The fortress of St. Gregory
April 9, 2011
George Pachymeres, in his History of the reigns of Michael VIII and Andronikos II Palaiologos, relates the death of John Bekkos in the following way (Pachymeres, History IX, 29 [Failler, ed., Georges Pachymérès: Relations Historiques, III (Paris 1999), pp. 296 f.]):
At that time John, the chief of the Lazoi, who was also the emperor’s brother-in-law as was indicated earlier, finished his life, leaving two children: of these, the one, Alexis, inherited his father’s authority, while as for the younger of the two, his mother Eudokia took him and went to live in the city near her brother, the emperor.
John Bekkos, the former patriarch, died as well, in the prison at the fortress of St. Gregory, at the end of the month of Kronion [March]; he was buried there in some random spot in his own cell. But the news of this was one more source of grief for the emperor, since there had not been time to bring about what had been arranged between the emperor and Bekkos’s party, that is, to organize discussions aiming at agreement and peace, with wise and spiritual men (not your everyday random know-nothings) serving as judges. As for Meliteniotes, he was brought back from there and given lodging with Metochites, who was in the city; since it proved impossible to make peace with them, they were incarcerated, in keeping with the demands of the emperor’s people and the leaders of the Church, at the Great Palace, there where John Tarchaneiotes also, later, would be locked up.
In 1926, Vitalien Laurent published an article (V. Laurent, “La date de la mort de Jean Beccos,” Échos d’Orient 25 , 316-319) in which he conclusively argued that the year in which John Bekkos died was 1297. But what about the place of his death and burial, the prison where he spent the last twelve years of his life, this fortress of St. Gregory? Where was it? Does it still exist?
Those are questions I have been asking for a long time. Using Google Maps, I some time ago came to the conclusion that the prison must have been located on a jetty of land that sticks out into the Gulf of Nicomedia, close to the present-day town of Hersek (the ancient Helenopolis); Pachymeres notes that the fortress was on your right as you enter the gulf, and this, it seemed to me, was the landmark he must have been referring to. Recently, however, I came across an article that convinces me that I was wrong about this: not about this being the landmark to which Pachymeres was referring, but about the fortress being located directly on this strip of land. The article, written by the editor of the critical edition of Pachymeres’s History, Albert Failler, collects the various known facts about the fortress provided by the three Byzantine historians who mention it (Pachymeres, George Metochites, and Nikephoros Gregoras). The reason why the fortress is not to be sought on this strip of land is a simple one: this jetty is quite flat, while the fortress of St. Gregory is described by Metochites as having perched upon a high, inaccessible hilltop. Still, it could not have stood too far from this neck of land, since this is clearly the landmark that marks the entrance to the Gulf of Nicomedia. Failler narrows down the site to a hilly area between Hersek and the modern-day town of Karamürsel (a town whose old, Greek name was Prainetos).
Thanks to the wonders of the internet, Failler’s article, published in 1990 in the scholarly journal Revue des études byzantines, is now easily available: a link to it is given here in the following bibliographical information: Albert Failler, “Chronologie et composition dans l’Histoire de Georges Pachymérès (livres VII-XIII),” Revue des études byzantines, 48 (1990), 5-87. Below, I have translated a passage from pp. 21-23 of this article, where Failler discusses the location of the fortress.
Before continuing, it would be opportune to determine the site of the fortress of St. Gregory, which has often been confused, without any reason, with the fortress of St. George on the lake of Nicaea. George Pachymeres alone names the citadel, which, in George Metochites’ narrative, is indicated by the general name of “fortress”; moreover, he alone indicates its approximate location. The fortress of St. Gregory is mentioned three times in [Pachymeres’] History as a place of detention, of Irene-Eulogia Palaiologina and her daughter Theodora, in the first case, and of John Bekkos and his archdeacons in the two other cases. Only the second passage contains data concerning the site of St. Gregory’s: “the fortress that is found on one’s right when one enters the Gulf of Astakos”. The information is precise: the fortress is found upon the southern shore of the Gulf of Nicomedia (the Astakenos Gulf, or Gulf of Astakos). Without expressly stating this, George Pachymeres suggests that it was situated alongside the sea. In an otherwise vaguer account, Nikephoros Gregoras confirms the coastal position of the unionists’ place of detention: “Bekkos and his companions were sent to a fort on the shores of Bithynia”. Though he never supplies a name for the fortress, George Metochites stresses five times the isolation of this veritable eagle’s nest, perched upon a steep height, inaccessible to travelers. The itinerary followed by the emperor, who debarked at Helenopolis and went towards Nicaea, serves to complete these scant pieces of information. The fortress must necessarily be sought to the east of Helenopolis, upon the portion of mountainous coastline that extends for some twenty kilometers, of which about five are on this side of Prainetos (the contemporary Karamürsel) and fifteen beyond it, since the rest of the coastline is bounded by a coastal plain. In fact, the town of Prainetos is the normal point of call for the voyager who is traveling from Constantinople to Nicaea by crossing the Gulf of Nicomedia off the coast of Helenopolis. To my knowledge, one can supply no more precise location for the fortress of St. Gregory, which was doubtless situated to the west of Prainetos, even if one were not, properly speaking, “at the entrance of the Gulf” of Nicomedia.
In chapter 15 of Book III of his Dogmatic History, George Metochites gives a more detailed account of the negotiations between the emperor and the unionists and better traces the steps of the imperial journey, after having nevertheless omitted the first of these steps, that is, the emperor’s visit to John IV Laskaris; once he had arrived in the East, Andronikos II spent some time in a village which was found at the foot of the fortress in which John Bekkos and Constantine Meliteniotes were imprisoned. From there, he sent a message to the prisoners; they were allowed to come down to the village; they met with Theodore Mouzalon and with the emperor, and gave their agreement to the holding of a conference at Lopadion, to which George Metochites, from his own place of detention, would also come; after having received a kind reception, they returned to the fortress. As for Andronikos II and Theodore Mouzalon, they made their way to Nicaea, then to Lopadion, where the conference which had been expected did not take place. At the start of chapter 17, George Metochites notes that, after their stay at Lopadion, the sovereign and his entourage arrived at Nymphaeum.
 Bonn ed., vol. II, p. 15, line 9; p. 103, lines 4-6; p. 270, line 15.
 τὸ κατὰ δεξιὰ εἰσπλέοντι τὸν Ἀστακηνὸν κόλπον φρούριον (Bonn ed., vol. II, p. 103, lines 4-6). The formula may be an echo of Thucydides (I, 24,1): Ἐπίδαμνός ἐστι πόλις ἐν δεξιᾷ ἐσπλέοντι τὸν Ἰόνιον κόλπον.
 ἔς τι πολίχνιον τῶν περὶ Βιθυνίαν παραλίων πέμπεται ἅμα τοῖς ἀμφ’ αὐτῷ (Bonn ed., vol. I, p. 171, lines 2-3).
 Metochites, I, 122; III, 6, 9, 11, 15: Cozza-Luzi, p. 175; p. 322, 324, 325, 327. He is very specific that the prisoners were brought there by boat upon their departure from Constantinople (I, 122; p. 175), even if this information does not, of itself, imply that the trip was made by boat all the way to the end and that the fortress was situated alongside the sea.
 It is true that that the verb εἰσπλέειν does not have so precise a meaning. To preserve for it a sense of greater indetermination, one might translate it using some vaguer form of words: “to go in” (to the gulf), “to sail forward” (into the gulf). The fortress of St. Gregory probably was found in the vicinity of Kavak Iskelesi.
 Metochites, III, 15 (Cozza-Luzi, p. 327): κατασκηνοῖ μὲν ὁ κρατάρχης ἔν τινι κώμῃ περὶ τοὺς πρόποδας τοῦ ὄρους ἐφ’ οὗ περιίστατο τὸ πολίχνιον.
 Metochites, III.17; Cozza-Luzi, pp. 328-329.
Linked is a picture of some of the hill country above Karamürsel: indeed, the foreground of the picture shows what looks like the foundations of an old building. Perhaps the fortress of St. Gregory stood on some such spot as this.