The Hungarian Settlement of Transylvania

In light of Transylvania's geographic endowment, it is understandable that Anonymus 'led' the Hungarian conquerors to the place that best suited their traditional way of life, the small plain known as the Mezőség. The Hungarians' presence in this region is attested by archaeological finds from the 10th century, 11th-century written sources, and toponyms that in some cases can be traced back to the 10th century.

Since Anonymus' purpose in writing his chronicle was to relate the 'origins of the kings and nobles of Hungary', the events concerning Transylvania had to serve this end. Judging from his accurate placement of Gyalu castle, the village of Esküllő, and the Almás, Kapus, and Szamos rivers, he had reliable information about the Kis-Szamos region, and indeed the text reveals his sources: members of the Zsombor and Agmánd clans, who were happy to pass on their family legends and thereby help the author to authenticate their ancient property rights.

Unfortunately, it is not clear who the Zsombors identified as their ancestor at the time of the conquest, for Anonymus embellished their ascendance, tracing it back to Tétény's grandson Zombor, though he fails to indicate who were the latter's sons; as for Tétény's other grandson, Gyula, Anonymus affirms that his sons Buja and Bonyha were banished by King St. Stephen from Transylvania. In fact, as has been noted, the Maglód (or Gyula-Zombor) clan lived in the diocese of Vác and had nothing in {1-372.} common with the Transylvanian Zsombor clan. The similarity of the family names Zombor and Zsombor helped Anonymus to concoct a fictitious family tree. With regard to Agmánd clan, it is clear that they traced their ancestry to Apafarkas and his son Agmánd; the former inspired their coat of arms, which sported a wolf (farkas = wolf), the latter their family name. In all likelihood, it was information provided by the Zsombor and Agmánd clans that led Anonymus to associate the names Gyula and Gyalu, and to link his account of Transylvania's conquest to the person of Tétény. To be sure, the two clans, according to their family legends, descended from original settlers in the Kis-Szamos valley. The Zsombors probably claimed to be descended from one of the seven chieftains, for their coat of arms held a lion; this animal had been the totemic emblem of tribal leaders and figured on the crests of the ruling élite. Thus Anonymus, helped by the similarity in the names Zombor and Zsombor, obliged the clan by making the chieftain Tétény their ancestor.

Anonymus's story, cobbled together from disparate elements, would be of little relevance were it not for the fact that the Zsombors' and Agmánds' properties were all located in Transylvania, north of the Maros, indicating that these families were indeed descendants of the original settlers in the region. The question remains, when did this settlement actually occur? According to evidence from other parts of Hungary (e.g. Komárom county), when the feudal state was established, two thirds of the land (and people) belonging to the clan chiefs (who were called bő) was taken away and converted into the megye (a word that came to mean 'county') of royal castles. Research has shown that as late as the 13th century, the Zsombor family still owned around a third of the land in Doboka county, divided in two parts. The first was situated to the left of the Kis-Szamos and extended to the Almás valley; its principal locality was (Magyarnagy)zsombor, and it also encompassed Drág, Milvány, Ugruc, Esküllő (cf. Anonymus's {1-373.} story), Adalin, Szótelke, (Ördög)keresztúr, (Réce)keresztúr, Szentkatolna, (Magyar)fodorháza, and Sólyomkő (also called Elefánt), as well as many villages that no longer exist. The second property, fifty kilometres east of the Kis-Szamos, straddled the Mélyes creek in the Mezőség; the villages there included (Szász)zsombor, Mányik, Kékes, (Szász)móric, and (Dellő)apáti. Most of these toponyms (including some of Slavic derivation) are personal names, rendered in the nominative and without suffix or article, which was the Hungarians' practice between 900 and 1270. Equally Hungarian are two toponyms that date from the 13th century and identify the patron saint of the local church: Keresztúr (= Holy Cross) and Szentkatolna (= Saint Catherine).

The many toponyms that date from after the Mongol invasion and contain the word telek indicate that the land owned by the Zsombors had been settled prior to the 13th century; the addition of 'telek' to an earlier toponym signified that the site, once settled, was now deserted. Most of these deserted villages bear old-style Hungarian toponyms: Póstelke, Lőrincszigete, Váma, Mikustelke, Kéthorpágy, Örkénytelke, Fikacstelke, Tibatelke, Berketelke, Palota, Rigótelke, and Boroszló. It is likely that the original centre of the Zsombor clan's domain was the castle later known as Doboka, and that the salt-mine at Szék also belonged to them. When the royal county of Doboka was established, the castle and the salt-mine became royal property, as did the proximate villages (preexisting or newly-settled) of Kend, Poklos(telke), Lózsárd, and Inak. The soldiers and servants of the royal county organization were settled nearby — the former at Cseh and Jenő, the latter at Udvarnok and Kovácsi. The episcopal properties that lay farther east, near Harina and between the Lekence and the Sajó rivers, were probably owned originally by the Zsombor family, and had been appropriated by the king before being transferred to the Church.

{1-374.} Similarly, the royal castle of Belső-Szolnok county, at Dés, and the Désakna mine lay on land once owned by the Agmánd clan; the latter still owned property to left of the Szamos (Keménye, Szekerestörpény, Gyékényes, Kodor, Szükerék, Ormány) as well as on the right bank (Péterháza, Boncnyíres, Hesdát, Bodzástelke). The Agmánds also had property farther afield, around the Lekence valley in the Mezőség: (Uzdi)szentpéter, (Nyulas)néma, Mező(rücs), and a few other villages that have disappeared.

Thus it appears that the Zsombor and Agmánd clans settled along both banks of the Kis-Szamos in the 10th century, well before the establishment of the royal county system, and that they were subsequently compelled to surrender two thirds of their land, including castles and salt-mines, to the king. By extension, one can assume that other, exclusively Transylvanian clans whose properties were divided up in this fashion also descended from the original settlers. There are three such clans, the Kalocsa, Borsa, and Mikola, the latter being named after its earliest known ancestor.

The Borsa clan, who in the 13th century also acquired land in Bihar county, possessed ancestral holdings around Kolozsvár and the Kolozs salt-mine, at Kolozsborsa, Kolozsgyula, Macskás, Szentpál, and Saság (to the left of the Kis-Szamos), as well as at Ajtony, Novaj, Mezőszentgyörgy, Mezőszentmárton, and Majos (on the right bank). Wedged between them lie royal properties: Lomb, Pata, Szopor, Szovát, along with two settlements that appear to have been reserved for the military, (Mező)őr and (Mező)keszü, and another, Udvarnok, whose name indicates that it was inhabited by servants.

The northern branch of the Kalocsa clan was called Szil. In the 13th century, Szil properties stretched along the northwestern edge of the Zsombors' domain and included Kalocsa, Szentpéter, and Mikó, as well as Szili and other extinct villages. This branch's name died out in the mid-1300s, but the southern branch of the family, called Tyukod, survived. The latter's ancestral holdings, clustered {1-375.} around Torda castle and the Tordaakna salt-mine, included Detrehem, Mindszent, Egerbegy, Kók, Tóhát, Szentkirály, Szentmárton, Füged, and Bányabükk, as well as the now extinct Tordalaka and Szarkad. In 1270, Székelys from Kézd were resettled amidst these villages on twenty-one sites that were part of the royal estate of Torda. The Székely villages all bear Hungarian names, such as Dombró (a 10th-century borrowing from Slavic), Örményes and Igrici — both names evoke royal servants, for the first comes from örlőmalmos, meaning 'miller', while the second means 'minstrel' — and, farther east, Megyer, where lived the royal castle's serfs.

Finally, there is a clan whose common name is not known, but which had many branches, including the Mikola, Gyerő, Kemény, Kabos, Radó, Tompa, Vitéz, and Veres. Their estates were spread around Gyalu castle, a royal property that had been transferred to the Church to become an episcopal seat. The clan owned properties in the valleys of the Kapus and the Nádas rivers, at Gyerőmonostor, Gyerővásárhely, Pányik, Kapus, Derite, Bedecs, Erdőfalva, Inaktelke, Szamártelke, Boc, Andrásháza, Tamásháza, Topa, Szentmihály, Szentkirály, and Berend; and east of Kolozsvár, along the bend of the Kis-Szamos River, at Dezmér, Szentmiklós, and Szamosfalva, which lay close to the salt-mine at Kolozsakna.

Hungarian clans often had a common monastery to serve all their branches. Of the monasteries that belonged to the five clans in northern Transylvania, one only has been clearly identified, that of the Mikola clan. Built prior to the Mongol invasion, it was eventually named Gyerőmonostor, after one of the clan's families. A common coat of arms symbolized the unity of this clan, which had many branches. The monastery at Almás, mentioned in sources dating from 1294–1320, was probably owned by the Borsa clan; the buildings were later destroyed, and never rebuilt, but the locality continued to bear the name Monostor-Almás well into the 15th century.

{1-376.} Although the five Hungarian clans established between the Szamos and the Maros cannot all be linked to known monasteries, each of them had a coat of arms that characterized a clan at the time of the Conquest. Each of these crests bore the figure of an animal: a lion for the Zsombor clan, a wolf for the Agmánds, a deer for Gyerős, a fish for the Borsas, and a bird for the Kalocsas. All five clans owned large blocks as well as scattered parcels of property, spread around royal (or episcopal) castles and salt-mines on both sides of the Szamos; judging from the location of these properties, the royal counties and episcopal holdings had been carved out of the clans' original estates.

The evidence also suggests that in the early period of settlement, clan families, or at least the shepherds, periodically moved their flocks from one pasture to another along the Kis-Szamos and other rivers. This pasture-rotation is indicated by pairs of toponyms, identifying sites some 50 km. apart, on both sides of the Kis-Maros: (Magyarnagy)zsombor — (Szász)zsombor, Buda (today's Bodonkút) — (Buda)telke, (Szarvas)kend — Ken(d)telke, Kecsed — (Mező)kecsed, (Oláh)péntek — (Szász)péntek, Néma — (Nyulas)néma, (Szekeres)törpény — (Szász)törpény, and (Kolozs)gyula — Gyulatelke. These toponyms are derived from old Hungarian and Slavic proper names, and were evidently applied before the early Hungarian toponymic custom died out around 1270; and many of them are found on estates of the clans that may be counted among the original settlers. The two places called Zsombor were in the ownership of the Zsombor clan as late as the 14th century, and the location of the two Budas, as well as that of the two Kends, points to the likelihood that these sites, too, belonged to that clan. Since Kecsed and Szekerestörpény belonged to the Agmánd clan, it is likely that the latter also owned the two corresponding sites on the right side of the Kis-Szamos; similarly, since the Agmánds owned (Nyulas)néma, on the right bank, it can be assumed that Néma, on the left bank, was also their property, {1-377.} particularly since Néma was situated close to other Agmánd holdings. The location of the two places called Péntek also points to Agmánd ownership. Apparently, the Agmánd clan had been granted land on the northern and eastern fringes of the original settlement area. This conclusion is consistent with Anonymus' contention that Agmánd had acted as scout for the leader of advancing Hungarians; his clan continued to perform this function on the periphery of the conquered land.

It is likely that the paired sites served to demarcate pastures belonging to the several family heads in each clan; this would explain why they are found on the periphery of the clans' original area of settlement. The paired sites became isolated from each other when the king expropriated the most valuable, central parts of these clan estates, along with their castles and salt-mines. (There may well have been additional paired sites, belonging to the three other clans; countless villages were destroyed by the Mongol invaders and disappeared from memory.

The date at which the estates were partially expropriated can serve as one reference point for estimating when the five clans originally settled in the region. Archaeological finds and Anonymus' chronicle point to the 10th century, but there is a more precise indicator: the toponyms (Szamos)jenő and (Mező)keszü, in the Kis-Szamos valley, and Megyer (later extinct), on the north bank of the Maros, which bear the names of three out of the seven Hungarian tribes. As was known long ago, and recently confirmed by research, the ruling prince's army consisted of soldiers recruited from the tribes, and the settlement of these soldiers occurred in the 10th century, ending during the reign of Stephen. The conquering tribes' names were not preserved in Hungarian legend; our knowledge comes from the notation made by the Byzantine Emperor Constantine in around 950. By the early 11th century, the names had faded from memory, for the Árpádian ruling princes generally drew tribal warriors — the jobbágy (later to mean serfs or villeins) — i{1-378.} nto their personal military service, and then settled the soldiers on land that they had taken away from the tribes; much of this land was inhabited by Slavs, either native-born or brought in as slaves.

When, at the end of the 10th century, Hungary was reorganized into a kingdom, this warrior class provided the officers — the 'castle warrior' — at the castles that served as royal county seats. Around these centres, there were generally several settlements bearing the name of one of the tribes, but only three were found in Transylvania: Jenő near Doboka castle, Keszü near Kolozs castle, and Megyer, which may have been attached Torda castle. This indicates that at the time counties were formed in Transylvania, most of the tribal soldiers who served in the princely retinue had already settled elsewhere; and that, in turn, implies the tribes in Transylvania were subjected to the authority of the supreme ruler or king rather late, probably during the early part of Stephen's reign. This hypothesis is reinforced by the fact that the first ispán (count) of Doboka county was closely related to King St. Stephen (according to the chronicles, the count's son, Csanád, was a nephew or cousin of Stephen). Earlier, the castle had probably served as the Zsombor clan's stronghold, and it acquired the name of its count when the royal county was established, just as Marosvár was renamed after Csanád.

Since the five clans in Transylvania retained a third of their estates, it is very likely that they had submitted peacefully to the expropriation mandated by the king. And in that case, a member of the Zsombor clan may well have been appointed royal count of Doboka — a post that was rewarded with one third of the taxes collected in the county. The case of Torda county reinforces this hypothesis. As noted, the castle at Torda may have been the political centre of the Kalocsa (a.k.a.Tyukod) clan's original settlement area. One of the earliest-named properties retained by the clan after expropriation was Tordalaka (in the vicinity of today's Aranyoslóna). Since the suffix laka indicates very early toponyms, {1-379.} it is most likely that at the time of King St. Stephen, a prominent member of the Kalocsa clan was called Torda (a name that occurs in other parts of Hungary as late as the 13th century), that he became the royal count of the clan's expropriated land, and that he gave his name to the castle taken from his clan. The earliest documentary reference to a Transylvanian royal castle dates from 1075, and it is to Torda. One of the properties retained by this count, probably his personal residence, is located near the castle and bears the owner's name.

Thus Doboka and Torda counties (and probably [Belső-] Szolnok and Kolozs counties as well) were established at the end of the 10th century, and, in the process, the five clans had to transfer to the king two thirds of their estates and many of their subjects. The clans obviously must have settled in Transylvania at an earlier date. According to the plausible legend related in several chronicles, the Hungarians, under Álmos's command, conquered Transylvania. Their next move is not entirely clear. Maybe the Hungarians proceeded en masse to the Danube-Tisza plain, leaving behind some soldiers to guard the hunting grounds, pastures, and mines of Transylvania. The legend, recorded by Anonymus, holds that the conquering tribes initially set the eastern boundary of their domain at the Meszes Pass; perhaps they wanted to keep a buffer zone, several days' march wide, between themselves and the Pechenegs, one that would absorb the brunt of the latter's attacks. The danger of Pecheneg incursions was very real, and as late as the 11th century, eastern nomads launched raids across northern Transylvania, all the way to the Nyírség. In the 10th as much as in the 11th century, these threats warranted the military occupation and defence of Transylvania.

Thus Anonymus' assertion is plausible, that the Hungarian tribes occupied the Mezőség as well the region between the Maros and Nagy-Szamos rivers in Transylvania in the first wave of settlement, and that the Zsombor and Agmánd clans played a leading {1-380.} role in this conquest. There are several indications of their links with the tribes that settled farther to the west. First, the vast county of Szolnok, which encompassed the Nagy-Szamos valley and the region around the Meszes Pass, was eventually subdivided into Outer, Central, and Inner parts, but it remained under the governance of a single count; it was the site of the salt-route, which passed through Szalacs to Szolnok, from where the salt was shipped on the Tisza River. Second, the bishop of Transylvania was from the start given authority over Szatmár and Kraszna, two counties that lay beyond that region but were on the salt-route. Numerous toponyms in the region between the Maros and Nagy-Szamos in Transylvania correspond to place-names in Szatmár (e.g. Kend — Kékkend, Kapjon — Kaplony), on the western fringe of Szabolcs (Zsombor — Zombor), and along the Szamos in Szatmár county (e.g. Bogát — (Nyír)bogát and Ogmándlovamezeje, to which there is only one documentary reference, dating from 1342, and which is the sole toponym bearing the name of the Agmánd clan). This evidence indicates that when the Hungarians moved westward from Transylvania, at least the major part of one of the tribes remained behind to supervise the salt-route, and this tribe kept contact via the Szamos valley and the Meszes Pass with those who had settled down between the Danube and the Tisza: the other six tribes and their Kabar allies (who had one, or up to three tribes).

There is further, corroborative evidence. Transylvania is the only region of Hungary where estates belonging to a 'ruling class' clan (the lion-crested Zsombors) could be found scattered among the holdings of ordinary clans. Moreover, the castles of four of clans in question — at Dés, Doboka, Kolozs, and Torda — were converted into royal county seats; and if the fifth, Gyalu, is an exception, it is because that castle soon became the seat (probably the first) of Transylvania's bishop, before the latter moved to Gyulafehérvár. This may be the reason why, of the seven bishoprics {1-381.} founded by King St. Stephen, the Transylvanian is the only one named after a province, and not a castle. The close links between the five clans are also indicated by the fact that five deaconries were created in the region between the Szamos and the Maros, at Szolnok, Doboka, Kolozs, Uzd, and Torda; although they do not correspond exactly with the four counties (Belső-Szolnok [Dés], Doboka, Kolozs, and Torda), their number and territory probably indicate an earlier administrative structure. When the Székelys of Marosszék were resettled in the region, the southern boundary may have been modified, leading to the abolition of Uzd county; that county's centre had been the castle and church which were excavated at Malomfalva.

In keeping with the above analysis, the first Hungarian settlement of the region between the Szamos and the Maros should be regarded as an occupation by elements linked to other Hungarians situated farther to the north and west. This occupation did not originally extend south of the Maros River. There a wholly different social and political structure emerged, and no traces were found of estates belonging to the conquering clans; the eventual settlers were probably propertied families who moved here from the western parts of the country. A further hypothesis is that the clans which occupied the region between the Maros and the Szamos all belonged to the same conquering tribe and were led by the Zsombor clan. In Transylvania, the process of adapting the ancient tribal structures to the new political forms, first the principality, then the monarchy, was slower than in the rest of the country, and the older type of property relations endured longer. The settlement in Transylvania of the tribe that consisted of the Agmánd, Kalocsa, Borsa, and Mikola clans under the leadership of the Zsombor clan, was thus part and parcel of the original conquest; these conquerors stayed, and for a time they preserved their tribal structure.