1Both the epithet Hippocratic and the unity and historicity of the Corpus as a whole are subject to increasing scepticism. Still in the 19^th and early 20^th centuries, the Hippocratic Corpus tended to be viewed in a similar way as other textual corpora of classical antiquity to which author names are attached, e.g. the Platonic, Aristotelian or Galenic Corpus. In these corpora, there is a core of undoubtedly genuine writings, accompanied by pseudepigraphic treatises from later times. Similarly, 19^th and early 20^th century scholars tried to trace the “genuine” Hippocratic treatises within the Hippocratic Corpus, allegedly written by the great Hippocrates himself. For it was common opinion that the Corpus, or rather its nucleus, existed in pre-Alexandrian times and went back to Hippocrates, his closest family circle or at least the Coan school of medicine. In the late 20^th and early 21^st centuries, researchers have become increasingly sceptical about these claims. Hippocrates or the “Hippocratic author” is now usually addressed with a uniform disclaimer stating that the name of Hippocrates is only used for matters of tradition and convenience and that no allusion to the historic Hippocrates and his possible doctrines are intended.1 Lorenzo Perilli (2009) has argued for dismembering the apparent unity of the Corpus upon consideration of its sources. According to him, the source nucleus of the technical treatises may have originated from archives of medical writing tablets that were kept in temples of healing gods, regarded as common propriety and used by a variety of physicians of different backgrounds2. A comparable3, slightly more radical view on the Corpus is expressed in Philip van der Eijk’s (2016) opening paper in the proceedings of the Austin Colloque Hippocratique (2008). Van der Eijk argues that even the notion of corpus should be abandonned with reference to the Hippocratic writings. Instead, he suggests, single treatises should be contextualized with theories of other contemporary medical and philosophical writings. Hippocratic studies, he concludes, should rather be called “studies of Greek medical thought and writing in the 5^th and 4^th c. BCE”. Is such a radical view justified? When, in the following, the “unitarian” roots of the Corpus are being recalled, this is by no means intended as polemic against such “non-unitarian” views, as I may term them. The unitarian roots do not hinder from contextualising Hippocratic writings beyond the boundaries of the Corpus. They do, however, underline that the Corpus is by no means a haphazard accumulation of disparate writings, but an intricate philological unity.

I. What is a Corpus? What is the Hippocratic Corpus? – The Corpus and its unifying traits

2Any attempt of defining or examining, to what degree the Hippocratic Corpus constitutes a corpus, needs to start with a definition of the term “corpus”. According to the Oxford English Dictionary, a corpus (3.a.) is “A body or complete collection of writings or the like; the whole body of literature on any subject.” (as examples the Corpus of the Civil Law and the Corpus of the Greek Poets are given.)

3To put it more generally, one could state that a corpus is a group of treatises sharing certain features, a tertium comparationis, which allows to bundle them. There is a wide range of possible criteria for formation of literary corpora, the most important of which are:

I.1. Common authorship

4As far as corpus formation is concerned, the criterion of common authorship presupposes a nucleus of writings whose authorship is beyond doubt, as in the case of the abovementioned Platonic, Aristotelian or Galenic corpus. In case of the Hippocratic corpus this criterion is, of course, hardly convincing, since we lack evidence from without the Corpus which treatises go back to a common author, be it “Hippocrates” or someone else.4 If all treatises that bear the name of Hippocrates or that were at some stage of their transmission attributed to Hippocrates were to be united in a corpus, a number of pseudepigraphic late antique and medieval treatises would have to be included5 along with a list of titles of treatises lost today.

I.2. Editorial choice, process of canonisation: Hippocrates, the “other Hippocrates” and still two “other Hippocrateses”

5The second possibility of corpus formation is the result of a process of canonisation or a choice of editors. This is the case for the actual Hippocratic Corpus. Both the corpus of the 1526 Aldine edition and the one of Littré’s 19^th century edition are nothing else than an attempt to get as close as possible to the most extensive list of Hippocratic writings transmitted, the index of the Vaticanus Graecus 276 (= V, 12^th c.). Since V and its apographs lack a number of treatises included in V’s index, other manuscripts had to be used to fill these gaps, primarily the Marc. Ven. 269 (= M, 11^th c.) and its apographs.6 The result is a corpus slightly smaller than the corpus of the index of V7 and slightly larger than the sum of Hippocratic treatises known to Galen.8 These divergencies only concern minor and somewhat spurious treatises (with the exception of the lost surgical treatise Wounds and Missiles).

6Hereby, a certain bias for the study of “Hippocratic” writings was established, since the aforementioned pseudohippocratic writings of late antiquity and the middle ages were excluded. Consequently, they have received little attention by scholars and have not even been edited, in most cases. Jacques Jouanna has used the term “the other Hippocrates”9 to refer to these little studied pseudepigraphic treatises. There are two more instances in which one might speak of an “other Hippocrates”. Firstly, with reference to lost Hippocratic treatises that were still known in Imperial Roman times, late antiquity and Early Byzantium. Titles of such lost treatises are transmitted in the Vat. Gr. 27610 and the Brussels Vita of Hippocrates. Only one of these lost treatises can be partly reconstructed from fragments (Wounds and Missiles)11; as to the second one (a treatise on twins), it is doubtful whether a surviving fragment indeed belongs to this treatise.12 Max Wellmann wrote a study on the Hygieinon, an allegedly lost treatise, the title of which is included in the index of the Vat. Gr. 276.13

7But even in a third case the term “other Hippocrates” might be used, i.e. with reference to lost treatises or source material that was quarried by those authors of the Hippocratic Corpus who produced the early compilations of the Corpus: Aphorisms, Coan Prenotions and Crises. This third “other Hippocrates” or rather “Proto-Hippocrates” is to be explored in the following.

I.3. Same doctrine or subject matter

8A third possibility of corpus formation would be a close coherence of doctrine or ideas suggesting common authorship or at least the possibility to ascribe treatises to a common school of medicine. The same holds for sub-corpus formation. As is well known, it has been tried in the past to differenciate between Coan and Cnidian treatises, a distinction that was finally abandoned.

9In order to answer the question, whether the Hippocratic Corpus has such ties of common doctrine, van der Eijk takes up a passage by Vivian Nutton in which the latter summarises the characteristic features of Hippocratic medicine, according to the common opinion. For each of these features, van der Eijk looks for similar doctrines in other medical or philosophical writings contemporary with the Hippocratic Corpus. The eight features and their counterparts in other physicians and philosophers are:

• Rational, yet religiously founded medicine opposing superstition – this feature can be found in Diocles, Aristotle and Plato as well
• Scientific nature of medicine: as well in Diocles, Praxagoras, Mnesitheus, Democritus
• Belief in healing power of nature: Aristotle
• Avoidance of extreme: Diocles, Aristotle, Plato
• Consideration of the environment: Diocles, Praxagoras, Aristotle
• Admission of medical failure: Plato, Aristotle
• Lack of financial interest: Diocles, Plato, Aristotle
• Importance of good communication doctor/patient: Plato

10From this survey van der Eijk concludes that “none of the allegedly typical attributes mentioned … are unique to the Hippocratic writings as a whole and are also found in ‘non-hippocratic’ authors of the same time frame.” One could object to his method, he continues, that he was “just picking and choosing from a wide variety of authors”. However, in his opinion, this method “is not so different from the habit of arbitrarily picking and choosing from the [Hippocratic Corpus] those passages that provide answers … or that best fit”. Thus, he concludes, there is no coherence of the so-called Hippocratic writings justifying their unification into a corpus. The Hippocratic Corpus should therefore be dissolved, not longer be regarded as a unity and treatises rather contextualized with material from without the Corpus. Hippocratic studies should rather be referred to as “study of Greek medical thought and writing in the 5^th and 4^th c. BCE.”14 Quod erat demonstrandum.

11But are things really so straightforward? Any attempt at proving or disproving the doctrinal coherence of the Hippocratic Corpus tends to be a self-fulfilling prophecy, since it is and has always been a well established fact that there is no such thing as a coherent Hippocratic lore. Any resemblances in doctrine can therefore necessarily just be resemblances of a certain group of treatises.

I.4. Philological coherence

12It is well known that there is a complex network of intertextual connections throughout the writings of the corpus.15 Some treatises that are now separate (Fractures/Articulations and Haemorrhoids/Fistulas) once cohered. Other texts are amalgamations, reelaborations and/or summaries of more substantial treatises: Mochlicon, Nature of Bones, Anatomy, Humours, Use of Liquids and Surgery quarried and summarized Fractures, Articulations, Prognostic and Prorrhetic 2. The last four of these treatises have close textual interdependencies with Epidemics 2, (4) and 6. At times, parallel passages in the Epidemics can even help to illuminate obscure passages in the derivative texts whose obscurity is due to excessive abridgement or condensation.16

13These so-called “parallel redactions” are passages of one treatise that reoccur either almost verbatim or more or less modified in another treatises.17 Parallel redactions are the most unique feature of Hippocratic intertexuality and allow to establish historic mappings of groups of Hippocratic treatises. They suggest that the nosological treatises like Affections, Internal Affections, Diseases 2 and 3 derive from a common source lost today.18 The same holds for the gynaecological treatises like Diseases of Women 1, 2 and Nature of Woman19 as well as for the gynaecological and embryological treatises Generation, Nature of the Child, Diseases 4, Diseases of Girls, Glands, Diseases of the Women 3 and Nature of Woman.20 For some surgical treatises, Pétrequin21 and some modern editors list a number of fairly specific parallels in therapy (wounds must not be moistened with anything except for wine, usage of laxatives in wound care, wounds that are not cleansed are prone to hypergranulation, round wounds must be made oblong so that they can heal, etc.).22 These marked parallels seem to suggest that the treatises draw on common sources. Despite of certain differences in style, these and other common traits made Littré postulate (III,270 L.) that the surgical treatises once belonged together, forming one comprehensive surgical treatise in serveral “books” that was later divided – and possibly reelaborated or rewritten by several different authors, as one might add. Parallel redactions and such specific parallels in content will certainly not be found by comparing any Hippocratic treatise with writings outside the Hippocratic Corpus, e.g. by Diocles, Plato or Aristotle. In the following it will be argued that the philological coherence of the treatises is the core feature, if not the very criterion that makes the Hippocratic Corpus a corpus.

II. The compilations within the Hippocratic Collection

14The term “compilation” derives from the Latin verb compilare (“to plunder from different sources”). The noun compilatio is first used by Cicero (ep. fam. 2,8,1) to designate a collection of files or documents. Since the Middle Ages compilatio was used for derivative texts, excerpted and/or summarised from preexisting sources. At first sight it might seem fairly easy to differenciate excerpts from summaries. However, this is not always the case, since compilations may contain both excerpts and summaries, while more important passages are excerpted more or less verbatim, less important passages are summarised and the whole is then rearranged.23

15Medical literature is, in its essence, Gebrauchsliteratur, in which practical knowledge for everyday professional use and teaching is transmitted. Through the ages, this type of literature was and is particularly prone for being subject to compilation. A possible claim, that the history of medical literature was basically a history of compilations might be a little overstated, but proves to be true in many instances. In cases when a core of received knowledge is transmitted and/or reelaborated for didactic purposes, as it virtually always applies for medical literature, a certain degree of compilation is unavoidable. This is especially true for the Hippocratic treatises. Langholf 2004 observed that the notion of intellectual property did not yet exist when these works were compiled.24 Their partly ill-structured and chaotic format is likely due to the fact that there was, at that period of time, no tradition or convention on how to compose and arrange scientific treatises.25

16All in all, two types of compilations may be distinguished: “monographic” compilations, as one might call them, and collection-like compilations. As it was stated above, the surgical and gynaecological writings of the Hippocratic Corpus respectively constitute a kind of subcorpus within the Corpus, in that they draw on a core of common material that was later reelaborated several times, as it seems. This is particularly evident in the gynaecological writings in which, by means of philologic analysis, different strata of different age can be differenciated.26 Clearly, older texts were quarried and their excerpts were embedded in new material. For those surgical writings which show many parallels in content, but not the same style of writing one might suppose that they go back to the same source material which later was drawn on and reshaped by different authors. Such works of monographic appearance which nevertheless contain compilated material might be called “monographic” compilations. Celsus’ On Medicine is certainly one, as well, even if it is more likely to contain bits and pieces from different sources, rather than being a complete translation plagiate from a Greek work, as Max Wellmann (1913) claimed. The monographs of Galen’s vast oeuvre are certainly compilations to a certain degree, as well. However, due to the fact that most of his sources are lost this can hardly be proven.

17There is yet another type of compilations, those who are collection-like. Such compilations are created whenever excerpts are not embedded into a continuous text, but just juxtaposed, one after another. This holds for the Aphorisms, Coan Prenotions and Crises that are in the focus of the present paper.

18The source of a compilation, either monographic or collection-like, may be a library, an archive or working files27 in form of e.g. wax tablets from which the excerpts were taken: there have been speculations about libraries of the Coan and Cnidian schools of medicine,28 libraries of temples and shrines;29 Galen (On Difficulty in Breathing 3, VII,890,15ff. K. and Commentary on Hippocrates’ Epidemics 6, XVIIA p. 796,12 seqq. K.) relates a common opinion of his time according to which Thessalus, Hippocrates’ son, found his deceased father’s unfinished workfiles in form of written skins and wax tablets. He is said to have compiled Epidemics 2 and 6 using these working files.30 Langholf 1989 found out that a number of Hippocratic writings has text units of similar length (in series or as appendices – προσκείμενα), corresponding to one standard writing tablet or approximately 100 epic verses. This may hint to the method of composition, that in these instances indeed writing tablets have been copied.31

19A source for compilation may also constitute a canon, like, many centuries later, the teaching canon of the Alexandrian school of medicine for the Alexandrian Summaries. It may also be an ad-hoc choice, whenever for the very purpose of the compilation the literature available to the compilator was perused and the best or most useful of it selected for taking excerpts (as in case of Oribasius’ Medical Collections or in case of the Roman Digests). Every single compilation thus reflects a certain underlying “corpus” of sources, some of which got lost after the compilation was made.

III. The “Circle of the Aphorisms”

20Karl Deichgräber was very well aware of the fact that the early compilations of the Hippocratic Corpus contain fragments of writings or sources lost today. At the end of his well-known monograph “Die Epidemien und das Corpus Hippocraticum” (1933) he reaches the following conclusion:

“Actually, the following remains to be done: the tree compilations of the Corpus would have to be analysed, the well-known compilation ‘Aphorisms’ the mass of excerpts π. κρισίων (IX,276 L.) and the κῳακαὶ προγνώσεις – or rather the writings that have been drawn on in these compilations.
The author who collected the ‘Aphorisms’ primarily used four writings of the Corpus … [Epid. II, VI, Progn., Aer., Liqu.] ... Apparently, to a Coan physician, these treatises were available as a closed circle of writings. Besides, he used other treatises … and several ones that are not transmitted, but which he also must have counted to this circle. It is an urgent task and a prerequisite for a definite study on the Coan school of medicine to reconstruct the lost writings as far as possible. If it is not possible, these excerpts should at least be examined as to whether and in what way they belong together.”32

21What Deichgräber calls a “closed circle of writings” (“fester Schriftenkreis”) is termed “circle of the Aphorisms” (“cercle des Aphorismes”) by Caroline Magdelaine. According to her, its core consists of the five treatises Acut., Progn., Aer., Epid. 2, 4, 6, Hum., to which were added excerpts from Mal. 1+2, Nat. hom., Us. Liq., Gyn., plus excerpts from a number of unknown sources.

22Interestingly, none of the writings which are traditionally thought to constitute the authoritative core of writings from the Coan school of medicine, writings that in the past used to be connected with the historical Hippocrates (Epidemics 1 and 3 and the surgical writings) have been excerpted by the authors of the early compilations in the Corpus (save for a few allegded excerpts from Wounds in the Head which – at a closer look – do not necessarily derive from this treatise, see below).

23To sum up, the early compilations of the Corpus reflected, according to Deichgräber, the oldest core of the Hippocratic Corpus, a core that was a kind of “proto-Hippocratic corpus” and included some writings (or, more generally speaking, sources) lost today. This alleged ancient pool of writings which served as the point of cristallisation for what later became the Hippocratic Corpus can however not be very easily reconstructed. For a number of interferences might have taken place by which the original pool of writings got obscured.

24First of all, one should consider that compilations like the Aphorisms are not necessarily the product of one single compiler, but could have been written, or rewritten at different periods of times by different authors, so that the material excerpted might go back to not just one, but several source corpora. Furthermore, seeming parallels to works that still exist could be due to the fact that not these very works were excerpted, but rather one or more intermediary source upon which these works draw. Consequently, the compiler could have had at hand quite another work than the one that seems to have been excerpted. Maybe the author did not even draw from treatises proper, but from an archive of files, wax tablets or the like (Langholf 1989, see above). Even oral transmission could have been included to some extent.33 On examining the parallel redactions in Prorrhetic I and Coan Prognoses, Elisabeth Craik observes that in many instances metrical patterns can be found (dactylic, anapestic, iambic etc. sequences). These metrical patterns seem to have been a mnemonic aid for oral transmission of these respective aphorisms (Craik hypothesised that this material may have come from shrines in Thessaly).34 A systematic metrical analysis of the aphoristic works of the Corpus is still a desideratum in scholarly research.

25Works like the Aphorisms that appear to be compilations do not necessarily have to be compilations. They could be original compositions, that were used by later authors to compose “monographic compilations” like Humours and Airs, Waters and Places. A seeming source treatise could therefore be the compilated work, not the one in aphoristic form. Of course, this question of priority is complicated and can hardly be completely solved.35 Of two parallel passages, the one that seems to be “more complete” does not necessarily have to be the source but could also be the result of contamination. Furthermore, an aphoristic work could consist of both compilated passages and passages of independent composition, so that passages of unknown origin might indeed be original contributions rather than excerpts from lost works. Finally, older works in aphoristic form could have been used rather than monographic treatises.

26In this context the mysterious “Cnidian Sentences” (Κνίδιαι γνῶμαι) can be stated as a possible example for earlier works in aphoristic form. This treatise is polemised against by the author of Acute Diseases which hence has the alternative title “against the Cnidian Sentences” – Πρὸς τὰς Κνιδίας γνώμας. The Cnidian Sentences were still available in Imperial Roman times (whether this was indeed the ancient original, a later redaction or reelaboration is debatable) and two passages are quoted by Rufus of Ephesus (Names of the Parts of the Human Body 190 = p. 159,13-160,2 Daremberg/Ruelle = frg. 14 Grensemann 1975) and Galen (Commentary on Epidemics VI,29 = XVIIa,888 K. = p. 54,1-4 Wenkebach = frg. 13 Grensemann 1975) respectively. Whether γνῶμαι in the title is to be translated by “sentences”, “opinions”, “signs” or “symptomatology” is unclear and cannot be answered from the two surviving fragments.36 However they do have metrical patterns once again, a fact that could be a hint to aphoristic form and orally transmitted material (see Craik’s observation above). The fragment quoted by Galen is entirely dactylic (11 dactyls), the one quoted by Rufus begins with four dactyls and continues with a more complicated metrical pattern.

27The Cnidian Sentence seem to have been transmitted along with Hippocratic writings (due to there textual ties to Acute Diseases) without ever having had the name of Hippocrates attached to them. In antiquity there was, of course, no such thing as a Hippocratic Corpus, the notion of corpus being a modern, anachronistic term. Only writings with the author name Hippocrates existed, some of them thought genuine, some of them spurious with different people having different views about this question. Consequently, a work with the local epithet “Cnidian” can never have had the name of the Coan Hippocrates attached to it!37 Furthermore, Acute Diseases is transmitted with the author name Hippocrates. The treatise this author polemises against can, under no circumstances, have been deemed to be by Hippocrates as well!

IV. Inventory of sources

28Anyone wishing to explore the sources of the aphoristic treatises of the Hippocratic Corpus has four ressources at hand: 1. Littré’s Hippocrates edition (Paris 1839-1861) which indicates many parallels to other Hippocratic works (this inventory is, all too correctly, called “excessively inflated” by Elisabeth Craik (2015, p. 50), since there is no differenciation between parallels proper and different degrees of vague ressemblances). 2. Also Daremberg’s edition of select Hippocratic writings (Paris 1843) lists these parallels. Daremberg largely seems to draw on Littré. Loci paralleli are furthermore indicated by 3. Ermerins (1832, 1859) and 4. Jones (1923-1931) in their respective editions.

29Since Deichgräber, critical editions and/or textual studies have appeared for all the Hippocratic compilations that are discussed here: the PhD theseis by Preiser for Crises and Days of Crisis (Kiel 1957), by Poeppel (Kiel 1959) and Ferracci (Paris 2009) for Coan Prenotions and by Magdelaine (Paris 1994) for the Aphorisms.

30In all of these studies extensive chapters about the parallels are included. All of these scholars draw from Littré and/or the other sources indicated, adding own observations. There are, however, several problems involved, which prevent from using their material directly for further analysis.

31First of all, the parallels are always given as lists, not in synoptic form. It is therefore not immediately visible to what degree parallels exist (not all resemblances are indeed parallels) – whether these parallels are verbatim, in the sense of a paraphrasis or whether they are completely vague. Even if a difference is made by the scholars between parallels and near parallels, it is inevitable for further research to restart with a synopsis that contains the complete wording of the aphorisms to be analysed and their respective sources, by underlining or highlighting what material is verbatim or parallel, since every scholar tends to consider the divergences differently (e.g. in the Loeb edition 68 parallels between Aphorisms/Coan Prenotions are mentioned, whereas Ferracci lists 85).

32In preparation for the present paper, synopses for all aphorisms of the collections discussed here and their (possible) sources were made. Due to limitations of space, the integral synopses cannot be printed here. Only a sketch of this survey can be made in which the most remarkable findings are highlighted.

33All in all, there are three types of aphorisms: little problems are connected with those which have 1. verbatim parallels in other Hippocratic treatises or which have only slightly modified text, compared with their sources. 2. Aphorisms that have no parallels in the surviving Corpus at all do not present a problem, either. Most difficult to assess are those aphorisms which appear to be 3. either a close or remote paraphrasis of passages otherwise known. In these cases, one can never be sure whether really these passages were excerpted or another intermediary source that is not available any more. “There is a vast penumbra of material, similar enough to appear linked yet not quite the same”, as Elisabeth Craik (2015, p. 33) observed.

34In the following, a very rough and sketch-like overwiew of the three early compilations of the Corpus will be given. Only in the Aphorisms the clusters of quotations will be analysed in more detail. While the Aphorisms focus on therapeutics, aetiology and dietetics, Coan Prenotions and Crises are uniquely devoted to prognostics.

IV. 1. Aphorisms

35Since at least Galenic times, the Aphorisms are divided into seven sections38, the first of which deals with general diaetetics and purgation. The first 18 aphorisms of this section (as well as aphorism 1,25) have no known source or counterpart in other works of the Hippocratic Corpus. Due to a marked resemblance of doctrine and thought to Acute Diseases, Littré (IV,397) thought the first 18 aphorisms to be a derivate from this treatise, in the form of a free paraphrasis.39 Magdelaine (p. 23-25), on examining these parallels in doctrine in more detail, considers two more possibilities as for the underlying source (hereby following Lonie 1965): it could either be a lost treatise that was close to Acute Diseases or a version of it that was more complete than the one we have it today. The first 18 aphorisms form a coherent unity with a clearly recognizable line of argument (“ensemble cohérent” … “redigé sous une forme continue”), a fact that is underlined by the grammatical particles used. Magdelaine hence concludes that the model must indeed have been a continuous treatise, lost today, the author of which is likely to be the same as the author of Acute Diseases.

36Section 1,19-1,24 is an almost verbatim excerpt from a coherent section of Humours (mostly excerpt type 1, only in the case of 1,24 type 3, see the following synopsis)40. This fact makes Magdelaine’s thesis the more plausible. Why should a compilator at one point paraphrase Acute Diseases very freely, to a degree that it is even doubtful whether these excerpts are from this work – and at another point take excerpts from another treatise, Humours, almost word by word? However, it must be kept in mind that Humours is by no means an original treatise. Rather, it is a derivative work as well, a “collection of personal memoranda”,41 close to Epidemics 2, 4 and 6. Littré (V,470) remarked that it could be called the “eighth book of Epidemics”. The compiler of the Aphorisms might therefore not have drawn on Humours, but on a precursor source, common to both Epidemics 2, 4, 6 and Humours. In the following synoptic tables parallels are marked as follows: litteral accordances are printed in italics. Underlined are words or passages with literal accordances where the word order has been changed or where words have been transposed to another place. Passages which have been reworded or where synonyms are used are printed in bold.

37The 54 aphorisms of section two deal with regimen and prognosis. They have no counterpart in other Hippocratic works, save for the four aphorisms 2,11-2,14 who parallel passages in the Epidemics. These do not form a cluster of excerpts from one single paragraph (as could be above in the case of Hum.). They rather parallel passages from different places of Epidemics 2, 4 and 6, apparently in random order, unless one assumed that they reflect an earlier textual stage of the Epidemics 2, 4 and 6 in which these passages were indeed neighbouring ones. A wording similar to aphorism 2,12 occurs in five different places in the Epidemics.42

38The third section of the Aphorisms (31 aphorisms) deals with prognosis and prognostic signs, particularly in fevers.43 The first half of this section consists of three consecutive clusters of excerpts, from the treatises Humours (3,1-3,7), Epidemics (3,8-3,10) and Airs, Waters, Places (3,11-3,14) respectively. Whereas in Airs, Waters, Places one continuous passage of one single chapter has been fragmented, excerpts from Humours were taken from six, mostly consecutive chapters. Excerpts from both treatises were taken fairly verbatim. Between these two clusters, three short excerpts from the Epidemics were inserted, two excerpts from consecutive propositions of Epidemics 4 and one from Epidemics 6.

39The second half of Aphorisms, section 3 (3,15-3,31) consists of excerpts taken from one or several source(s) unknown today.

40The forth section of the Aphorisms opens with aphorisms on artificial evacuations (4,1-4,20) for which no source is known. The following aphorisms on black excrements (4,21-4,26) have no known source, either, nor aphorisms 4,27-4,35 on fevers, save for the short cluster 4,31-4,33 which is an excerpt from chapter 7 of Humours:

41Aphorisms 4,36-4,68 on sweating are, again, of unknown source, except for the cluster 4,52-55 that is taken from Epidemics 2, 4, and 6. The aphorisms 4,56-59 and 4,61 have remote parallels to Diseases and Epidemics 2 and 6.

42In aphorisms 4,69-4,81 on urines, only the sources of 4,74 and 4,76-4,79 can be identified. 4,74 was either taken from Humours or the verbatim parallel passage in Epidemics 6. The aphorisms 4,76-4.79 are a cluster taken from chapters 13,2-4 of Nature of Man.

43Section five of the Aphorisms focuses on prognostic signs, the effects of heat and cold as well as on gynaecology (5,28-63). Whereas the source of aphorisms 5,1-5,15 remains unknown, 5,16-22 and 5,25 were taken from chapters 1,2, 1,4, 2,1, 6,2, and 6,4 of Use of Liquids, 5,24 and 5,26 from the Epidemics 2 and 6.

44The aphorisms 5,28-63 have a common focus on gynaecology. Within this section, the parallels to other Hippocratic writings are scarce: only aphorisms 5,32-34 appear related to Dieseases 1,7, 1,8 and 5,43-46 and have some common wording with Dieseases 1,3 and Nature of Woman 12,17,19 and 20.These aphorisms are, however, not excerpts. The parallels are rather remote and seem to suggest the use of common intermediary sources. Ann Elis Hanson has devoted a comprehensive study to this gynaecological secion of the aphorisms,44 comparing wording and content to other gynaecological writings of the Corpus. She came to the result that the content of the gynaecological aphorisms is more progressive than the substantial Hippocratic gynaecological treatises, suggesting that this section was updated once or several times after it had been written.

45In contrast, 5.65-5,71 are quite verbatim excerpts from the Epidemics 2 and 6 (the last aphorism of this section, 5,72, has no counterpart).

46As section 5 ended with excerpts from Epidemics 2 and 6, the beginning of section 6 continues to do so.

47Finally, the aphorisms 6,15-20, 6,24-25 and 6,50-51 show remoter parallels to Diseases 1,3, 1,4, 1,7 and 1,8, sect. 7, while there are no real, verbatim parallels. This again suggests that the treatise Diseases was not directly excerpted, but rather, that common source material was used.

48All in all, there are 250 parallels of the Aphorisms with other writings of the Corpus, all of them listed by Magdelaine. In the synopseis printed above only the verbatim parallels of the quotation clusters were included, in order to illustrate the interdependencies of the corpus. The Aphorisms are called the “cornerstone of the Corpus” by Magdelaine since they are both a receptacle of older knowledge and a source for other writings (i.e. other compilations).

49In the following two short paragraphs a very succinct summary of the sources of Coan Prenotions and of Crises will be given, based upon the findings of Ferracci and Preiser. A synoptic comparison cannot be given here due to lack of space. Furthermore, a synopsis seemed less instructive, since both writings do not have clusters of excerpts as in case of the Aphorisms.

IV. 2. Coan Prenotions

50The treatise Coan Prenotions is a collection of 640 aphorisms, the half of which (348 according to Ferracci) has closer or farther parallels with passages of other writings of the Corpus: Aphorisms, Prognostic, Prorrhetic 1, Epidemics 2, 4, 6, 7, On Sevens, On Diseases 1-3, Acute Diseases (B, spurious part) and Crises. Ferracci also regards Epidemics 1,3, Wounds in the Head and On Days of Crises as sources of some aphorisms in Coan Prenotions. However, on comparing these four works referred to by Ferracci to Coan Prenotions, one notices that the parallels are far too remote to suggest any direct use of material. This may be illustrated in the following synopsis for the alleged parallels with Wounds in the Head:

51348 aphorisms of Coan Prenotions have no known source and are thus likely to be excerpts from lost treatises, files or source material of whatever format. In contrast to the Aphorisms, there are no clusters of excerpts from only one single treatise.

52There are multiple parallels with other compilations of the Hippocratic Corpus, a fact that led to various theories concerning the interdependencies of these treatises. Aphorisms and Coan Prenotions have approximately 85 parallels which are mostly less verbatim and show numerous contaminations). Prorrhetic I was completely incorporatey in Coan Prenotions, Prognostic almost completely; only 4 chapters of this treatise were not, for unknown reasons. There have been varying theories, which treatises are older and which ones younger, which one was the source and which one composed from excerpts45.

Galen: Prognostic, Aphorisms, Epidemics > Prorrhetic and Coan Prenotions

Ermerins 1 (1832), Littré: Prorrhetic 1 > Coan Prenotions > Prognostic

Ermerins 2 (1859): X > Coan Prenotions/Prorrhetic 1 > Prognostic

Daremberg: Progn. > Prorrh. 1 > Coan Prenotions

Jones, Poeppel, Magdelaine, Ferracci: Aphorisms, Prognostic, Prorrhetic 1, Epidemics, Acute Diseases, On Diseases > Coan Prenotions.

53The last thesis is common opinion among modern scholars and is supported by the following arguments: In Coan Presotions a systematic replacement of vocabulary, as opposed to Aphorisms and Prognostic, can be observed. Older and rather obsolete terms are replaced by more modern ones. Coan Prenotions is in better order than Aphorisms and Prorrhrtic 1 which makes it more likely that Coan Prenotions is the younger treatise. The fact that passages in Coan Prenotions seem to be “more complete” than parallel passages in its alleged source treatises is explained by textual contamination46.

IV. 3. Crises

54Crises is neither mentioned by Erotianus nor by Galen. It is a compilation from Aphorisms, Coan Prenotions, Prognostic, Epidemics 2, 4, Acute Diseases (B, spurious part), Sevens and a lost work.47 The date of this treatise is uncertain, ranging between the end of the 4^th c. B.C. and the 5^th c. A.D.48 The content of the excerpted sources is reproduced without any modifications, save for the fact that the vocabulary was modernised.49 This lost work was excerpted in §§ 9, 11a, 11d, 11e, 38, 39e, 46 of Crises, passages dealing with bilious remittent fever (καῦσος) and icterus. That these passages are indeed from the same treatise is evident by similar vocabulary, style and doctrine.50 According to Preiser, Crises consists of nothing else than excerpts. The author is likely not to have contributed a single sentence.51

V. Summary

55After this succinct overview, Deichgräber’s claim needs to be taken up again, to reconstruct the lost sources of the “proto-Hippocratic corpus“ as far as possible or rather to examine how excerpts of unknown sources might belong together.

56The results are somewhat decieving. There is no evident coherence of unknown material from different compilations in a way that the same lost writings were provenly used by authors of all or only of more than one compilation of the Corpus. This may either suggest that a number of sources got lost – or that the “proto-Hippocratic corpus” was a rather open pool of writings to which, at different times, different sources were added. Not all compilers may have had access to the same sources or may not have deemed all of them equally important for their purposes. The following lost source material was used by the compilers of the three compilations Aphorisms, Coan Prenotions and Crises (this may either have been books, notes, files or other sources):

A source on kausos and icterus used by the author of Crises

A diaetetic source (the author of which seems to be the same as the author of Acute Diseases) (Aphorisms, sections 1 and 2)

A source on seasons/diseases (Aphorisms, section 3)

A source on prognostics/excretions of the body (Aphorisms, sections 4-6, Coan Prenotions?)

and one or several sources on gynaecology (Coan Prenotions 524-542; there are no correspondencies in this section with the gynaecological section of the Aphorisms, 5,28-63)

Furthermore an unknown quantity of oral material may have been included.

57Even a rough survey on the sources of the compilations of the Corpus, as the present paper, once again reveals the complex network of parallel redactions throughout the Corpus, ranging from verbatim parallels to free paraphraseis and punctual allusions, a network that seems impossible to disentangle. Textual parallels, as illustrated above in the synopses, can certainly not be found between the Hippocratic treatises and any fragments or works by Diocles, Mnesitheus, Aristotle and others.

58It is regrettable that there is no manual on all the parallel redactions and parallel passages within the Corpus, a kind of Anastassiou/Irmer comparing the writings of the Corpus to one another and indicating different forms of references (from verbatim to a vague paraphrasis). These informations still need to be pieced together from different critical editions, studies and commentaries. Such a manual (either in print or electronic format) remains a desideratum in scholarly research. It would need to list the parallel passages in synoptic form (with resemblances marked or underlined), rather than giving mere numeral text references in list format.

59Due to limitations of space, the present paper cannot provide a thorough study of the issue raised by Deichgräber. It is rather a sketch of the main problems involved and at the same time a manifesto, arguing for keeping the traditonal denomination of a “Hippocratic Corpus”. For the Corpus is undeniably a corpus by virtue of its multiple textual interdependencies which firmly unite the different treatises and create a philological coherence of the Corpus, even if treatises may vary in doctrine and scope and even though they were written and possibly re-written by a multitude of authors at different times. Dropping the name of “corpus” would renegate these evident philological bonds, so unique and characterisic of the Hippocratic Corpus. There is indeed no other textual corpus with such a multitude of textual interdependencies. Dropping the epithet “Hippocratic” would be an anachronism, as well, and detach the treatises from their historic perception and reception. It would be equally advisable to drop the epithet “Homeric” in the Homeric epics and rather speak of “Ionic epic poems in the 8^th c. BCE”52. That even with the epithet “Hippocratic” attached sound research on parallels in Diocles, Mnesitheus, Aristotle and others can be made, indeed a worthwile endeavour, is clear from studies like the ones by Hanson 2004 and Oser-Grote 2004.