Some Notes on Andrápolis, the Royal City: Apocryphal Acts of Thomas 3 

Notes on Andrápolis


Introduction
Many issues come up when one tries to critically analyse the 3 rd century Apocryphal Acts of Thomas (AATh), a long narrative that describes the evangelical mission of the apostle Thomas in India, where he is sent from Jerusalem by his alleged twin brother Jesus when the apostles divide up the regions of the earth for evangelisation. 1One major problem is determining whether or not the original language of the text was Syriac, a hypothesis postulated by Burkitt, 2 Klijn 3 and Attridge, 4 and generally accepted by most scholars, 5 but recently contested by Roig Lanzillotta, who highlights the superiority of the Greek version, arguing that ‚it preserves the general tenor of the primitive text more accurately, it reflects the different parts of the text in a better way, and it shows fewer traces of editorial intervention‛. 6ven if an examination of this issue is not the main goal of this paper, the question is nevertheless significant, and the results of our analysis have direct implications in it.Indeed it may determine our preference for, on the one hand, the Greek or the Syriac text, and on the other hand, for one of two main textual groups within the Greek tradition of AATh, relating to the archetypes Δ and Γ in Bonnet's edition.It all boils down to a debate around a simple word, appearing in AATh 3,2, which differs depending on whether we are dealing with the Syriac version (Sanadrūk, Sandarūk, Sandrūk), or the Greek, in which we commonly find two different toponyms depending on the manuscript tradition.While most of the manuscripts reflect the variant Andrápolis -᾿Ανδράπολισ, in the Bonnet's Δ 7those relating to the family Γ present the textual variants Enadróch (Ἐναδρὼχ, H), Enadóch (Ἐναδόχ, G), or even Edrón (Ἐδρὸν, Z), which have been observed as a poor translation of the Syriac Sandarūk. 8

AATh and Historicity
Nonetheless, before beginning with the arguments concerning the meaning and identification of Andrápolis in the Greek manuscripts, it is worth devoting a few words to the consideration of AATh as a historical document, which in our opinion is important for tackling the aforementioned issue.Many attempts to reconstruct the history of Christianity in the East, particularly in India, were based on this text. 9In our view, however, even if the locations and proper names transmitted by AATh suggest a historical episode from the apostolic era, we cannot grant any truthfulness to the events narrated in the apocryphal text.It is clear that the narrative of Thomas's journey to India enjoyed great popularity in early Christianity, and there is a remote chance he undertook such a trip.However, in our 6 See L. Roig Lanzillotta, ‚A Syriac Original‛, p. 105.However, regarding the Greek version exclusively, the variegated and complex textual transmission represents a second serious problem, since at least two different variants are contained in the manuscripts, as can be seen in Bonnet's edition of the text; Maximillianus Bonnet, Acta Philippi et Acta Thomae (Hildesheim -Zürich -New York: Georg Olms Verlag, 1990 2 ), p. 104; see also XIX-XXX.7 See M. Bonnet, Acta Apostolorum, pp.XIX-XX.We follow Bonnet's nomenclature of the manuscripts (at p. XVI), with exceptions made for those that would have been unknown to the German editor in his time.For all variants, see below.As George Nedungatt highlights in ‚The Apocryphal 'Acts of Thomas ' and Christian Origins in India‛,  Gregorianum 92.3 (2011), pp.533-557, espec.556, we still do not have relevant studies of the monuments and/or archaeological sources relating to the Apocrypha.Therefore, our most reliable source is the text; see below for the last advances in the field of numismatics.
view, fiction and reality should be separated, especially so when we are dealing with sources of this kind, namely narratives created some time after the historical context in which the action takes place and whose narrative's primary concerns are moral instruction and good storytelling, rather than historical accuracy.
Regarding this, we agree with Andrade that AATh is extremely vague in terms of geographical, historical, and ethnical descriptions. 10At the time in which AATh was supposedly written, the author had sufficient geographical information at his disposal, collated from diverse Greek and Roman authors.11India was a vast area and descriptions of different peoples, tribes and cities were common among historians and well-educated people of the Hellenistic and Roman Imperial periods.However, it is also true that during the Christian era there was significant confusion regarding the settlement of Indian populations, and peoples of the Arabic gulf area and North Africa were commonly alluded to as ‚Indians‛. 12t first sight, we can only find generic mentions of India (Ἰνδία) usually referring to the vast territory in which Thomas allegedly performed his evangelistic mission.Hence, in the narrative of AATh this territory seems to be divided into different kingdoms, as can be inferred from the existence of two kings, Goundaphor and Misdaeos. 13The former appears as 'king of the Indians' (c.2: Γουνδαφόρου τοῦ βαςιλέωσ τῶν Ἰνδῶν), whereas no royal title is given to the latter.
Of particular interest for our study is the alleged historicity of Goundaphor, linked with Northern India.Klijn 14 states the Iranian origin of this anthroponym and, following historical and archaeological evidence -mainly numismatic 15 -posits that the ‚legendary king‛ may have reigned from about 30-15 BC. 16 The keyword for understanding the whole debate, as well as our particular approach, is the term ‚legendary‛ as employed by Klijn.The author of AATh clearly resorted to legendary proper names, connected either historically or geographically with the popular knowledge of India that they may have had, or at least suggesting an Indian provenance or identification.Thus, in our view, the description of exotic or remote lands follows the classical technique of novelistic literature: the use of common topoi to refer to imaginary communities or places, albeit on the basis of historical traces. 17

Andrápolis, The Royal City
Having established the context of AATh as largely fictional, even if it evokes the alreadylegendary age of the Apostolic era at least one or one and a half centuries earlier, we are going to deal with the case of Andrápolis and its variants throughout the Greek manuscript tradition.The first time the term is mentioned is at the end of chapter 3.2 of AATh, when the apostle Thomas and his owner Abban stop for a little while on their journey by boat from Jerusalem to India: 18 Ἤρξαντο οὖv καταπλέειν• ἔςχον δὲ ἐπιτήδειον ἄνεμον, καὶ προθύμωσ ἔπλεον ἕωσ ὅτε κατήντηςαν εἰσ Ἀνδράπολιν, πόλιν βαςιλικήν.
So they began their voyage.They had a favourable wind, and sailed prosperously until they arrived at Andrapolis, a royal city.
As has already been pointed out, the setting of this journey indicates that it should be understood in novelistic, fictional terms.Accordingly, with regard to the point of departure, as Andrade highlights, ‚Jerusalem is an inland city.Even if it had been located Asiatic Society of Bengal 23 (1854), 679-719; André-Jean Festugière, Les Actes Apocryphes de Jean et de Thomas. Traduction Française et Notes Critiques (Genève: Patrick Cramer, 1983), p. 45, n. 1. 16 See also N.J.Andrade, The Journey of Christianity, pp.42-44, who concludes: ‚The author of the text, knowing virtually nothing about India (other than a king named Gudnaphar/Goundaphores), simply interwove existing traditions about the apostle with new material pertaining to his interactions with King Mazdai‛, p. 44. 17This stylistic strategy of blending historical (or even pseudohistorical) information is one of the bestknown characteristics of AAA.For instance, it is worth mentioning the city of dog-headed men, where the apostles are sent on their evangelizing mission in the Apocryphal Acts of Andrew and Matthew (IVth CE) -AAMT 1, and the Apocryphal Acts of Andrew and Matthew (Vth CE) -AAAB 1,1-2.However, the location and historical veracity of this place are as doubtful as they are in the Indika of Ctesias the Cnidian -Ctes.F45 Jacoby = Phot.Bibl.72, p. 45A21-50A4. 18See AATh 3, ed.M. Bonnet, Acta Apostolorum; transl.by Han J.W. Drijvers, ‚The Acts of Thomas‛, in Wilhelm Schneemelcher (ed.), New Testament Apocrypha, vol.II (Cambridge-Louisville: WJK, 1992), pp.322-411.
on the Mediterranean coast, one would not have been able to sail to India from it‛. 19 With regard to destination, it should be noted that what we have here is only a brief mention of a supposedly Indian or near-Indian city, so-called Ἀνδράπολιν, πόλιν βαςιλικήν.However, as stated previously, the question regarding the original language of AATh is also of significance in determining the name of the city.Scholars who argue for the Syriac version over the Greek give preference to the variants Sanadrūk / Sandarūk / Sandrūk.Consequently, two possibilities stand out among others. 20The first is that Sanadrūk / Sandarūk / Sandrūk may be a reference to a certain site in Bahrain where the Sasanian king Ardashir I (224-242) killed a king named Sanatruk / Sanatruq.Since Bahrain was one of the main stopovers between Mesene and the Indian Ocean, it is possible that the fortress acquired the eponymous name of the king killed there. 21The second possibility is that the name is related to the island of Kharg / Kharak / Harak, based on the testimonies of Claudius Ptolemy and Pliny, who attest that in the second century the island was known as Ἀλεξάνδρου / Ἀρακία or Kassandra. 22he Greek equivalent of the toponym has deservedly received scholarly attention.᾿Ανδράπολισ traditionally relates to the Andrha, a people in South India. 23Guttschmidt proposed this identification on the basis of the testimony of Pliny the Elder HN VI 22, concerning Andarae, 24 which may have formed an independent kingdom. 25Its capital may have been a settlement, whose name demands our attention: Amārāvatī, also known as Andhranagari, could be designated as a 'royal city', which can perhaps be identified as the Andrápolis of AATh.Could the Greek ᾿Ανδράπολισ possibly refer to the Indian Amārāvatī?
The linguistic evidence may indeed support this hypothesis.Even if the Greek toponym ᾿Ανδράπολισ can be understood to be a Hellenic creation, meaning the ‚City of the Andhra‛, this is most likely a derivation from either Amārāvatī or Andhranagari.
In the first hypothesis ᾿Ανδράπολισ may represent a sort of hybrid of the compound toponym Amārāvatī, whose meaning in Sanskrit is ‚place or garden (-vatī) of immortals or divine men (amārā-)‛.The initial amārā-, which shares with Greek the Indo-European root 19 N.J.Andrade, The Journey of Christianity, p. 44. 20The possible relation to the santalum, 'sandal-wood', or the sandarach, 'red sulphuret of arsenic', phonetically related to this Sandarūk, does not offer a convincing explanation.ἀμβροςία (‚ambrosia, food of Gods, divine‛), may have evolved phonetically into ἀνδρα-, whether within the Indian context, as with Andhranagari, or in the Greek.Here, an evolution such as amārā-> *amra-> andra, with the introduction of an epenthetic [d] to ease the pronunciation of the two syllables -since the combination [m] + [r] is not attested in Greek -is plausible.The -πολισ compound would here be understood to be a Greek translation of the Sanskrit -vatī.
Attending now to the second hypothesis, namely, the derivation of ᾿Ανδράπολισ from Andhranagari, the Sanskrit toponym may in this case be translated as ‚the city of the Andhra‛, although in our opinion an etymological relationship of this ethnonym with the aforementioned amārā, and an internal evolution towards *and(h)ra, cannot be ruled out.Thus, in the Greek adaptation, the final part of the word does not change with regard to the schema presented by the first hypothesis, although -πολισ in this case would perfectly match the Sanskrit nagari, meaning ‚city‛, whereas ἀνδραmay be straightforwardly transliterated.
Either way, the original semantic meaning of the place name gets completely lost in the Greek term ᾿Ανδράπολισ, which paradoxically would have been interpreted by the Greek readers of AATh as 'the city of men' rather than 'the city of the Andrha'.This Amārāvatī is located in East India, on the bank of the river Krishna in the current Andhra Pradesh region, and was in fact a royal city, as is alluded to by AATh's author.Furthermore, it is located on the navigable part of the river, so the travellers could have reached the city by boat, as in the text.If this identification is right, it implies that Thomas and Abban may have travelled from a harbour in nearby Jerusalem across the Arabian Gulf and arrived in East India by sea, then headed north to the kingdom of Goundaphor in Northern India, identified in the text simply as 'India'.
In any case, one should not forget that we are dealing with a literary text, a fictional document.The point is that if we assume that AATh cannot be read as history, but only interpreted in its literary context, then this ᾿Ανδράπολισ was used by the author of AATh to represent a city of India, as it would have been somehow known and recognized as such within his community, perhaps as an imperial city among one of the many ‚Indias‛ of the Greek conception of this territory.
Therefore, the name ᾿Ανδράπολισ used for the imperial city mentioned in AATh can be considered as an attempt to describe a mythical geography of the East, even if the city did indeed happen to exist.Other questions emerge at this point: was the author of AATh conscious of the location of this city?Not necessarily.He probably knew that it was an imperial city in India, but had no further information to go on other than, at best, its location near a navigable but unspecified place.This idea is supported by the scarce details provided in the text, as in the case of characters with a historical basis, such as the aforementioned Goundaphor.

The Variants of Ἀνδράπολισ in the Greek Tradition
Whichever possibility we accept, a new problem emerges with the variant Enadróch, attested by an important group of Greek manuscripts of AATh, otherwise known as family Γ in Bonnet's edition. 26Actually, the presence of this variant is one of the main features distinguishing family Γ from family Δ, and is cited as alleged proof of the fact that the Greek is a translation from the Syriac version.In our view, the inclusion of the variant Enadróch in the Greek manuscript tradition is a contamination proceeding from the Syriac transmission, rather than evidence of the priority of the Syriac original, as has generally been assumed.New textual witnesses transmitting AATh reveal a well-established textual tradition within the texts comprising Γ in Bonnet's edition -in reality, a summary created from the version of the story found in family Δ -and show that the variant Enadróch is exclusive to the manuscripts comprising family Γ in the Bonnet's work.
In our view this abridged version follows Bonnet's text quite closely and seems to be the result of a progressive, step-by-step reduction of the content of the full version. 30In fact, we can trace a path from earlier to later manuscripts, on the basis of the transmission of the city's name: 37 > 7 | B | 11 | > G-H-36-54.37 would here constitute the first testimony of Ἐναδρὼχ, from which different rewritings could be done independently (7, B  and 11).From these would be created a final group, which would be the most stable and homogeneous in terms of textual uniformity.
This variant is a characteristic of family Γ, but is not an exclusive feature of it, as it appears similarly in codex 43 (Athon.Karakallou 8 [Lambros 1521], s.XIII, ff.40v-49r), a manuscript clearly related to family Δ, but where in chapter 3 we read εἰσ πόλιν καλουμένην Ἐναδρὼχ.The witness makes the Bonnet's distinction between the families Γ and Δ much more complex, since this variant Enadróch is the only similarity codex 43 shares with the group represented by Γ.In any case, this manuscript points at a contaminatio of different textual traditions relating to the apostle Thomas.It contains chapters 1-29 of AATh, as in the manuscripts of family Δ, but the text is followed by the final part of the unedited commentary of Symeon Metaphrastes on Thomas (BHG 1835).The section of one and a half folio included in this manuscript shows how different textual traditions were sometimes used to create versions of the story differing from the alleged original, especially when these versions were abridged texts, such as those comprising family Γ.With this phenomenon in mind, one may propose contact or familiarity with the Syriac textual tradition to explain the adaptation Enadróch in these manuscripts, without the assumption of a Syriac original, which is based on linguistic premises as doubtful as this Enadróch in the Greek textual tradition.
Our current investigation does not allow us to speculate much further.The only certainty is that the variants Andrápolis / Enadróch indicate two undeniable facts.Firstly, Andrápolis was the preferred term in the Greek textual tradition; Enadróch only comes up in manuscripts containing abridged versions of AATh (family Γ), which are derived from the most stable family of Greek manuscripts (family Δ), and the particular case of codex 43 proves that this version was created from two different texts: AATh, and the commentary of Symeon Metaphrastes on Thomas.Secondly, as these abridged versions were dependent on the most complete and stable Greek textual tradition represented by the family Δ, the inclusion of Enadróch in the Greek versions would be secondary to and linked with the contact between the Greek AATh and the Syriac version, which is quite difficult to date on the basis of the data to hand.Nevertheless, the inclusion of Enadróch in the version of the family Γ cannot be used as evidence for the alleged Syriac origin of the text, but rather as confirmation of the eventual contact between the two textual traditions.
Linguistically, the Syriac influence over the Greek version, resulting in the inclusion of Enadróch, seems the most likely explanation.The transformation from Andrápolis into Enadróch may have come about the Syriac Sandarūk.To the scribe or writer reworking the Greek AATh, Andrápolis would here mean simply 'city of men'.Besides this, they may perhaps have believed that this place name sounded too Greek and decided, either by his own initiative or with the help of a Syriac interpreter, to check the Syriac version, in which he found the place name Sandarūk.As this name contains non-Greek features, such as the consonantal ending, Enadróch was a better candidate with which to suggest a city in India, or at least somewhere beyond the limits of the Hellenistic world.
The writer would then have phonetically adapted the name according to the Greek spelling of his time.Although the loss of the initial s-can be understood as an incompatibility, we have found a curious testimony that may shed some light on this.A certain γή Ἐδρὰχ is attested in Theodore of Mopsuestia's Commentary on the Twelve Minor Prophets.In interpreting the Book of Zechariah 9, the author mentions the Septuagint Ἐδρὰχ, which is in fact a reference to Sedrách (Σεδράχ), near Damascus.This was perhaps the toponym that the scribe of codex 11 had in mind when he wrote Ἐδρὼχ.In any case, what is more interesting, in our opinion, is the appearance of the same phonetic pattern, i.e. the loss of the initial s-during the adaptation into Greek of a non-Greek place name beginning with this phoneme; that is, if we assume -as many have done so -the relationship between the Syriac Sandarūk and the Greek Enadróch.
As far as we know, few linguistic studies on this topic exist that would allow us to collect other testimonies in support of this hypothesis.This field of study, i.e. the behaviour of loanwords from Eastern languages in Late Antique or Byzantine Greek, is one of the most neglected points of current research.

Conclusion
It is time to draw some conclusions.We agree with almost all scholars that the level of historicity in AATh's narrative is very low.It seems that the author is not interested in offering a historical account at all, but in providing an edifying story.In doing so, the author resorts to the topoi of the genre, mixing into the scenes of his Christian novel some new material that is hard to trace back to source, but with echoes of Scripture, 31 as was usual in early Christian literature, in order to frame the adventures of the apostle within a plausible world.Consequently, the attempt to reconstruct Thomas's itinerary seems as illadvised as the many other proposals of scholars up to the present day. 32In our view, the ᾿Ανδράπολισ of AATh may or may not be the historical Amārāvatī or Andhranagari, since the implications of this identification change nothing with regard to either the historicity of the text or the poor knowledge of the Indian region that we have attributed to the author of AATh.
Besides this, a much more interesting interpretation can be formulated, since the inclusion of this Indian toponym with all the necessary caveats may support the hypothesis that Greek is the original language.Such an identification with a real location in India suggests the higher quality of the Greek version over the Syriac; the latter's toponym Sandarūk seems to us much more obscure.It is not a minor question, since, as we have already highlighted, this reference to Andrápolis is the only time the author of AATh uses a single name of an Oriental city.The Indian origin of the Greek ᾿Ανδράπολισ points to a slightly higher attempt at geographical accuracy in order to make the story trustworthy.
Concerning the use of proper names, if a foreign author were to write a novel situated in 19 th -century Spain, he might situate the story in a place generally well known to his readers such as Seville, as did Merimée in his work of 1845, Carmen.The realistic background exists, but historical accuracy is not the intended purpose.The same narrative technique is displayed, in our view, by the author of AATh, who uses proper names to evoke India, in a general sense, to his readers.This would therefore be a narrative technique shared by the rest of the Apocryphal Acts of the Apostles, in which the use of generic names is very widespread, since these texts are basically edifying narratives with novelistic and propagandistic purposes, rather than historiographical or biographical works aiming at historical accuracy or lack of bias.

Abstract:
The following paper aims to explore the meaning and significance of the so-called royal city Andrápolis (Ἀνδράπολισ) within the narrative of the Apocryphal Acts of Thomas's (3.2).The identification of this toponym and the attempt to extract the few reliable historical traces supposedly transmitted by the apocryphal text have attracted the attention of many scholars during the last century, as well as recently.After a short introduction, in this paper, we deal with the alleged historicity of the text.We then assess the different variants of the location in the Syriac or Greek transmission of the Acts of Thomas and propose a new interpretation.Finally, we check the remaining Greek variants among the manuscriptscollating new discoveries and those of by Bonnet -, in an attempt to explain the differences between them and the better attested Andrapolis.We close with some final remarks.
Keywords: Apocryphal Acts of Thomas; Andrápolis; Textual criticism; Historicity of the apocryphal writings.