The Byzantine Orientation and its Supporters
17. DAI 37 = FBHH 41
18. Ioannes Skylitzes Synopsis 5. = FBHH 85-86.
sian m anuscript — repeats, and fills out, this momentous report. The es
sence of the addition runs as follows: "A nd the Greek high-priests had still not been able to plant their feet firmly on their land (on the land of the Paeons or Magers = Hungarians) or to teach them the words of the Scrip
tures, when one of the two princes (knaza = the harka and the gyula), whose nam e was Stephen, died in the holy Christian faith and having had pleased the Almighty with many good deeds, entered the Kingdom of Heaven in peace." W e also learn that no holy books had been written in the Paeons' (=
Pannons') own language, and that "the heretic Latins from Rom e ... bring
ing along their books and writings''19 exploited this situation. Since the Latin mission only began in the autumn of 972, the above-mentioned must have occurred some time earlier.
In order to understand the above text, we must first recall that there were three leaders in the Hungarian tribal hierarchy: the kende (the chief duke) and the gyula (the military commander); the third was known as the harka (perhaps he was the chief justice). W hile the titles of kende and harka later disappeared into oblivion, gyula became a proper name after the m id
dle of the eleventh century, and subsequent chroniclers projected it back into the past as such. Arab sources describing the Hungarians before the time of the Conquest, however, were aware of the fact that the di.la or jila (gylas) was not a name but a rank, and so was Constantine Porphyrogennetos, w ho learned about the second leadership rank in Turkia from his H ungar
ian visitors.
But to get back to the story: the Hungarian attacks on Byzantium were renewed in April 959, and the Hungkrians once again reached the gates of Constantinople under the leadership of a commander called Apor (Opour), whose memory and erstwhile habitat is what is probably preserved in the nam e of a now vanished village near Mindszent by the Tisza River. The Hungarian army plundered Thrace and Macedonia in 961, and after raid
ing the environs of Constantinople and Thessaloniki in 968, returned home w ith a large num ber of captives. Two years later, in 970, the H ungarian raids came to a definite end following the defeat of the allied H ungarian- Russian-Bulgar army at Arcadiupolis.
The evidence that helps in the location of the base for these cam paigns against Byzantium is quite spectacular. During the joint reign of Romanus I and his sons, Byzantine coins suddenly came pouring into Hungary. W hile only fifteen early tenth-century coins and one golden solidus have been found, there is a marked increase after 934: twenty-two coins, five of which are gold. The peak can be dated to a relatively short period: the rules of Constantine VII and Romanus II (948-959) — twenty-eight coins, of which twenty-four are gold! —, and it perfectly matches the events we have been describing. The archaeological finds show a small decline for the period between 963 and 970 (sixteen coins, eight of which are golden solidi), but this is misleading, since it was precisely in 965 that the astonished Ibrahim Ybn Yakub noted that the "Tu rk" merchants of Hungary were paying with 19. Facsimile in G. Fei ie r, A nagyszentmikl6si kincs-rejt61y megfejtesenek utja. (The Way to Solving the Nagyszentmikl6s Treasure-Mystery.) Archeologiai trtesito
1950, 45. (Based on an old edition, Moscow.)
122
gold coins at the market in Prague. The year 970, however, marks the last year for these coins.
N ot counting the single solidus found in Sirmium, all these Byzantine coins, mostly gold pieces dating from the period between 934-969/970, have com e to light in the Tisza region and east of the Tisza from Tokaj down to Orsova. More specifically, the authenticated graves containing golden coins (eleven sites) have all been found in the area stretching from the Beretty6 and Koros rivers down to the Temes region by the Lower Danube. The distribution area of these Byzantine coins is precisely the same as the area from where Byzantine "lion-buckles" — five of the six occurrences —, gold and silver earrings, and a sword have been recovered. The latter was un
earthed at Kunagota together with some Byzantine silver coins — hardly a coincidence. With the exception of Sirmia, tenth-century Byzantine gold and silver pieces and jewellery have not been found in the Carpathian Basin on the right side of the Danube. That absolutely none have been found in Tisza region, the area between the Beretty6 and Koros rivers and the Lower Danube, who participated in the campaigns against Byzantium. This, of course, is not to say that the ransom Byzantium paid for non-aggression and the presents given to the horka and the grand duke did not find their w ay to other centres, though no traces of their having done so have been uncovered. All the finds have come from warrior graves in the Tisza re
gion. Since the soldiers brought the booty back here with them, the prison
ers freed by theg y u la between 959 and 969 must also have been taken to this sam e area.
All these details add up to the following: the land of the "patrikios and gyula" Stephanos was probably located east of the Tisza, between the Koros and Maros rivers. It is probably the name of his headquarters that is pre
served in the name of the town of Gyula (previously Julam onostora). Since Bishop Hierotheus did his missionary work among the people in the gyula's land, his permanent residence (assuming he had one) must also have been at the gyula's court.20
20. It is hard to imagine that his seat would have been in Sirmium (Sz&vaszent- demeter/Sremska Mitrovica), which counted as the border territory of Bulgaria, where, on the other side of the Sava in Sermon (Macvanska Mitrovica), by this time, a Bulgarian-Serbian bishop was active, whose cathedral, built on the ruins of an early Christian basilica, is well known from the middle of the tenth cen
tury (D. Minic, Le site d'habitation medieval de Macvanska Mitrovica. Sirmium XI. Beograd 1980; cf. with the text written about early mediaeval cemeteries:
Sirmium XII, Beograd 1980). This latter Sirmion/Sermon came under Byzan- tinian rule in 1018, and is not identical with the Hungarian Szer6mseg, as his
torical literature thinks. Also, a diocese is unimaginable at the far away Gyula- feh6rv6r, which, before 950 could not have been the seat of the gyulas.
In 970, for the first time in 300 years, the Byzantine em peror Ioannes Tzim isces reached the Lower Danube with the Byzantine army in pursuit of the allied Russian-Pecheneg-Bulgar forces. In 971, he set up a Byzantine thema w ith Dorostolon (Silistra) at its centre. This event may have been be
hind the gyula Stephanos's unexpected removal east to Transylvania. Since the defeated Pechenegs of Wallachia had temporarily disappeared from the historical stage in the tenth century, the gyula, in southern Transylvania in effect became Byzantium 's neighbour in the Lower Danube region. In the Hungarian Old Gesta, the event is given the usual epic aura. W hile on a hunting trip "in Erdeel", the great and mighty commander, Gyula (Gula dux m agnus et potens) is supposed to have discovered the fort of Alba (Civitatem Albam) that had been built in ancient times by the Romans. There is a seed of truth in the hunter-legend, insofar as the gyulas had, indeed, not resided in the Roman town for some time following the Conquest and only moved there from Hungary at a later date. Hierotheus could not very well have been alive by then, so it is more likely that the gyula was accompanied by his successors,21 if indeed the bishop had a successor. The twelfth-cen- tury polemical writing, however, vaguely supports this possibility. In any case, the Byzantine presence on Transylvania's borders was soon a thing of the past. The Bulgars had had a bad year in 971, so much so that they had been pushed back into Macedonia. Their envoys appeared in the court of Otto I in Quedlinburg in 973 together with Grand Duke G 6za's Hungarian delegation in the hope of securing some western support. But the Bulgars soon recovered, probably with G6za's help. In 976, they succeeded in oust
ing the Byzantines from the Lower Danube region for another quarter of a century. It is not very likely that the pious Stephanos lived long enough to see these events, although his putative death much earlier, in 956, is not supported by documents. But in the 970s at the latest, the aged gyula, a contem porary of Prince Fajsz and Taksony, was succeeded by the "second"
gyula, who was a contemporary of Gaza's. His daughter, Sharolt, became Grand Duke Gaza's wife and, later, the mother of Vajk (afterwards King Stephen I).
The marriage of Grand Duke Geza and Sharolt was undoubtedly politi
cally motivated. It was the gyula who needed this marriage in order to find favour with the prince. Although Sharolt, born east of the Tisza River in the second half of the 950s, must have been christened by Hierotheos on in
structions from her pious grandfather, she was still given a Kabar/Khazar- type Turkic name: Sharaldy = W hite weasel, or Sharylty = W hiteness. Her nam e was later translated into Slavonic form as Bele-knegini, which has the sam e meaning: W hite princess/lady. [Her supposed sister "C arold u " is
21. Inscription "Antonios bishop governor (proedros) of Turkia" on a lead stamp of unknown origin, dated to the 11th century and "Theophylaktos bishop (episkopos) of the Turks" on a similar one: FBHH 253. — It is not likely that these refer to Hungarians, they should be considered rather as alluding to high priests of the
Turks near Vardar.
124
again, an invention of Anonymus's or rather, one of his misreadings.)22 In the same way that he created a Bulgar commander called Shalan from the Hungarian name, Calan (Calan —» Chalan -> Shalan), he similarly, contrived the version "C arold " from the name Sharolt/d (Sharolt/d <-C harold—>
Carold). Since the name Sharolt, spelled with an "S h ", was fam iliar from other sources, "Caroldu" became the "sister" for lack of a better alterna
tive.]
The actual marriage ceremony took place in the 970s. If V ajk-Stephen really was bom in 977 — the latest research indicated that he was definitely born after 975 — then his birth came at a time when the gyula was in serious trouble. It was exactly in 976 when the Bulgars cut off his direct links with Byzantium. Sharolt had been taken to Esztergom from Transylvania, as evi
denced by the name of the village of Sarold, formerly her estate near Segesvar by the Nagy-Kuktilld River. (The villages featuring "D ecse" in their names, for instance Marosdecse, the salt port at Torda, and the village of Magyar- d6cse located in the salt mining region of the Szamos River, may all have belonged to Stephen's father G6za, but they are more likely to have been nam ed after King Geza I.23 The original form of Geza was Gyecse.) There is not m uch m ore that we know about the rule of the gyula, Sharolt's father, although this much is certain: he tried to maintain his court in Feherv^r in a princely fashion. The court chapel, rotunda, was probably built during his rule, and the "gyepu" in the Kukiill6 Valley was pushed further east, be
yond Segesvar.
Sharolt's brother — who must have been younger than Sharolt, since w hen his two sons, Boja and Bonyha, died a violent death at the time of the second plot against King Peter in 1046, they were both still in the prime of life — probably took over the rank of the "third" gyula of Transylvania at the beginning of the 980s, as the author of the Old Gesta could still vividly recall. His contemporary, the extraordinarily energetic Byzantine emperor, Basil II — later known as Bulgaroctonus = "Bulgar Slayer" — began his forty- year-long incessant and, in the end, successful fight to restore the Byzantine Em pire to its former glory shortly afterwards, in 985. He launched his Eu regnum manu tenuit”.2i Geza's death, Stephen's elevation to the throne and the crushing defeat, in 997, of the pretender to the throne, Koppany, all 22. The written Hungarian sources from the twelfth and early thirteenth centuries
spell the s (s), cs (c) and k sounds, sometimes the h, too with an ch. Cf. Chimon = Simon, Choma = Soma, Chanad = Csanad, Chatar = CsatSr, Chemey = kemej, Chemen = Kemeny, Chupa = Kupa. (K. F e h £ r t 6 i , Arpadkori kis szemelyn6vtSr /Personal Names of the Arpad Period/, Budapest 1983.) Charold as the "wife of Doboka" was also mentioned in the history of Transylvania.
23. On his coins: Geuca, on his charter: Geisa, according to the Byzantine script on the Crown: Geovitza(s), which has to be pronounced as Gieitcha-Gievitcha; the harder Decse-Devicse version developed from this.
125
24. G o m b o s , Catalogus ... Ill, 2203-2204.spelt trouble for the younger brother, whom the Slavs, for some obscure reason, nicknamed "P roku j", meaning "the son of Prok". (Prok — "rem ain
d er", "su ccessor" or possibly "descendant".) The quarter of Kopp&ny's drawn and quartered body sent to Transylvania must have gone to th e gyula in Feh6rv£r, which can be interpreted in many ways, except as a gesture of friendship.
The gyula's unexpected and rapid fall was initiated by Basil II's attack on Bulgaria in 1001. By the end of 1002, the emperor had occupied Vidin and restored the Byzantine thema Thrace south of the Lower Danube. Stephen could not afford to wait for the emperor to offer help to the power-hungry gyula, who was in any case already far too much under the influence of his Greek priests.
The Annals o f Hildesheim an d Altaich n ote briefly in 1003 that "K ing Stephen led an army against King Gyula, a maternal uncle of his, and took him to
gether with his wife and two sons into captivity, forcibly converting his whole country (regnum) to the Christian faith".25 The sources m ake no m en
tion of there having been armed resistance or a clash of some sort, let alone a "Rom anian-H ungarian w ar", and the aftermath of the action certainly suggests nothing of the sort.
W ith this, the unbalanced "separate history" of Transylvania, covering a quarter of century came to an end. Stephen unprudently did not hold the prince in captivity who had assumed the role of a king. He escaped from Esztergom some years later and changed sides to join the Polish king Boles
lav I the Mighty (992-1025). Stephen generously allowed his wife to follow him freely, without requiring a ransom. The best evidence how deeply the (senior) Lord "Prokuj" deplored his lost "kingdom " was that he had the face to take arms against Stephen and his country. Thus, Stephen felt obliged to expel him from the border fortress which was entrusted to him by King Boleslav. All these happened before 1018, when Bishop Thietm ar of M erse
burg, who related the late reports of Prokuj, died. The former prince must have finished his life somewhere on Polish soil. Boja and Bonyha, his sons, remained faithful to Istvan and their people.
The action which was over-estimated by both the Hungarian and the Rom anian historiography, i.e., the defeat of "Leader" Ajtony, took place after the organization of Transylvania was established when the royal trans
port of salt on the Maros River to the Tisza River had become already regu
lar. The exact date of this event, possibly ranging from 1003 to 1028, has been extremely contested up to the present.
It is unlikely to ever be determined down to the year. At the tim e of G eza's rule and at the beginning of Stephen's reign, all aspects of life in the M aros-Tem es region followed the pattern being set in Hungary (see the cem eteries of the conquering Hungarian warriors and, later, the swords and the burial places of Gaza's militia). Graves from the second half of Stephen's rule, containing royal coins, have come to light from Hodony all the way down to Mehadia.
The incident itself had left no written trace in the contemporary histori
cal sources. Of the close to ten written sources, only the greater Legend of
2 5 . G o m b o s , Catalogus ... 1 , 1 4 1 , 1 3 9 a n d I, 9 2 , 2 0 5 .
126
Saint Gerald (Legenda maior S. Gherardi) and Anonymus' Gesta mention any
thing, the two accounts providing two radically different versions. The leg
end had it that Prince Achtum/Ohtum was baptized in Vidin according to the Greek rite. (His name originated from the Turkish word Altun, m eaning
"g o ld ", and through regular patterned linguistic change, took the H ungar
ian form "A jtony" in the same way as "Falis" was transformed into "Fajsz".) Later, on G reek (read: Byzantine) instruction (authority), he founded a m onastery at Marosv&r, dedicated to Saint John the Baptist, and invited Greek monks as well as an abbot to settle there. This event could only have taken place after 1002, for by the end of 1002, Basil II had taken Vidin from the Bulgars and extended the frontiers of the Byzantine Empire up to the Low er Danube region, that is, to the borders of Ajtony's domain.
A jtony's personal habits, in spite of his flirtation with Byzantium, re
m ained that of a heathen as he took "seven w ives". Though this precise num ber sounds suspiciously like a folk-tale element in the legend, there can be no doubt that he had more than one wife. After Ajtony's defeat, one of these wives was given to comes Csandd, another to comes Bees. (On this, the two reports confirm one another, and the information is undoubtedly au
thentic.) Vast herds of horses and cattle grazed on Ajtony's land "n o t count
ing those which were put up by the herdsmen in stables".26 (Only a few years ago, this later information was still regarded as one of the later added elem ents of the legend, on a par with the descriptions of Ajtony's manors and mansions. Today, however, there is growing archaeological evidence indicating that the practice of stabling animals was known in the H ungary of the tenth and eleventh centuries.)
Ajtony probably had a sizeable army of the same type as G eza's militia, and was a man of overwhelming self-confidence. Originally this choleric, w ealthy and braggart lord was probably the governor of the then recently built Marosvdr (urbs Morisena) who underestimated the balance of power, and switched his loyalty to his mighty southern neighbour. H e succeeded in subjugating the heathen Hungarians living around Bekesv^r and as far as the KOrOs River, and made himself lord of the Temes region. W ith this, he began to pose a threat to communications between the royal seat and the recently acquired Transylvanian territories. Ajtony's unbridled covetous
ness — he was insolent enough to raid the royal salt consignments —, his heathen practices and, above all, his flirtation with Byzantium infuriated King Stephen.
To vanquish the rebel, King Stephen sent his kinsman, Doboka's son, Csanad (Chanad, Sunad), and — according to the "Csanad Saga", an inser
tion of dubious historic value in the Gerald Legend — the gyula, w ho hap
tion of dubious historic value in the Gerald Legend — the gyula, w ho hap