ARMAGH ARMENIA, GEOGRAPHY record a vigorous campaign, particularly in the eighth century, to extend its area of jurisdiction. There are numerous references to its abbots and other ecclesiastics throughout the early medieval pe¬ riod. Frequent references to its scribes and teachers mark it as an academic as well as religious center of note. Armagh’s prominence made it a target for raid¬ ing; its churches were periodically plundered and burned. From the late tenth to the early twelfth centuries {ca. 965-1134), the abbacy of Armagh was held by the powerful Airgialla family Clann Sinaigh. In the early twelfth century under Abbot Cellach O’Sinaigh and his successor, St. Malachy, Armagh became a center of the reform movement that brought the Irish church into conformity with the Continent, es¬ tablishing a diocesan system and introducing the newer monastic orders into Ireland. At the Synod of Rathbreasail in 1111, Ireland was divided in half with the northern sees under Armagh and the southern ones under Cashel. The Synod of Kells in 1152 modified this arrangement by making it a fourfold plan, adding the centers of Dublin for the east and Tuam for the west. Under both these divisions, Armagh retained primacy. In the early twelfth century the Augustinians, Do¬ minicans, and Franciscans established themselves in Armagh, though the Dominicans did not have a community there. Two communities of nuns, Tem¬ ple-breed and Temple-na-Ferta, became communi¬ ties of Augustinian canonesses around 1144; the Abbey of Sts. Peter and Paul had become Augustin¬ ian about five years earlier. These communities were finally suppressed in 1562. The Franciscan friary founded by the Dominican Archbishop Patrick O’Scanlain around 1263-1264 lasted until it was raided and destroyed in 1587. One of the older houses, originally a group of Culdees established prior to 921, had disappeared by 1600. BIBLIOGRAPHY Aubrey O. Gwynn, The Medieval Province of Armagh, 1470-1545 (1946); Katherine Hughes, The Church in Early Irish Society (1966); R. Buick Knox, James Ussher, Arch¬ bishop of Armagh (1967); Hugh J. Lawler and Richard I. Best, “The Ancient List of the Coarbs of Patrick,” in Pro¬ ceedings of the Royal Irish Academy, 29c (1911); J. Ryan, ed., Essays and Studies Presented to Professor Eoin MacNeill (1940), 319-334, 394-405. Dorothy Africa [See also Irish Church; O’Sinaich, Cellach.] ARMAGH, BOOK OF. See Patrick, St. ARMENIA, GEOGRAPHY. The physical outline of the Armenian highlands is clearly identifiable. The region that has been called the “Armenian For¬ tress” or “Armenian Island” is a high plateau with an area of some 300,000 square kilometers, formed as the east-west Syrian and Pontic chains draw closer together between 37° 30/ and 41° 30r north latitude. Towering over both the South Caucasian plain in the north and the lowlands of Mesopotamia in the south, it is linked with Asia Minor to the west and Iran to the east, primarily through the valleys of the upper Euphrates and Araks rivers. It is bounded in the north by the Pontic chain and in the south by the Taurus and the mountains of Kurdistan south of Lake Van. The general altitude of the plateau ranges from 1,000 to 2,000 meters, and its fortresslike character is reinforced by the higher mountains of the periph¬ ery, which average 3,000-4,000 meters. Within the plateau the “spine” of Armenia (consisting of a series of west-to-east chains culminating in Mt. Ararat [about 5,205 meters], then turning southward by way of Mt. Tendiiriik [3,548 meters] and running down between Lakes Van and Urmia to lose itself in the Zagros Mountains) divides the Armenian high¬ lands into northeastern “Araksine” and southwest¬ ern “Vannic” Armenia. Within these broad divi¬ sions, chaotic mountain formations with numerous volcanos (Sipcan, 4,176 meters; Nemrut, 2,910 me¬ ters, near Lake Van; and the fourfold Aragats, 4,180 meters, north of the Araks), higher plateaus, and iso¬ lated, restricted plains (of which the largest is formed by the valley of the Araks) create small units for which communication even with close neighbors is extremely difficult. The rivers, except for the Araks (flowing from west to east to join the Kura before both empty into the Caspian Sea), are normally mountain torrents rushing through all-but-inaccessible canyons. This is true even of the mighty Euphrates and Tigris within Armenia. The approaches to the major Armenian lakes, such as Sevan in the northeast (1,916 meters altitude, 1,416 square kilometers) are equally re¬ stricted by formidable mountain formations. Finally, the harshness of the relief is intensified by a severe climate with temperatures oscillating between ex¬ tremes of +40°C. and —40°C. Winter snows close the mountain passes up to eight months of the year, 470 ARMENIA, GEOGRAPHY ARMENIA, GEOGRAPHY but the burning summer heat, aided by the fertile volcanic soil, allows the raising of considerable crops where the relief of the land permits and the rainfall (greatly reduced by the Pontic Mountains’ halt of rain clouds carried by winds from the Black Sea) is adequate. The sharply accented and austere character of the highlands has considerably influenced Armenian his¬ tory. The lofty “Fortress” rising above the Mesopo¬ tamian plain has repeatedly provided refuge from at¬ tacks, and its strategic position has not escaped the notice of Armenia’s powerful neighbors over the centuries. The fragmentation of the country into small, isolated units has continually impeded at¬ tempts to create a centralized state. The relative ac¬ cessibility of the “Armenian Fortress” from the east and the west both increased its vulnerability to in¬ vasion and enhanced its cultural variety and its role in international trade. Finally, the close proximity of high mountains and plains within the country fos¬ tered the simultaneous development of two nearly ir¬ reconcilable socioeconomic systems: sedentary agri¬ culture and nomadic pastoralism. Whereas the physical elements of Armenian ge¬ ography are clearly identifiable, its political and ad¬ ministrative aspects still present numerous problems. The sources are often unclear: classical, Armenian, and Muslim authors use different names for the same area, while the same term has often been used, as we shall see, for different areas at different times. Con¬ sequently, extreme care must be taken to avoid seri¬ ous misunderstandings. From antiquity the term Armenia has included several subdivisions. 1) Greater Armenia/Armenia Magna or Maior, east of the Euphrates River, ex¬ tended between the latter and the confluence of the Araks and the Kura to the east, and between the mid¬ dle Tigris in the south and the Corox and the Kura in the north. This portion reached its maximum ex¬ pansion under the Armenian Arsacids. According to the seventh-century Armenian Geography /As\- arhaccoycc, formerly erroneously attributed to the historian Movses Xorenacci, Greater Armenia (.Mecc Haykc in Armenian) consisted of fifteen countries (asx^rhs) with imprecise boundaries, con¬ taining various numbers of districts (gawafs). In a generally counterclockwise order, beginning in the northwest, they were Barjr Haykc/Upper Armenia, on the upper Euphrates (nine districts); Corrord Haykc/Armenia IV (borrowing a Byzantine term discussed below), in southwestern Greater Armenia, east of Melitene (eight districts); Aljnik/Arzanene (in Greek), north of the upper Tigris (ten districts); Tu- ruberan, east of Armenia IV, extending from the upper Euphrates-Arsanias to the north and west of Lake Van (sixteen districts); Mokkc/Moxene, south of Lake Van (nine districts); Korcekc/Gordyene (in Greek), south of Mokkc (ten or eleven districts); Parskahaykc/Persarmenia (in Greek-Latin), around Lake Urmia (nine districts); Vaspurakan/Basprak- ania (in Greek)/Busfurradjan (in Arabic), from south of Lake Van, east of the lake and up to the Araks (thirty-five districts); Siwnikc or Sisakan/Sunitai (in Greek), southeast of Lake Sevan and south to the Araks (twelve districts); Arcax, northeast of Siwnikc (twelve districts); Pcaytakaran/Baylaqan (in Arabic), east of Arcax (twelve districts); Utikc/Otene (in Greek), north of Arcax and of Lake Sevan (seven districts); Ayrarat (the central royal domain), in the valley of the Araks (sixteen districts); Gugarkc/Go- garene (in Greek)/Somxetci (in Georgian), northeast of Ayrarat on the upper Kura (nine districts); Taykc/ Tao (in Georgian), northwest of Ayrarat on the Corox (eight districts). This Greater Armenia formed the kingdom of the Armenian Arsacids until the par¬ tition of 387. 2) Lesser Armenia/Armenia Minor, west of the Euphrates (not to be confused with Cilicia, to which this name was also given in the eleventh century). It gradually spread south to Melitene and west to Se- baste/modern Sivas and Caesarea of Cappadocia. In¬ corporated into the Roman Empire in the first cen¬ tury of the Christian era, it became the separate province of Armenia Minor with Melitene as capital under Diocletian at the end of the third century. A century later Theodosius I divided this province in two: Armenia I, with capitals at Sebaste and, later, Salala, and Armenia II, with its capital at Melitene. 3) The Armenian satrapies/ethne (in Greek)/ gentes (in Latin)—also called “the Other Ar¬ menia”—between the Euphrates-Arsanias and the Tigris in the southwest. Their territory, status, and number varied. Five or six are recorded in the sixth century: Anzitene/Anjit (in Armenian), Ingilene or Angelene/AngeTtun (in Armenian), Asthianene/ Hasteankc (in Armenian), Sophene/Copckc Sahuni or Sahe (in Armenian), Sophanene/Mec Copck or Supcan (in Armenian), Balabitene/Balahovit (in Ar¬ menian). These territories, which were autonomous until the end of the fifth century, shifted from the Iranian to the Byzantine sphere of influence, and dis¬ appeared as separate units in Justinian’s reform of 536. In 387, Arsacid Greater Armenia was partitioned 472 ARMENIA, GEOGRAPHY ARMENIA, GEOGRAPHY between Byzantium and Persia along a line stretch¬ ing from Theodosiopolis/Karin in the north to Dara, west of Nisibis, in the south. To the west the new Roman territories had three subdivisions. In ad¬ dition to its former territories—the two provinces (Armenia I and II) created from the earlier Armenia Minor, and the satrapies—Byzantium now acquired eleven districts from Arsacid Greater Armenia. These lay just east of the Euphrates and consisted of the entire Upper Armenia/Barjr Haykc of the Ar¬ menian Geography plus two districts taken from Armenia IV. This new unit was, confusingly, also called Greater Armenia/Armenia Magna (like the much larger Arsacid unit from which it had been taken), although writers of the period also refer to it as Armenia Interior. On the Persian side, to the east of the border, former Arsacid Greater Armenia, still called Mec Haykc by the Armenians but now known to Western sources as Persarmenia, lost most of its borderlands to its neighbors. Of its fifteen asgarhs listed above, it retained only the central six: Ayrarat, Turuberan, Mokkc, Vaspurakan, Siwnikc, and Taykc. The sixth century brought major changes to the administrative subdivisions of Byzantine, or West¬ ern, Armenia. In 536 a new law of Emperor Justinian (Novella XXXI) totally reorganized his Armenian holdings into four provinces: Armenia I was composed of the previous Greater or Interior Armenia plus the north¬ western territories of the old Armenia I (in Armenia Minor) and some Pontic lands. It lay between the upper Euphrates and the Black Sea, and included Theodosiopolis/Karin. Its capital was the new city of Justinianopolis, the location of which is still dis¬ puted. Armenia II (southwest of Armenia I) was composed of the rest of the old Armenia I and other Pontic districts. The capital was Sebaste/Sivas. Ar¬ menia III (south of Armenia II) coincided exactly with old Armenia II. Its capital remained at Meli- tene. Armenia IV (east of Armenia III) coincided with the former satrapies. The capital was Martyr- opolis/Npckert (in Armenian)/Mayyafariqln (in Ar¬ abic). A new partition of Armenia in 591 moved the frontier east to a line running from Garni (southeast of modern Erevan) in the north to Arest at the north¬ east tip of Lake Van in the south. The imperial dis¬ tricts were again reorganized to include the newly acquired Armenian territories in six or seven prov¬ inces. The new Armenia I was equivalent to Justini- anic Armenia III. Armenia II remained unchanged. Armenia III consisted of Justinianic Armenia I plus new territories including most of Arsacid Ayrarat and Turuberan. The new Armenia IV presents prob¬ lems. Some scholars equate it to the Justinianic Ar¬ menia IV, although it was now occasionally given the name of Upper Mesopotamia. Others, like Ni¬ kolai Adonts, postulate a second Armenia IV: Ar¬ menia IV altera, also known as Justiniana IV, though this hypothesis has been disputed. Finally, two more provinces were created from the newly ac¬ quired territories: Armenia Interior, between Karin and Kars, and Lower Armenia/Armenia Inferior, to the east between Kars and Lake Sevan. These admin¬ istrative divisions were to last about half a century. The Arab conquests of the mid seventh century brought other changes to Persarmenia, the western boundary of which moved back toward the Eu¬ phrates although its precise position fluctuated. The new Arab administrative term Armfniya should not be taken as an equivalent for Armenia, even in its widest extension. Armfniya included most of the lands of Caucasia, from the Caucasus chain in the north to the mountains of Kurdistan in the south; from the Euphrates, Corox, and Black Sea in the west to Lake Urmia, the lower Kura, and the Cas¬ pian Sea in the east. Some Arab writers, such as al-Baladhurl and Ibn Khurradadhbih, follow the Byzantine practice of subdividing the region into four parts, but these di¬ visions are composed of totally different territories. For them Armlniya I consisted of Caucasian Albania (that is, Utikc and Siwnikc) and eastern Caucasia; Armlniya II, of Georgia and western Caucasia; Ar¬ mlniya III, of the Armenian heartland (Sirak in the northwest, the Araks valley, and Vaspurakan); and Armlniya IV, of western Armenia from Karin to the north of Lake Van. Other geographers speak of In¬ terior Armlniya (Karin, Dwin, Naxcawan) and Ex¬ terior Armlniya (around Lake Van), or of Greater/ Armlniya al-Kubra and Lesser/Armlniya as-Sughra, around Xlatc/Akhlat and Tiflis, respectively. The autonomous Armenian territories recognized by the caliphate still centered on Ayrarat, but west¬ ern districts remained Byzantine while southern ones were absorbed into Mesopotamia. With the creation of the Bagratid kingdom at the end of the ninth cen¬ tury, however, a number of lost borderlands in the north were recovered. The new kingdom conse¬ quently extended over ten of the original fifteen Ar¬ sacid asgarhs—Turuberan, Mokkc, Vaspurakan, Siwnikc, Arcax, Pcaytakaran, Utikc, Ayrarat, Gu- garkc, Taykc—and some territory in Parska- 473 ARMENIA, GEOGRAPHY ARMENIA:HISTORY OF haykc, although its borders were unclear and much territory in the southwest around Lake Van and even the valley of the Araks was controlled by Muslim emirates. After the fall of the Bagratid kingdom in the mid eleventh century, the name Armenia passed to Cili¬ cia, or Armenia Minor. It was also perpetuated in that of the Byzantine theme of Armeniakon, created in the late seventh century from most of the lands of the original Armenia Minor west of the Euphrates up to the Halys, but subsequently reduced to a cen¬ tral strip running south from the Black Sea around Sinop. Additional, though short-lived, themes came into being on Armenian territory as a result of the Byzantine reconquest: Taron, in the southwest, in the late ninth century and, in the eleventh century, Theodosiopolis, around Karin; Mesopotamia, be¬ tween the main Euphrates and the Arsanias, west of Taron; Iberia, from Taykc to the vicinity of Ani in Sirak; and Basprakania, in Vaspurakan. With the Seljuk invasions of the mid eleventh cen¬ tury, the Armenian highlands could no longer be separated politically or administratively from the ad¬ jacent lands of Azerbaijan or Asia Minor. No fixed boundaries could be maintained in the prevailing chaos of the twelfth through fifteenth centuries; and in general only temporary local units, some limited to a single city, can be identified. Exceptions are the administration of the Zakcarids, who held much of north-central Armenia during the early thirteenth century, and the twelfth-century realm of the Shah, stretching north of Lake Van on either side of the capital of Xlatc. The eastward expansion of the Ot¬ tomans in the late fifteenth and early sixteenth cen¬ turies once again divided the Armenian highlands, this time between Turks and Persians along a north- south line that moved sharply with the fortunes of war, but often came to rest west of present-day Ere¬ van and east of Lake Van. BIBLIOGRAPHY Nikolai Adonts, Armenia in the Period of Justinian, N. G. Garso'ian, ed. and trans. (1970), pp. 7-182, 234-251, *103-*246; R. Blanchard, “L’Armenie,” in P. Vidal de la Blache and L. Gallois, eds., Geographie universelle, VIII (1929), for the best succinct account of the physical geog¬ raphy; Sowren T. Eremyan, Hayastan ast “Asxarhacc oycG (1968), for the fullest discussion of the seventh-cen¬ tury Armenian Geography and the best map of seventh- century Armenia; R. Hewsen, “Armenia According to the Asxarhacoyc,” in Revue des etudes armeniennes, n.s 2 (1965), for a summary of Eremyan’s work; and ‘On the Date and Authorship of the Asxarhaccoycc,” in Revue des etudes armeniennes, n.s. 4 (1967); Ernst Honigmann, Die Ostgrenze des byzantinischen Reiches von 363 bis 1071 (1935), with maps; Heinrich Hubschmann, Die altarmen- ischen Ortsnamen (1904; repr. 1969), for place names and a useful map; J. Laurent, L’Armenie entre Byzance et I’ls- lam depuis la conquete arabe jusqu’en 886 (1919; new rev. ed. by M. Canard 1980), pp. 39-75, for a useful summary and updating; Hakob Manandyan, The Trade and Cities of Armenia in Relation to Ancient World Trade, N. G. Garso'ian, trans. (1965), pp. 23, 47-52, 90-116, 155-172, 187-202, with maps (for the main transit routes); Josef Markwart, Eransahr nach der Geographie des Ps. Moses Xorenacci (1901); Sudarmemen und die Tigrisquellen (1930); and “La province de Parskahaykc,” ed. G. V. Abgaryan, in Revue des etudes armeniennes, n.s. 3 (1966); Arsene Soukry, ed. and trans., Geographie de Molse de Corene (1881), for the text and a translation of the seventh- century Armenian Geography. In addition to the maps indicated in the above works, see H. F. B. Lynch, Armenia: Travel and Studies, 2 vols. (1901; repr. 1965)—the map in the original edition is far more legible than its reproduction. Also see William M. Calder and George E. Bean, A Classical Map of Asia Minor (1957); Haykakan SSR Atlas (1961); A. H. M. Jones, The Cities of the Eastern Roman Provinces (1971), map of Mesopotamia and Armenia for the satrapies; and Heinrich Kiepert, Karte von Kleinasien in 24 Blatte (1902). Nina G. Garso'ian [See also Ani (in Sirak); Ararat (Mt.); Armenian Muslim Emirates; Ayrarat (Province); Caesarea (Capadocia); Theo¬ dosiopolis; Van, Lake.] ARMENIA: HISTORY OF. From antiquity, Arme¬ nia’s geographical position at the meeting point of the Greco-Roman and Iranian worlds created a sit¬ uation that favored the country’s cultural life, en¬ riching it with two major traditions but playing havoc with the continuity of its political history. As a general pattern, therefore, Armenia flourished only when the contending forces on either side were in near equilibrium and neither was in a position to dominate it entirely. This instability, characteristic of the entire medieval period, was compounded by a number of factors and has led to historiographic misconceptions likewise deriving in large measure from the geographical conditions, at least in the ear¬ lier period. The term “Armenia” has been used loosely to in¬ dicate any portion of the Armenian plateau and even adjacent areas. In the fourth century this indefinite “Armenia” consisted of three main sub-divisions. 474 ARMENIA:HISTORY OF ARMENIArHISTORY OF Armenia Minor (Lesser Armenia), was a Roman province west of the Euphrates that remained polit¬ ically and largely culturally, though not demograph- ically, outside the sphere of Armenian history. Ar¬ menia Magna or Maior (Greater Armenia) was the buffer kingdom east of the Euphrates ruled by a ju¬ nior line of the Irano-Parthian Arsacid royal house but under Roman protectorate since Nero’s corona¬ tion of the first Armenian Arsacid king, Trdat (Tiri- dates) I, in a.d. 66. It is this portion that is most com¬ monly equated with the general term “Armenia” by contemporary sources and modern scholarship. The autonomous Satrapies of southwestern Armenia (north of the Arsanias River) consisted of five or six separate units that had in large part entered into the Roman sphere of influence at the end of the third century as autonomous political and fiscal units (ci- vitates foederatae liberae et immunes), which were required only to coordinate their foreign policy with that of the empire. Nevertheless, Armenian sources ignore them and their autonomy, treating most of this region as part of the Arsacid kingdom of Greater Armenia. This geopolitical fragmentation was greatly in¬ tensified by those internal divisions of the Armenian plateau that favored the centrifugal tendencies of the magnates (naxarar) who asserted their hereditary prerogatives against any centralizing policy. Thus al¬ though Armenia Minor increasingly followed a di¬ vergent path from that of the eastern portion, and the Arsacid kingdom disappeared in 428-429 and the Satrapies in 536, Armenia can rarely be treated as a single unit. The internal fragmentation of the pla¬ teau, intensified through the mutual antagonism and perpetually centrifugal aims of the naxarars, may not be ignored, since the individualism of northern and southern districts continually asserted itself in the particularism of Vaspurakan, Taron, Siwnikc, and Taykc. Centrifugality and fragmentation are constants in Armenian medieval history, and they impeded all attempts, internal or external, to achieve a politically centralized state. Equally serious for Armenian historiography has been the question of sources. Not only are Armenian texts unavailable before the fifth century a.d., insuf¬ ficient or lacking in a number of cases, and riddled with problems of dating and attribution, they also normally present very definite and limited points of view. Characteristically they manifest an inalterable antagonism to all manifestations of Iranian culture or Byzantine Chalcedonian Orthodoxy; with few ex¬ ceptions they seek to obliterate the political, reli¬ gious, and cultural importance of sourthern Ar¬ menia; they overstress the importance of the particular noble house—Mamikonean, Bagratuni, Arcruni—that they support, while overlooking or denigrating others; and they disregard the important demographic changes brought about by the Arab, Iranian, and Turkish invasions of the Armenian heartland. Thus, they present a unified but greatly oversimplified view of the complex development of Armenian history. As a result of these distortions and of the paucity of supporting archaeological evidence, much of the history of early medieval Armenia, especially its chronology, still requires considerable investigation. The overthrow of the Parthian Arsacid dynasty in Iran by the Sasanians (around a.d. 224) made the Ar¬ sacid Armenian kings the enemies rather than the kinsmen of the Persian ruling house, at the same time that it rekindled the war between the Roman and Iranian empires. Battered between the two world powers, and probably divided for a time, Greater Armenia did not regain a measure of stabil¬ ity until the enthronement with Roman support of the Arsacid heir Trdat III/IV over the whole of this region and the transfer of the allegiance of the au¬ tonomous Satrapies to Rome. These events most likely occurred as a result of the Peace of Nisibis, concluded in 298 between Rome and the Sasanian king of kings, Narseh, whom inscriptions name as the former ruler of a portion of eastern Greater Ar¬ menia. During the reign of Trdat the Great, Christianity became the official religion of the Arsacid kingdom with the conversion of the king and the consecration of the patriarch, St. Gregory the Illuminator, at Cae¬ sarea of Cappadocia in 314. This conversion is tra¬ ditionally presented as the inauguration of Armenian Christianity, although the new faith had undoubt¬ edly penetrated into the southern area of the auton¬ omous Satrapies considerably earlier. This is attested by a passage in the Ecclesiastical History of Eusebius of Caesarea, by the persistent memory of the Apos¬ tolic or Edessene origin of the Armenian church, and by the assertion that “Armenia” had been the first officially Christian realm. Consequently, no single date seems warranted for the Christianization of all the Armenian lands, but the conversion of the Ar¬ sacid kingdom of Greater Armenia unquestionably marked a watershed in its history. It turned the country sharply away from its Iranian past. At the same time, the increasingly powerful national church provided both a durable focus for the cultural 475 ARMENIA:HISTORY OF ARMENIA:HISTORY OF and even political unity of Armenia and, on occa¬ sion, a challenge to the authority of its king. Much of the political history of Armenia remains unclear during the fourth and early fifth centuries. Armenia Minor, divided into two provinces by the emperor Theodosius I, was totally absorbed into the Roman Empire. Almost nothing is known concern¬ ing the autonomous Satrapies beyond the fact of their existence. In Greater Armenia the allegiance of the Christian Arsacid rulers (314-428/429) vacillated between Rome and Persia, although the Roman his¬ torian Ammianus Marcellinus considered them “faithful friends” of the empire. Probably the most outstanding among them, Arsak II (350P-367) sought to maintain a precarious balance between the two great powers. His attempts to curb the autonomy of the Armenian magnates by making their hereditary prerogatives subject to royal control and by creating at least one urban center, as well as his support of the Arianizing religious policy of Constantine’s successors, earned him the antago¬ nism of the rigidly Nicaean Gregorid patriarchs and of much of the nobility. Abandoned by Rome, as a result of the unfavorable peace concluded by the em¬ peror Jovian in 363 and by the southern satrap Me- ruzan Arcruni, who sided with Persia, Arsak II was defeated by the king of kings Sahpuhr II, and de¬ ported to Iran, where he commited suicide in the “castle of oblivion,” while the Persians overran Ar¬ menia and destroyed its cities. With Arsak’s disappearance the Arsacid dynasty, dominated by the powerful Mamikonean house, rap¬ idly declined. His son, Pap, was murdered in 374 by the very Romans who had reinstated him on the throne a few years earlier, and with the probable connivance of some of the nobles outraged by the king’s assassination of the great Gregorid patriarch, St. Nerses I (353-373). Seemingly by 387, although the date has been disputed, Rome and Iran agreed to a partition of Greater Armenia along a line running east of the Euphrates from the northern city of Karin, soon to be renamed Theodosiopolis, down to Dara in western Mesopotamia. This division left the western fifth of the kingdom in the hands of a powerless Roman client, Arsak III, while the far greater eastern portion became a Per¬ sian protectorate under another Arsacid candidate, Xosrov III/IV, who kept the royal capital of Dwin. Even this compromise solution was not lasting. In the western portion, the Arsacid dynasty ended with the death of Arsak III, whom the Romans eventually replaced by a primarily civilian official known as the comes Armeniae. In eastern “Persarmenia,” it lin¬ gered until 428-429, when the self-serving Armenian magnates themselves urged the Sasanian king of kings to put an end to the autonomous Armenian kingdom. With the disappearance of the independent Arsa¬ cid state, the lands of Armenia entered into an inter¬ mediate period of some two centuries, known in Ar¬ menian historiography as the marzpanate, during which they continued to be divided and disputed be¬ tween Byzantium and the Sasanians. In the absence of an internal political focus, the increasingly inde¬ pendent church provided the only element of unity and continuity, while the naxarars pursued their in¬ dividual interests, forming continually changing al¬ liances that coalesced periodically around the Ma¬ mikonean leaders of the proimperial party or the supporters of Persia, the princes of Siwnikc, and eventually the Bagratunis. The earlier contemporary sources, Tazar Pcarpecci and ETise, presented the Ar¬ menian opposition to Sasanian Iran in the light of a crusade; but from the sixth century on, the political and religious policy of Byzantium aroused a similar antagonism in the native historians, and recent in¬ vestigations have tended to show a highly complex situation. Fundamentally, the Armenian social structure fit¬ ted far more satisfactorily into the similarly aristo¬ cratic Iranian “feudalism” than into the centralized and bureaucratic Byzantine state; and most of the privileges of the naxarars were maintained by the Sasanians, who merely sent a viceroy, or marzpan, to Dwin as governor of their Armenian territories. Yet the national historians systematically stress the constant rebellions of Armenia against Iran, brand¬ ing as archtraitors the leaders of the pro-Persian party, Vasak and Varazvalan Siwni. There is, to be sure, no question that Sasanian pol¬ icy was oppressive at times and that the ill-advised attempt of Yazdgard II to reimpose Zoroastrianism on Armenia provoked the violent reaction that united the clergy and most of the magnates around the hereditary commander in chief of the Armenian army (sparapet), Vardan Mamikonean, in the great revolt of 450. Not even the decimation of Vardan and his supporters at the memorable battle of Ava- rayr (2 June 451) and the martyrdom of the religious leaders halted the guerrilla opposition of the coun¬ try, which exploded again in the successful rebellion of Vardan’s nephew, Vahan Mamikonean (481-484), at a time when the Sasanians were distracted by war with the Hephthalite Huns on their eastern border. 476 ARMENIA:HISTORY OF ARMENIA:HISTORY OF In 571 Vardan II Mamikonean rose again against Xusro I Anosarwan until the absence of help from the emperor Justin II forced him to take refuge in Byzantine territory. Nevertheless, Persarmenia often flourished under the Sasanian marzpans, especially under the native princes Vahan Mamikonean and his successors Vard Patrik and Mzez Gnuni (485- 548?), who followed the successful outcome of the rebellion of 484, and a vigorous national culture grew apace. As the Armenian church increasingly turned against the Chalcedonian Orthodoxy of Constanti¬ nople, it became more acceptable to the Sasanians, who no longer viewed it as a fifth column and a focus of hostility to their state. The Armenian patri¬ archs, or katcotikoi, preferred to reside in the marzpan’s capital of Dwin rather than on imperial territory; and the benevolence of later Sasanian rul¬ ers toward the Armenian church is reflected in the apocryphal stories of their secret conversions to Christianity found in contemporary Armenian sources. Simultaneously, the numerous favors shown by Xusro II Aparwez to Smbat Bagratuni, marzpan of Hyrcania, known as Xusro snum (“the Joy of Xusro”), and to his son Varaztirocc won over much of the Armenian military nobility, which had already helped Xusro II defeat the usurper Vahram Coben in 591. Hence, with the passage of time, the weakening of the Sasanian dynasty in the seventh century and the development of a deep-rooted Christian culture after the creation of the Armenian alphabet early in the fifth century ensured that the naxarars would preserve a considerable part of their prerogatives and that Persarmenia would not be absorbed into the Ira¬ nian world. The pressure of Byzantium on the lands it con¬ trolled as a result of the partition of 387 rarely led to explosions of violence, but it was perhaps more in¬ sidious and ultimately more damaging. Little is known of the precise functions of the comes Arme- niae, but the constant policy of Byzantium seems to have been the total integration of its Armenian ter¬ ritories into the empire, as had been the case for Ar¬ menia Minor. Before the end of the fifth century, the autonomous Satrapies that had backed an unsuccess¬ ful candidate for the imperial throne lost their right of hereditary succession, but the major transforma¬ tions were introduced by Justinian I. The entire east¬ ern border of the empire, from its main anchor point, Karin (Theodosiopolis), in the north to Dara, was heavily fortified, thus reinforcing the division of former Greater Armenia. More important, in 536 the autonomous Satrapies were abolished altogether; and all of Byzantine Ar¬ menia, including Armenia Minor, the Satrapies, and the newer territories east of the Euphrates, was transformed into four imperial provinces, all called Armenia and subject to Roman law and customs. The rights and privileges of the magnates were abol¬ ished, and the entire region was set on a rigid path of denationalization. Critical as was Justinian’s legislation for Arme¬ nian social and administrative history, the religious policy of the empire was equally damaging. Like most of the East, the Armenian clergy had come to reject the doctrine promulgated by the Council of Chalcedon (451), which it considered “Nestorian,” but as head of the church the Byzantine emperor could tolerate no dissent within his realm. As a re¬ sult, religious pressure to ensure the adhesion of the Armenian church to the faith of Constantinople reinforced Byzantine policy and aroused a growing antagonism in the native clergy and in the popula¬ tion. Some attempts at compromise were made by the Justinianic Council of “The Three Chapters” in 553, but the situation of Armenia took a sharp turn for the worse after 591. In return for the support received from the em¬ peror Maurice in the recovery of his throne, the Sas¬ anian king of kings Xusro II ceded to the Byzantine Empire a large portion of Persarmenia. The line of demarcation between the two empires moved mark¬ edly eastward—to the Azat River just west of the capital of Dwin in the north, and to the northeastern tip of Lake Van in the south. The new lands were again incorporated into imperial provinces, and the Byzantine policy of de-Armenization applied to them as well. Armenians had long staffed the impe¬ rial armies and rebellious magnates had been de¬ ported to Cyprus; now, massive transportations of Armenian troops and settlers to the Balkans helped secure the imperial Danube frontier but drained Ar¬ menia of both population and leadership. Even more serious, Maurice’s attempt to force Chalcedonianism on Armenia split the country and precipitated a twenty-year schism (591-611), during which rival katcotikoi at Dwin and Awan confronted each other over the Azat frontier. Simultaneously the national Armenian church broke with the northern church of Iberia, which returned to communion with Constan¬ tinople (607), leaving Armenian and Syriac Chris¬ tianity isolated in the midst of the imperial eccle¬ siastical world. With the beginning of the seventh century, the 477 ARMENIA:HISTORY OF ARMENIA:HISTORY OF situation of Armenia became even more difficult. After the brief Persian reconquest following Mau¬ rice’s murder in 602, the rapid disintegration of the Sasanian dynasty left the country increasingly in Byz¬ antine hands. The emperor Heraclius used Armenia repeatedly as a base for his retaliatory campaigns against Persia (622-629). Disregarding the tradi¬ tional Armenian pattern of hereditary offices, he by¬ passed the Mamikonean commanders in chief to confer imperial titles such as curopalate and patri¬ cian and even the supreme internal dignity of prince (isxan) on the lesser houses of the Saharuni and the Rstuni. Exploiting the considerable Chalcedonian ele¬ ments still remaining in Armenia, he continued the oppressive religious policy of Maurice. Hence, as a result of the break in the international balance of power between Byzantium and Persia following Heraclius’ victories, Armenia in the early seventh century seemed destined to be absorbed into the po¬ litical and cultural world of Byzantium. This seemed inevitable even though the antagonism of the clergy and the remaining magnates had been exacerbated by the imperial policies, and even though a national consciousness centered on the church in the absence of internal political leadership, as well as on social and cultural institutions, had come into being within the country. Before Byzantium had succeeded in its attempted integration of the entire Armenian plateau, the in¬ ternational equilibrium was unexpectedly broken again by the Arab invasions, which apparently reached Armenia by 640. Numerous problems, pri¬ marily of chronology, are still connected with the Arab conquest; the contemporary Armenian sources (Sebeos followed by Tewond the Priest) often dis¬ agree with the basic Arab historians (al-Baladhuri or al-Tabari), and modern scholarship has been equally divided. The first attacks seem to have been hit-and- run raids from neighboring Mesopotamia and Azer¬ baijan, although the capture of the capital of Dwin on 6 October 640 was unquestionably a severe blow. The seriousness of the incursions rapidly increased, so that by 652 the isxan Theodore Rstuni thought it wise to accept the favorable terms, recorded by Se¬ beos, that had been offered by the future caliph Mucawiya, then governor of Syria. A brief return of Byzantine domination occurred under Constans II, who again antagonized the Ar¬ menian clergy by his forcible imposition of Chalce¬ donian doctrine during the winter of 652-653, which he spent at Dwin. The Arabs under Habib ibn Maslama then began the systematic conquest of Transcaucasia. Most cities were captured, including Karin (Theodosiopolis), the capital of Byzantine Ar¬ menia; Dwin, with whose population a new treaty (preserved by al-Baladhurl) was concluded; as well as Tiflis in Iberia and Derbent in Azerbaijan, which commanded the northern passes of the Caucasus. In 655 the Arabs named a new isxan, Hamazasp Ma¬ mikonean, and by 661 the magnates of Armenia, as well as those of Iberia and Caucasian Albania, had recognized the suzerainty of the Umayyad caliph. Like many of the disaffected Christians of the East embittered by the imperial religious policy, the Ar¬ menian naxarars do not seem to have offered a pro¬ longed resistance to the Arab conquest. The contem¬ porary historian Sebeos complains that the princes “made an accord with death and an alliance with Hell.” The submission of Theodore Rstuni seems to have been voluntary, and even the traditional leaders of the pro-Byzantine party, Hamazasp Mamikonean (655-658) and his brother Gregory (662-685), ac¬ cepted the title of isxan from the conquerors. The church in the person of the katccdikos Nerses III (641-661) acquiesced. In fact, the early period of Arab domination seems to have been relatively favorable for Armenia. The conquerors needed Armenian contingents, es¬ pecially the famous cavalry explicitly noted in the treaties for the continuing war with Byzantium and its threatening Turkic allies, the Khazars, who were raiding southward through the passes of the Cauca¬ sus. Southern Armenia was tied into the developing system of fortified border zones (tbughur). Hence, in