A Greek Source for the Treatise on the Composition of Man Attributed to Aḥūdemmeh Anṭīpaṭrōs?

I. Introduction

Here I will show that a lengthy portion of the Treatise on the Composition of Man attributed to “Aḥūdemmeh Anṭīpaṭrōs” (or simply “Anṭīpaṭrōs”--the text has two separate attributions) has a close relationship to the Greek pseudo-Hippocratic treatise of the same name first edited by Jouanna in 2006; one is certainly a translation of the other.1 In this Syriac work we have a new addition to the “collection of texts” in various languages attesting to the post-Galenic “development of the theory of the four humors” discussed by Jouanna.2 Firm conclusions about the nature of this relationship (the place and time in which the translation occurred, as well as the translation’s direction) must await an edition of the Syriac text that incorporates all relevant manuscripts; my aim here is simply to establish the relationship between these texts. Along the way I will make some suggestions about the underlying causes of their affinities, as well as some adjustments that may need to be made to the Greek text on the basis of the Syriac.

The Treatise on the Composition of Man (Mēmrā ʿal rūkkābā d-barnāšā) attributed to Aḥūdemmeh Anṭīpaṭrōs was originally edited and translated by Chabot in 1943.3 This edition, which will serve as the basis for this article, was produced using the East Syrian manuscript copied in the early 20th century now known as CSCO Syr. 21.4 Chabot already suspected that this work was based on a Greek model, citing the Greek origin of the name Anṭīpaṭrōs, and the “numerous Greek technical terms” that appear in the text.5 Chabot suggests that the work’s Greek Vorlage had simply been attributed to a certain Antipatros; the name Aḥūdemmeh was the addition of an overzealous copyist, who was familiar with ʿAbdīšōʿ's attribution of a work with a similar title, the Treatise on the Composition of the Person, or Persons (Mēmrā ʿal rūkkāb qnōmā/qnōmē) to Aḥūdemmeh, and, assuming that this must be the same as what he had in front of him, “corrected” the attribution accordingly.6 Chabot mentions two Roman physicians named Antipatros as possible authors of this Greek Vorlage.7 As Chabot also notes, we have a West Syrian treatise attributed to Aḥūdemmeh, first edited by Nau, with the all-but-identical title Treatise on the Composition of Man (Mēmrā ʿal rūkkābeh d-barnāšā), but completely different content than the treatise attributed to Aḥūdemmeh Anṭīpaṭrōs; Chabot does not spell this possibility out, but a copyist aware of the former work’s existence but unfamiliar with its contents could have made an analogous overcorrection.8 Reinink adduces further support for the notion that the attribution of the East Syrian On the Composition of Man to Aḥūdemmeh is late and erroneous, in pointing to the work’s curious double attribution: the text’s very first line, after the title “Treatise on the Composition of Man, pronounced by Mār Aḥūdemmeh Anṭīpaṭrōs,” includes a greeting addressed to the “explorers of secret things... from Anṭīpāṭrōs.”9

Gignoux, meanwhile, has advanced a dissenting view, saying that this work is “difficult to attribute to a Roman physician,” but is rather an original Syriac composition, likely to be attributed to the same Aḥūdemmeh who wrote the West Syrian On the Composition of Man.10 Gignoux believes that this Aḥūdemmeh is to be identified with the East Syrian bishop “Aḥūdemmeh of Nisibis” who is named as a participant in a 544 church council.11 Gignoux makes two points to support his contention that this treatise is an original Syriac work, neither of which is particularly convincing. First, he floats the possibility that both Aḥūdemmeh and Anṭīpaṭrōs are not simply proper names in the attribution of the East Syrian On the Composition of Man, but are rather to be interpreted in their Syriac and Greek root meanings--“brother of his mother” and “in the place of the father,” respectively. For Gignoux, these terms could be working in concert to specify the role the author of this work played in his family--an uncle who served as a kind of “guardian” for his sister’s children.12 This explanation, while ingenious, does nothing to get around the difficulty the unmistakably Greek name Anṭīpaṭrōs poses for the identification of this text as an original Syriac composition; indeed, if “Anṭīpaṭrōs” was actually meant to mean “in the place of the father” here, rather than simply functioning as a name (given how well-attested this name is in the Roman Empire, the latter possibility is much more likely), this would actually be a strong indication that the bearer of this name lived among Greek speakers.13

Second, Gignoux maintains that this work’s content is “not well understood except within the Syriac tradition,” and that there is “no need to appeal to a Greco-Latin tradition” in order to account for its origins.14 Curiously, the support for this claim looks forward to a thirteenth-century didactic poem on the theme of “the human being as microcosm” authored by George Wardā, which, Gignoux convincingly demonstrates, has clear parallels with the treatise attributed to Aḥūdemmeh Anṭīpatrōs, and likely drew on this work, whether directly or indirectly.15 This has no clear relevance for the question of the origins of the Treatise on the Composition of Man.

As it turns out, a large section of the treatise attributed to Aḥūdemmeh Anṭīpatrōs, beginning on page 54 of Chabot’s edition, closing on page 57, and immediately followed by the words “thus said the great Hippocrates,” corresponds closely to the Greek pseudo-Hippocratic treatise On the Composition of Man edited and translated by Jouanna in 2006.16 As will be discussed below, multiple scenarios could explain this correspondence: the basis for the Greek could be an excerpt from the Syriac treatise attributed to Aḥūdemmeh Anṭīpatrōs, or one of the Syriac treatise’s component parts that circulated separately; or the basis for part of the Syriac treatise attributed to Aḥūdemmeh Anṭīpatrōs could be this Greek pseudo-Hippocratic treatise—the part of the Syriac not corresponding to this pseudo-Hippocratic treatise would then consist of other translations from Greek (or other languages), or original Syriac compositions.

Jouanna’s edition is based on four manuscripts: the 15th-century Ambrosianus gr. 331 (which I will hereafter designate as A., following Jouanna’s shorthand); the 16th-century Parisinus gr. 985 (P.); the 16th-century Parisinus gr. 2320 (Par.); and the 15th- or 16th-century Parisinus gr. 2494 (Pa.).17 Judging P. to be the manuscript preserving the treatise in its “most complete” form, Jouanna makes it the basis for his edition, but he leaves open the possibility that the version in P. is “in some respects” a more “developed” form of an originally “shorter version” and that Par. may preserve such a shorter version, rather than being a lacunary version of the text in P.18

In fact, as we will see, the section of the treatise attributed to Aḥūdemmeh Anṭīpatrōs corresponding to this Greek pseudo-Hippocratic work mostly agrees with Par., as well as Pa., over P., indicating that Jouanna’s suspicions about the chronological priority of Par. relative to P. were likely correct, and that the former manuscript should be prioritized accordingly for any Greek text. If the Syriac was the basis for the Greek, it is self-evident that the Greek manuscripts corresponding more closely to the Syriac should be prioritized. If, on the other hand, the Greek was the basis for the Syriac, it is probable that this translation was made sometime between the sixth and ninth centuries—the period in which medical translations are known to have been made from Greek into Syriac. Given that even the ninth century precedes our earliest Greek manuscript by some 500 years, in this scenario, too, the Syriac text should be assigned great importance in any edition of the Greek text.19 This agreement of the Syriac treatise with Par. and Pa. over P. extends to the work’s title, as the treatise in Par., along with Pa., is said to be on the “composition” (κατασκευή) of human beings, rather than on their “formation” (διάπλασις).20 Accordingly, I take On the Composition of Man to be the original title of the Greek treatise, and refer to it as such, rather than On the Formation of Man, as Jouanna calls it.

The Syriac On the Composition of Man opens with the aforementioned greeting from “Anṭīpatrōs” to the “investigators of secrets,” and a discussion of the great benefits the medical discussions of “the first philosophers” offer humanity.21 The final passage of this introductory section includes a list of some of these “philosophers,” who have helped to combat “subjection to harmful diseases,” in which Plato, Aristotle, Asclepiades, Chiron, Democritus, Hippocrates, and Galen are named explicitly.22

The section corresponding to the Greek On the Composition of Man comes next. I will first give a general overview of the contents of the text, and the relationship among the Syriac version and the four Greek manuscripts, before turning to a more in-depth comparison of the Syriac and Greek manuscripts’ respective renditions of selected passages and phrases. The text begins with a general description of “man,” outlining “his” relationship to God, and to the rest of creation (according to Jouanna’s numbering of the sections, which I will adopt here, this is section I).23 After this, it discusses the various attributes that help to define the human being, including a detailed accounting of the amounts of bones and “conduits” in the body (Jouanna’s II-III).24 The text then turns to the “places” or “homes” of the various humors in the body, and the places in which they have “dominion” (IV),25 followed by a discussion of the influences various parts of the body and their associated humors have on human seed (V),26 the ways in which the hour of conception influences the individual’s constitution, appearance and character (VI-IX),27 and, finally, the four different types of seed, and the combined influence of seed and “surrounding air” on the development of any creature (X).28 As discussed above, Jouanna bases his text on P., but Pa. and Par. are considerably closer to the Syriac than P. (or A.). Jouanna’s text generally corresponds to the Syriac quite closely, and discrepancies between Jouanna’s text and the Syriac can often be resolved by selecting a reading from Pa. or Par. over one from P. or A. Jouanna’s text diverges most from the Syriac in the sections concerning the influence of one’s time of conception (VI-IX); while Jouanna’s text and the Syriac continue to share general structural features here, the two texts’ specific contents diverge to a considerably greater extent than they do in the surrounding sections. Two features of the Greek manuscript heritage may help to explain this: Pa., perhaps the closest of the Greek manuscripts to the Syriac, breaks off just after the beginning of section V; while Par.’s rendering of sections VI-IX is considerably more laconic than that of P. (the basis of Jouanna’s text, of course), and is in some respects closer to the Syriac.

The Syriac On the Composition of Man continues after the conclusion of the section corresponding to the Greek pseudo-Hippocratic text. In the Syriac text, the latter section closes with the words “thus said the great Hippocrates,” almost certainly referring to the sentence immediately before it, which discusses the effect the surrounding air has on the constitution of the individual, in line with what is laid out in the Hippocratic work Airs, Waters and Places.29 After this, we have an enumeration of the parts of the body and their respective functions, including a broader description of the entire human body (introduced as a “return to our theme”) from the inside out: “the human body on the outside is skin, and inside the skin [there is flesh], and inside the flesh there is fat, and inside the fat there are arteries…”30 In Chabot’s text, the Treatise on the Composition of Man attributed to Aḥūdemmeh Anṭīpatrōs is directly followed by something entitled From The Book of Medicine; this latter is now recognized to be an entirely independent work from the treatise attributed to Aḥūdemmeh Anṭīpatrōs.31

Aside from the awkward double attribution at the very beginning, the treatise attributed to Aḥūdemmeh Anṭīpatrōs is seamless and coherent, proceeding from one topic to another with little redundancy or inconsistency: from a general history of medicine, to a more detailed discussion of the constituents of the human body, known thanks to the progress medical science has made.32 There is therefore no reason to suppose, purely on the basis of a reading of the Syriac, that we are dealing with some kind of compilation here; only a demonstration that the Greek is the basis for the section of the Syriac text to which it corresponds would lead us to that conclusion.

II. Comparative Analysis of Selected Passages

I now turn to an in-depth analysis of selected passages and expressions paralleling one another in the Greek and Syriac. This analysis will illustrate the general degree of correspondence between the Greek and Syriac, and the tendency of the Greek manuscripts Pa. and Par. to agree with the Syriac against P. and A. It may also shed light on the direction of the translation, though on this front my results remain preliminary and speculative. I will start from the very beginning of the text; as the following table shows, the profound similarities between the Greek and Syriac are already apparent here.

Syriac33	Greek (Jouanna text)34	Greek text (Pa. only)35
netnē mādēn w-nēmarʿal barnāšā. ak d-īt. w-hākanā bīdā bīdāʿal ḥšaḥtā d-sammmānē. barnāšā hākēl alāha-w d-arʿānē. brīta-w d-tahrā. mārā da-knawāteh. esārā d-tartēn brītē. dmūtā d-bāryeh. māna-w d-tedmūrtā. wa-ṣbūta-w da-b-tahrā qaymā.	Ὁ ἄνθρωπος οὗτος· θεός ἐστιν ἐπίγειος· ἐπεὶ καὶ πλάσμα θεϊκόν ἐστι· κύριος δὲ κατέστη ὑπὸ τοῦ δημιουργοῦ τῶν κτηνῶν καὶ ἀλόγων, καὶ πάσης τῆς ὁρατῆς κτίσεως· ἐζωσμένος τέ τοῖς δυσὶ μέρεσιν· εἰκὼν δέ ἐστι τοῦ πλάσαντος αὐτόν· ἐνδιαίτημά ἐστι θαυμαστόν· καὶ θέλημα ἐπιθυμητόν.	Ὁ ἄνθρωπος θεός ἐστιν ἐπίγειος· θεϊκὸν πλάσμα· κύριος ἐστὶν παντων τῶν κτηνῶν καὶ τῶν ἀλόγων ζώων· ἐζωσμένος τέ τοῖς δυσὶ μέρεσιν· εἰκὼν ἐστι τοῦ πλάσαντος αὐτόν· ἐνδιαίτημά ἐστι θαυμαστόν· καὶ θέλημα ἐστιν ἐπιθυμητόν.
Thus we will narrate, and we will speak of man, what he is; and likewise, little by little, about the use of drugs. Now, man is a god of terrestrial things, a wonderful creation, master of his fellow [beasts?]. bond [or girdle; esārā] of the two creations, the image of his creator. He is a marvelous instrument [or “vessel”; mānā], and is something [or a will; ṣbūtā] that consists of wonder [or, perhaps, “exists in wonder”].	Man is such. He is a terrestrial god, for precisely [his] formation is divine. He was made master of the beasts and of those lacking reason, and of every visible creature, by the craftsman. And he is girt with two parts. He is the image of the one who created him, and [his] dwelling-place is marvelous, and [his] willis desirable.	Man is a terrestrial god, a divine formation. He is master of the beasts and of the animals lacking reason. And he is girt with two parts. He is the image of the one who created him, and [his] dwelling-place is marvelous, and [his] will is desirable.

Even within these first few lines, then, we have several clear parallels between the two texts, on the most noteworthy among which I will now briefly comment. To begin, we have two particularly striking bits of nearly-equivalent phrasing. In the Syriac and Greek texts “man” is called a “god of terrestrial things” and a “terrestrial god,” respectively, in a nearly parallel construction, which finds the copula after the noun, but before the adjective (elāha-w d-arʿānē; θεός ἐστιν ἐπίγειος ). Further, the Syriac text refers to “man” as “the image of his creator” (dmūtā d-bāryeh) while the Greek has the almost identical “the image of the one who created him” (εἰκὼν…τοῦ πλάσαντος αὐτόν).

There are also some curious partial parallels here, which may point to the direction of the translation. First, there is the Syriac phrase “bond of the two creations” (esārā d-tartēn brītē) corresponding to “girt with two parts” in Greek (ἐζωσμένος... τοῖς δυσὶ μέρεσιν ). In this particular case, it is easier to imagine the Greek being a translation from the Syriac than vice-versa. This Syriac phrase “bond of the two creations” has a close parallel in an anonymous Syriac homily initially attributed to Ephrem, which Jansma believes to be of Greek origin.36 In this text, the human being is designated as the single “bond” (esārā) which God made in order to “gather and fasten together the creation” (neqoš w-neḥzoq l-brītā); humans are uniquely “related to” both the “invisible” and “visible” parts of creation, insofar as they possess both soul and body.37 The phrase “bond of the two creations” in the Syriac On the Composition of Man is surely an elliptical reference to the same notion, the unique link between the “invisible” and “visible” that humanity constitutes. As for the Greek “girt with two parts,” meanwhile, I have been unable to find such a clear parallel elsewhere in this language’s corpus, though here it could mean something analogous but not equivalent to the Syriac phrase: the “two parts” could be the body and soul, which the human being has “put on” like a piece of clothing. Furthermore, the Syriac word used for “bond” here (esārā) can also mean “girdle,” whereas the Greek word corresponding to “girt” (ἐζωσμένος) cannot be construed in the wider sense of “bound.” It is most likely that the discrepancy between the Syriac and Greek on this point stems from a translator’s incorrect choice among the meanings of the Syriac word esārā, rather than his invention of an entirely new meaning for the Greek ἐζωσμένος.

The apparent correspondence between Syriac “thing” or “will” (ṣbūtā) and Greek “will” (θέλημα) may be an analogous case. Although the possibility that ṣbūtā in Syriac means “will” rather than “something” here should not be entirely discounted, here too we may have an instance of a Greek translator choosing incorrectly among the various meanings of a Syriac word, translating ṣbūtā as “will” rather than “something.”

Finally, as noted in Chabot’s edition, the Syriac text refers to the human being with a phrase from the Peshitta version of Ecclesiasticus 43:2, “marvelous instrument” (mānā d-tedmūrtā).38 Incidentally, precisely the same phrase, mānā d-tedmūrtā, is used as a descriptor for humanity near the beginning of Aḥūdemmeh’s On the Composition of Man.39 This phrase seems to correspond to “marvelous dwelling-place” (ἐνδιαίτημά… θαυμαστόν ) in the Greek text. Jouanna deems this to refer to paradise, adducing a parallel passage in ps.-Gregory Nyssa’s On Paradise, in which the same phrase meaning “marvelous dwelling-place,” ἐνδιαίτημα… θαυμαστόν , refers to paradise, and, as in the Greek pseudo-Hippocratic treatise, occurs just after an assertion of humanity’s superiority to animals. There is a real possibility of a translation going either way here, as both “marvelous instrument” and “marvelous dwelling-place” make sense in context, and have clear parallels elsewhere in Syriac and Greek, respectively. It is also conceivable that neither of these phrases is a translation of the other, despite their similarity to one another, and corresponding place in each text. Still, it is worth noting that as in the two cases we have just examined, the Syriac word mānā has a broader range of meanings than the Greek word that seems to correspond to it, ἐνδιαίτημα; in addition to “instrument,” mānā can also mean “vessel.” A Greek translator may have had difficulty understanding how human beings could be “vessels” in themselves, and accordingly construed mānā as the “vessel” containing humanity: their “dwelling-place” (ἐνδιαίτημα). In this case it is not too hard to see the misunderstanding going the other way, however; it is conceivable that a translator working from Greek to Syriac could have construed mānā in the passage from Ecclesiasticus as something more like “vessel.” Seeing a corresponding equivalence between mānā and ἐνδιαίτημα, he could have rendered “marvelous dwelling-place” (ἐνδιαίτημα… θαυμαστόν ) as “marvelous vessel” (mānā d-tedmūrtā).

A broader stylistic comparison between the Greek and Syriac texts may also be relevant for the direction of the translation. The Greek On the Composition of Man makes a somewhat abrupt transition, from initially describing “man” directly, as “girt with two parts” and “the image of the one who created him,” to describing his attributes--“he is the image of the one who created him” is immediately followed by, literally, “the dwelling-place is marvelous” and “the will is desirable.” The corresponding passage in the Syriac text lacks this abrupt transition, as the descriptors there are consistently predicated of “man” himself, not his attributes. Of course, this discrepancy between the two texts could be the result of a translator working from Greek to Syriac smoothing out an awkward transition he found in his Vorlage. Still, it is somewhat easier to see the valiant attempt of a translator working from Syriac to Greek to make sense of an incompletely understood Syriac Vorlage here.

The Syriac text may also help to better establish the hierarchy among the Greek manuscripts. As the above table shows, the Syriac text follows the Greek manuscript Pa. significantly more closely than P., the basis for Jouanna’s text. For example, the Syriac has no equivalent for the conjunction ἐπεὶ (“for”) following θεός ἐστιν ἐπίγειος (“he is a terrestrial god”) in Jouanna’s text, which Pa. and Par. also lack; and the Syriac has no equivalent for the phrases κατέστη ὑπὸ τοῦ δημιουργοῦ (“made by the craftsman”) or καὶ πάσης τῆς ὁρατῆς κτίσεως (“and of every visible creature”), each of which is unique to P.40

Both the Syriac and the Greek then turn to the enumeration of various body parts. Here again the close parallels between the texts are unmistakable.

Syriac41	Greek42
kolleh gūšmā da-treʿsar haddāmaw. w-rukkābā da-tlātmā wa-štīn wa-štā minyānaw. w-quyyāmā d->ḥamšā <d-regšaw>.43 māʾtēn gēr w-arbʿīn wa-tmānyā garmīn mdabbqīn b-barnāšā.	Στάσις ἐστὶ σεπτὴ ὅλον σῶμα. Σῶμα δέ ἐστι δωδεκάμελον ματηρτισμένον εἰς τριακόσια ἑξήκοντα· ὁ ἀριθμὸς αὐτοῦ οὗτος· στοιχεῖα δ´·αἷμα, φλέγμα, χολὴ ξανθὴ καὶ χολὴ μέλαινα· αἰσθήσεις αὐτῷ πέντε· ὅρασις, ὄσφρησις, ἀκοή, γεῦσις καὶ ἀφή. Κεκόλληνται δὲ τῷ ἀνθρώπῳ ὀστᾶ σμη´
And the whole body is of twelve members, and the composition three-hundred sixty-six. [As for] his numbers: his state is of five <of his senses>. For two-hundred forty-eight bones are put together in man.	[His] state is venerable with respect to the whole body. His body is twelve-membered, compounded into three-hundred sixty-six; its number is such: 4 elements: blood, phlegm, yellow bile and black bile. Five senses are there: sight, smell, hearing, touch and taste. 248 bones are put together in man.

Many parallels are immediately apparent here. The Syriac and the Greek agree in assigning the body “twelve members,” some kind of further subdivision into 366 parts, and 248 bones. In all likelihood, they also both declare “man” to have five senses: Chabot’s text has d-reglaw (<drglwhy>), “of his feet,” but this is one letter away from d-regšaw (<drgšwhy>), “of his senses.”44 A discussion of the bones in the body follows this passage, starting with the bones in the foot, and it is not hard to imagine an inattentive copyist jumping the gun here, writing “feet” for “senses.” Finally, as in the case of the passage examined before this one, the Greek manuscripts Pa. and Par. are in a certain respect closer to the Syriac than Jouanna’s text, which prioritizes P. The most obvious difference between the Syriac and Jouanna’s Greek text here is Jouanna’s text’s enumeration of the four “elements” of the body, the humors, which is entirely lacking in the Syriac. As it happens, Par. entirely omits this section, while Pa. has “four elements,” but also lacks the specific detailing of the four humors.45

A detailed discussion of the bones in the body follows, in both Chabot’s Syriac text and the Greek. For this section, P., Pa. and Par. agree with Chabot’s Syriac against A., which has something almost entirely different.46 The following table shows the numbers of bones the Syriac On the Composition of Man and P., Pa. and Par. situate in each part of the body, in the order in which these figures occur in each manuscript (an empty cell designates that nothing in the given manuscript corresponds to whatever figures occur in the columns alongside it):

Syriac47	P.	Pa.	Par.48
30 [total] in the feet, “5 and 5 again” in the sole of the foot	“5 and 5 per toe”; 5 in the sole of the foot
3 in the knees	3 in the knees
2 in the thighs	2 in the back of the knee49
1 in the chest	1 in the thigh50
5 in the loins	9 in the [region of?] the kidneys	5 in the [region of?] the kidneys
	8 in the legs
9 in the head		7 in the head	9 in the head
8 in the neck	5 in the neck	8 in the neck
5 in the palm of the hand	5 in the palm of the hand51
5 in each finger	5 per finger	5 [lit. “five five”] per finger52
3 in the back	3 in the elbow
	2 in the arm
	1 in the shoulder	1 in the “muscle”
101 in the right side	5 in the right side	101 in the right side	102 in the right side
100 in the left side	5 in the left side	101 in the left side	118 in the left side
18 “ribs-vertebrae” in the back, “18 ribs are joined to them, these vertebrae”	18 vertebrae in the lower back53
6 in each shoulder	6 in each shoulder
3 in each arm	3 in each arm	[something else]54	3 in each arm
5 in the openings of the heart	5 in the opening of the heart		5 in the opening of the heart
	“The ribs are 6. The false [ribs] are 5. And for the vertebrae of the lower back 14. For the right side 101. And for the left side 101. So that altogether the bones are 360.”

Here the broad trends we saw in the passages previously examined continue: the Syriac agrees closely with the Greek in general, and particularly closely with the manuscripts Pa. and Par. As can be seen in the table above, there are several instances in the discussion of the bones in which the Syriac agrees partially or completely with Pa. and Par. over P.; and there are no instances in which the Syriac agrees with P. over either Pa. or Par. These instances of the Syriac agreeing with Pa. and Par. against P. include: the Syriac, Pa. and Par. assigning five bones to the “loins” (in the Syriac) and “kidneys” (in the Greek), where P. assigns nine bones to this region; Pa., Par. and the Syriac addressing the bones of the head after those associated with the kidneys and loins, where P. discusses the bones in the legs; Pa., Par. and the Syriac assigning eight bones to the neck, as opposed to P.’s five; and Pa., Par. and the Syriac assigning a hundred or more bones to the right and left sides of the body, while P. locates only five in each side in the corresponding passage (though P. does register a figure more in line with Pa., Par. and the Syriac later in the text).

I will now discuss one recurring set of corresponding expressions in the Greek and Syriac, whose Greek component has already been puzzled over by Jouanna. Several times in the text, we have the corresponding phrases, perplexing in the Greek and Syriac alike, “transformation [or, once, “form”] of the back of the head [?]” (μεταμόρφωσις [or, once, μόρφωσις] τοῦ ἐγκροτάφου ) in Greek, and “form [?] of the brain” (<ṣwrtʾ> [the vocalization is uncertain—perhaps ṣūrtā?] d-mūḥā) in Syriac.55

The Greek μεταμόρφωσις/μόρφωσις τοῦ ἐγκροτάφου is a difficult expression to understand. Ἐγκρόταφος is only otherwise attested in the Etymologicon Gudianum, where it is glossed as ἀντικέφαλος, “back of the head,” though Jouanna opts for “part of the head between the temples” on the basis of the word’s derivation (ἐν-, “inside” + κρόταφος, “temple”).56 As for μεταμόρφωσις, Jouanna simply observes that this word “cannot have its usual sense of transformation here,” and leaves it untranslated in each of its attestations, rendering the entire phrase μεταμόρφωσις/μόρφωσις τοῦ ἐγκροτάφου simply as “the part of the head between the temples.”57 Probably because of the strangeness of the phrase, the Greek manuscripts have significant discrepancies in their renderings of it; κρόταφος (“temple”) often occurs as a variant for the scantily attested ἐγκρόταφος (“the back of the head”?), as does μορφή (“form”) for the more complex μεταμόρφωσις (“transformation”), or for the rarer μόρφωσις (“form”). It is only in P. that the phrase consistently occurs as μεταμόρφωσις (or, once, μόρφωσις) τοῦ ἐγκροτάφου . The overall tendency of Pa. and Par. (in which variants such as κρόταφος and μορφή often occur) to agree with the Syriac against P. should raise some questions about whether P.’s reading of μεταμόρφωσις/ μόρφωσις τοῦ ἐγκροτάφου is the correct one here.58 Nonetheless, in my estimation here the principle of lectio difficilior should override whatever general fidelity to the original Pa. and Par.’s broader closeness to the Syriac implies; it seems likely that a difficult phrase like this would have been simplified and clarified in the course of its transmission, and that P. preserves an older reading that has been adjusted in other manuscripts.

The Syriac phrase <ṣwrtʾ> d-mūḥā presents its own difficulties. As for its distribution in the text, its three attestations correspond directly to the last three attestations (out of four) of μεταμόρφωσις/μόρφωσις τοῦ ἐγκροτάφου in the Greek--the first attestation of this phrase in the Greek corresponds to “upper entrance” (tarʿā ʿillāyā) in the Syriac.59 In his translations of <ṣwrtʾ> d-mūḥā, Chabot consistently renders <ṣwrtʾ> as cervix, “nape”, claiming that the word’s literal meaning is collum, “neck,” and suggesting that the whole phrase <ṣwrtʾ> d-mūḥā designates the “cerebellum,” the part of the brain in the back of the skull.60 Chabot’s suspicion that <ṣwrtʾ> means something other than simply “neck” here finds support not only in the obscurity of the phrase “neck of the brain,” but also in the fact that the most common Syriac word for “neck,” ṣawrā (spelled <ṣwrʾ>), is attested earlier in the text (directly corresponding to τραχήλος, also “neck,” in the Greek).61 But Chabot’s reading of <ṣwrtʾ> d-mūḥā, while persuasive to a certain extent, is on somewhat shaky ground in its own right; I have been unable to find anything in the lexica for <ṣwrtʾ> as “nape,” “neck” or, indeed, a word referring directly to any part of the body. The most common meaning of <ṣwrtʾ> is “form” (in the vocalization ṣūrtā) though we also have the identically-spelled ṣāwartā recorded on its own with the meaning of “insanity” or “confusion” and “ringing of the ears” in Mannā’s Syriac lexicon.62

Neither the Greek nor the Syriac phrase, then, has a particularly obvious meaning in itself; hence, neither is obviously a translation of the other. The correspondence between the most common meaning of <ṣwrtʾ>, “form” (with the vocalization ṣūrtā) and those of the Greek words μεταμόρφωσις and μόρφωσις, “transformation” and “form,” respectively, is surely significant, as is the correspondence between the meanings of Syriac mūḥā, “brain,” and Greek ἐγκρόταφος, which clearly has to do with the head; but what was translated into what? We have signs pointing both ways. On the one hand, the fact that <ṣwrtʾ> is so similar to the Syriac word for “neck” (ṣawrā), and that a word with this spelling, ṣāwartā, has two further meanings having to do with health problems (“ringing of the ears”; “insanity” or “confusion”), could sway us to take <ṣwrtʾ> d-mūḥā as a scantily attested phrase with an anatomical significance in Syriac. The sense of this phrase could have eluded a Greek translator, who simply did the best he could with the only meaning of <ṣwrtʾ> known to him: “form.” Speaking for a Greek to Syriac direction, on the other hand, is the obscurity of the word ἐγκρόταφος: why would a translator working from Syriac to Greek use such an arcane word to render mūḥā, “brain,” with the more straightforward ἐγκέφαλος also available to him?63 Moreover, it is conceivable, of course, that in the context in which it occurs μεταμόρφωσις could have an anatomical sense in Greek as well, as a technical term referring to a specific part of the body; <ṣwrtʾ> d-mūḥā could be an attempt at rendering an unfamiliar Greek expression on the part of a Syriac translator.

III. Concluding Discussion

As discussed above, definitive conclusions about the direction, place, and time of the translation that the profound correspondences between these texts indicate will have to await an edition of the Syriac text that incorporates all relevant manuscripts, as well as, perhaps, a revision of Jouanna’s text prioritizing the Greek manuscripts that tend to be closer to the Syriac. As far as these details go, at this point we only have a definite terminus ante quem for the Syriac text: the composition of George Wardā’s poem on “man as microcosm” in the thirteenth century, which, as Gignoux has convincingly demonstrated, draws upon the Syriac On the Composition of Man attributed to Aḥūdemmeh Anṭīpaṭrōs.64 Consequently, the following suggestions remain equivocal and provisional.

I will begin with the direction of the translation. The limited collection of textual comparisons I have presented here does not decisively point either way (nor have I been able to find anything conclusive in the rest of the text), and this question is worth a comprehensive treatment once the Syriac On the Composition of Man is available in a better edition. At this point, several indications suggest that the original language of the text was most likely Greek: the likely original attribution of the entire Syriac On the Composition of Man to “Anṭīpaṭrōs,” as convincingly argued by Chabot and Reinink; the many Greek philosophers and physicians cited within the Syriac text; and, perhaps most importantly, the fact that many more translations are known to have been made from Greek into Syriac than from Syriac into Greek. If it is the case that the direction of the translation is Greek to Syriac, this would fix the likely date of composition for both the Greek and the Syriac between the sixth and ninth centuries, within the period of known medical translation activity from Greek into Syriac.65 In this scenario the Greek text would have likely been initially translated into Syriac on its own, and then subsequently incorporated into On the Composition of Man, alongside what were originally separate texts.

If, on the other hand, the direction of the translation is Syriac to Greek—the less likely possibility, though some of the results of my preliminary comparative analysis of the Greek and Syriac texts suggest that this may have been the case—the range of plausible dates for each text’s composition widens considerably. Translations from Syriac to Greek are firmly attested from the third century through the eleventh century, and while the overall number of such translations is quite low, and nothing medical is included among them, the possibility cannot be excluded that this work went from Syriac to Greek at some point during this stretch.66 Touwaide, meanwhile, has noted several Byzantine Greek medical manuscripts that are either attributed to a “Syrian” author or designated as “Syriac” themselves, construing them as part of a broader “transfer of knowledge from the Arabic to the Byzantine world,” taking place from the tenth century through the fourteenth, and ultimately resulting in “Arabic medicine” being “massively present in late thirteenth- and early fourteenth-century Constantinople.”67 The linguistic origins of these texts have yet to be thoroughly investigated, but it is possible that this corpus includes some translations of Syriac medical texts into Greek, which would have been made sometime between the tenth and fourteenth centuries, according to Touwaide’s reckoning; the hypothetical Syriac-to-Greek translation now under discussion could have been part of this trend. The scenario here would be either of the following, each somewhat involved, but not impossible: a pseudo-Hippocratic work was initially composed in Syriac, incorporated into the Treatise on the Composition of Man attributed to Aḥūdemmeh Anṭīpaṭrōs, and independently translated into Greek; or the Greek On the Composition of Man was based on an excerpt from the longer Syriac work attributed to Aḥūdemmeh Anṭīpaṭrōs. In the latter scenario, we can hypothesize that the Greek translator misconstrued the concluding sentence of the portion of the Syriac text that corresponds to the Greek pseudo-Hippocratic treatise, “thus said the great Hippocrates,” as referring to the entire section of the text leading up to it, rather than simply the preceding sentence (as it almost certainly does).68 Of course, neither scenario, in itself, would imply any particular terminus post quem for the composition of the Syriac text; and the Greek translation could have happened after the thirteenth-century terminus ante quem (discussed above) for the Syriac text’s composition.

As for what the philological data say about the Greek and Syriac texts’ respective dates, any conclusions on this basis must await an updated edition of the Syriac text (as well as, perhaps, the Greek), but it is worth noting that I could find nothing conspicuously late in the Syriac text: the text contains no loanwords from Arabic, nor was I able to discern any other linguistic features (such as a proliferation of certain adjectives in -āyā, or nouns in -ānā and -ānūtā) suggesting that the composition of the text postdates the sixth century.69 Interestingly, the spelling of “Hippocrates” in the Syriac text, <ywpqrṭys>, finds no exact parallel in the Syriac corpus aside from the Syriac Epidemics, a translation of an Alexandrian super-commentary on Galen’s commentary on Hippocrates’ Epidemics.70 The Syriac Epidemics is quite early; our sole manuscript was copied no later than 705 CE, in Khuzestān, and in large part on these grounds Kessel has argued that Sergius of Rešʿaynā translated the text from Greek into Syriac during the sixth century.71 <ywpqrṭys> is clearly closer to Sergius’ favored spelling, <ypqṭrys>, than it is to Ḥunayn’s <ʾyppwqrʾṭys>, which follows the Greek much more closely, but it should be noted that it is not identical to Sergius’ spelling.72 In any case, this spelling of Hippocrates suggests that “Aḥūdemmeh Anṭīpaṭrōs’s” Treatise on the Composition of Man was composed before Ḥunayn’s floruit in the ninth century, and perhaps in the same milieu (or even by the same writer) as the Syriac Epidemics.

Jouanna has singled out the Greek text’s usage of συγκερασμός for “mixture,” as opposed to the more usual μίξις, κρᾶσις, and derivatives thereof, as an unusual feature of the text. Jouanna observes that this word is quite rare; it is recorded in the Corpus Glossarium Latinorum, but is only otherwise attested in a twelfth-century Byzantine chronicle, referring to a mixture of colors. For Jouanna this suggests a late date of composition for the Greek, though he holds off on drawing any firm conclusions on this basis.73 But given that συγκερασμός also occurs in a passage from the Sayings of the Desert Fathers, which was almost certainly composed prior to the sixth century, an attestation overlooked by Jouanna, there is little reason to posit a late date for the Greek text on the basis of its use of this word.74 Still, the rarity of συγκερασμός is noteworthy nonetheless; and it is worth mentioning in this connection that two of the very few known attestations of a close relative of συγκερασμός with the same meaning of “mixture,” συγκέρασμα, apparently occur in purported Greek translations of Syriac texts: the Homily On Virtue and Vice attributed to Ephrem, which may simply be an original Greek composition; and (probably) the Greek version of a homily of Isaac of Nineveh, whose roots in Isaac’s original Syriac are not in doubt.75

As for the Syriac text’s place of origin, the nature of the manuscripts in which it occurs likely speaks to this. The Treatise on the Composition of Man attributed to Aḥūdemmeh Anṭīpaṭrōs is known to have been transmitted in three East Syriac manuscripts, all of which also contain a likely extract from the Book of Medicine authored by Šemʿōn of Ṭaibūteh, and two of which have a portion of the anonymous Syriac Book of Medicines (eventually edited by Budge).76 While texts could and did circulate between West and East Syrian communities in late antiquity, the fact that this treatise is consistently attested in East Syriac manuscripts, as well as its association with two texts whose origins lie in the territory of the former Sasanian Empire--the anonymous Book of Medicines and a work probably to be attributed to Šemʿōn of Ṭaibūteh, an East Syrian monk based in seventh-century Khuzestān--suggest that this treatise, too, was initially composed or translated by an East Syrian Christian, and circulated among members of this community from an early stage, whether in the Sasanian Empire itself, or in the Empire’s erstwhile territories in the centuries following the Islamic conquest.77 78*