τρίτην ἢ τετάρτην ἡμέραν οἴσει τιν[alpha] παροξυσμόν. (Do not be surprised still at why many doctors are not ashamed to shout about the diatritus, even though they have seen us a thousand times making prognoses for patients as to whether the fever is daily, or whether it will bring a paroxysm on the third or fourth day).

28It should be noted that, as perhaps with this passage, Sextus’ following sophism (2.239-240) is certainly taken from a Methodist context (cf. terminology such as στέγνωσις, πύκνωσις, ὁ χαλαστικὸς τρόπος τῆς ἐπιμελείας, the references to relevant conditions (τ[alpha] προσεχῆ πάθη, i.e. the κοινότητες), etc.). It is possible that Thessalus in particular is lurking somewhere in the background here, if not necessarily consciously or directly. The roundabout references to the κοινότητες, with the qualification that they are προσεχεῖς, recalls Thessalus’ definition of the Method as γν[omega with tilde]σις ϕαινομένων κοινοτήτων with the specification προσεχ[omega with tilde]ν καὶ [alpha]ναγκαίων πρὸς ὑγίειαν (cf. Galen, De Sect. Introd. 6 (Scr. Min. Vol. 3 p. 14.4-8 Helmreich); also POxy. LII 3654 fr. 8. 3-6). Sextus was plainly familiar with the details of Methodist doctrine (cf. PH 1.236-41), and was ready to use aspects of this in support of his own arguments, but scholars have overstated the nature of his ties with Methodism. Cf. the discussions of this relationship in M. Frede, ‘The method of the so-called Methodical school of medicine’, in id., Essays in Ancient Philosophy (Minneapolis, 1987), 261-78, at 276-8; G. E. R. Lloyd, Science, Folklore and Ideology (Cambridge, 1983), 182-200; and A. Bailey, Sextus Empiricus and Pyrrhonean Scepticism (Oxford, 2002), 97-9.

29For Antipater the Methodist, see M. Wellmann in RE, s.v. Antipater (32) and (33), col. 2517; and M. Tecusan (2004) (n. 13), 45-51.

30Caelius does not always employ the Greek term when he refers to the diatritus, and for him diatritus and tertius dies are often synonymous (see above). We have also seen how he retrojects the term in discussing previous authors where he means simply ‘the third day’.

31For directions to give food ‘after the diatritus’, cf., e.g., ps.-Gal. Intro. 13 (14.731 K); Anon. Paris. On Acute and Chronic Diseases 1.3.3; 2.3.3; 14.3.6, 20 etc.

32For directions to keep the patient without food ‘until the diatritus’, cf., e.g., Sor. Gyn. 3.28.4; 4.37.3; Anon. Paris. On Acute and Chronic Diseases 6.3.5; 7.3.2; Orib. Ecl. Med. 56.2-3; 65.1; 84.1; 149.6.

33For the necessity to ‘pass the diatritus’ without food, cf., e.g., Gal. Meth. Med. 10.536, 542, 559 K.

34Le Clerc (n. 5), 473 notes the contradiction between Caelius Aurelianus’ and Galen’s accounts, without mentioning Galen’s own internal contradictions. He seeks to account for this firstly by positing a scribal error in the transmission of the Galenic work, or alternatively with the suggestion that Soranus, as Caelius’ source, gave food on the third day in opposition to the practice of other Methodists to whom Galen refers. There is, however, no evidence for either hypothesis.

35Note especially the repetition of οἶμαι and the use of δηλαδή in the first passage, which would seem to cast some doubt on Galen’s full commitment to his account.

36Caelius elsewhere warns against giving food in the evening, e.g., at Tard. Pass. 2.215-6 (p. 674 Bendz); 3.146 (p. 766 Bendz).

37διάτριτον suppl. Garofalo (n. 4).

38Cf. above n. 5. Certainly, Galen focuses mainly on the enforced period of fasting involved in the diatritus, but this is the aspect of it to which he objects most strongly, and with which he can attack it most effectively.

39It is important to note the distinction among Methodists, with regard to chronic diseases, between the form of therapy used during periods of attacks of the disease (in which the diatritus was used, as for all acute diseases), and that used between such periods when the disease was dormant (in which the metasyncritic cycle was used). Cf. e.g. Cael. Aur. Tard. Pass. 1.96 (p. 486 Bendz), on the two parts of therapy used for chronic diseases, here with reference to epilepsy; also Sor. Gyn. 3.28.6 (p. 111.21-30 Ilberg = p. 29.95-105 Burguière-Gourevitch-Malinas Vol. 3).

40Exercise appears to have been deemed unsuitable for patients being treated according to the diatritus. We might compare the various comments made by Galen and his interlocutors, which characterise the immobile state, and indeed boredom, of the patient awaiting the diatritus: e.g., at Meth. Med. 8.6 (10.582.11 K), the patient should be lying down watching the clock; at De Praen 12.8 (14.663-664 K = p. 132.15 Nutton), patients are to be forced to lie down.

41For the overall structure employed by Caelius in his individual sections on each disease, cf. A. M. Urso, Dall’autore al traduttore. Studi sulle Passiones celeres e tardae di Celio Aureliano (Messina, 1997), 23-37, and on the structure of the therapeutic sections, 33-6.

42Cf., again, the discussion by Caelius at Tard. Pass. 1.80-1 (p. 476 Bendz).

43Aside from, or rather in connection with, the issue of his identification, the doctrinal background of the Anonymus Parisinus has been the subject of some discussion among scholars, whether emphasising the Methodist or Pneumatist elements of his work, but there seems to be no good reason to consider him a Methodist: cf. esp. P. van der Eijk, ‘The Anonymus Parisinus and the doctrines of “the ancients”’, in id. (ed.), Ancient Histories of Medicine (Leiden-Boston-Köln, 1999), 295-331, at 324-9, with references to earlier literature.

44Knowledge of the diatritus was apparently not restricted to its medical practitioners. Galen also provides some interesting anecdotal evidence for the reactions of certain members of the public to its clinical application. Cf. Meth. Med. 8.6 (10.582 K), where a patient’s friend, a non-medical person, jokingly styles physicians who adhere to the diatritus ‘οἱ διατριτάριοι’. This may be a pun on diatretarii, carvers or artisans of open-work decoration (cf. Cod. Iust. 10.64.1; Cod. Theod. 13.4.2), its derogatory connotations presumably lying in its association with a non-scientific, manual trade - one might also compare Galen’s constant jibes at Thessalus’ father’s supposed occupation as a wool-worker (cf. n. 8). Annia Faustina, a close relative of the emperor Marcus Aurelius, also discusses the use of the diatritus with some of its adherents in Galen’s presence at De Praen. 12.5-9 (14.663-664 K = pp. 130-132 Nutton).

45The lack of surviving context in the two papyri mentioning the diatritus, PSI inv. CNR 85/86 and PGolenischeff (see n. 3), does not allow us to say definitively whether or not these authors were Methodists, but the PSI fragment employs other terminology which is consistent with Methodism, and the advice found in both papyri on the use of phlebotomy parallels the general recommendations in Caelius.

46At 14.731.13, 14 and 15 K, where phlebotomy and the giving of food are under discussion, it seems likely that where the obscure phrase τῆς δι[alpha] τρίτης is printed, we should in fact have a reference to the diatritus, reading τῆς διατρίτου. For certainty we shall have to await C. Petit’s forthcoming edition in the Budé series.

47Namely ch. 1 (phrenitis), ch. 2 (lethargy), ch. 3 (epilepsy), ch. 6 (synanche), ch. 7 (spasm), ch. 8 (pleurisy), ch. 9 (pneumonia), ch. 13 (cholera), ch. 14 (ileus), ch. 16 (satyriasis), ch. 21 (paralysis), ch. 26 (haemoptysis), ch. 33 (jaundice), ch. 41 (lientery), ch. 46 (tenesmus), ch. 47 (gonorrhoea), and ch. 50 (sciatica).

48For a more nuanced examination of the Methodists’ position with regard to the relevance of the causes of disease, drawing attention to certain discussions in Caelius where causes do indeed seem to be important for therapy, cf. P. van der Eijk, ‘The Methodism of Caelius Aurelianus: some epistemological issues’, in P. Mudry (ed.), Le traité des Maladies Aiguës et des Maladies Chroniques de Caelius Aurelianus. Nouvelles approches (Actes du colloque de Lausanne) (Nantes, 1999), 47-83, at 63-75 (repr. in P. van der Eijk (ed.), Medicine and Philosophy in Classical Antiquity (Cambridge, 2005), 299-327, at 311-21). It should be noted, however, that the majority of these discussions of cause are introduced in order to determine whether or not there is a disease in the first place, and therefore whether therapy is needed at all. The consideration of cause cannot be said to influence directly how one treats an illness, but pertains merely to the question of whether there is one to treat.

49Cf. also perhaps Gal. De Cris. 2.3 (9.655.17-656.2 K), where he derides those who still wait for the diatritus in order to be able to diagnose a tertian fever.

50As V. Nutton, Ancient Medicine (London, 2004), 193, notes, drawing a contrast between this process and the Hippocratic approach associated with a pre-determined, fixed regimen concurrent with diagnosis. Cf. also V. Nutton, ‘Therapeutic methods and Methodist therapeutics in the Roman Empire’, in Y. Kawakita, S. Sakai & Y. Otsuka (edd.), History of Therapy: Proceedings of the 10th International Symposium on the Comparative History of Medicine - East and West (Sept. 8th-15th 1985, Susono-shi, Shizuoka, Japan) (1990), 1-35, at 21-4.