(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "The Origin of the Naxarar System"

NICHOLAS ADONTZ 



ARMENIA 
IN THE PERIOD OF JUSTINIAN 

THE POLITICAL CONDITIONS 
BASED ON THE NAXARAR SYSTEM 

TRANSLATED WITH PARTIAL REVISIONS 
A BIBLIOGRAPHICAL NOTE 

AND 

APPENDICES 



BY 



Nina GL GARSOlAN 

COLUMBIA trarVBRSHTV 



Chapters 9-15 (pages 165-371), their Notes (pages 433-529), 
Appendices l-V (pages 1*-246*), and full Bibliography 
(pages 247*-3Q3*) ' 



CALOU3TB GTJLBENKIAN FOUNDATION 
LISBON 

WO 



J 



EDITOK'S PEEFACE 



Eor more than half a century since its publication in 1908, Nicholas 
Adontz's monumental thesis on Armenia in the Period of Justinian 
has proved to he both a landmark and a guidepost in the field of 
Armenian studies although its general inaccessibility, either from the 
rarity of procurable copies, or from linguistic difficulties, has made 
of it far too often a semi-legendary document rather than a useful 
tool. Perhaps as the result of this fortuitous isolation as well as of 
external circumstances, Adontz's first and probably greatest work 
did not lead to an immediate proliferation of studies along the lines 
that he had traced. He, himself, was to develop a number of them 
in later works such as his articles on the Armenian Primary History, 
Mesrop Mast'oe, Koriwn, P'awstos Buzand, and Movses Xorenaci; 
on the date of the Christianization of Armenia ; on the Iranian aspects 
of Armenian society ; and, as late as his postumously published History, 
on pre-Achaemenid Armenia 1 , But it is only relatively recently 
that the works of such distinguished contemporary armenologists 
as Gerard Garitte, Cyril Toumanoff, and the late Hakob Manandian 
have developed a number of problems in mediaeval Armenian history 
significantly beyond the point reached by Adontz at the turn of the 
century, and these scholars have not failed to acknowledge their 
indebtedness even where they have outstripped him s . Not even a 
Marxist presentation which of necessity challenged many of Adontz's 
premises and interpretations prevented A.G, SuMasian from admitting 
that " „. the admirable work of N, Adontz ... remains to this day one 
of the most authoritative works on Armenian feudalism" 3 . Such 
tributes are all the more impressive if we remember that they are 
addressed to the first major work of a young scholar composed at a 
time when a number of crucial studies on Late-Koman, Byzantine, 
and Iranian history as well as on the historical geography of eastern 
Anatolia were still to be written, 

The scope of Adontz's encyclopaedic work is not conveyed adequately 
by even a full quotation of his title, since, far from restricting himself 
to the reign of Justinian, or to an investigation of the nayarar system, 
he went on to scrutinize nearly every aspect of ancient and mediaeval 

1 A bibliography of Adontz's works can be found in the commemorative article in 
BA, LXI (May, 1947), pp. 313-318, and in A1PHO, IV (1936), pp. 991-993. 

2 JSr,gr,, Toumanoff, Studies, p. 108, See also helow n. 4. 

3 Sukiasian, Armenia, p. 36, Also Yuzbasyan's recent article in PBS (1962), 



XVI 



EDITOR'S PREFACE 



Armenia — geographical, political, religions, administrative, social, 
and intellectual — while giving simultaneously an extensive analysis 
of all the available sources. Perhaps the clearest index of the breadth 
of Adontz's information is the all too clear incompetence of a single 
individual to edit his work ; a team of specialists — historians, geo- 
graphers, archaeologists, philologists, anthropologists, and ethno- 
graphers — would have been necessary to do it justice. 

The value of Adontz's work for a new generation of scholars is not, 
however, limited to being a source of rare information to be exploited 
for reference; his methods and insights into the crucial problems of 
early Armenian history may yet prove more useful than even the 
enormous material accumulated by him. His application of critical 
scholarly methods to Armenian studies, and particularly his recognition 
of the dangers inherent in purely literary sources, have led to consid- 
erable work on the re-evaluation and re-dating of many Armenian 
historical documents, a task in which he continued to participate 
energetically, and which is by no means completed. His simulta- 
neous use of the techniques of varied disciplines while stressing the 
maintenance of the historian's rigorous chronological criterion, and 
his comparative method of juxtaposing the information of all relevant 
sources, Classical, Armenian, and Oriental, provided a workable 
blueprint for attacking the difficulties characterizing Armenian 
historiography. His ground breaking qualitative and quantitative 
analyses of Armenian social structure, reaching beyond superficial 
generalities, provided us with some of the first detailed information 
and with a framework for further research. 

Particularly illuminating is Adontz's constant refusal to be led 
astray by the conscious or implicit assumptions of his sources that 
ancient Armenia was a simple, undifferentiated, and unchanging 
entity, rather than the complicated aggregation of varied components 
whose geographic, political, and even religious particularism must 
be recognized even in periods of seeming unification, and whose 
characteristics and interests must be accounted for and balanced 
anew in each successive period. On numerous occasions Adontz's 
hypotheses have required development or rectification, but his basic 
conclusions repeatedly reached beyond the theses then current to 
what would prove to be the crux of a problem : beyond the familiar 
division of Armenia between the Graeco-Eoman and Iranian worlds 
to the paramount importance of the elaborate nexus of family traditions 



EDITOR'S PREFACE 



XYH 



and loyalities, " dynastic " as well as " feudal ", as shown in Ton- 
manoff's recent Studies ; beyond the double strain of Armenian Chris- 
tianity, Syriac as well as Hellenic, to the relationship of the ecclesi- 
astical hierarchy to the nayamr structure, and its influence on the 
pohtical evolution of the country, as I hope to demonstrate in a 
forthcoming work. Professor Garitte already observed the value of 
Adontz's inspired guesses when his own publication of the new Greek 
version of the Life of Si. Gregory repeatedly vindicated Adontz's 
hypothetical corrections of Marr's readings in the Arabic version 4 , 

It is self evident that a book written more than sixty years ago 
should now be superseded in a number of instances: Armenian 
archaeology was all but non-existent at the time, so that the Urartian 
aspects of Armenian history were perforce ignored, though Adontz 
himself rectified a considerable part of this lacuna in his Hisioire 
d'Amtenie ; new epigraphic material both in Armenia and in Iran has 
added significantly to our knowledge of both countries, and new 
editions of Iranian texts have altered a number of etymological 
derivations ; the Erwandian-Orontid dynasty identified by Manandian 5 
has altered radically our knowledge of the Hellenistic period; the 
lengthy survey of Diocletian's administrative reforms while perhaps 
still useful to Adontz's Russian contemporaries, now seems superfluous ; 
and a number of his conclusions as to the « feudal » nature of the 
Armenian naxarar system rest on antiquated interpretations of 
European feu-dalism. 

The entire book bears the marks of hasty publication, whether in 
the more superficial details of faulty proofreading, insufficient and 
often exasperatingly inadequate references, as well as the absence 
of the indispensable map, whose omission was regretted by the author, 
or in the far more fundamental aspects of occasionally confused, 
repetitive and contradictory organization, dubious etymologies, 
overstatements, and premature conclusions. The involutions of 
Adontz's style in a language not native to him add nothing to the 
clarity of the presentation. 

Yet Adontz himself anticipated much of the criticism which must 
attend a pioneer venture by disclaiming any pretension to a definitive 
study. " ... in publishing this work we are very far from any illusion 
as to its perfection, Armenian philology is still at a stage where the 

4 Garitte, AgatJwnge, pp. 351-353, 

5 See belmv Chapter XIV, n, 1. 






xvm 



EDITOR'S PREFACE 



presentation of any interpretation or theory as unchallengeable correct 
is out of the question. Students of Armenian antiquity can only 
grope their way toward many historical problems by way of more or 
less successful hypotheses; some of these may be corroborated at a 
later date, others will fall by the way, .,. . Our clarification of the 
nayarar system should bring a ray of light into the darkness which 
hangs oyer the Armenian past ... and should prove a starting point 
for a scholarly analysis of the extensive subsequent period of Armenian 
history ... " 6 , On these terms, the value of his work has di m inished 
but little in the intervening half-century, notwithstanding the necessary 
alterations, It remains a mine of information for the specialist, and 
a source of seminal ideas for those re-interpretations and further 
investigations the author had requested. As such it is a fitting 
reminder that in every generation it behoves dwarfs to take advantage 
of the shoulders of the giants who have preceded them. 



The instinct of every translator running the ominous gauntlet 
between the Charybdis of inaccuracy and the Scylla of unreadabihty 
is to open with his own apologia. This temptation is all the stronger 
in the case of Armenia in the Period of Justinian, since, as I have 
already indicated, Kussian was not Adontz's native language, Unlike 
Armenian, which has three steps in the demonstrative-relative system 
(hie, isie, ille), Kussian shares with most European languages a two 
step system. As a consequence of Adontz's shift from the one to 
the other, his writing abounds with cases of ambiguous antecedents, 
not all of which can readily be resolved from the context. His 
complicated and often awkward sentence structure is particularly 
foreign to English usage; the paragraphing is often erratic. Never- 
theless the text has been consistently respected, and alterations held 
down to a minimum even where some awkwardness ensued, Aside 
from the introduction of occasional elucidations such as " Xosrov II 
of Armenia " for " Xosrov ", the subdivision of unmanageable sen- 
tences, the clarification of antecedents, and the correction of minor 
misprints, no liberties have been taken with the original. 

The only significant difference between this edition and the Russian 
one lies in the realm of quotations from primary sources, Following 
the fashion of the day, Adontz often gave lengthy paraphrases rather 

6 Introduction pp. 6 and Chapter XV, p, 371. 



EDITOR'S PREFACE 



XIX 



than direct quotations. In several instances where this method 
seemed awkward or unnecessary, the original quotation has been 
ie~introduced, each case being duly recorded in the notes. To facilitate 
the reading, all extensive quotations in foreign languages have been 
shifted from the text to the notes and replaced by their English 
translations. Since so much of the value of Adontz's work lies in 
his vast collection of sources, many of which still remain extremely 
scarce even for the specialist, it has seemed useful to include in the 
notes the texts of a number of passages to which Adontz merely 
referred, all such additions being set off by square brackets. Further- 
more, a series of Appendices containing in extenso, or in their relevant 
portions, the main documents, Classical and Armenian, used by 
Adontz, has been added to this edition to allow the reader to draw 
his own conclusions from the material. 

In many instances the editions used by Adontz were either super- 
seded or, in the case of some Armenian documents, unobtainable; 
these have been replaced by more recent or accessible ones. All such 
substitutions have been noted in the Bibliography. Similarly, the 
English versions of Classical sources found in the Loeb Classical Library 
have been used wherever possible for the sake of convenience, but 
any significant differences between their translations and the ones 
given by Adontz have been recorded. Additional notes by the editor 
are indicated by letters as well as numbers eg. la, 

A full scale re-edition of Adontz's book to bring its manifold aspects 
in line with their modern scholarship would have entailed a major 
re- writing of the book, and would consequently lie well beyond the 
scope of this edition and the competence of its editor, Consequently 
it has seemed best to leave Adontz's text substantially as he composed 
it, adding only, wherever possible, some indication in the notes as 
to the agreement or disagreement of subsequent investigators, new 
material, need for rectification, or corroborative evidence. The new 
Bibliographical Note attempts to provide some, albeit cursory, indica- 
tion of the relevant works published since 1908. Finally, it is hoped 
that the Bibliography, which follows Adontz's lead in reaching beyond 
the limits of Justinianic Armenia to include a number of problems 
implicit or explicit in his text, will provide still more comparative 
material and criteria for a further re- evaluation of some of his conclu- 
sions. 

All those who have had the occasion to experience it will Teadily 



XX 



EDITOR'S PREFACE 



recognize the eternal nightmare of inconsistency in transliteration, 
especially in the case of proper names which haye reached us in multiple 
versions. In the kaleidoscopic world of eastern Asia Minor is a locality 
to be identified by its Classical, Armenian, Persian, Syriac, Arabic, 
or Turkish name? Which is the preferable transliteration system 
to be used for the name of an author writing both in Armenian and in 
Russian ? The most that this edition can hope to claim is an attempt 
to bring a little order into what can only be called Adontz's systematic 
inconsistency. Wherever possible, Armenian terms have been given 
according to the prevailing Hlibschmann-Meillet system, Arabic ones 
according to the spelling of the Encyclopedia of Islam, the Persian 
ones according to Christensen's IS Iran sous les Sassanides, 2nd edition 
(Copenhagen, 1944) with minor alterations, Russian ones according to 
the system of the U.S. Library of Congress, Georgian ones according 
to Toumanoff's Studies in Christian Caucasian History (Georgetown, 
1963), and Turkish toponyms according to the Office of Geography, 
.Department of the Interior, Gazetteer No, 46: Turkey (Washington, 
1960), For the sake of convenience, author's names have been given 
a single form, e,g, Manandian, irrespective of the alterations required 
by the diverse languages in which they wrote, the form selected being 
wherever possible the one more generally familiar, In all cases of 
ambiguity alternate versions have been given, Por Armenian topo- 
nyms, the Armenian form has generally been preferred for localities 
in Persarmenia, and the Classical (preferably Greek rather than Latin) 
for the western section of the country which was part of the Eastern 
Roman Empire, except in the case of familiar names where such a 
procedure would entail unwarranted pedantry, Por all the occasions 
on which these guide lines have failed, as they needs must, I can only 
appeal to the sympathetic indulgence of my colleagues. 

The precious geographical sections of the book carry their own 
particular series of problems. The map envisaged by Adonta was 
never published, and nearly every locality in eastern Anatolia has 
experienced at least one name change since 1908. Consequently 
Kiepert's and Lyneh's maps to which Adont^ normally refers are of 
but limited value to the modern reader, since no concordance of 
earlier and contemporary names exists to my knowledge. The 
identification of many ancient sites remains controversial in spite 
of the extensive investigations of Markwart, Honigmann, Eremyan, 
and many others. In Appendix Y some attempt has been made to 



&-■ 



EDITOR'S PEERAGE 



XXI 



coordinate the information on toponyms, giving where relevant and 
possible their ancient Classical and/or Armenian name, the modern 
equivalent, the coordinates given in the TJ.S, Office of Geography, 
Gazetteer No, 46 ', and a reference to the appropriate sheet of the USAF 
Aeronautical Approach Chart (St, Louis, 1956-1958) and the Turkish 
General Map, Where this has proved impossible, the available 
information will be found in the relevant notes. 

Finally, I should Eke to express my thanks to my friends and 
colleagues, professors Seeger Bonebakker, Associate Professor of 
Arabic Studies, Tibor Halasi-Ktm, Professor of Turkic Studies, Karl 
H. Menges, Professor of Altaic Philology, and Ehsan Yar-Shater, 
Hagop Kevorkian Professor of Iranian Studies, all of Columbia Uni~ 
versity, as well as professors Gerard E, Caspary, Associate Professor 
of Mediaeval History at Smith College, Wendell S, Johnson, Associate 
Professor of English Literature at the University of the City of New 
York, and Norma A, Phillips, Assistant Professor of English Literature 
at Queens College of the City of New York, for their help and patience 
on the many occasions when I was forced to turn to them for assistance, 
I am most grateful to Professor Emeritus Sirarpie der Nersessian of 
the Dumbarton Oaks Center for Byzantine Studies, both for her 
suggestion that I undertake this edition and for the help and encou- 
ragement she has so often given me, To my constant advantage, 
I have also benefited from the vast knowledge and inexhaustible 
kindness of Monsieur Haig Berberian of the Revue des Etudes Arrn§- 
niennes. Finally, my thanks are also due to Dr, Robert Hewsen for 
his help with questions of Armenian geography, and to my students 
Dr, Linda Eose, Messers, Krikor Maksoudian and Jack Yartoogian 
for the endless hours they spent in the thankless tasks of verifying 
references, hunting out copies of rare works, and proofreading, Eor 
the many flaws which such an edition must perforce still contain, the 
responsibility remains of course mine alone. 



Nina G, Gaesoiast, 



New York, July 3, 1967, 



ABKEYIATIONS 



AASS Ada Sanctorum Bollandiana (Brussels). 

AAWB Ablmndlungen der Ahademie der Wissenschaflen zu Berlin, 

AB Analecta Bollandiana (Brussels). 

ABAWM Ablmndlungen der bayerischen Ahademie der Wissenschaflen zu Munchen, 

AGO Ada Conciliorum Oecumenicorum, Schwartz, E, ed, (Berlin, 1914). 

AEHE Annuaire de VScoh des Hautes Etudes (Paris). 

AIPHO Annuaire de VInstiiiit de philologie el d'histoire orientales et slaves (Brussels). 

AJSLL American Journal of Semitic Languages and Literatures (Chicago). 

AKGWG Ablmndlungen der honiglischen Gesellschaft der Wissenschaflen zu Gbttingen. 

AO Ada Orientalia (Copenhagen). 

AQ Armenian Quarterly (New York). 

AE.BBL Academie Boyale de Belgique, Bulletin Classe des Lettres (Brussels). 

ASGW Ablmndlungen der sachsischen Gesellschaft der Wissenschaflen, 

B Byzantion (Brussels), 

BA Bulletin armenologique, Melanges de VUniversite de Saint-Joseph (Beirut). 

Ber Berytus (Beirut). 

BGA Bibliotheca geographorum arabicorum, de Goeje, M.J, ed, (Leiden), 

BIM Bulletin de VlnstUui Marr (Tbilisi), 

BK Bedi Karthlisa* Revue de Karthvelologie (Paris). 

BM Banber Matenadarani (Erevan). 

BNJ Byzaniinisch-neugriechische Jahrbucher (Berlin). 

BSL Bulletin de la Socieii Linguistigue de Paris, 

BSOAS Bulletin of the School of Oriental and African Studies (London). 

BZ Byzantinische Zeitschrift (Leipzig), 

Ca Caucasica (Leipzig). 

CAH Cambridge Ancient History, 

CHA Collection d'historiens armeniens, Brosset, M.P. ed. (St, Petersburg, 1874- 

1876), 

CHAMA Collection d'Mstoriens anciens et modernes de VArminie, Langlois, Y. ed, 

(Paris, 1967-1869), 

CHK The Catholic Historical Meview (Washington), 

CIG Corpus Inscriptionum Graecorum. 

CIL Corpus Inscriptionum Latinomm. 

CJC Corpus Juris Civilis, Mommsen, T,, Kxuger, P., et al., edd, (Berlin). 

CMH Cambridge Medieval History, 

Cod. Th. Codex Theodosianus, Mommsen, T., et al,, edd. (Berlin). 

CP Classical Philology (Chicago). 

CR Classical Beview (London- Oxford), 

CSCO Corpus Scriptorum Christianorum Orientalium (Louvain), 

CSHB Corpus Scriptorum Historiae Byzantinae (Bonn, 1828-1897). 

DHG Didionnaire d^Histoire et de GiograpMe JScclesiastique (Paris). 



ABBREVIATIONS 



XXIII 



DTC Dictionnaire de TMologie Oaiholique (Paris). 

EHR English Historical Beview (London). 

EI- Encyclopaedia of Islam (Leiden, 1913-1948). New edition (1 954-), 

EO Echos d'Orient (Paris). 

EGH Fragmenta Historicorum Graecorum, Muller, C. ed. (Paris, 1841-1883). 

G Georgica (London). 

G46 Office of Geography, Department of the Interior, Gazetteer No. 46 : Turkey 
(Washington, 1960). 

GGM Geographi Graeci Minores, Mutter, C. ed. (Paris, 1855-1861), 

HA Handes Amsorya (Vienna). 

IAFAN Izvestia Armianshogo Piliala Akademii Nauk SSSB (Erevan). 

IAN A Izvestiia AJcademii Nauk Armianskol SSB (Erevan). 

IANS Izvestiia Akademii Nauk SSSB (Moscow), 

IKIAI Izvestiia Kavkazskogo Istoriko-Arkheologicheskogo Instituta (Tbilisi). 

IZ Istoricheskie Zapiski (Moscow). 

JA Journal Asiatique (Paris). 

JEH The Journal of Ecclesiastical History (London). 

JHS Journal of Hellenic Studies (London), 

JKAS Journal of the Royal Asiatic Society of Great Britain (London), 

JR.GS Journal of the Boyal Geographic Society (London). 

JKS Journal of Boman Studies (London). 

K. Klio. Beitrdge zur alten Geschichie (Leipzig), 

KSINA Kratkie Soobshcheniie Instituta Narodov Azii Akademii Nauk SSSB (Mos- 
cow). 

KV KhristianskU Vostoh 

L Loeb Classical Library (Cambridge, Mass.-London). 

LTK Lexikon fur Theologie und Kirche (Freiburg i/B), 

Mansi Sacrorum Conciliorum Nova et Amplissima Gollectio, Migne, J.B. ed. (Floren- 
ce - Venice, 1759-1798). New edition (Paris, 1901). 

MAIP Memoires de VAcademie Impiriah des Sciences de St, Piiersbourg. 

MBAK Monatsberichte der berlinischen Akademie der Wissenschaften, 

MDGKO Morgenlandische Darstellung aus Geschichie und Kultur des Ostens (Berlin). 

MVG Mitteilungen der vorderasiaiischen Gesellschaft. 

NT Nord Tidsskrifi for Sprogviden (Oslo), 

OC Oriens Chrisiianus (Leipzig). 

OS Orientalia Suecana (Uppsala). 

P Pazmaveb (Venice). 

PBA Proceedings of the British Academy (London). 

PBH Patma-banasirakan Handes (Erevan). 

PG Patrologiae cursus completus. Series graeco-Iatinu, Migne, J.P. ed. (Paris, 
1857-1866). 

PL Patrologiae cursus completus. Series laiina, Migne, J.P. ed. (Paris, 1844- 
1855). 

PO Patrologia Orientalise Graff in, K, and Nau, !F, edd. (Paris, 1903). 

PP La Parola del Passaio, Bivista di Studi Classid (Naples). 

PS Palestinskil Sbornik (Moscow). 

PW BeaVencyclopadie der classischen Aliertumsurissenschaft, Pauly, A., Wisso- 



XXIV 



ABREVIATIONS 



wa, G., and Kroll, W. edd, (Vienna, 1837-1852). New edition (Stuttgart, 

1893). 
REA Bevue des Etudes Arminiennes (Paris, 1920-1932). New series (Paris, 

1964-)- 
REAnc Bevue des Etudes Anciennes (Bordeaux). 
REB Bevue des Etudes Byzantines (Paris), 

REIE Bevue des JJJtudes Indo-JSuropeennes. 

RH Bevue Historique (Paris). 

RHE Bevue d'JBistoire Ecclesiastique (Lou vain). 

RHR Bevue de VHisioire des Beligions (Paris), 

ROC Bevue de VOrient Chretien (Paris), 

RSJB Becueils de la Societi Jean Bodin (Paris). 

S Syria (Paris). 

SAW Sitzungsberichie der philologisch-historische Classe der kaiserlichen Akademie 

der Wissenschaften (Vienna). 
SBAWM Sitzungsberichte der bayerischen Ahademie der Wissenschaften zu Munchen, 
SI A Studia Instituti Anthropos (Vienna). 

SMM SaFarfvelos Muzeume Moambe (Tbilisi), 

SV Sovetshoe Vostokovedenie (Moscow). 

T Traditio (New York), 

USAEM USAF Aeronautical Approach Chart (St. Louis, 1956-1958). 
UZL Uchennye Zapiski Leningradskogo Universiteta. 

VBAG Verhandlungen der berlinischen anthropohgischen Qesellschafi* 

VDI Vestnik Drevnel Istorii (Moscow). 

VI Voprosy Istorii (Moscow). 

VIA Voprosy lazykoznaniia (Moscow), 

W Vizantilskil Vremmenik (St. Petersburg, 1894-1928). N.S, (Leningrad, 

1947). 
WZKM Wiener Zeitschrift filr die Kunde des Morgenlandes, 
ZDMG Zeitschrift der deutschen morgenlandischen Gesellschaft (Leipzig). 

ZE Zeitschrift fur Mthnologie, 

ZKO Zapishi Klassicheskago Otdeleniia Imperatorskago Busskago Arkheologi- 

cheskago Obshchestva (St. Petersburg). 
ZMNP Zhurnal Ministerstva Narodnago Prosveshcheniia (St. Petersburg). 

ZNW Zeitschrift fur neutestamentliche Wissenschaft, 

ZVO Zapiski Vostochnago Otdeleniia Imperatorskago Busskago Arkheologicheskago 

Obshchestva (St. Petersburg). 
ZVS Zeitschrift fur vergleichende Sprachforschung. 



E AS TEEN ARMENIA 



IX 



ARMENIA — THE MARZPANATE 



Eastern or Pers-Armenia as the homeland of the nayavo/rs — The naxarar system 
as a factor in Armenian history — The position of Eastern Armenia in the Persian 
Empire — The administrative system of the Sasanian monarchy: the Tmslahs and the 
sahrs — Eastern Armenia as one of the sahrs of the Caucasian kustak — Eastern 
Armenia as a Marzpanate — Her relations with the neighbouring lands and the boun- 
daries of Marzpan- Armenia — Comparison of the territory of Marzpan and Arsacid 
Armenia — The appearance of the border districts — The division of Marzpan- 
Armenia according to the Armenian Geography — The origin of Siwnik'j Tayk% and 
Mokk' as given in the Armenian Geography — The central provinces; the Tanuier 
lands and Vaspurakan until 591 — The Iranian gunds and the Byzantine themes — 
The origin of the provincial divisions adopted by the Armenian Geography. 



The Eastern or Persian portion of Armenia differed markedly from 
the Western as a result of its position and importance in the history 
of the Armenians, It played the dominant role in the development 
and preservation of the innermost foundations and principles which 
sustained the historical life of Armenia. All of Armenia had been 
drawn from antiquity into the political world of the Persian states 
and had shared for centuries in the cultural sphere of Iran, From the 
time that the Arsacids seized Persia while their younger branch 
consolidated itself in Armenia, the traditional relations of the two 
countries had drawn even closer. It is in this period of Arsacid 
domination that the nayp/rm system, the socio-political pattern emi- 
nently characteristic of -Ancient Armenia, developed. It is true that 
the naxcirar system evolved from the core of pre- Armenian society, 
and that its root lie deep in the ethnic and geographical setting peculiar 
to the country, Nevertheless, the final shaping of its peculiar cha- 
racter unquestionably took place in the Arsacid period. 

The naxarar system is such a characteristic and such an amazingly 
stable institution that no serious understanding or interpretation of 
Ancient Armenian life and history is possible without it. Although 



166 



CHAPTER IX 



modified in some of its aspects, this system survived in Armenia 
until the fall of the Bagratids, and its final destruction came only 
with the Mongol invasions a . During the whole of this lengthy period, 
the nayarar system was an important factor and one might even say 
the moving force in Armenian history. Scattered throughout the 
country, it provided the cradles of political liberty. Its significance 
is derived from the fact that it provided the means for reconciling 
external subjection with internal independence, and thus for the 
preservation of the individual character of the country. 

It is well known how often religion is mentioned as the outstanding 
factor in the history of Armenia. Some scholars have even been 
willing to reconstruct the entire historical life of Armenia on this 
basis. This approach, inherited from our ancestors, is one of the most 
hackneyed ones in Armenian historiography, and it originated in the 
period following the disappearance of the nayarar pattern in the 
country. It is correct insofar as it reflects the situation of a later 
period; it is incorrect when archaized and applied to earlier times 
as well. As long as the nayarar system functioned in Armenia, 
the Church was important only insofar as it adapted itself to the 
nayarar pattern. The bitter struggle for the nationalization of the 
Church in Armenia was in reality for its nayarariz&tion, for the transfer 
of nayarar customs into the ecclesiastical sphere. The nayarar 
pattern was so deeply ingrained as a mode of life, and so reinforced 
by the complicated setting in which Armenia found herself, that 
any new system was acceptable only on the inflexible terms that it 
be compatible with the existing nayarar structure. Wherever the 
Church was successful in accomplishing this, it became nayarar- 
national, but in the parts of the country where, under the influence 
of Imperial policy, it failed or did not see the necessity of adapting 
to local forms, the Armenian Church remained a part of the 4 common 
ecclesiastical structure. This political framework for ecclesiastical 
events is a fundamental factor in the isolation of the Armenian Church 
from the Catholic Church, regardless of dogmatic principles or dis- 
agreements. In the nayamr period, the importance of the Church 
must be measured not by its Christian content, as many have thought, 
or very little by it, but rather by its feudal or nayarar structure. 

Given the importance of the nayarar institution, it is evident that 
a detailed knowledge of it is indispensable for an interpretation of 
the vast period of Armenian history which precedes the Mongol 



ARMENIA — THE MARZPANATE 



167 



invasions. This institution, which was born under the Arsacids, 
continued to develop throughout their possessions even after theix 
downfall. Primarily from the time of Justinian, Western Armenia 
began to lose her na^arar aspect, but this pattern persisted for a long 
time in the eastern part of the country. The information about 
naxarar principalities preserved in Armenian historical works refer 
almost exclusively to Eastern Armenia, hence the main focus of a 
study of the constant features found in this part of the country must 
concentrate on the nayarar structure, Oriental Armenia is usually 
called Persarmenia by Byzantine historians. Together with the 
provinces of Iberia, Albania, and Atrpatakan, it formed one of the 
large administrative units of the Sasanian Empire. Under the 
Sasanians, the Persian realm was divided into four parts: east, west, 
south, and north, the Fu^t i Xordsun, Xorwaran, NemroJ, Apaytar x . 
The northern quarter was more often called hust~ah i Kapkoh, the 
name used by the Armenians, or, after its central province, kusi-ak i 
Atrpatakan, As for the name Apa^tar, it was used for the Scythian 
or Sarmatian tribes living further to the north who made their presence 
known through frequent raids on Persian and Eoman possessions. 
Some scholars attribute this division to X usro [Khusro] I Anosarvan 
(531-579). They believe that up to his time Persia was split into 
many provinces whose satraps or governors were directly subject 
to the crown ; but because of the difficulties in supervision entailed 
by this multitude of subordinates, 

[Xusro] conceived the plan of forming four great govern- 
ments, and entrusting them to four persons in whom he had 
confidence, whose duty it should be to watch the conduct 
of the provincial satraps to control them, direct them, or report 
their misconduct to the crown 2 , 



This opinion is not altogether correct, Xasro I can hardly be 
the initiator of the administrative division. The Arab historian 
Tabari, who had valuable Pehlevi material at his disposal, and who 
had, incidentally, made use of the Khudhay Nameh, the prototype 
of the famous Sahnameh, relates that when Xusro Anosarvan 
mounted the throne, " he sent letters to the four padhghospan who 
ruled the four regions of Persia " ; one of which was addressed to 
" Zadhoe Nakhveraghan, padhghospan of Aderbaygan, Armenia and 
the neighbouring lands " 3 . These four padhghospan were the heads 



168 



CHAPTER IS 



of the four hustahs, and Zadhoe was the governor of the hustak of 
Atrpatakan, or of the Caucasian border. Judging from this infor- 
mation, the divisions of Persia must have existed before Zusro 
Anosarvan, and cannot be attributed to him, 
The same historian relates that, 



Before the accession of Zusro the office of spahbadh that 
is to say of commander in chief was held by a single man for 
the whole empire. .Zusro, however, as soon as he became 
king divided this office among four persons, the spahbadh 
of the Orient, i.e. of Khorrasan and the neighbouring lands, 
the spahbadh of the Occident, the spahbadh of Nimruz, i.e. of 
Yemen, and the spahbadh of Aderbayjan and the neighbouring 
lands, i.e. of the country of the Khazars 4 . 

The only thing attributed to Xusr5, therefore, is the placing of a 
military commander (spahbadh) in the four regions or Jcustaks, side 
by side with the pddhghospdn. The reform consisted in the separation 
of military and civilian power, the pddhghospdn serving as the instru- 
ment of civilian authority, and the sjpahbadh of the military 5 , 

According to the Armenian sources, the division of the army into 
four groups dates not from the period of Zusro but from that of his 
grandson Yazdgard, the last of the Sasanians [sic] 5a . Under him, 
says Sebeos, 

... the Persian armies were divided into three parts of which 
one corps was toward Persia and the Orient, and another 
corps of Xoream toward Asorestan, and one corps in the 
province of Atrpatakan, But the seat of his kingdom was 
in Tizbon [Ctesiphon] and all together honoured him in common 
agreement 6 , 

Persia (Fars), the main province of the southern region, is used here 
instead of the entire JcusiaJc as pars pro ioto. In reality, therefore, 
the Armenian historian also distinguishes four armies, evidently 
having the four spdhbadhs in mind. The Armenian writer sees in 
their creation merely the weakening of the royal power under Yazd- 
gard, while, according to the information of the Arab author cited 
above, this was an important reform inaugurated by Xusro I Ano- 
sarvan, the grandfather of Yazdgard. Neither version is apparently 
correct, and the four spdhbadhs as well as the four pddhghospdns 
existed in Persia long before either Xusro I or Yazdgard 6a . 



s~ 



ARMENIA — THE MAMZPANATM 



169 



We know that according to ancient Iranian cosmology, the entire 
heavenly sphere was also divided among four spahbadhs: the star 
Tistar watched over the East, Sataves, over the West, Vanand, over 
the South, and Haptoring, over the North 7 , It is true that the book 
of BundaMsn, if om which we obtain this information, has reached 
ns in a version not older than the end of the Sasanians, but its origin 
goes back to more ancient times, There is no doubt in this case 
that these cosmological concepts of the Persians were a direct reflection 
of the administrative divisions of Persia, of its division into four 
commands. 

The offices of padhghospan and spahbadh may in effect be compared 
to those of praefectus pvmtwio and magister miliium in the Roman 
Empire, Originally the praetorian prefects also held both full civilian 
and military powers, and it was only in the fourth century that the 
military command was put under a separate authority. We have 
seen that four prefects, standing at the heads of the four prefectures 
into which the Empire was divided were known to the Romans. 
Two of these belonged to the eastern part of the Empire, and two ' 
to the western, There were as many military commanders [magistri 
milihim]: of the Orient, Illyria, Thrace, and Gaul, if we do not count 
the four court commanders [praesentales], two to each court in the two 
capitals. Such a similarity between the divisions of the Roman and 
Persian Empires can hardly be fortuitous; there is evidently some 
connexion, Unfortunately, the problem of the genetic interrelations 
of political institutions in the two neighbouring realms has not been 
touched by scholars, who evidently cannot conceive that proud Rome 
borrowed anything from a barbarian state 7a , Perhaps this is true, 
and if so, the borrowing was on the Persian side. In any case, there 
is no doubt that the Persian Jcmis or Icustaks are as ancient as the Roman 
prefectures. 

The prefectures were divided into dioceses, which in turn were 
subdivided into provinces. The Persian JcmtaJcs, however, were 
directly split into smaller units, according to the historical development 
of territorial or ethnic conditions. Each of these units was a more 
or less independent country, a SaJir with a historical past. According 
to the Armenian Geography, the Caucasian hust consisted of thirteen 
such countries or sahrs; 



170 



CHAPTER IX 



1. Atrpatakan 

2. Arum i.e. Armenia 

3. Varjan i.e. Iberia 

4. Ran i.e. Alovania 

5. Balasakan 

6. Sisakan 

7. Are 



8. Gelan 
[9. Sancan] 

9. Dlmunk* 

10. Dmbawand 

11. Taprestan 

12. Rwan 

13. Amis. 



In the shorter version of the same work, only ten countries aTe 
indicated, only Atrpatakan being given in the first column, while 
the entire second column is listed with the following alterations, 
Mukan is given instead of Sancan, and Amadan, which is lacking in 
the first version but undoubtedly belonged here, has been added. 
It is curious that Arab sources also occasionally leave the Caucasian 
countries out of their description of the Northern region of the Sasanian 
Empire 9 , 

One of the early and well informed Arabian geographers confirms 
that, 

Djarbi or the countries of the North form the fourth part 
of the Persian Empire under the rule of a spahbadh who is 
known under the name of Adarbayjan-spahbadh, this quarter 
includes in itself Armenia, Adarbayjan, Rey, ... Alania and 
others 10 . 

The same author, speaking of the titles of the kings, says that Ardaslr 
honoured the following kings with the title of sah: " Buzurg-Armenan 
sah, Adarbayjan sah, Alan sah, who is in Mukan, Balasakan sah, and 
Sisajan sah" 11 , There is no need to take these words literally; 
their meaning is that the countries named enjoyed the rule of their 
own kings, a situation associated here with the name of the founder 
of the Sasanian dynasty, but which prevailed in general throughout 
the period of its rule. Perhaps the relative independence of these 
countries explains their omission from the list of the provinces in the 
Caucasian Jcustah which we have already noted. Royal power did 
in fact exist in Buzurg- Armenia, i.e. Greater Armenia, and in Arran- 
Albania; but the Arab author also lists as kingdoms, Sisajan and 
Balasakan, provinces which in reality belonged to Greater Armenia. 
What we obviously find reflected here are the conditions existing 
between the sixth and the eight centuries, when these provinces 
were separated from Armenia. As for the absence of Iberia, it is 
to be explained by the fact that the author included it in Armenia, 
as was the custom among Muslim writers, 



ARMENIA — THE MAJRZPANATM 



171 



In the last chapter of the Chronicle of Zaeharias of Mitylene, the 
following passages is added to a brief survey of countries according 
to Ptolemy, 

And besides these there are also in this northern region five 
believing peoples, and their bishops are twenty-four, and their 
Catholic fives at D'v/in, the chief city of Persian Armenia. 
The name of their Catholic was Gregory, a righteous and a 
distinguished man. 

Further Gurzan, a country in Armenia, and its language is 
like Greek; and they have a Christian prince, who is subject 
to the king of Persia, 

Further the country of Arran in the country of Armenia, 
with a language of its own, a believing and baptized people; 
and it has a prince subject to the king of Persia, 
Further the land of Sisagan, with a language of its own, a 
believing people, and there are also heathens living in it, 

The country of Bazgun, with a language of its own, which 
adjoins and extends to the Caspian Gates and the sea, the 
Gates in the land of the Huns. And beyond the Gates are 
Bulgarians ... 12 . 



At first glance, it might seem as though there were five christian 
peoples in addition to the five countries listed further on, but, in 
reality, these christian peoples are the Armenians, Gurzan, i.e. Iberia, 
Arran, i.e. Albania, Sisakan [Siwnik'], and Bazgun, The last name 
is usually identified with Abasgia. But Abasgia, the present Abkhazia, 
lies, as is well known, on the shore of the Black Sea, while Bazgun 
according to its description adjoins the Caspian Sea and the Caspian 
Gates. There are no grounds for taking Abasgia in a broad political 
sense to include all the territories to the Caspian Sea, especially since 
this broader sense was given to Lazika on the Black Sea during the 
sixth century, to which this description belongs. The rise of Abasgia 
on the political scene was to come at a slightly later date. We believe, 
therefore, that Bazgun is a deformation of Barasakan or Balasakan 
[bazgun for ba(ra)zgan), a province closer to the countries listed than 
is Abasgia. Balasakan lay in Atrpatakan between the cities of 
Berzend and Vardana-kert, Berzend still exists at present and stands 
on one of the tributaries of the Bolgara cay on the Eusso-Persian 
frontier ; from this the position of Balasakan on the lower Araxes can 
be determined. Vardan, the Vardanakert of Armenian writers, 
whose position is not exactly known, is next to it, and is given in the 



172 



CHAPTER IX 



list of district in the province of Paytakaran, and obviously Balasakan 
was also numbered among them 13 , Balasakan is also found in 
Koriwn together with SiwniF, Armenia, Albania, and\Iberia; these 
are the countries covered by Mesrop in trie course of Ms evangelizing 
activity 13a , 

According to one indication, the separation of Sisakan from Armenia, 
occurred in the period of the Armenian rebellion and of the murder 
of the marzpan, the Suren in the year 571, 

,,, a little before this rebellion the isocan of Siwnik', named 
Vahan, separated himself from the Armenians and asked 
Zosrov, king of Persia, to shift the diwan of Siwnik' from 
Dwin the capital of Armenia to the city of P'aytakaran and 
to put this city in the sahrmar of Atrpatakan so that in the 
future the name of Armenians should no longer be given to 
them 13 *> 

The country of Sisakan was again reunited with Armenia after the 
fall of the Sasanians 14 , In the description of Zacharias of Mitylene, 
Sisakan is already given as a country separate from Armenia and on a 
par with Albania and Iberia in 555, Should we, therefore, suppose 
that it was already separated from Armenia at that time ? Because 
of geographic and ethnic circumstances, Sisakan stood a little apart 
from Armenia, and this division may sometimes have given the 
impression of a completely autonomous country, Procopius too 
thought that the Suniton i.e. Siwnites were a nation separate from 
the Persarmenians 15 . 

Sisakan, together with Paytakaran, and, therefore, with Balasakan 
belonged, according to Sebeos, to the sahrmar of Atrpatakan, We 
would expect Sisakan and Balasakan also to have had their own 
sahrmar, as independent units of the Caucasian kustak, but apparently 
they were not considered the equals of Armenia and Atrpatakan from 
an administrative point of view, Sisakan belonged to the sahrmar 
of Armenia, and Balasakan to that of Atrpatakan, Similarly, they ' 
did not have marzjpans but were ruled by their own princes, The 
Armenian provinces of Aljnik* and Nosirakan held approximately 
the same position in the western Knstah The first of these apparently 
belonged to the sahrmar of Arabastan, and the second to that of 
Adiabene or Nohadra, according to the Armenian Geography, bui 
they enjoyed a certain independence since they were still ruled by 




ARMENIA — THE MABZPANATE 



173 



local titular princes called bdesxs. In the Geography both provinces 
are listed in the western region as independent sahrs similar to Sisakan 
and Balasakan 16 , 

Thus, the administrative units equivalent to Armenia seem to have 
been Iberia, Albania, Atrpatakan Adiabene and Arabastan. With 
the fall of the Sasanians the territorial relations of Armenia to these 
countries were significantly altered, We must therefore determine 
what these alterations were, in other words, how the Armenia of the 
royal period differed from the Armenia of the Marzjpanaie, from a 
geographical point of view. 

We speak of Marzpan Armenia since after the Arsaeids a represent 
tative of the Persian king called marzpan had his seat in the country 17 , 
His capital was the city of Dwin, which had the same significance for 
Oriental Armenia as did Theodosiopolis, the residence of the Count 
of Western Armenia, which served at the same time as an international 
center for Asia 18 , In the north, Marzpan Armenia stretched to Kan- 
gark\ According to the narrative of Lazar P'arpeci, king Va^tfang of 
Iberia sent a messenger to Armenia saying 

... a powerful Persian detachment has reached the land of 
Iberia, and I, seeing that I was unable to withstand them, 
have put my hope in the Armenian mountains, near the bor- 
der of Iberia, and I await you. 



Vahan Mamikonean set out to Va^t'ang's assistance and " the Ar- 
menian army came to the Iberian king and camped in a locality in the - 
district of KangarF ", where Va^t'ang assured the Armenians that 
the Persians had returned home 19 . From all this we see that the 
Armenian mountains bordering Iberia, in which Ya^t'ang was hiding, 
were near or even in Kangar. Moreover, since Kangar was considered 
to be in the southern corner of the province of Gugark', the latter 
must have stretched beyond the limits of Armenia, Further south, 
the province of Tayk' bordered on Iberia, and its northenmost district 
was named Kol, Tayk', itself, remained entirely within Armenia 20 , 
On the Albanian side, the Kura had ceased to be the frontier, 
as it had been under the Arsaeids, The frontier line shifted from the 
river to the city of Xal^al, which tazar P'arpeci places in Albania 
and Etise calls the winter residence of the Albanian kings, Xal^al 
stood on the right bank of the Rura, since the Persian army on its 
way from Albania to Xal^al ^as forced to cross the river. After 



174 



CHAPTER IX 



the battle of Xal^al, the Armenian cavalry was conveyed across 
the river, to move toward the Albanian gates 20 a . It is entirely possible 
that the modern Tartar village of Xil^il on the Zegam river is indeed 
the ancient ^Tal^al, The latter lay in Uti, and since it is called an 
Albanian city by Lazar P'arpeci, we must conclude that Uti was 
considered to be part of Albania at that time 21 . Faustus of Byzan- 
tium knows of Haband, with the village of Amaraz where the Albanian 
Kat'olikos Gregory, the grandson of St. Gregory the Illuminator, 
had been buried, as of a border province between Armenia and Albania, 
The position of Haband is not precisely known, The Armenian 
Geography lists two Habands, one in Siwnik' and the other in Arca^, 
Movses Xorenaei, in transmitting Faustus' story replaces Haband 
by Lesser Siwnik' 2a , by which he means the whole of Arca^, or perhaps 
— which is less likely — one of its districts which bears in the Armenian 
Geography the Persian name of Sisahan i kotak, Lesser Sisakan, 
Haband of Arca^; must have adjoined Siwnik', forming part of Siwnian 
Haband and belonging to Albania together with all the districts of 
Arca^ 22a . 

Armenia was separated from the territory of Atrpatakan by the 
province of Her and Zarewand on the northern shore of Lake TJrmiah, 
between Salamas and Xoy. The Persians carried on operations 
against the Armenians from this impregnable base at the time of the 
rebellion of the fifth century. Having heard of the arrival of the 
Persian host in Her and Zarewand, Vardan Mamikonean hurried 
to meet it, thinking that, 

.,, matters might be settled there, in Her and Zarewand, 
and no one would be left to hinder the Persian army from 
reaching the land of Armenia to strike and harm it with murder 
and captivity. 



Similarly, when, Vahan Mamikonean, received the news that the 
Persian army had entered Her and Zarewand, he gathered his forces 
and set out to strike against them in those parts " .., so as not to 
allow the Persian army to move into Armenia ". When the Persians 
thereupon decided to settle peacefully with Vahan, and sent the 
ambassador Ni^or with that purpose " .,, he came to Armenia and 
could not make up his mind to penetrate into the Armenian territory, 
but remained in the province called Her " 23 , The region of Her and 
Zarewand, with which we are concerned, later formed, together with 



ARMENIA — THE MARZPANATM 



175 



the adjoining distincts, the province called Parska-hayk', that is to 
say Persarmenia, in the Armenian Geography, and the examples ]ust 
cited demonstrate that this region did not enter into Marzpan Armenia. 

In the south, Armenia ended with the provinces of Albak, Anjewaeik' 
and Mokk £ , The rebellions of the fifth century against the Persians 
arose exclusively within the limits of Marzpan Armenia, and these 
provinces seem to be the southernmost ones whose representatives 
took part in the rebellion, Albak occupied the upper course of the 
river Zab, and the neighbourhood of Basjkale still bears this name. 
West of it lay the district of AnjewaeiF, at the source of the Bohtan-su, 
in it stood the settlement of Kangowar and the monastery of Hogeae- 
vank\ Still further west, beyond AnjewaciF, lay Mokk', the present 
Moks or Mokus, also in the valley of the Bohtan-su, Beyond Mokk £ , 
stretched Aljnik', which already belonged to the Western quarter, 
to the Jcustah of Xorwaran and was consequently excluded from 
Marzpan Armenia, though it adjoined it at Balaleison (Armenian 
Bales) the present Bitlis 24 , 

Within these limits, Marzpan Armenia was noticeably smaller than 
the Armenia of the royal period, A strip of royal border provinces 
remained outside the later frontier, namely: 1.) Gugark', 2.) TJti, 
3.) Arca^, 4.) P'aytakaran-Balasakan, 5.) Parskahayk'-Persarmenia, 
6,) Korcek', and 7,) AljniF. The Armenian Geography includes these 
provinces in Armenia but notes that Gugark' had been taken from 
the Armenians by the Iberians, Uti and Arca^, by Albania, and 
P'aytakaran by Atrpatakan [Atropatene], It is not difficult to guess 
that the same fate befell Persarmenia and Korcek' 24a . The separation 
of these lands from Armenia dates from the period of the downfall 
of Arsak II, in the second half of the fourth century. According to 
the testimony of Faustus of Byzantium, the following lands fell away 
from Armenia at that time: L)the bde§x of Aljnik', 2.) the bde§x of 
Nosirakan, Mahkert-tun, Ni^orakan and Dasn, 3,) the bdesx of Gugark', 
4.) the prince of Gardman, 5.) the province of Arca^, 6.) the land of 
Korduk'-Tmorik'-Korde, 7,) the domains of the Armenian kings 
adjoining Atrpatakan, 8.) the land of the Mar, 9.) the land of Kaspe 241 \ 
Under Pap, the son and successor of Arsak II, an attempt was made 
to win back the lost territories, and the following were recovered, 
according to Faustus 25 : 1.) the possessions of the Armenian king in 
Atrpatakan, 2,) the country of Nosirakan, 3,) the province of Korduk'- 
Korde-Timorik', 4,) the land of the Mar, 5.) the land of Arca^, 6.) the 



176 



CHAPTER IX 



provinces of TJti, Sakasen, Gardman, Kolt e — " and the Kura river 
as before was made the frontier between the land of Albania and 
theirs [Armenia] " 25a — , 7,) Kaspe, with the city of P'aytakaran, 
8.) Gugark 4 — and there too " the ancient boundary between the 
land of Armenia and the land of Iberia, which was the great Kura 
river itself, ,., was restored * 25T) — , and finally, 9,) Aljnik*. In short, 
precisely the nine provinces which had been lost under Arsak II, 
After the treacherous murder of king Pap. the kingdom of the Axsaeids 
declined irreversibly to its destruction, and the division of Armenia 
between the two pretenders Arsak III and Xosrov III followed soon 
thereafter, " In their time ", complains the historian, " many 
provinces were gnawed away and cut off here and there, and only 
an unimportant part of the country remained in the hands of the two 
kings " 26 . This " unimportant part " made up the Armenia of the 
post-Arsaeid, Marzjpan period, 

The Armenian Geography, in agreement with Faustus, acknowledges 
the falling away of the Armenian border provinces ; but it disagrees 
with the historian in that it does not mention among them "the 
royal district in Atrpatakan " and the lands of Nosirakan. The 
repeated assertions of Faustus that Ganjak was the frontier of Armenia 
on the side of Atrpatakan and that the guard of the Armenian king 
stood there 27 , testify to the fact that the Armenian crown did possess 
great domains in Atrpatakan, Ganjak of Atrpatakan, not to be 
confused with the Armenian city of the same name, lay south-east 
of Lake TJrmiah, Maragha and Zenjan, on the ruins of Ta^t-i Sulaiman, 
and was a very important religious center in the epoch of the -Sa- 
sanians 28 , 

At the time of the restoration of the Arsacids, at the end of the 
third century, the city of Zintha was considered to be the border 
point between Armenia and Persia, according to the treaty of 298, 
One- of the clauses proclaims, "the border of Armenia shall be the 
fortress of Zintha lying on the frontier of Media " 29 , Unfortunately, 
the position of Zintha is not known ; might one perhaps associate it 
with the modern city of Sinna south of Ta^t-i Sulaiman 29a ? Accord- 
ing to Armenian accounts, the Persian king Sahpuhr ceded Atrpatakan 
to the Armenians, He -also promised that, as a reward for the help 
given him by the Armenians in his war against the Emperor, " he 
would give him [Arsak II] a great territory, of such size that in going 
from Armenia to us [Persia] he should ride continually over his own 



ARMEOTA — THE MABZPANATB 



177 



land all the way from Armenia to Ctesiphon " 30 , These words, 
despite their legendary coloration, confirm the treaty of 298 as to 
the limits of Armenia on the side of Atrpatakan, 

The frontiers of Armenia also stretched far to the south, The 
provinces of Nosirakan, Mahkert, Dasn, and Ni^orakan were all 
subject to the Armenian Arsacids. These lay south of the Armenian 
province of Korduk', on the border of Adiabene, The city of AIM, 
which is still in existence, lay in Korde or Tmorik', one of the districts 
of Korduk* 31 , Consequently, the provinces just listed lay south 
of AIM in the valley of the Khabur, which empties into the Tigris 
below the city of Jazirah-ibn- ( Omar, Nosirakan is listed among the 
provinces in ecclesiastical relations with the Armenians under the 
KatfoKkos Bagben, and is given as part of the province of Nineveh 32 , 
Nineveh was one of the five Nestorian eparchies more often known 
to the Syrians under the name of Hedayab, the ancient Adiabene, 
Six dioceses were counted in Adiabene besides the metropolis of 
Arbela: Bed NoMdre, Bed Bagas, Bed Dasen, Remmonm, Bed Mahqert, 
Dabarinos( 1) 33 , Strangely, Nosirakan is not listed among them, 
as it is in the Armenian sources, We must suppose that Nosirakan 
was a secondary name for one of the Syrian districts above, which was 
current primarily among the Armenians, most likely for the one which 
bordered on Armenian territory. Nohadre occupied the left bank 
of the Khabur river, up to the ruins of Eski-Mosul, north of the city 
of Mosul, the ancient Nineveh, Bagas lay along the Zab between 
Zerran and Diza and seemingly coincided with the present Gever 34, 
Dasn lay in the neighbourhood of the city of Amadia, in the vicinity of 
Djelu and Baz. A Yezidi tribe known under the name of Tasani 
ot Dasani is known in the sanjak of Hakkari. The province of 
Mahqert, the land of the Kurds, called ahmdxarddn by Arab writers, 
was to be found in the same region. Hence these provinces lay south 
of the Korcek s and Lesser Albak of the Armenian Geography, along 
the southern limits of the present sanjak of Hakkari, 

The fourth district in which we are interested, Ni^orakan, lay further 
east, closer to Lake TJrmiah. The Armenian- Geography mentions 
the mountains of Ni^orakan (JcoM NihoraJcan) as a spur of the Taurus 
in Atrpatakan 34a , Arab writers are acquainted with the locality of 
ddxerraMn (ie, deh Na-^rahdn) the present Deh Zargan, on the eastern 
shote of the lake south of Tabriz 34b , 

The Armenian Geography lists Nohatra, Sirakan, and Arzon-ostan 
among the lands of Xorwaran, or the Western quarter, adjoining 



178 



CHAPTER IX 



Armenia 35 . According to Syrian ecclesiastical divisions, Arzon 
together with Qardu, Zabde, Kehime, and Moksaje, made 
tip the eparchy of Arabastan, next to Hedayab 35a . The 
Armenian Geography apparently takes Aljn to mean the whole 
of Arabastan, Similarly, Nohatra and Sirakan are equated with 
Hedayab. Moreover, it is altogether possible that Sirakan is the Ar- 
menian synonym for Nohatra and that is has been introduced into 
the Armenian Geography to clarify the name Nohatra, unfamiliar 
to Armenians. However, according to Faustus of Byzantium, Nosi- 
rakan, Mahkert, Ni^orakan, and Dasn were under the authority of 
a single Mes^ 35b . Since Ni^orakan lay on the other side of the lake, 
it is more probable that Nosirakan was a political term designating 
the districts subjects to the Arsacids on the frontier of Adiabene, 
while Ni^orakan was a similar general term for their possessions in 
Atrpatakan. 

The territories listed mark the maximum limits of the Arsacid 
realm. The memory of this was still alive among the Armenians 
in the Seventh century. In order to win the Armenian princes over 
to his side, the famous Vahram Tchoben swore to them that in the 
case of his victory and the overthrow of the Sasanians, he would re- 
establish the Armenian kingdom and would give to the Armenians, 

... all of the Armenian land to Kapkoh and to the Albanian 
gates, and on the side of Syria — Arabastan and Nor-Sirakan 
to the frontier of the Tacik\ which belonged to them also in 
the days of their forefathers, and on the western side [every- 
thing] as far as Caesarea of Cappadocia 36 . 



Armenian accounts assert that the territories of Adiabene and Atro- 
patene fell away from Armenia at the same time as the other boundary 
provinces. If this is the case, it is not understandable why some are 
attributed to Armenia in the Armenian Geography, and others are not. 
The lands of Nosirakan and Ni^orakan, which are not included in 
the description of Armenia, must have fallen away earlier ; they were 
lost forever after the imprisonment of Arsak II. The successors 
of Arsak ruled over the whole of the districts which the Armenian 
Geography gives as part of Armenia ; they were the ones which made 
up the Arsacid realm par excellence. The subsequent division of 
Armenia between two pretenders, and after that, between two king- 
doms, was accompanied by a new curtailment of the border lands. 



AEMENIA — THE MARZPANATE 



179 



As a result, Armenia was reduced to the limits of the Marzpanate. 
These two periods explain to us the origin of the border districts: 
GugarF, Uti, Arca^, P'aytakaran, Pers- Armenia, Korcek' and Aljnik', 
in the form in which they appear in the Armenian Geography. We 
are not speaking, of course, of the names themselves, which are very 
ancient, many of them being of pre- Armenian origin, but of the terri- 
torial content of these terms according to the Armenian Geography. 
The frontiers of these provinces were determined from without by 
the boundaries of Armenia in the royal period, from within, by the 
boundary line of Marzpan Armenia, The belt of territories found 
between these limits produced the provinces of 6ugark e , Uti, Arca^, 
P c aytakaran, Pers-Armenia, Korcek', and were distinguished from 
one another by the fact that they belonged to different administrative 
units: Gugark' to Iberia, Uti and Arca^ to Albania, P'aytakaran 
and Pers-Armenia to Atrpatakan, and finally, Koreek' and Aljnik* 
to the kustak of Xorwaran, one to Hedayab and the other to Arabastan. 
Consequently we must admit a politico- administrative rather than 
an ecclesiastical origin for the border provinces, as might have been 
expected. Their population was always of mixed composition: 
Armeno-Iberian, Armeno- Albanian, Armeno-Persian, Armeno-Syrian, 
Armeno-Korcek' (Kurdish), so that the indicated division into provinces 
was also justified oil an ethnic basis. 

As a result of its loss of the periphery, Marzpan Armenia kept 
a territory subdivided into six countries, according to the Armenian 
Geography: Ayrarat, Taruberan, Vaspurakan, Siwnik', TayF and 
Mokk*. On what was this division based ? The last three provinces 
are distinguished from the first three in that they were subject to 
single princely families: TayF to the Mamikonean, Siwnik' to the 
princes of Sisakan, and Mokk' to the princes of Mokk', while Ayrarat, 
Taruberan and VaspuTakan were fragmented among numerous princely 
houses 36a . It is probable that the Armenian Geography means by 
Siwnik* and Mokk* the territories of the corresponding princely houses 
in its period, in other words, that their origin was based on landed 
property, Taruberan was separated from Vaspurakan along the 
new line of demarcation between Roman and Persian Armenia esta- 
blished by the treaty of -591, In fact, the same line divided Ayrarat 
from Vaspurakan. Taruberan and Ayrarat had the same sort of 
origin as Fourth and Upper Armenia. The last two provinces sprang 
from the territories assigned to the Empire at the time of the division 



180 



CHAPTER IX 



of Armenia in 387, while Tarabexan and Ayrarat came from the new 
territorial acquisitions of the Empire in 591 361 \ 

Up to 591, that part of Persian Armenia in which so-called Tanuter 
custom prevailed was occasionally called by the Armenians the Tanuter 
Land {Tanuierahan tun). After that date, part of the Tanuter land 
was parcelled off to the Empire, In the time of troubles at the 
Persian court, when the legitimate heir, Zusro II asked for the help 
of the Emperor Maurice to regain his father's throne, he promised 
in return, 

,.. to cede him the Syrian region, all of Arwastan to the city 
of Nisibis, and from the Armenian lands, the land of Tanuter 
power to Ayrarat and the city of Dwin to the shores of the 
Lake of Bznunik' (Le. Van) and the city of Af est. 

When JTusro had consolidated his position on the throne with the 
Emperor's help, he fulfilled his promise, 

... he gave him all of Arwastan to Nisibis and the Armenian 
lands which were under his power: the Tanuterakan tun all 
the way to the Hurazdan river with the district of Kotek' up 
to the village of Garni and to the sea of Bznunik*, and the town 
of Af est, and the district of Gogovit to Haciwn and to Maku. 
While the region of the gund of Vaspurakan remained under 
the domination of the Persian king 37 . 

The boundaries of the lands ceded by the Persians are very clearly 
indicated in these words, All of western Pers- Armenia went over 
to the Empire. Garni in Kotek' marked the upper point of the line 
of division, Aiest on the shore of Lake Van, the lowest point, and 
Haciwn and Maku on the side of Kogovit, the middle points. This 
means that the frontier line passed from Garni through Haciwn and 
Maku to Af est, West of this line lay the Tanuter land, east of it 
the gund of Vaspurakan, and the historian seems to be opposing 
these two terms, The true meaning of the word Vaspurakan appears 
in the expression VasypumJcan hamarakar, " the reckoner of Vaspura- 
kan " or the " collector of tribute ", where Vaspurakan means Persian. 
The members of the great Persian families were known as Yasfuhrs, 
and it seems to us that Vaspurakan might replace, the more ordinary 
expression Persian in a high flown style 37a , It is, however, also 
possible that Vaspurakan used in relation to the Armenian territory 



ARMENIA — THE MARZPANATM 



18] 



^ 



i 



is the Iranian equivalent of the Armenian term Tanuterakan. All 
of Pers- Armenia tad in fact been Tanuter land up to 591 and was 
evidently so called. With the abandonment of part of the Tanuter 
land to the Romans, Vaspurakan designated that part of Armenia 
which had remained Peisian as opposed to the Roman share 38 , 

The Tanuier lands axe sometimes called a gund as was Vaspurakan. 
The gund coxxesponds in fact to the Byzantine Theme (Be^a) and 
designates a particular division of military forces. The origin of the 
Theme organization in Byzantium cannot be considered as finally 
clarified even after the detailed investigations of Diehl and Gelzer 39 , 
Historical tradition traces the origin of the Themes to the period of 
the Emperor Heraclius, while some scholars suppose that the initial 
period of this institution was the creation of the Exarchates of Italy 
and Africa by the Emperor Maurice, We believe that the Themes 
as a military system replacing a civilian administration must be 
studied in relation to the analogous institution of the gunds in Persia. 
According to scholarly studies, the word di\xa designated both the 
district and the army corps stationed within it; moreover, it was 
unquestionably attached to the corps before it began to be used to 
indicate the province 40 , In this double meaning the word Bijxa can 
also be compared to the Persian gund, Muslim writers testify that 
a system of military encampments existed in Persia and that the 
might and military glory of the Pexsians was based on them. These 
wexe called Gundjdigdh or JRamm. According to some authors, 
there were five such stations for the gunds, according to others four 41 , 
The gunds were obviously those forces which stood under the command 
of the spahhadhs. The military encampments survived in Persia 
up to the time of the Mongol invasion which destroyed them first 
in order to bring final destruction to the power of the Persians, Scho- 
lars do not deny that the beginnings of the Theme organization were 
already visible in the sixth centuxy although it xeached its full ex- 
pxession only under Leo III (716-741) 41a , The gund in the sense of 
an army division, a regiment, is a common word in Ancient Armenian 
documents ; in its geographical meaning it appears for the first time 
in Lazar P'ar]j>eci, and after him in Sebeos, who are authors of the 
sixth and seventh centuries. This date also points to the affinity 
of the Theme and the gund. 

The Tanuier and Vaspurakan lands probably served as stations for 
certain army corps, either Armenian or Persian, and, therefore, 






182 



CHAPTER IX 



likewise bore the name of gund. Traces of the over-all administrative 
pattern are often visible in the structure of Armenia, and in general 
in that of the lands under Persian domination. 

The Tanuter Lands consisted of two parts which were called, 
according to their main provinces ; Ayrarat and Taron or Taruberan. 
Vaspurakan or Sepuhrakan was also divided into the districts of Vas- 
purakan in a narrower sense and Mardpetakan. The probable cause 
for the division of the provinces was the re-organization of Koman 
Armenia. Justinian's Armenia IV underwent alterations in 591 
as a result of the acquisition of new territory. With the shift of 
Arzanene to the Empire, Sophanene, was joined to it and to the 
district of Amida, to form a new province under the name of Upper 
Mesopotamia or Armenia IV, while the former Armenia IV, together 
with Chorzane and Muzuron, districts drawn from Interior Armenia, 
formed an eparchy called Justiniana, or the Other Armenia IV, The 
metropolis of the former was Amida, that of the latter Dadimon. 
This division has been preserved in the work of George of Cyprus 42 . 
In the Armenian Geography, Armenia IV is different from the province 
given the same name by George of Cyprus, in that Muzuron is not 
included in it, In the Geography Armenia IV or Upper Mesopotamia 
bears the name of its chief district Arzanene [Aljnik'], Furthermore, 
all of Sophene is not included in it, but only the part lying on the 
left bank of the Tigris and called Np'ret ; the right bank of the Tigris 
to Bnabel is excluded. All of these alterations and differences found 
in the Geography as against George of Cyprus occurred in Armenia 
after 591. 

This is the form in which the genesis of the provincial divisions 
given by the Armenian Geography appears to us, Its final establish- 
ment belongs to the period following the sixth century. Up to that 
time Armenia consisted of a network of larger and smaller districts 
conforming to natural boundaries and other conditions. All of them 
together formed a single administrative unit from the point of view 
of the Persian state: Marzpan Armenia, Its interior divisions derived 
from the existence of a dominant politico-social structure in the 
country — the na^arar system. 



A QUANTITATIVE ANALYSIS OF THE NAXARAED01B& 



Limits of the problem under consideration — Historical and literary sources dealing 
with the number of naxarar houses — The belief in the existence of 900 or 400 houses, 
and its lack of foundation — Actual evidence concerning the naxarar dynasties, the 
data in Conciliar Lists and their comparison, The Throne List, or Gahnamak, and the 
Military List, evidence relating to them in ancient sources — Their literary analysis 
— The content of the Gahnamak collated with that of the Military List — Variants 
in the Gahnamak and their significance — Information concerning the naxarars in 
Movses Xorenaci — Evidence of his familiarity with them — Historical analysis of 
the documents — Hierarchical precedence as the social basis for the Gahnamak, its 
existence in the VII century before the downfall of the Sasanians — Military service 
and the resultant census of Armenian cavalry before the begining of the VIII century 
as the social basis of the Military List — The basic features of the List, its points of 
contact with the Histories of Zenob Glak and Movses Xorenaci — ■ Historical evidence 
concerning the size of the Armenian cavalry — General conclusions: The literary origin 
of the analyzed documents, the sources of the Gahnamak, the Military List and the 
JRamakan nama. 



The nayarar system existed in Armenia from antiquity until the 
Mongol invasions. Like any institution developing in accordance 
with conditions of place and time, the nayamr system often changed 
in character and passed through several phases. But nayamr customs 
once developed in the period of the Arsacids, continued to function 
generally unaltered in the era of the Sasanians as well as in that of 
Justinian a . The process of disintegration of the nayamr system of 
Axsacid type began with the transfer of a significant part of Marzpan 
or Tanuter Armenia to the power of the Emperor, that is to say, 
from 628, when the Persians finally renounced it, The stern, and at 
first hostile, attitude of the Arabs toward the nayarars contributed 
to this disintegration. The catastrophe of Na^ijewan in 702 [sic] struck 
the nayarars like a bolt of lightning and destroyed the flower of 
Armenian nobility on the gallows and at the stake, This blow marks 
an absolutely decisive moment in the history of the weakening of the 
nayarars, and can probably be compared only to the ravages of the 
Turk, Bula, the Arab governor of the mid-ninth century. 



184 



CHAPTER X 



The History of Movses Xorenaci belongs to the pexiod of the dis- 
integration of Arsacid norms in the naxarar system, as is evident 
from the nature and content of the problems which plague the author *. 
The History of Armenia is the first attempt to present a history of 
the nayarms* It differs from other historical documents primarily 
because of the subjective attitude of the author toward his theme, 
Xorenaci is not a simple narrator guilelessly recounting the events 
of days long past, he is above all a critic and an investigator. He 
not only relates events, he tries to understand and interpret them, 
not always avoiding exaggerations and extraneous philosophizing 
in the process. Yet in his role as historian of the nayarars he looks 
upon the problem with perspective, seeking to discover their genealogies 
in the light of a historical past, 

Events and institutions nead no systematization or interpretation 
while they are still alive or capable of life, but they inescapably become 
the subject of research when have outlived their time, when they 
begin to fade from life and memory, gradually moving backward 
to become the property of the past. Thus, the critical attitude of 
the author of the History of Armenia indicates a period of decay, 
when the bases of nayarar life tottered and historical practices dis- 
integrated and passed from living reality into the realm of legend and 
reminiscences. This stage is truthfully described by Xorenaci, 
himself. Through the lips of the first Arsacid ruler, the historian asks, 

,,, whence originate the nayamr estates which exist here in 
Armenia? No order can be discerned here, and it is not 
known who is the first among the leaders of the country, and 
who is the last. Nothing is established, but everything is 
in disorder and disorganized x . 



According to this passage, these were the conditions found in Armenia 
by the founder of the Arsacid dynasty when he decided to familiarize 
himself with the country entrusted to him, and with the naxarar 
customs prevailing in it. Naturally, what we have here is not the desire 
of the royal newcomer, but the curiosity of the historian himself, and 
the situation depicted in no way belongs to the time of the Arsacid 
accession ; it is an exact picture of the reality contemporary with the 
historian. In view of the close relation between the History of Ar- 
menia and the period which we have set as a limit for the present 
study of the naxarars, its evidence could serve as our starting point 



A QUANTITATIVE ANALYSIS OF THE NAXARASDOHSB 



185 



in the solution of this problem, and we shall see later how the author 
of the History answers his own questions. But first we shall attempt 
to answer them independently and without his help, on the Basis 
of material unknown to him, Let us, therefore, first consider the 
concrete side of the problem: the number of nax^rar houses, and their 
distribution through the country. 

Some scholars have supposed that there were up to 900 princely 
families in Armenia at the time of the Arsacids, This opinion is 
based on a misunderstanding, or on an insufficient grasp of the his- 
torical evidence, Faustus of Bysantium writes that when Arsak II 
ascended his father's throne, he began to put the country in order 
and to revive the life of the state which had been shaken under his 
predecessor. Thanks to his activity, the country seemed to be reborn 
and to recover its former aspect, " the magnates again found them- 
selves each on his throne, the officials each in his rank" 1 *. The 
responsible function of hazampet was entrusted by the king to the 
house of Gnuni, and that of sparapei to the Mamikonean, Here are 
the words of the historian exactly translated, 

„, and the others of these houses (i,e* of the nobility such as 
the Gnuni and the Mamikonean) and of the lesser ones, who 
in their quality of gorcakal sat before the king on cushions with 
a diadem on their heads. Not counting the great nahapets 
and ianuiers, only those who were gorcakals made up the 900 
cushions which were brought out at the hours of the invited 
palace feasts, this without counting the persons from the same 
gorcakal service who remained standing on their feet s . 



The historian distinguishes the nahapets and tanuters from the gorcakals 
(i,e. the servants or officials). The Gnuni and Mamikonean princes, 
having accepted the functions of hazarapet and sparapei, became as 
a result, gorcakah, or officials in the bureaucracy. There were then 
900 such gorcakals with cushions at the court of Arsak, in addition 
to those who remained standing. It is true, as is obvious from the 
words of the historian, that these officials were selected from the 
nobility, and that one and the same person could be simultaneously 
a nahapet and a gorcakal, but it does not follow from this that there 
were as many gorcakal — officials as there were nax^rar families, 
since several officials might be drawn from the same family, In 



\ 



186 



CHAPTER X 



such an interpretation of the historian's words, the hypothesis of the 
existence of 900 princely families loses all foundation 2a , 

Another important Armenian literary document is the Life of 
St Nerses, which, if it is not an extract from the work of Faustus 
of Byzantium, is directly derived together with Faustus from a common 
source. The Life differs from Faustus in giving the leading role not 
to Arsak II, but to the patriarch Nerses I, Arsak's contemporary 
and opponent, around whom it groups the events. In the Life of 
St Nerses, the revival of the nd-gamr gahs is, therefore, attributed, 
as we might expect, to the patriarch Nerses, and only 400 individual 
cushions are mentioned instead of 900, " Nerses established the 
following 400 cushions at the table of king Arsak " 3 . At this point 
the author of the Life attempts to give a list of all the gahs, but having • 
listed 132 names (and 13 names outside the gahs), he breaks off with the 
excuse that " there are many more gahs but it is hard for him to list 
them all " ; though he then affirms again that, 

they were altogether 400 in number ... they were re-esta- 
blished by king Arsak ... at the order of the Great Nerses 3a . 



Stephen Orbelean, a relatively late writer of the thirteenth century, 
is also familiar with the 400 cushions, but, according to his information 
the initiator of the system was not Nerses but Gregory the Illuminator, 

He ordered to Trdat the Great that he should put his realm 
in order according to the example of the Greek emperors and 
grant to the princes gah and jpaiiw on the right and left side 
according to their rank; and the admirable custom was esta- 
blished that four hundred princes should sit on cushions at 
the royal table 4 . 

A curious remark is found in the History of Tovma Arcruni. Tovma 
relates that king Smbat Bagratuni granted to one of the princes of 
Vaspurakan, 

the title of marzban according to the regulation of the ranks 
of cushions established by the Armenian kings and particularly 
by king Trdat the Great 5 , 

From this reference we can conclude that even in this time, i.e. in the 
tenth century, the origin of the list of cushions, or GahnamaJc, was 
still associated with the name of king Trdat. 



A QUANTITATIVE ANALYSIS OF THE NAXABABDOMB 



187 



It would seem likely, therefore, that we are dealing here with an 
ancient tradition to which attention should be given. In reality, 
however, all of it is founded on a misunderstanding. The historian 
Zenob Glak relates that in the days of Trdat, the king of the North 
made a raid on Iberia and devastated the entire land to Karin. At 
the request of the Iberians, Trdat sent to their assistance a contingent 
of Armenians who defeated the enemy and chased away the moun- 
taineers. At the same time the Armenians captured three princes 
and four hundred nobles, whom they brought to the king. The 
prisoners were put into the fortress of Olkan, and when Trdat set out 
from Taron to Apahunik', 

.„ he took thither with him the captive princes, and ordered 
to include the four hundred men into the state diwan 8 . 

From a careless reading of this passage, the words of the historian 
might be understood to mean that all four hundred prisoners had 
been raised to princely rank, and from such a conclusion it is but a 
step to the legend of the four hundred cushions and of king Trdat as 
their originator Such a misunderstanding is not surprising. 
The historical development of Armenia was subject to such interrup- 
tions due to foreign invasions, that even the next generation occasion- 
ally found itself as helpless before many questions of its immediate 
past as we are now. Interpretations and commentaries became 
necessary, and in such work errors and omissions are not only possible 
but unavoidable. Thus the question of the 900 or 400 najarar clans 
vanishes altogether. These figures are not supported by any other 
data as to the number of naxarar families in Armenia 6a , 

The Arab writer Yaqubi, who lived in the ninth century, relates 
that the number of principalities in Armenia reached 113. Although 
Yaqubi was a native of Isphahan, he says himself that he " had lived 
a long time in Armenia and was even the secretary of many kings 
and rulers there " 7 . Hence, his information acquires a particular 
value, although we must remember that Armenia for Muslim authors 
meant not only Armenia proper but likewise Arran and Iberia. For 
instance, Yaqubi includes the principality of Sahib-as-Serir, between 
the Alans and Bab-al-Abwab into his 113 principalities, and Arran 
is listed as the first principality of Armenia 8 . The number of prin- 
cipalities which were found in Armenia proper remains open to discus- 
sion, In any case, the figure 113 must be reduced when applied to 
Armenia alone. 



188 



CHAPTER X 



This indication of Yaqubi is very close to the truth. We aie 
convinced of this by the concrete aspects of the subject,* i.e. by the 
study of the evidence found in historical materials which are drawn 
primarily from works of history and from special documents such 
as the Oahnamah [Throne List], The first rank among these must 
be given to the works of tazar P'arpeci and of Ehse which contain 
a wealth of material on nayarar nomenclature. Several .lists of 
princes are found in the History of tazar, to wit : 

I, The participants in the Council of 450 summoned to 
compose an answer to the Persian king Yazdgard II 9 , 
II, The personages summoned to Yazdgard's court 10 , 

III. The supporters of prince Vasak of Siwnik' n . 

IV, The supporters of Vardan Mamikonean fallen at the 
battle of 451 ™ 

V, The participants in the rebellion captured by the Per- 
sians 13 . 

All these lists, with the exception of the first, are also found in the 
History of EHse, and this with such precision that even the order 
in which the names follow each other coincides in both histories, 
Elise likewise appends a special list of the adherents of Vardan which is 
lacking in Lazar 14 , and also gives a second listing of the partisans 
of Vasak of Siwnik' according to a different version 15 . We should 
note in passing that these variants, as against the text of tazar, 
are of the utmost importance for the critique of the text of Elise. 
For a summary of the material given, let us take as a basis Lamar's 
first list, and complement it with the remaining catalogues: 



1, The prince [lord] of Siwnik* Vasak 



10, 



Prince Arcruni 
» Mafyaz 
» Mamikonean 
» Vahewuni 


Nersapuh 

Vriv 

Vardan [sjparapet of Armenia] 

Giwt 


» MokF 


Artak 


» Anjewaei 


Smawon 


» Apahuni 
» Vanand 


Manec 
Arawan 


» Arsaruni 


Arsawir 


» Amatuni 


Vahan 


» Gnuni 


Atom 


» Paluni 
» Asoc 


Varazsapuh 
Hrahat 



A QUANTITATIVE ANALYSIS OF THE NAXARABVOM.S 
15. 



189 



» 


Dimak'sean 


Hmayak 


» 


Abelean 


Gazrik 


» 


Afawelean 


P'ap'ag 


» 


Jiwnakan 


Vren 



All the most famous princely houses are listed here, although the 
representatives of the Rstuni and of the Bagratuni are missing, These 
were not present at the Council of 450 because they did not support 
the movement. The Rstuni are mentioned in the list of princes 
summoned to Persia: 



Prince Rstuni 


Artak 


From the third list we obtain: 




20. Prince Bagratuni 
» (G)abelean 
» Urc 


Tiroc 

Arten a6 
Nerseh 



From the fourth Hst: 



Prince Gnduni 


Tacat « 


» K'ajberuni 
25. » Hncayni 
)> Srwanjit 


Nerses 

Arsen 

Garegin 



From the last list: * 8 

The prince of Tasir 
Prince Arcruni II 
» Mandakuni 
30. » Rop'sean 

From the independent lists of Ehle: 19 



Vren 

Aprsam 

Sahak and P'arsman 

Babik and Yohan 



The prince of Ake 


Hnjut 


Prince Sahafuni 


Karen 


» Slkuni 


Ayruk 


» K'olean 


— 


» Trpatuni 


— 



35. 



In addition to the names common to both lists, the following are also 
mentioned in the History of Yahan Mamihonean by Lazar P'arpeci: 20 



190 





CHAPTER X 


Prince Erwanduni 
» Artakuni 
» Arsamuni 
» Afawenean 
40, » Yovsepean 
» k-Ark e ^ 
» Albewrik 
» Maxdpetakan 


Nerses 

Pap 

Xurs 

K'ont' 

Ners 

Vasawurt 

Arawan 

Pacok 


llowing apparently also belon 


g among the n\ 


Prince Arsakan 
45. » Zandalean 
» Sa^orapet 


Atxoxmizd 



The protocols of the Councils of the sixth century and the historical 
work of Sebeos, in the seventh, contain considerable data on the 
patronymics of the naxarars, and several new families can be added 
to our list. The Protocol of the Council of 505, held under the presi- 
dence of the kat'olikos Babgen I, has preserved the names of fourteen 
princes, among whom the following should be noted 2a : 



Prince Dastakaran 



Varaznerseh 



At another, moxe widely attended Council, held under the kat'ohkos 
Nerses II in 555, thirty-two princes were present, among those men- 
tioned for the first time are: m 



Prince h-Awenuni 
» Vaxaznuni 
50, » Spanduni 



Astwacatux 

Hmayeak 

Manuel 



From a rather large circle of naxarar houses found in the History 
of Sebeos, the ones missing from the above list are: 24 



The prince of Tayk s 
The prince of Golt'n 
Prince Abrahamean 
The prince of Basean 



Sargis 
Vram 
K'ristap'or 



We should also note the family of the 
55. Princes Mehnuni 



A QUANTITATIVE ANALYSIS OF THE NAXAKAEDOM3 



191 



mentioned by Znob Glak 25 , since Its spiritual representatives 
were present at the two Councils mentioned, Tiae bishops of Bmunik' 
and Zarehawan were also invited to the Councils, hence the princely 
houses of that name must unquestionably have existed; 

56, The prince of Bznunik 4 — 

57, The prince of Zarehawan 

Thus, according to the trustworthy evidence of Armenian historical 
documents, we can count up to half a hundred princely houses of 
greater or lesser renown a5a . 

The so-called Gahnamaks [or Throne Lists], which are a sort of cata- 
logue of noble ranks, have a direct relevance to' our problem. Because 
of their outwardly unpretentious appearance which has not inspired 
confidence, they have not received sufficient attention from scholars. 
They present, however, a great historico-literary interest, and in view 
of their limited size we think it proper to give them in externa: 851) 



GAHNAMAK 

(I Sa)hak sought from the court of king Artases, that which 
was spoken in Tispon, the famakan nama of Artasir which 
I saw in the diwan (on the 17th day of the month Kaloe). 
And to Yf am, Bang of Bangs and Benefactor, I wrote a letter, 
I Sahak, Kat'olikos, [saying] let Your Beneficence give the 
order to make for Your diwan a list of the Armenian freemen 
and magnates, just as it was formerly in the Armenian nation, 
so that henceforth the gahs of the Armenian freemen and 
magnates be known. Likewise, at the order of Nerseh, King of 
Kings, I also (Sa)hak, Kat'oKkos of the Armenians, signed 
[sealed] the Gahnamak, and we affixed the seal of the King 
of Kings and our own, and thus it is correct and true, 

(The first Prince of the Armenians and Ma^laz [sic]) 

1, The Prince [lord] of Siwnik' 

2, The .4^ 

3, Prince Arcruni 

(4. Mal^aznuni) [Malxacuni] 

4, Prince Mamikonean 

5, Sahapn Prince of Oop'k* 

6, The Prince of Mokk c 

7, Prince Rstuoi 

8, Prince Yahuni 

9, The Prince of Kaspe 



192 



CHAPTER X 



10. Prince Anjawaci 

11. Prince Apahuni 

12. The Kamsarakan 

13. Another Apahuni 

14. Of the Vanandean 

15. Prince Amatuni 

16. Prince Golt'n 

17. Prince Gnuni 

18. Another Anjawaci 

19. OftheTayk* 

20. The Judge of Basean 

21. Prince Gnt'uni 

22. The Varjawuni 

23. The Prince of Gardman 

25. The Saharuni 

26. Prince Gabelean 

27. Prince Abelean 

28. Siwnik' II 

29. Arcruni II 

30. Arcruni III 

31. Maroikonean II 

32. TheEop'sean 

33. TheAsocean 

34. The Dimak'sean 

35. The Bu^a Dimak'sean 

36. Another Abelean 

37. Another Dimak'sean 

38. The Paluni 

39. The Ai awalean 

40. The Asahmarean 

41. The Hambuzean 

42. The Varaspakean 

43. The Jiwnakan 

44. The Akeaci 

45. The Zarehawanean 

46. The Hneayeci 

47. TheMandakuni 

48. The Sliuni 

49. The Taygrean 

50. The Ermant'uni 

51. The Spanduni 

52. The Afawenean 

53. The Truni 

54. The Mamberaci [Tambexaci] 

55. The Hawnuni 

56. The Bznuni 

57. The K'ajberuni 



193 



A QUANTITATIVE ANALYSIS OF THE NAXABAEDOMS 

58. The Mehnuni 

59. The Na^-eeri 

60. The Keeper of the Koyal city 

61. The Keeper of the Koyal hunt 

62. The Artasesean 

63. Vanandean II 

64. The Cul 

65. The Vizanu(ni) 

66. Ak'aei 

67. The Dimak'sean of Sirak 

68. The Gazrikan 

69. Pr(inee) Marayean 

70. The Vaagraspu(ni) 26 

We possess another document which, although it too is called 
Gahnamak by one of the ancient historians, differs in content from 
the first. This is a list of nayavo/r families having next to each name 
the size of its cavalry contingent. According to the form of this 
List, all the naxarars were divided into four armies to defend the 
country from the north, south, east, and western sides, or, as the 
document expresses it, gates. To distinguish this document from 
the Gahnamak, we shall call it the Military List 26a . 

MILITARY LIST 

Western Gate: 

1. Angel tun . 3,400 

The Bdes x of AljniF ....... 4,000 

Beznunakan 3,000 

Manawazean ......... 1,000 

5. Bagaxatuni 1,000 

Xor^ofuni 1,000 

Cop'aei 1,000 

Vahuni 1,000 

Apahuni 1,000 

10. Gnuni 500 

Basenaei 600 

Paluni ...... 300 

Hneak'i. . 4,000 

Mandakuni 300 

15. Sa&uni .......... 300 

Varaznuni 300 

Aycenakan .......... 100 

Aiwenean .......... 300 

Varznunean ......... 100 



194 



CHAPTER X 



21, 



Span(d)uni 
Rap'sean 



300 
100 



'Eastern Gate: 



1. Siwni 19,400 



Amaskoni 
Awaeaci 
Varjawuni 
Tamxaraei 



Colkepan . 

Grzcnni 

Yarnuni ... 
10. Bak'an .... 

Keruni .... 

Gukan .... 

Patsparuni . 

Gazrikan . . . 
15. Yizanurd . . . 

Zandalan . . . 

Sodaci . . . . 

Ak'aceci 

AscSnean . . . 

Kinan .... 
21. Tagrean . . 

Northern Gate: 

1. The Bdes x of Gngark' 

Kamsarakan 

Kaspeei. 

Uteaci . 

5. Cawdeaci 

- Tayeei . 

Mamikonean 

Vanandaei 

Gardmaneei 
10. Orduni 

Axweleank' 

Asocean 

Dimak'sean 

Gent'uni 
15. Bo^ayeci 

Gabelean 

Abelean 

Ha-wnnni 

Saharuni 



200 

200 

200 

100 

100 

100 

50 

50 

50 

50 

50 

50 

50 

50 

50 

50 

50 

50 

50 

50 



4,500 

600 

3,000 

1,000 

1,000 

1,000 

1,000 

1,000 

1,000 

700 

500 

500 

300 

300 

300 

300 

300 

300 

300 



A QUALITATIVE ANALYSIS OE THE NAXABABDOMS 



195 



20. Jewnakan . 
As^adafean 
22. Varazartikean 

Southern Gate: 

1, Kadmeaei 

Kordwaci 

Areruni 

Rastuni 
5, Mokaci , 

Golt'neci 

Anjewaci 

Haruzean 

Trpatuni 
10. Mehnuni 

Akeaci . 

Zarehwaneay 

Er(w)ant c uni 

Hamastunean 
15. Artasesean . 

Sagratuni . 

Abrahamean 

Truni . . 

Buzuni . 
20, K'ajberuni . 

Boduni . 
22, Muracan 



300 
100 
100 



13,200 
1,000 
1,000 
1,000 
1,000 
500 
500 
100 
100 
100 
300 
300 
300 
100 
300 
100 
100 
300 
200 
100 
100 
300 



... and certain others occupied in other lands; and the number 
of men from the nations [clans] was 84,000 besides those who 
serve the royal court, that is the Ostan, who go forth to war 
with the king, and the Mardpetalcan, who are the inner guard 
oyer the queen and the treasure, and in all the number of the 
Armenian forces is one hundred and twenty thousand 27 , 

Both documents are known from single highly incorrect manuscripts, 
they are full of inaccuracies and errors, some of the names are distorted 
beyond recognition. The opening note, " wnwgfa fefuwh fajng £/. 
Swfuqwqh " should be taken as an example of the inaccuracies in the 
Gahnamah The first name listed is that of the prince of Siwnik', 
the list then continues accurately to fa (23), then jumps from 23 to 
fa (25) leaving out fa (24), It seems to us that this omission of a single 
number occurred because the principality of Mal^az had originally 
had a place in the list, but, for one reason or another, it had later 
been put at the head of the list instead of within it. It would be reason- 



196 



CHAPTEE, X 



able to suppose that the missing number 24 was precisely the point 
at which the Malxazuni were listed, but other evidence, of which 
we will speak later, compels us to give a higher position to the Mal^az 
clan. 

This hypothesis which was originally purely a priori in character, 
was demonstrated when we came across an interesting passage hitherto 
overlooked by scholars in the history of U^tanes of Urha, The 
Gahnamah was known to TJ^tanes, since he says that Valarsak, in his 
address to Arsak, asked him specifically to " seek out in the diwan 
of the Persian kings the QaJmamak of the Armenian w^arars", 
Arsak opened the royal diwan before Maraba, and the latter found 
the Gahnamah which he brought to Valarsak. Guiding himself by 
the Gahnamah, the king set up the na%arar clans, granting to each of 
them, 

a gah [cushion] and a patiw [honour], and according to the 
cushion the power corresponding to his dignity. 

Repeating the words of Movses Xorenaei, "Octanes relates that 
the king rewarded prince Bagarat in particular, then continues as 
follows, 

he established also other na^arar clans and gave them names: 

first: the prince of Siwnik* 

second the asjpet Bagratuni 

third: the Areruni 

fourth: the Mal^azuni 

fifth: the Mamikonean, and others likewise the king appointed 
and confirmed each in his principality 28 » 



We see from this passage that "Octanes was acquainted with and 
made use of the Gahnamah His words bring an important correction 
into the text which has reached us, by showing the original place of 
the Mafyaa in the list. Moreover, the survival of the Gahnamah 
to the tenth century, when it was used by TJ^tanes, is demonstrated, 
and the historian's tracing of the Gahnamah to the days of Valarsak 
shows that even in the tenth century a respectable antiquity was 
attributed to it. 

Information on the other document, that is to say on the Military 
List does not go further back than the thirteenth century. The 
historian Stephen Orbelean, speaking of the na%arar ranks established 



/ 



A QUANTITATIVE ANALYSIS OF THE NAXABAXDOMS 



197 



by king Trdat, adds that the same king appointed four military 
commanders for the army and the country. On the eastern side, 
the Prince of Siwnik' was appointed with twenty-one princes, on the 
north, the Bde$x of Gugark' with twenty-two princes, on the west, 
the Prince of Korduk' with twenty-one princes, and on the south, 
the Prince of Angeltun with twenty- two princes. The historian 
notes, 

this we found in this form in the Gahnamak of the Armenian 
princes which was written by Lewond and which is also more 
briefly indicated in the GaJinamaJcs of Agat'angelos and Ner- 
ses 39 . 



J 



From these valuable words of the historian we learn that the document 
with which we axe concerned was attributed to a Lewond who must 
be identified with the well-known historian and author of The Arab 
War, not only is he the only historian named Lewond, but the fact 
that the brief reference of Stephen Orbelean presupposes a common 
knowledge of the Lewond cited supports our hypothesis ; only Lewond 
the historian could be so considered. This circumstance raises the 
insufficiently studied problem of the defective character of our version 
of Lewond's History, Judging from the full title of the manuscript, 
the work of Lewond presumable began with the appearance of Mu- 
hammad, but at present it opens with a narration of the events follow- 
ing the death of the prophet 30 . The opening of the work, lacking 
the preface customary among Armenian historians, gives the impression 
of a defective and incomplete text. 

Even if we admit that the beginning of Lewond's work has been 
lost, the attribution of the Gahnamak to it can hardly be taken as 
definite, for what, in fact, can be the relation of a list of Armenian 
nayarars to the lost section of a history which dealt with the life of 
the prophet? To be sure, Armenian literature does provide the 
example of a compilation of fragments in the fusion of the Anonymous 
History with the History of Heraclius by Sebeos 30a , and the Gahnamak 
may have similarly been put in front of the work of Lewond. in con- 
nexion with some sort of historical preface. It is also possible that 
the Gahnamak was an accidental addition to the version of Lewond 
used by Stephen Orbelean, and that the later historian took it to 
be an authentic part of Lewond's work, while in reality it bore the 
same relation to Lewond as the Anonymous History has to Sebeos, 



.-^r 



198 



CHAPTER X 



in some scholars' opinion: Simeon of Aparan, an Armenian author 
of the sixteenth century, made use of a version of the History of 
tazar P'arpeci to which was added an Original History of Armenia ; 
Simeon took this to be Lazar's work, but it is evident from his retelling 
that this is the very same history that has reached us as an addition 
to Sebeos 31 , This example serves as a warning that something of 
the kind may have happened in Orbelean in relation to the Gahnamak 
as an addition to the History of tewond, hence the evidence of Orbelean 
unfortunately adds nothing positive to our knowledge of the fate 
of the document which concerns us. 

Seventy naya™? elans are listed in the Gahnamak, and eighty-six 
in the Military List. Since the original text of the Gahnamak consists 
of a single sheet covered to the very end, the problem of a defective 
text remains open, but the comparison of the two documents shows 
that this defect, if it exists, is altogether negligible. The Gahnamak 
lacks thirty-two names as against the Military List, but, on the other 
hand, it has a few additional names not found in the latter. From 
the sense of the introductory words of the Gahnamak in which the 
document is connected with the KaVoKkos Sahak I (t 439), the 
following thirty-two clans could not have been part of it : 

1, Bznuni 

2, Manawazean 

3, Orduni, 

from a purely chronological point of view since these houses had 
already disappeared at the beginning of the fourth century; 

4, Angeltun 

5, Aljnik' 

6, Gugark* 

7, Uti 

8, Cawdeaci 

9, Kadmeaci 
10, Korduk',' 

because these lay outside the boundaries of Persian Armenia ; 



11. (Amaskoni) 33 

12. Awacaci 

13. Colkepan 

14. Vafnuni 

15. Ascsnean 

16. Kinan 



A QUANTITATIVE ANALYSIS OE THE NAXAJtAXDOm 199 

17, Aycenakan 

18, Hamastunean 

19, Sagratnni; 

these 9 names are found nowhere else and are apparently badly distorted. 
There are occasional references to the remaining names among the 
missing thirty-two: 

20, Bak'an 

21, Kcruni 

22, Gukan 

23, Patsparuni 

24, Boduni, 

all districts in Vaspnrakan ; 

25, Maza^aci, 

a district in Ayrarat; 32a 

26, Varazmmi 

27, Yarlnunean, 

districts respectively in Vaspnrakan and AyraTat; 32b 

28, Zandalan, 

in the list of osianik? of Lasar P'arpeci ; 32c 

29, Sodaci, 

from Sodk c , a district in Srwnik* ; 32d 

30, As^adarean, 

from As^adar, the name of the father-in-law of Trdat III in MoYses 
Zorenaci 32e , 

31, Trpatnni 

32, Abrahamean 

known from Sebeos 32f . 



200 



CHAPTER X 



The additional names found in the Gahnamah as against the Military 
List, are ten repeated names — the junior lines of certain famous 
families ; 

Siwnik' II 
Areruni II 
Arcruni III 
Mamikonean II 
Anjewaci II 
Apahuni II 
Abelean II 
Dimak'sean II 
Dimak'sean of Sirak' 
Vanand II 

Finally there are six names which are distortions with one or two 
exceptions : 

Asahmarean = As^adarean 

Na^ceri 

Qui 

Vaagraspuni 

The Keeper of the royal city 

The Keeper of the royal hunt 32 s, 

These names make up the dubious element which has discredited 
our document. Only with the discovery of a new list can we hope 
to clear it of doubt. The Pseudo-Gahnamah preserved in the Life 
of Nerses suffers from still greater defects and is not suitable for this 
purpose. Nevertheless, it is instructive to compare it with our 
document, "We give it in extenso, italicizing the names also found in 
our Gahnamah a2h : 



10, 



ANONYMOUS PSEUDO-GAHNAMAK 

Haykaznik* 

Part'eweank 5 

Ayrarateank' 

BagratuniF 

Trdatunik* 

Aspetunik' 

Ma-xazunih* 
.Xor^orunik' 

ArcruniF 

Mamikoneank' 



A QUANTITATIVE ANALYSIS OF THE NAXAXAEDOMS 



201 



Siwnecik* 
Amatunik' 

Angeleank* 

Vracik' 
'Cofk* 
VaraznuniF 
Mardpetunik* 
Yahewunik' 

Pahlawunik* 
20. Kazbk 9 

Sisaneank' 

Kadmeank* 
Manawazeank* 

Edeseank* 

Gamreank' 
Bznunik' 

Sasaneank* 

Gisonk' 

Ekeleank' 
30. K'awpetunik' 

Anjteayk* 

Sebasteank' 

Astisateank' 

SrwanjteayF 
Anjawacifc 

Aspagnunik' 

RstuniF 
Vaharumik* 

AtyneanW 
40. K'oleantt 

Kazbunik' 

KamsarakanF 

Mokacik' 

Slkuni¥ 

Atrjpatuni¥ 
GoltneayW 
GazrikanW 

Jolkertk* 
50. Ma^azeank* 

Mrowtmik' 

Razmunik* 

Gabelean¥ 

Sparunik' 
YahuniW 

Vronjunik' 



202 



CHAPTEK X 



SureanF 

Dimah$ean¥ 

Snmik e 
60. Darbandeank' 

Aragaeeank' 

Kogovteank' 
Ajpahuni¥ 
Sncaynoc¥ 
Har¥ean¥ 

Kordowayk' 
Afawelean¥ 

Hasteank' 

Vreank' 
70, Vanandean¥ 

P'afacimik' 
Tasraci¥ 
Urceay¥ 
MandaJcuni¥ 
Tay¥ 

Meliteank' 

Dastlcarin¥ 
BasenW 

Calkimik' 
80. MamiJconean II 

P'erezunik' 

Tlk'eank' 

Bagwank' 

As(t)ocean¥ 
Abeluni¥ 

Xalteank' 

Sahafuni¥ 
Asxagorean¥ 
Gnuni¥ 
90. Hamazgunik' 
Aheay¥ 
Yizuni¥ 
Afo / p i suni¥ 
Sahofapet¥ 

Cayt'iwnik' 

(Za)naxoirapet¥ 

Varazatahean¥ 

Sncayeni¥ 

Mtnmik' 
100. Arsimik' 

K'afeank' 

Mardaleank* 

Lekandeank' 



A QUANTITATIVE ANALYSIS CXF THE NAXAMAJRDQMS 

Xorjerrik* 

Jiwnakank* 

Hamuceank* 
Kort'eank* 
Klundik* 

Zareivhank* 
110. Turberaneank' 

Bzunik* 

MehruniW 

K'alaFajpetF [Keeper of the city] 
Kayuseank' 

SpanduniV 
Artasateank' 

Orsaypeik? [Keeper of the hunt] 

Ark'acoc tear¥ 
120, Rap'seank' * 

Bagrasjpunik? 

Parspunih' 
123, Abrahameank' 33 , 



203 



This long Est presented by the author as the Gahnamak of King 
Arsak II, ox of his contemporary the Kat'olikos Nerses I, is in reality 
nothing but a late and poor version of a Gahnamak, Led astray by 
the literary tradition of the existence of 400 gaM under Arsak II, 
the anonymous chronicler tried to stretch the Gahnamak found in the 
literary sources up to the corresponding number. This attempt 
proved beyond the powers of the author because of his scant knowledge 
of his native literature, He was so ignorant that he introduced 
indiscriminately into his list of gahs a series of geographical names, 
The repetitions of the clans of the Bagratuni-Aspetuni, -Xof^oruni- 
Mal^azuni, Kazbe (sc. Kaspe)-Kazbuni, etc., must also be attributed 
to his lack of knowledge, What is important for us is that the unknown 
author undoubtedly had a copy of our Gahnamak before him as he 
carried out his task, A clear proof of this is to be found in the very 
characteristic listings which are common to both documents: 



Mamikonean II 

The Keeper of the city 

The Keeper of the hunt 

Na^ceri 

(Za)na^cirapet' 



Xfwdfilinhg hpljpnpq. 

fjpuutiqkm 
Xfiuju£kpfi 
Qui hwfu£ft pwiuhm 



204 



CHAPTER X 



The Gahnamak has Vahunik' ^wCnriili g\ instead of Vah[ew]unik s 
<lw£\kL\m.hfi g\ this accidental error has been included in the list 
in spite of the fact that VahewuniF [^ui^AunJi^], in the correct 
form is also found there, listed separately. 

Traces of the Gahnamak are also found in the famous Account of 
the Armenian Monasteries in Jerusalem, A description, or rather 
a list, of the churches of Jerusalem supposedly built by Armenian 
princes composed by the monk Anastasius at the request of Prince 
Hamazasp Kamsarakan, who intended to visit the Holy Land 33a . A 
total of seventy churches are given in this List, a number which 
immediately brings to mind a possible influence of the Gahnamak 
and compels us to look for a relation between these churches and 
the seventy gahs. It is unfortunately impossible to verify the version 
in which the Gahnamak was taken over into the work of Anastasius, 
since not all the churches bear the family names of the princes who 
founded them in the surviving text of the Jerusalem Account ; in many 
cases, they are listed under the name of the saint to which they were 
dedicated 3 . The account is unquestionably legendary, although 
it is not devoid of some authentic archaic traits, It must date in a 
period preceding the History of Albania since the work of Anastasius 
is mentioned in the History and the passages referring to the Albanian 
churches have been included in it 35 . The particular significance for 
us of this document, as of the preceding one, lies in its demonstration 
that the Gahnamak was a common document familiar to Armenian 
writers. 

Of the seventy names of the Gahnamak, only the following ten 
are missing from a place in the variant of it just given: 



Varjawuni 

Gardman 

Paluni 

Taygrean 

Ermant'uni 



Afawenean 

Truni 

Hawnuni 

K'ajberuni 

Cul 



In comparison with the Military List, however, more than thirty out 
of eighty-six listings are missing in the variant; the anonymous 
author evidently did not have it at his disposal. He unquestionably 
made use of the Gahnamak, however, and the points of difference 
between them, are to be explained by inaccuracies either in our version 
or in the one used by him. This aspect is clarified in some degree 



A QUANTITATIVE ANALYSIS OE THE NAXABAEDOMS 



205 



by a comparison of the Oahnamak with the information of Movses 
Xorenaei. At present, the History of Armenia is the touchstone foi 
determining the date and, in general, the significance of monuments 
of ancient Armenian literature. The parallel between the two Gahna- 
maks and the nayarar names found in Movses Zorenaci demonstrates, 
therefore, the historico-literary value of the documents under investi- 
gation. 

Movses Xorenaci devotes the seventh and eighth chapters of his 
second book to a survey of nayarar families and of their origin. The 
circle of nax^rar clans known to him is as follows ; 



L Bagratuni 

Gnt'uni 

Xor^ofuni 

Varaznuni 
5, Gabelean 

Abelean 

Arcruni 

Gnuni 

Spanduni 
10, Hawnuni 

Jiwnakan 

Mardpet 

Muracean 

Kadmeayn 
15, Sisakan 

Uteaci 

Gardmanaei 

Sodeaci 

Gargaraci 
20. Asocean 

Tasir 

Gugark' [bdesxi 

Orduni 

Angel-tun 
25. Cop'k' 



Apahuni 

Manawazean 

Bznuni 

Slkuni 
30. Mandakuni 

Vahnuni 

Afawenean 

Zafehawanean 

Aljn 
35. Mokaci 

Kordwaci 

Anjewaei 

[Akeaci] 

Jlstuni 
40. Goltneci 

Vanand 

Dimak £ sean 

Truni 

Amatuni 
45. Arawelean 

Kop'sean 

Mamikonean 

Kamsarakan 
49. As^adarean 35a 



Comparing this list with the Gahnamalc, we find that all the names 
here are also found in the Gahnamah with the exception of twelve : 



Kadmeayn 
Uteaci 
Sodeaci 
Gargaraci 



Angeltun 
Aljn 
Orduni 
Manawazean 



206 



CHAPTER X 



Tasix 
Qugaxk* 



Bznuni 
Koxdowaei 



Moxeovex, the Vaxasnuni of Movses Xoxenaei is equivalent to the 
Keeper of the xoyal hunt (f}puwiqkm wpgnihfi) in the Gahnamah; 
as fox the Maxdpet, as we shall see, he belongs to one of the two seeon- 
daxy blanches of the Axcxuni, Xoxenaci, howevex, lacks twenty-two 
families, not counting duplications as against the Gahnamah: 



1. Kaspe 
Tayk< 
Basean 
Vaxjawuni 
5. Sahaxuni 
Palnni 
Hambuzean 
Vaxaspalean 
gpacayeci 
10, Taygxean 
Exmant'uni 



Tambexaei 
Bzmrni 
K'ajbexuni 

15, Mehnxini 
Na^cexi 
Axtasesean 
Vilanuni 
Cul 

20. Ak'aei 
Gazxikan 

22, Yaagxaspuni 



The Gahnamah includes only Marzpan Armenia; it is, thexefoxe, 
undexstandable that the twelve naxara/r clans indicated should be 
missing from it; of them, one had ceased to exist befoxe the period 
in question, while the othexs lay outside the boundaries of Marzpan 
Axmenia, The one exception is the pxincipality of Kaspe, which 
is included in the Gahnamah, though it did not belong to Marzpan 
Axmenia but was paxt of the Satxapy of Cop'k* 35b , 

As fox the naxarar clans which axe missing from Zoxenaci in compa- 
rison with the Gahnamah, even thexe we find a connexion between 
Xoxenaei and the Gahnamah. A careful study reveals the familiarity 
of Zoxenaci not only with the Gahnamah but with the Military List 
as well. Let us give the Gahnamah once again, but now with the 
indication of the numb ex of knights belonging to each family according 
to the Military List : 

1, The Prince of SiwniF ..,,,,, 19,000 

The Prince Aspet 1,000 

Prince Axcxuni , 1,000 

» Malxazuni ...,,. 1,000 

5, » Mamikonean 1,000 

» Cop'F 1,000 

» Mokk'. 1,000 



A QUANTITATIVE ANALYSIS OE THE NAXAJRABVOMS 



207 



» Rstnni 

» Valmni 
10, The Prince of Kaspe 

Prince Anjewaci 

» Apahiud 

» Kamsaxakan 

)> Apalrani II 
15, » Vanandean 

» Amatnni 

» Golt'n 

» Gnuni , 

» Anjewaci II 
20. Tayk* . * 

The Judge of Basean 
Prince Gnt'nni 

» Varjayxtni 

» Gaxdman 
25, Prince Sahaxniri 

» Gabelean 

» Siwni II 

» Axcxuni II 
30, Prince Axcxnni III 

» Maxoikonean II 

» Rop'sean , 

» Asocean . 

» Dimak'sean 
35, Prince Dimak'sean of Bu^a 

» Abelean II 

» Dimak'sean III 

» Paluni 

» Axawelean 
40, Prince As^adaxean 

» Hambnzean 

» Vaxaspalan 

» Jiwnakan, 

» Akeaci 
45, Pxince Zaiehawan 

» jjncayeei . 

» Mandakuni 

» Slkani 

» Taygxean 
50, » Exmant'nni 

» Spandnni 

» Aiawenan 

» Txnni . , 

» Tambexaci 
55, Pxince Hawnnni , 



1,000 
1,000 
3,000 

1,000 
600 

1,000 

500 
500 
500 
600 
600 
300 
200 
1,000 
300 
300 



100 
500 
300 
300 



300 
500 
100 
100 
100 
300 
300 
300 
4,000 
300 
300 
50 
300 
300 
300 
300 
100 
300 



208 CHAPTEB X 

» Bznuni 200 

» K'ajberuni 100 

» Mehnuni ....... 100 

» Na^ceri — 

60, The Keeper of the royal city ..... — 

The Keeper of the royal hunt 300 

Prince Artasesean ...... 300 

» Vanadean II — 

» Qui ....... . — 

65. Prince Vizanu(ni) 50 

» Ak'aci , 50 

» Dimak'sean of Sirak ... — 

» Gazrikan ....... 50 

» Maracean 300 

70. » Vaagraspuni 100 



It is not difficult to see that military power has been taken as the 
basis of the Gahnamah The nayarar clans have been listed in descend- 
ing order, according to the number of knights at their disposal. This 
is the circumstance which first sugggested to us that the Mal^azuni 
were not in their proper place, as later proved to be the case. The 
variations from this order are probably to be attributed to the defec- 
tiveness of our version. A comparison with Movses Xorenaci shows 
that some of the errors go back to earliest antiquity, that is to say 
to a period earlier than his History of Armenia, The nayarm clans 
missing from Xorenaci, as against the Gahnamah are precisely those 
which interrupt the proper sequence of gahs or those whose contingent 
numbered less than 300 knights. Among the former are the princes of; 



Kaspe , , , . . 3,000 

Tayk' 600 

The Judge of Basean. 600 

Prince gpacayeci = Anja^ ,,,..,. 4,000 

Among the latter are the lords of: 

Varjavuni 200 

Hambuzean 100 

Varaspalan 100 

Taygrean , 50 

Tamberaci 100 

Bznuni .*.... . 200 

K'ajberuni 100 



A QUANTITATIVE ANALYSIS OjF THE NAXAXAXDOm 209 

Mehnuni 100 

Viisanuni . 50 

AFaci , 50 

Gazxikan 50 

Vaagraspuni = Sagratuni 100 

The exceptions are the princes Eop'sean and As^adarean, who are 
mentioned by JTorenaei, although they have but 100 knights apiece, 
and on the other hand, the lords Sahafuni, Paluni, Artasesean and 
Erwant'uni, who have 300 knights each but have been left out by the 
historian. We have already seen that Paluni and Erwant'uni were 
not mentioned in the version of the Gahnamah included in the Pseudo- 
Gahnamah of Nerses I, It is possible that they were also missing 
from the version of Xorenaei. The silence of the historian on the 
lords Saharuni and Artasesean may also be attributed to the defec- 
tiveness of the list used by him 36 , 

The particular attention given by Xorenaci to the houses of Eop'sean 
and As^adarean is to be explained by theix exalted origin. According 
to the information of the historian, the Eop'sean wexe descended 
from queen Eop'i, the wife of king Tigran, while the As^adarean 
wexe the descendents of the fathex of queen As^en, the wife of king 
Txdat III, Concexning the elevation of the Eop'sean to the dignity 
of naxamrs at the time of Tigran, Xorenaei makes a few comments 
which shed light on his relation to the Gahnamah: 

[Tigxan] also established othex minor clans which wexe found 
hexe ox in the xegion of Koxcek*. These wexe people of no 
importance because of the insignificance of theix forces, but 
they had signaled themselves by theix actions and had fought 
against the Greeks fox the liberation [of Tigxan]. Paxt of 
them came from Koxcek* and some from oux side, they wexe 
from the close descendants of the original inhabitants, and 
from the family of the Haykids, and some wexe newcomexs 
from outside. We will not call them by name, paxtly because 
thexe axe some things which we do not know, paxtly because 
we avoid weaxisome investigations, and finally because the 
lack of cextainty about many (clans) would compel us to in- 
vestigate them from all sides, Because of this we will say 
nothing about those naymar clans which wexe cxeated by the 
last Tigxan, although thou hast insistently asked us to do so, 
but we will speak only of the subsequent events which we 
know with certainty, Insofax as possible we have avoided 
superfluous or elaborate tales and all that would give the 
impxession of a doubtful account ox judgement; we have 



210 



CHAPTER X 



striven, insofar as we had the strength, toward the true and 
the authentic, whether this was derived from others or from 
ourself. Following the same goal here, as well, I lead my tale 
away from all that is unsuitable and all that might awaken 
disbelief 37 , 



The remarks of Xorenaei are significant. He seems to have the 
Oahnamah before him and to be giving a detailed account of his 
attitude toward it. He is omitting those naxarar s " npg Sfnuh^wS 
whui^qftg ifib ji i[uizm ", i.e. who had insignificant forces at their 
disposal — less than 300 knights. About some he has no knowledge, 
" i[wuh jivjuilifj n£ ifihipj Skii ". Others, although known, require an 
investigation burdensome to the historian. This may be a hint at 
the secondary branches of certain naxarar clans, and also at those 
naxarar houses possessing 300 knights which are not mentioned by 
him. Finally, many names are either incredible or uncertain. The 
explanations given by the historian for his omission of the minor 
naxarar houses corresponds exactly to the numbers left out by him 
as against the GahnamaJc, and are justified both singly and as a whole. 
It is also true that some of these naxarar clans originated from Korcek*, 
and some from other districts. 

Zorenaei likewise did not overlook the fact that the mighty houses 
of the Kamsarakan and the Amatuni occupy far from honoured places 
in the Oahnamah The Kamsarakan are given the thirteenth place, 
and the Amatuni the sixteenth. The forces of the former are given 
as 600 knights, while the Amatuni are not even listed in the Military 
List. Even if we admit that Amaskoni or Hamastunean is a corruption 
of Amatuni, the forces of or 200 or 100 knights assigned to them do 
not correspond to the might and renown of the Amatuni 37a . In the 
opinion of the Armenian historian, the Kamsarakans, as descendents 
of the house of the Pahlawuni, deserved an incomparably more honour- 
able gah among the naxarars. The insult to the Kamsarakans was 
felt all the more by the historian that the Mamikoneans, whom he 
dislikes, have been assigned the fifth place. The perplexity of Zorenaci 
and his solution is reflected in the following of his tales: 

After the death of the Aspet Sahak Bagratuni, the Kat'ohkos 
Sahak I tried to have his son-in-law, Hamazasp Mamikonean appointed 
in his place. Giving in to the prayers of his daughter, he journeyed 
to the Persian court with this goal in mind, and petitioned king Ardasir 
II, the successor of Sahpuhr II, the Long Lived ; at the same time he 



A QUANTITATIVE ANALYSIS OF THE NAXABAEDOM.S 



211 



begged the king to lighten the lot of the princes Kamsarakan and 
Amatuni who were in disgrace. The king received the kat'olikos 
most cordially and acceeded to his request 



concerning his son-in-law Hamazasp and also concerning 
the families of the guilty princes Kamsarakan and Amatuni. 
He granted their children their lives and ordered the return 
of the estates of both princes, which had been confiscated 
by the fisc. Only, he did not confirm them in the gah of their 
fathers, but lowered them below many na^arar to a place 
among the minor ones. As for the house of Hamazasp, that 
is to say the Mamikonean, it was raised so that it would have 
the right to the fifth gah among the Armenian nobles, and this 
was to be written down in his diwan 38 . 



The historian meant to allay his doubts by the recognition that the 
Kamsarakan and Amatuni princes were in a position due to their 
disgrace. He goes on to tell us the source of their guilt. When the 
Persians seized the Armenian king, Xosrov, they also took with them 
to Persia the powerful prince Gazawon Kamsarakan, who was an 
Arsacid through his mother. The brother of Gazawon, allied with 
prince Amatuni, and at the head of 700 men, lay in wait on the road 
of the Persian caravan and fell upon it with the intention of freeing 
Jfosrov. This bold attempt was unsuccessful, however, since the 
king being laden with chains was unable to escape, and the two princely 
houses paid for their audacity with their estates which were confiscat- 
ed 39 . For this reason the Kamsarakans and the Amatunis were 
out of favour with the king and lost their family gafis ; and in spite 
of the intercession of Sahak I, were not reinstated in their dignities 
but transferred to the rank of minor naxaraTS. As for the Mami- 
konean, according to the explanation of Zorenaci, they obtained 
a more honourable position, not because they deserved it, but merely 
thanks to the intercession of the kat'oKkos Sahak. 

Critics have already noted that the remark of the Armenian historian 
concerning the Mamikonean gah is based on his knowledge of the 
GahnamaJc, where the Mamikonean are specifically listed in the fifth 
place 40 . Concerning the Mamikonean gah and to clarify the problem 
it presents, the historian speaks of traditions supposedly existing at 
the Persian court. These considerations are unquestionably suggested 
to him by the opening words of the Gahnamak itself. According to 



212 



CHAPTER X 



the Persian custom, says the historian, when a new king ascended 
the throne, the coins found in the treasury wexe immediately re-cast 
and stamped with his likeness ; the diwan was transcribed in the name 
of the new king with slight alterations, but without obliterating the 
name of the old king. If, however, the king remained in power 
a long time, and a new transcription was made, the older one was 
set aside so that the king's name should appear only in the new one 40a , 
Ardasir II, because of the shortness of his reign, did not have the 
time to have a new transcription made, consequently he had all these 
matters, i,e. all that concerned the disgrace of the Kamsarakans and 
the elevation of the Mamikonean gah, added to the old transcription 
which he had had transferred from the name of his precedessors to 
his own. At the same time, the Persian king wrote a letter to the 
Armenian king Vramsapuh ordering him to do the following, 

.,. appoint Hamazasp sparapet of the army and grant to his 
house the fifth gah among the na^arar dignities. And allow 
them to have the villages and estates granted to their fathers 
by thy ancestors. Likewise release the estates of the guilty 
houses confiscated by the treasury to their children as their 
inheritance without prejudice, but do not honour them with 
the gah of their house 41 , 

Ardasir II died soon afterwards, leaving the throne to Vahram 
IV 41a , Zorenaci then has Sahak I address the latter with the same or. 
a similar request. Summoned to court over a matter concerning 
the patriarchal see, the kafoftkos, seizing an opportune moment, 
begged the king to, 

,„ grant him the gah of the Armenian m^arars as it was 
instituted by Ardasir, and order that as the matter had stood 
hitherto so it should continue in the future, and that the Persian 
Marzpans should not dare to alter the gahs at will by promul- 
gating independent orders about this. 

At the same time the kat'olikos begged him to return to Hrahat, 
the son of Gazawon, his father's estates and to include him in the 
list on the same basis as the other naxarars, granting him, if not his 
own rank, then at least some other one pleasing to the king, As for 
the Amatunis deprived of their ancestral dignity, he should assign 
them at least a place among the lesser ranks 42 , 



*— 



A QUANTITATIVE ANALYSIS OF THE NAXABAXDOMS 



213 



In this passage, JTorenaci is directly indebted to the Gahnamah 
and to its Preface which is rather confused in content and awkward 
in style. The account of X orenaei is an attempt to intexpxet the 
confusion of the Preface, whose defects in the version which has 
reached us wexe unquestionably alxeady pxesent in the copy used by 
the historian, if they do not go all the way back to the oxiginal version, 
Txanslated literally, the Preface states, 

(I Sa)hak sought from the couxt of king Artases that which 
was spoken of in Tispon [Ctesiphon], the ramakan nama of 
Artasir which I saw in the diwan (on the 17th of the month 
Kaloe), And to Vxam King of Kings and Benefactor, I wxote 
a letter, I Sahak, KaVolikos [saying], let Your Beneficence 
give the oxdex to make fox youx diwan a list of the Armenian 
freemen [azaiF] and magnates [ianuter ¥] just as it was foxmexly 
in the Armenian nation, so that henceforth the gahs of the 
Armenian freemen and magnates should be known. Likewise, 
at the oxdex of Nerseh, King of Kings, I also, (Sa)hak, Kat'olikos 
of the Armenians, signed [sealed] the Gahnamah, and we 
affixed the seal of the King of Kings and our own, and thus 
it is correct and txue 43 . 



It is not difficult to see the existence of a close connexion between 
these woxds and the refexences of -Xoxenaci to ox about the Gahnamah 
in the passages whexe Sahak I, Ardasir and Vahf am make their appear- 
ance, First, we must decide who is the Artases at whose court the 
Kat'olikos sought the Gahnamah, and who is the Artasir in whose 
diwan he saw it or whose JRamahan nama he had seen. Possibly 
one and the same person is intended, namely the Armenian king 
Artases, and for this reason Xoxenaci waxns us that the Persians had 
altered the name of the Armenian king Artases to Artasir 44 , It is 
also likely, however, that the Artasir with whose name the Mamakan 
nama is connected was the king of Persia, a possibility admitted by 
Zorenaci, as can be seen from his commentary about some sort of 
traditions of the Persian court. He acknowledges the existence 
of a transcription made in the name of Ardasir II. The request of Sahak I 
to Vfam mentioned by -STorenaci also has our Preface as its source. 
The difference between them, is that in the Preface Sahak inquires 
about the existence of the list, whereas Jfoxenaci's stoxy concexns 
the xatification and legalisation of the Gahnamah of Artasir, 

Thus, in our opinion, ^Corenaci unquestionably made use of the 
Gahnamah The opposite conclusion, is, that the author of the 



214 



CHAPTER X 



Gahanamah relied on .Xorenaci, is impossible, first of all because there 
was no need to concoct such a sorry work as the Gahnamak after 
JCorenaci and the work he had already accomplished. After the 
appearance of the History of Armenia, in which all the problems 
concerning the na^arar system had found an authoritative answer, 
the necessity for any further re- working disappeared of its own 
accord. 

The conclusion which we have reached significantly increases the 
value of the Gahnamak as a historical document. In the present 
state of the problem concerning X orenaci, however, now that the re- 
jection of the traditional point of view has removed him from the 
ranks of the first enlighteners of Armenia, and until his position in 
the subsequent two or three centuries is established, his familiarity 
with the Gahnamak unfortunately provides no conclusions as to its 
date or degree of reliability 44 a , Consequently, the problem of the origin 
and reliability of the documents under consideration: the Gahnamak 
and the Military List, first requires the investigation and determination 
of the historical setting in which such documents developed, We 
must become familiar with the specific historical raison d'etre of these 
documents, and this information brings out, among other evidence, 
important aspects of nayarar life. 



The historical or actual basis for the Gahnamak is the institutionalized 
concept of hierarchy, without which the appearance of such a document 
is impossible. The existence of a hierarchy among the Armenians 
is beyond question; it was accepted at the court of the Armenian 
Arsacids, as at that of the Persian kings, as evidenced by the national 
historians. Faustus repeatedly speaks of princes, " senior according 
to gah y senior according to cushion ", at the royal court; he likewise 
speaks of the nine hundred cushions intended for the nobility at the 
table of Arsak II. To thank Arsak for his services, the Persian king, 
Sahpuhr II, showered him with great honours, and among others, 
" ,„ seated him on the same gah as himself during a banquet " 45 . 
Soon, however, these friendly relations turned to enemity; Arsak 
was summoned by the king, but when the time came for supper, 
contrary to 



.,, the custom whereby the Armenian and Persian kings 
were seated on the same gah, that day, gahs were first placed 
for the other kings who were present and then a place was set 



asrryTftfaim* -- 



A QUANTITATIVE ANALYSIS OE THE NAXARAKDOM8 



215 



for Arsak below all the otters. ... first all were seated, each 
at the place to which he was entitled, aixd only then was Arsak 
brought in and seated 46 . 

The hierarchy of the na-^arar^ established by the Arsacid dynasty 
outlived their rule and was observed at the Sasanian court. In the 
History of Lazar P'arpeci, Vardan Mamikonean, in a speech before 
Yazdgard II, excuses himseK on the pretext that among the princes 
of the three countries, Armenia, Iberia and Albania, who are present, 
" there are many persons who are senior to him in gah and age and 
a sufficient mimber who are inferior ". The historian Ehse also 
knows something of the hierarchic customs: the Persian king now 
allowed the Christian princes to take their places at his table, and now 
deprived them of this right 47 . Smbat Bagratuni, the Marzpan of 
Vrkan [Hyrcania], was showered with honours by Xusro II Abharvez 
for his victory over the Hephthalites and " was ranked as the third 
nayamr at Xusro's court " 48 . 

The famous story of the quarrel between one of the Armenian 
princes and the Persian king over a gah is particularly interesting 
for this problem. In it we are told that the Persian king Sahpuhr 
desired to verify 

which nations and languages have cushions and honours 
(p.wp& ki. mwinfiL). He arranged a magnificent banquet for 
all the nobles from (the seven) 49 clans of the ancient Persian 
nax^rars. Designating all according to their gah, he wel- 
comed each of them with a cup and branch 50 , The Mohbadh 
of Mohbadhs was honoured with the highest dignity at the 
royal table. Taking council with his nobles, the king said, 
I am thoroughly familiar with the Persians and Parthians 
who are native Pahlawis 51 , and with the ranks of free men, 
but the noble houses of the Armenians and their ranks we were 
unable to discover either from the kings our ancestors or from 
writers. Consequently you Armenian naxarars must expect 
one of two things: either to show us the ancient document 
on the rank and dignity of each house, and to be showered 
by us with still greater honours, or, if you cannot show the 
order of your cushions before our eyes, we, who are the assembly 
of the Aryans, we shall bestow your honours, estates, earth, 
water, and all your possessions on members of the Persian 
nobility. The princes of Greater Armenia immediately con- 
sulted among themselves and presented to the king the desired 
History of Agat'angelos. He ordered it read and translated 
into the Persian script and language. Having learned that 



216 CHAPTER X 

the book began with his own ancestor Artasir, the king rejoiced 
still more, he praised the book, and moved to tears raised 
it to his eyes. He found in it the listing of seventeen cushions 
and began to apportion the seats at the royal table according 
to it. The fourteenth place was assigned to Andok, prince of 
Siwnik*. But Andok sulked and refused to eat. The king 
was informed of this but paid no heed to it. 

Later on we are told that Andok, having quarreled with Sahpuhr 
abandoned his country and went over to the service of the Emperor. 
His son Babik, upon his father's death, wished to return to his native 
land. He set out for the Persian court and once there displayed 
such valour that he made the king forget the insult once inflicted 
by his father, and was rewarded with great honours. For his services, 
the Persian king returned his native principality to him and gave 
him the right to honours equal to those of the Bagratids and the 
Mamikonean. This tale is primarily interesting as a picture of life 
in nax^rm society, as such it does not lose its importance even if it 
should prove to be pure fiction. Furthermore, and from a factual 
point of view, the story is not devoid of interesting elements ana- 
chronistically fused together. 

The hero of the story, Andok, is the famous prince of Siwnik 4 known 
to Faustus, He is the father of the fair P'aiaujem, and he really 
lived in the period of king Sahpuhr II, Made famous by the romantic 
fate of his daughter, Andok provoked a quarrel on her account between 
Arsak II and Sahpuhr II, and, together with Arsak, he waged extensive 
wars against the Persians, The fundamental motif of the quarrel 
between Andok and Sahpuhr has been preserved in the tale cited above, 
but it has been transferred to a different setting and another period 5ia . 
Babik is also a historical figure. In this episode he is presented as 
the son of Andok though in reality he was not his son but his close 
descendant. We can identify Babik with the famous contemporary 
and supporter of Vahan Mamikonean, the personage to whom the 
historian Lazar P'arpeci refers with great praise, calling him an *' excel- 
lent man " 52 . He belongs among the princes gathered around the 
kat'ohkos Griwt and Vahan Mamikonean, who held steadfastly to the 
beliefs of the dedicated heroes of Avarayr, and followed in their steps. 
Amidst the general despondency and despair which seized the country 
after the war of the Vardananians, they maintained a valiant spirit 
and did not lose hope of a better outcome for the still persisting struggle 
against the Persian oppressors. Following the example of their 



A QUANTITATIVE ANALYSIS OE THE NAXAJtAXDOm 



217 



predecessors they sought help from the Christian Emperor, and with 
this aim sent repeated embassies to the Emperor Leo I. It is 
altogether possible that Babik was one of the member of the embassy 
and that he had spent some time in the Byzantine capital together 
with this mission. Because of his outstanding qualities and services, 
his compatriot, Bishop Peter of Siwnik' dedicated a special panegyric, 
to him. This work has not yet been found in the manuscript collections 
and we owe our knowledge of it to Stephen Orbelean who refers 
several times to the work of Peter of Siwnik* as the source from which 
he took the history of Babik. In one of the references he notes that 
the Panegyric of Bishop Peter was dedicated to the valiant Babik 
whom the Persians called excellent in their own language {i.e. veh = 
law) 53 . Here Babik is given the very name with which tazar P'arpeci 
honours the Babik contemporary with Vahan, This fact is the best 
possible demonstration of our hypothesis of the identification of the 
two Babiks, the hero of the tale, and the contemporary of Vahan. 

Orbelean attributes to Peter's pen not only the history of Babik, 
but also the episode concerning Sahpuhr. Peter lived in the first half 
of the sixth Century and was present at the Council of Dwin of 555 53a , 
hence, the span of time separating him from Babik was brief, and his 
concern with the latter' s action is almost that of a contemporary, 
On the other hand, it is impossible that a man so close to Babik in 
time should have been guilty of the anachronism of combining the 
fates of Babik, who was his older contemporary, with those of Andok 
and Sahpuhr, figures belonging to the fourth century. The tale of 
Babik and Andok, in the version which has reached us, cannot be 
attributed to Peter in its entirety. It must be taken as a popular 
reworking of subjects taken from Eaustus and Peter of Siwnik* and 
its appearance in Armenian literature must be set down in a period 
subsequent to that of Peter, i.e. in one subsequent to the sixth century. 

The nature of the reworking of the tale is important for our present 
purpose. According to Eaustus, the incident with Andok occurred 
as a result of the Persian king's wish to marry his daughter to king 
Arsak II, who was already married to Andok's daughter P'aranjem, . 
The offended father decided to defend the honour of his daughter. 
Eor this purpose he destroyed the friendship of Arsak and Sahpuhr and 
succeeded in bringing about a war between them 53b , In the tale, how- 
ever, the incident has been shifted to a different setting, one of disputes 
over rank, and the conflict with Sahpuhr is presented as the result of 



218 



CHAPTER X 



the injured pride of prince Andok, displeased with the gah assigned 
to him in the naxarar hierarchy. 

Quarrels and dissatisfaction over gah and rank must have become 
more frequent with the weakening of the Arsacid tradition and the 
corresponding increase in influence of the Persian authorities. After 
the fall of the Armenian Arsaeids, matters of hierarchy came .under 
the supervision of the Persians, who transferred gahs according to 
the interests of their court, Movses X orenaei is not idly emphasising 
the danger threatening the naxarar hierarchy from the arbitrariness 
of the Persian Marzpans, in the words which he attributes to the 
kat'olikos Sahak I. The danger was felt particularly in the period 
of the revolts of Yahan Mamikonean, as is evident from Lazar P'ar- 
peci's description of princely society, Naxarar relations must have 
become especially strained in the sixth and seventh centuries, when 
Marzfans of exclusively Persian origin were appointed in Armenia, 
Consequently, documents such as the Gahnamak, insofar as they 
answer actual needs, can have appeared in Armenian literature only 
up to the fall of the Sasanian monarchy, since the institutionalized 
hierarchy kept its significance throughout this period, After that, 
the appearance of such documents can have no other motive than 
an interest in the past. 



As for our other document, namely the Military List, works of 
this type should be taken as closer reflections of society than the 
Gahnamah Whereas the significance of the latter decreased with 
the fall of the kingdom and of the royal court, the same cannot be said 
about the Military List. The historian Elise says that after the 
destruction of royal power in Armenia, " „, the ruling power was 
transferred to the Armenian naxarars ", and with it the duties of the 
Armenian crown toward the King of Bangs, One of the main features 
of Armenian subjection to Iran was the supply of military aid, as 
represented by the Armenian cavalry 54 . New conditions changed 
nothing here. As before, an accurate list of the cavalry forces, which 
the naxarars, singly and jointly, were required to keep and furnish 
at the command of the King of Kings, was a primary necessity. The 
cavalry was the main reason for the census which the Persian autho- 
rities took from time to time, Densapuh, the Persian agent sent to 
Armenia before the national rising of 451, had instructions from the 
King to carry out a census in the entire country of Armenia, " „, to 



A QUANTITATIVE ANALYSIS OE THE NAXARAXDOMS 



219 



remit taxes and to lighten the burden of cavalry ", Although the 
historian treats the mission of the Persian official with suspicion, 
the Marzjpan Atrormizd did, in fact " .,, remit the tribute and tempo- 
rarily reduce the cavalry force supplied to the King " 55 soon after 
the rebellion because of the destruction to the district, 

Cavalry regulations were inherited by the Arabs together with 
a number of other Sasanian practices. The historian Lewond relates 
that the Arab governor in 705, Mahmet [Mu'hammad ibn-Marwan], 
" ... summoned the Armenian nax^ars with their cavalry to Na^i] ewan, 
as though to include them in the official register and to distribute 
payments " 5f \ The important point here, of course, is not the 
validity of the historian's suspicions or the sincerity of the Arab 
governor's actions, but rather the indication that the Arabs took 
over the custom of keeping a list of the Armenian princes, and of 
paying them. The distribution of subsidies to the naxarars was 
also an inheritance from the Sasanians, Historians of the Sasanian 
period often speak of the pay {nn&j>l{) received by the Armenian princes 
from the Persian king. After the defeat of Zarehawan, Tazdgard II 
forgave the emprisoned princes and, " ,„ began again to give them 
their previously suspended maintainance ", He acted in the same 
manner toward the captive nax&nws "^ho had been the ringleaders 
of the revolt of 451 57 , Even before the partition of Armenia, Musel 
Mamikonean, foreseeing the gathering storm, considered the establish- 
ment of a permanent subsidy from the Empire for the Armenian 
nobility to counteract Persian inclinations 58 , According to Eaustus 
the Emperor received Musel's suggestion altogether favourably and 
promised to put it into effect, If this is not an echo of latexpxactices, 
we must conclude that both of the powers about to divide Armenia 
relied on a system of distributing subsidies to the Armenian nobles 
and army, both to ingratiate themselves by this method, and to obtain 
a ready army, 

This tradition was still alive in the seventh century. At the time 
of his expedition against the Arabs in 653, the Emperor Constantine 
[Constans II], invited the Armenian princes to Karin to plan their 
joint action against the enemy, and at the same time, he " „, promised 
them aid in the form of subsidies " 59 , The Persians wooed the 
Armenians with similar backing, At the time of the rebellion of some 
of the Armenian princes against the Persians, King Xusro II invited 
the princes who had remained loyal to him to his capital, honoured 



220 



CHAPTER X 



them in all ways, and " ... assigned them a salary from the royal 
treasury " 60 . 

In 724, the first year of the reign of the Caliph Hisham, the Arabs 
made a general census in Armenia, " ,., to increase the oppressive 
yoke of tributary obligations " 61 according to tewond's comment. 
Soon afterward Prince Asot presented himself before the Caliph and 
petitioned for a lightening of his country's heavy lot. He complained, 
among other things, that, " ,., for the last three years the official 
maintenance given to the Armenian princes and their cavalry had 
been withheld " 62 , Presumably the discontinuance of the subsidy 
paid yearly be the treasury was one of the unpleasant results of the 
recently taken census, In answer to the request of Asot, EQsham 
ordered that three years arrears be paid to him, calculating 100,000 
for each year 63 . After Hisham's death (f 743), a new decree was 
promulgated under his immediate successor, the Caliph \Abd Allah 
(750-775) 63a , According to this, 

,,, the flow of silver pouring yearly from the treasury for 
the benefit of the Armenian army was henceforth to be halted. 
The princes were ordered to furnish cavalry in specified num- 
bers and were required to maintain their forces at their own 
expense 64 . 

As long as these regulations concerning the Armenian cavalry were in 
effect, accurate information on the number of Armenian princes and 
their forces was indispensable. Because of this necessity, relevant, 
official or other, written documents similar to the Military List with 
which we are concerned, had perforce to exist. After its abolition 
by as-Saffah, the military subsidy was never renewed. There is no 
mention of financial assistance given to the naxarars by the Arabs 
in such later historians as John the Kat'olikos and Tovma Areruni 
With the abolition of this custom the need to keep of the naxarars 
and of their cavalry by means of a periodic census disappeared, and 
the census of 724 connected with the frightful catastrophe which had 
occurred in Armenia a short time before may be said to be the last 
taken for this purpose. 

The Arabs looked with disfavour upon the ruling princes of con- 
quered lands and persecuted them from the beginning. Thus, for 
example, hardly had they entered Gilan in 654 before they destroyed 
the local nobility and their cavalry 65 . In Armenia they pursued 
a similar goal. Already under s Abd-al-Malik (685-705), according 



A QUANTITATIVE ANALYSIS OF THE NAXAXABDOM8 



221 



to Lewond, the Arab governor of Armenia, " ... had conceived the evil 
plan of destroying the estate of noble families in the land of Armenia 
together with their cavalry ", but had met with a powerful rebuff 
from the Armenians. According to the same historian, *Abd-al-Malik's 
successor, the Caliph al-Walid (705-715), " ,,. plotted in the first 
year of his reign to uproot the najawir families and their cavalry 
from the land of Armenia " 66 . Pursuant to this plot of the Arab 
authorities, the Armenian princes with their troops were summoned, 
as for a census, to the city of Na^ijewan, the residence of the Arab 
governor, and were treacherously emprisoned, some in the church 
of Na^ijewan, and some in the nearby locality of Xram. The signal 
was then given for their total anihilation. The Byzantine historian 
Theophanes also mentions this event, though in his opinion the cause 
of the disaster was the rebellion of the Armenian princes against the 
Arabs 67 , This cruel reprisal cost many najmar lives. The tewond 
laments, 

I have not the strength to list them one by one. Having 
taken them all from this life, they bereft the land of its 
naxarars . , , The land of Armenia has been emptied of naxarar 
houses and the people are left like lambs among wolves 68 . 



This event took place in 705, the first year of al-Walid ; the general 
census of 724 followed shortly thereafter. The crime of 105 dealt 
a severe blow to the Armenian nobility; it marks one of the most 
sorrowful moments in the history of the ..naxarars. 

From all that has just been said, we can deduce that the need 
for a Military List disappeared after 'Abd al-Malik's abbrogation 
of the Sasanian regulations governing the naxarar cavalry, just as 
the downfall of the Sasanians in the mid-seventh century destroyed the 
institution providing a foundation for the Gahnamah 68a , Now that 
the time span within which this institution functioned has been 
determined, we can raise the question of the relation between the 
documents under consideration and the institution itself: were they 
created as historical documents from the start, or did they first appear 
in some other setting ? The answer to this question is to be found 
in an analysis of the internal evidence of the documents. 

The Military List shows many signs of contrivance, Its charac- 
teristic trait is the division of the entire naxarar cavalry into four 
groups each containing twenty-one or twenty-two houses, This 



222 



CHAPTER X 



division shows no sign of logic or of any criterion such as, for instance, 
a territorial one, For example, the Bagratuni and the Princes of 
Basean are listed together in the army of Angeltun, the Prince of 
Kaspe is listed in the army of Gugark', the Prince of Golf an in that 
of Kadme, etc, In addition, the List includes nayamr families such 
as the Orduni and the Manawazean, who ceased to exist before the 
appearance of this document, All of these circumstances point to 
a literary origin. 

The division of Armenia into four armies does not correspond to 
historical reality and is not justified by the evidence which we have 
on this subject. There is, for instance, no mention whatsoever of 
such a division in our earliest document, the History of Faustus of 
Byzantium 69 , During the royal period, the army fought under a 
single leader, the Sparapet, or commander in chief. This title was 
hereditary in the warlike house of the Mamikonean. Under King 
Zosrov II of Armenia, the Sparapet was Vace's son Artawazd; Under 
Arsak II, the famous Vasak gained renown in this office, and under 
Pap, the no less valorous Musel. Pap's successor, Varazdat, an 
imperial appointee, wished to weaken the mighty Mamikonean princes, 
by killing Musel and transferring the office of Sparapet to Bat from 
the house of the Sahafuni ; but the Mamikonean heir, Manuel, succeeded 
in defending the rights of his family and seized once again the office 
of Sparapet, which then passed to his son Artasir 70 . In the period 
of the Marzpan&te, this title was still born by the Mamikonean in 
the person of the famous princes, Vardan, Vahan, Vard, etc. 

Ancient documents speak of four bdefys, but these should not be 
confused with the matter under discussion. The concept of the bdefys 
as vassals of the king of Armenia whose duties were to guard the 
marches is based on an inaccurate interpretation of the evidence 70a . 
According to Faustus, the bdesys were ranked among the servants 
or vassals of the Armenian king, and were distinguished from other 
princes by the fact that they were, " ,.. senior to all at the royal court 
according to cushion and gah " 70b . But four persons cannot occupy 
the same place of honour in the hierarchy of gahs ; if one of them stands 
higher than another, the preceding passage is not applicable to all 
of them taken together. Furthermore, according to Faustus, the 
first gah belonged to the prince of Angeltun. Presumably the bdesys 
did not enter into the hierarchical framework established for the 
other naxamrs ; they stood, so to speak, hors Ugne, in a special position, 



A QUANTITATIVE ANALYSIS OF THE NAXAJUXDOm 



223 



as almost the equals of the king, In this sense, each of them could 
be considered as a qw£tp£ij — a senior gain. The bde$x& were lesser 
kings ruling over royal territories who had originally been independent, 
but who had acknowledged the hegemony of the kings of Armenia 
from the time of Tigran the Great. The concept of the bdesys as 
marcher lords developed much latex and was, we believe, suggested 
by the structure of the Persian kingdom, which was divided into 
four regions under the Sasanians. We have already seen that under 
Xusro I the civilian rulers of the four regions were replaced by the 
military spahbadhs, and the army was accordingly divided among 
them 70c , This transformation also affected the Armenians' concept 
of their own past, the tale of the division of the Armenian army into 
four, parts is nothing more than an echo of the Persian system. In 
imitation of the Persians and under their influence, the Armenians 
claimed that under their kings too the military forces of the country 
had been divided into four armies corresponding to the four parts 
of the world, with the purpose of defending the frontiers of the realm. 
Documents such as our Military List are the literary expression of this 
concept. Consequently the Military List, in the version which has 
reached us, belongs in the period following Zusro I Anosarvan 
(531-579) and cannot go back to a period earlier than the seventh 
century, Living in this relatively late period when the historical 
norms of the nayptrw system were falling into decay, the author of 
the GaJinamalc had to rely on literary sources to carry out his purpose. 
The same literary origin is reflected in the artificial grouping of the 
ndyamr cavalry. At the head of the four groups stand respectively 
the bdes-gs of Gugark' and Angeltun, and the princes of Siwnik* and 
Kadme. It is interesting that the house of Kadme, found in the Ano- 
nymous Pseudo-Gahnamalc, should be set out on a par with the ruling 
house of Siwnik* in place of the two other bdesys who have disappeared 
from the scene, The remaining nayara/r houses are grouped around 
these four names without any kind of jprincipium divisionis, Xorenaci, 
among other Armenian historians, mentions the division of the army 
into four corps, and furthermore as is his custom, attributes the initia- 
tive for this to King Artases. He was the one, according to the his- 
torian, who divided the Armenian army into four corps: eastern, 
western, northern, and southern, and entrusted their command to 
his three sons and to the general Smbat 71 , 

Traces of familiarity with the divisions of the army can also be 



224 



CHAPTER X 



observed in Zenob Glak, He knows that King Trdat, having 
repelled the invasion of the King of the North, left there, 

... as marcher lord the Prince of Aljnik' with 4,000 men, 
and before that, while awaiting battle and deploying his army, 
Trdat had also left two gates as passages and appointed Prince 
Bagratuni as gardian of the gates; he entrusted the right 
wing to the Prince of Siwnik c and the left one to the Prince of 
Angeltun, but he kept the semoi-gah princes as support for 
himself 73 , 

The Prince of Aljnik', with four thousand men, and the Princes of 
Angeltun and Siwnik' in their roles of commanders of the left and 
right wings of the army, are evidence of the connexion between Zenob 
and our Military List 

Zorenaci relates that King Valarsak established the order of the 
armies: first, second, third, etc., ... 73 . The subject being considered 
here is the military hierarchy, and the historian is undoubtedly referring 
to a document similar to ours. It is very possible that in this case 
he has the Military List in mind. To be sure, the period of Jforenaci, 
like that of Zenob Glak, is still controversial and cannot yet be 
established with certainty, nevertheless, their familiarity with the 
List enhances the value of the latter and permits a significant advance 
toward the elucidation of the period in which it appeared. 

The evidence on the size of the nayarar forces found in the List 
must also be considered in any attemp to determine its date. The 
numerical indications of the Gahnamak are its most interesting and 
original feature, There is no basis for supposing that the figures 
given in it are fictitious, since in the matter of figures the Gahnamak 
relies on relatively trustworthy historical materials 73a . 

The army corp of Siwnik* consists of 21 houses with 21,000 

knights, 
that of Gugark' — of 22 families with 18,000 knights, 
that of Angeltun — of 21 families with 24,000 knights, 
that of Kadme — of 22 families with 21,000 knights. 
A total of 84,000 knights. 



If, however, we do not count the princely families which lay outside 
Marzpan Armenia, and restrict ourselves to the circle of nayarar 
clans indicated in the Gahnamak of Sahak, we obtain: 



A QUANTITATIVE ANALYSIS OF THE NAX ABANDONS 



225 



In the first group — 7 families -with 19,000 knights, 

in that of Gugark' — 16 families with 7,800 knights (minus 

the 3,000 of Kaspe), 
in that of Angeltun — IS families with 8,200 knights (minus 
the 4,000 of Axeax), 

in that of Kadme — 15 families with 6,000 knights. 
A total of 51 families with 41,900 knights (exclusive of the 

7,000 doubtful eases). 



P 



Although we cannot verify the numerical evidence of the Gahnamah, 
insofar as we can judge from the scanty information scattered through 
historical literature, it is not far from the truth. 

According to the evidence of Agat'angelos, the army of the Ar- 
menians numbered more than 70,000 in the days of King Trdat III 74 , 
As many men were counted under his son and successor Zosrov II, 
whose general Databey took along an army of 40,000 in his campaign 
against the rebellious bdesx of Aljnik*, Because of the treachery 
of its commander, this army was defeated by the Persians, and King 
Xosrov set out against the latter at the head of 30,000 men 75 . It is 
of course unimportant whether all 40,000 men really perished or 
whether this is an exaggeration of the author, what matters is his 
estimate of the forces of the king as 40,000 plus 30,000, or altogether 
70,000 men, that is to say, exactly as many as had been in the army 
of Trdat. 

Commanders of the Armenian army had forces of between 10 and 
120 thousand men at their disposal. During his continuous wars 
against the Persians, Vasak Mamikonean attacked and defeated 
innumerable hordes of the enemy with now 12 hiwr'U (120,000) men, 
now 60,000, now 40,000, now 70,000, now 30,000, and occasionally 
11 or even 10,000 men at his disposal 76 , Another general, Musel Ma- 
mikonean, had only 40,000 picked men, In the battle of Mount Nipha- 
tes on the Euphrates, Musel succeeded in collecting up to 90,000 men, 
As many men were under his command at the battle of Ganjak. After 
he had ceded part of his army to the Mardjpet Gilak for the defense 
of the Armenian border on the side of Atropatene, Musel was left 
with 30,000 men 77 , In the time of troubles under Varazdat, Manuel, 
Musel's successor could barely gather 20,000 men, and on another 
occasion only 10,000 78 , If we compare these figures with the Gah- 
namak, in which the nayarar forces together with those of the king 
and the cavalry of Mardpetakan add up to 120,000, we can affirm 



226 



CHAPTER X 



with, assurance that under the Christian Arsacids, Armenia was able 
to put up to 120,000 knights in the field. 

In the period of Marzpan rule, the na-^arar cavalry shrank to 
30,000 men. Vahan Mamikonean, in spite of Lazar P'arpeci who 
deliberately diminishes the forces of this famous warrior for greater 
effect, had up to 30,000 admirably armed soldiers, as is admitted 
by the more truthful Sebeos 79 , Under Zusro II Abharvez, the 
Marzpan of Armenia also disposed of 30,000 men, according to the 
account of John Mamikonean 80 . It is interesting to note that 
in the version of Zenob, Trdat III had an army of 30,000 men as 
against the figure of 70,000 given by Agat'angelos. Here we have 
evidence of the late date of composition of Zenob' s work in comparison 
with that of Agat'angelos 81 . According to the information of 
the Byzantine historian Procopius, in 531, the Persian king Kavadh 
sent against the Emperor an army which, " ... was composed of 
Persarmenians and Sunitae whose land adjoins that of the Alani ", 
i.e. of forces from Siwnik' and other districts of Armenia 82 . This 
army, according to the same author, was composed exclusively of 
cavalry, and numbered up to 30,000 men, of whom 3,000 were Huns. 
This valuable information of a non- Armenian source serves as corro- 
boration for the thesis that the cavalry forces of Marzpan Armenia 
included up to 30,000 men, 

In the period of Arab domination, this number was reduced 
by half. At the time of the conquest of Armenia by the 
Arabs, a treaty was concluded between the conquerors and 
the Armenians, one of whose clauses required the Armenians 
to maintain a cavalry of 15,000 men 83 . The country found the 
maintenance of a large cavalry force burdensome, and its reduction 
by the Arabs was considered to be a concession on their part. This 
obligation apparently remained in force in the subsequent periods 
of Arab domination as well. According to the account of iewond, 
at the height of Marwan's struggle for the throne in the mid-eighth 
century, his enemies were defeated, when the Armenian prince Asot 
Bagratuni came to Syria, the scene of the war, to see Marwan on matters 
related to Armenia; this happened because the news had reached 
them that, the Armenian patrician with 15,000 picked knights 
had come to Marwan's assistance 84 . The size of Asot's cavalry 
contingent was undoubtedly determined by the terms of the treaty 
mentioned above, Thus, the military forces of Armenia were gra- 
dually declining from the time of the overthrow of the Arsacid dynasty. 



A QUANTITATIVE ANALYSIS OE THE NAXABAXDOM8 



227 



Consequently, we should conclude that the Military List is a faithful 
account of the conditions existing in the royal period of Armenian 
history, since it reckons the forces of the country at 120,000 knights. 
We do not have sufficient evidence to check the accuracy of the 
Military List's apportionment of knights among the various princes. 
References to small detachments under the command of this or that 
prince can be found in historical sources, but such information is 
insufficient to give us a clear picture of the subject 85 . According 
to Paustus, King Xosrov II of Armenia decreed after the episode of 
Prince Databey that, 

... the greater magnates : the nayarars who were possessors 
and lords an entire province, those with 10,000 down to 1,000 
men, should reside at court 86 , 

According to another historian, the Emperor Maurice, backing the 
Persian king Jfusro II Abharvez, sent to his assistance an Armenian 
detachment of 15,000 knights; this army was composed of naxarar 
regiments from 100 to 1,000 men, each of which was a separate corp 
with its own standard 87 , The earliest historian, Paustus, indicates 
that there were nax^rars disposing of 10,000 knights, while others 
had 1,000 men apiece. If we take the expression of Sebeos, 
" £wpfjLpwLnp^ iwqwpwLnp " in the same sense as Paustus (and not 
in the sense of the subdivision of the army into groups of 100 and 
1,000), we can take as proven the thesis that the distribution of the 
naxarar army into groups of 1,000, 100, and 50 knights, found in 
the Military List, was not due to the fantasy of a literary romancer 
but was based on a historical foundation. 



After this discussion, we can proceed to a discussion of the docu- 
ments themselves. The realistic and historically accurate features 
found in them are not in themselves sufficient to prove the official 
origin claimed by these documents. Let us even concede that both 
documents reproduce a known historical situation, yet are they, 
themselves, official documents or literary creations ? The previous 
discussion has already suggested the answer to this question, If 
we look closely at the order of naxarar families in the GahnamaJc, 
we note a close resemblance between it and the listing of princes in 
Lazar P'arpeci and EHse, Indeed, if we omit from the GahnamaJc 
the Satrap of Cop'k* and the prince of Kaspe, and compare the re- 



N/ 



228 



CHAPTER X 



maining list to that of the participants at the Council of 450 according 
to Lazar, we observe the following pattern: 



Gahnamak 

1. The Prince of Siwnik c 

2. TheAspet 

3. Arcruni 

4. Mafyaz 

5. Mamikonean 

6. Mokaci 

7. Estuni 

8. Vah(ew)uni 

9. Anjewaci 

10, Apahuni 

11, Kamsarakan 

12, The Prince of Vanand 

13, Amatuni 

14, Prince of Goltn 

15, Gnuni 



Lazar P'arypeci 

The Prince of Siwnik' 

(Bagratuni) 

Arcruni 

Mafyaz 

Mamikonean 

Vahewnni 

Mokaci 

(Rstnni) 

Anjewaci 

Apahuni 

The Prince of Vanand 

Arsaruni — Kamsarakan 

Amatuni 



Gnnni 



Several names have been added at this point to Lazar's list, but 
they do not coincide with those in the Gahnamak. This is to be 
explained by the incomplete state of Lazar' s list resulting from the 
absence of many princes from the Council, This is likewise the 
explanation for the absence of the Bagratuni and Estuni princes 
from Lazar' s list. Tiroe Bagratuni disapproved of the undertaking 
and did not come to the Council, but the very fact that he is named 
first in the list of renegades, however, indicates that he would have 
occupied the second place at the Council had he been present 88 , 
Prince Estuni was likewise absent. 

Since the two colums coincide mutatis mutandis, the Gahnamak 
reflects the hierarchical order of the fifth century, but it does not 
follow from this that the document itself belongs to this period. The 
Gahnamak might have been written on the basis of Lazar' s work. 
Such an interpretation is made impossible, however, by the fact 
that the Gahnamak cannot be derived from Lazar in ioto ; it contains 
features which cannot be found either in Lazar or in any other literary 
document 89 . It is more plausible that the author of the Gahnamak, 
interested in his country's past, composed a typical Gahnamak on 
the basis of the literary materials available to him at the time, the 
work of Lazar among other. Since the Gahnamak goes back to a 



A QUANTITATIVE ANALYSIS OF THE NAXABAXDOM8 



229 



period earlier than that of Zorenaei, it almost attains the importance 
of an official document as far as we are concerned. 

Some help in determining the period in which the Gahnamak made 
its appearance can be found in the tale of Andok cited earlier 89 a . In . 
this tale the name Gahnamak of Agafangelos is given to the list of 
princes who accompanied St. Gregory to Caesarea for his episcopal 
ordination. The author of the tale is not acquainted with our 
Gahnamak ; otherwise he would not have failed to refer to it in such 
a suitable contest as the inquiry of Sahpuhr about the gahs of the 
Armenian princes. Consequently, should we not assume a date for 
the creation of the Gahnamak later than that of the story? The 
tale itself cannot be accepted as the work of Peter of Siwnik 4 , in spite 
of the testimony of his compatriot Stephen Orbelean, so that for 
the time being it can only be interpreted in connexion with the History 
of Caspian Albania, The compilatory nature of this work is 
unquestionable, hence, the question of the relation of the tale to the 
History reduces itself in effect to the following point: to which stage 
in the History should we assign the tale? The core and older com- 
ponents of the History of Caspian Albania are found in the pages 
treating the journey of the Kat'olikos Viroy to the Prince of the 
Khazars in 628, which were written by an eyewitness of the events 
he describes 90 . These are narrated in Book II of the History, which 
opens with the story of Andok, It is natural to suppose that the 
tale also goes back to a period no later than the events described 
in this book, i.e. to the seventh century. The problem can also be 
approached from the other side, through the identification of the list 
mentioned in the story under the name of the Gahnamak of Agaf- 
angehs, Let us then determine the nature of this Gahnamak. 

In the History of Agafangelos there is still another list which 
contains the names of the princes who composed Trdat Ill's retinue on 
his journey to Kome 90a . Both these lists can be called lists of dignities 
only as the result of a misunderstanding. In the story of Andok, 
the first of these lists, called Gahnamak by the author, is made up of 
sixteen families. There are seventeen cushions in the story probably 
because the king was also included in the list, It is surprising that 
the Prince of Siwnik' is listed as eleventh in rank by Agat'angelos, 
while in the story he is assigned the fourteenth place. In the Arabic 
version of Agat'angelos, however, the kings of Georgia, Abkhazia, 
and Albania are listed ahead of the list, and the princes are enumer- 



230 



CHAPTER X 



ated thereafter. If we add the three kings to the list, the prince of 
Siwnik* does go down to fourteenth place. This fact suggests that 
the author of the story did not use our version of Agat'angelos, but 
rather an older form of the text corresponding to the newly discovered 
Arabic version 91 , 

The sixteen princes are listed here as follows: 



Armenian and Greek 
Agafangehs: 92 

1. The Prince of Angeltun 

2. The Prince of Aljnik', the 

Great bde$x 

3. The Mardpet 

4. The Aspet 

5. The Sparapet 

6. The Prince of Kordovit 

7. The Prince of Covp* 

8. The Prince of Gargar, the 

other bdes% 

9. The Prince of Rstunik' 

10. The Prince of Mokk* 

11. The Prince of Siwnik* 

12. The Prince of Cawde 

13. The Prince of Utik' 

14. The Prince of Zarawand 

and Her 

15. The Mal x az 

16. The Prince of Arcrunik' 



Arabic Agat angehs: 93 

The Prince of Angeltun 
The Prince of Aljnik 4 , the 

Great bde$x 94 
The Mardpet 95 
The Aspet 96 

The Sparapet (Mamikonean) 
The Prince of Kordovit 97 
The Prince of Covp' 
The Prince of Gargar, the 

other bdesx 
The Prince of Rstunik' 
The Prince of Mokk' 98 
The Prince of Siwnik' 
The Prince of Cawde " 
The Prince of Utik' 
The Prince of Zarawand 

and Her 
The Malxaz (Xor^ofunik') 
The Prince of Arcrunik* 



The only point at which this list coincides with our Gahnamak is 
at the listing of the Mamikonean, who occupy the fifth place in both, 
a coincidence which we believe to be accidental. If we look more 
closely at this list, we can observe that of 16 princes, 8 correspond to 
districts found in the Armenian Geography: Aljnik 4 , Korcek', Gugark', 
Mokk', Siwnik*, Cawdk', or Sawdk 4 (= Arca^), Uti, Zarewand and 
Her (= Parskahayk') 99a , The other 8 are representatives of the 
central provinces known jointly as Tanuterakan. We have already 
seen the subsequent partition of the Tanuter lands, but this list is 
still unacquainted with these subdivisions; it merely records the 
difference between Mardpetakan and Vaspurakan, in the narrow 
sense, in terms of the representatives of the Arcruni and Kstuni 
houses. The portion of the Tanuter lands later called Taruberan is 
represented here by the Mamikonean and the Jfor^ofuni, There 



A QUANTITATIVE ANALYSIS OF THE NAXAMAMDOMS 



231 



v 



are no representatives from Ayrarat; as the domain of trie Arsacids 
it is represented by trie king in person, Part of it, however, Bagre- 
wand with the adjoining Kogovit, has the Bagratid prince as a repre- 
sentative. 

The princes listed here reflect the situation present at the end of 
the sixth eentury, and coincides with the divisions which had existed 
in Armenia before the composition of the Armenian Geography. The 
list bears a tendentious stamp ; the author wished to associate it with 
the ordination, and in general with the activities of the Illuminator 
of most of Armenia, We learn from Faustus of other occasions on 
which the princes accompanied a patriarchal candidate to Caesarea 
for the same purpose, but we cannot find a similar selection among 
the attendants. The composition of Agat'angelos evidently belongs 
to the period when the name of the Illuminator served as a rallying 
slogan for the unification of the country under the leadership of the 
Armenian Church. The use of Gregory's name as a means of propa- 
ganda began with the Council of 555, where it served anti-Nestorian 
purposes, which shifted imperceptibly but logically to anti-Chalee- 
donianism, and to anti-dyophisitism in general The break in Armeno- 
Iberian church relations at the turn of the sixth to the seventh centuries 
marks the moment when anti-JSTestorianism was transmuted into 
anti-Chalcedonianism as a consequence of the altered political situation. 
In 612 a Council met at the court of the Persian king Zusro II to 
settle the quarrels of the Monophysites and the Nestorians. The 
contest ended in the defeat of the latter. Eepresentatives of the 
Armenian Church, such as the future Kat'oHkos BLomitas, here Bishop 
of the Mamikonean, and Mattew, Bishop of the Amatuni, were also 
present at this Council; according to an Armenian historian, " ,„ 
they took with them ready for use the Book of Saint Gregory " 10 °, 
Hence, by this time, the version of Agat'angelos, preserved in the 
Arabic version and reflecting an anti-Nestorian attitude as against 
the anti-Chalcedonian version of the Armenian Agat'angelos that 
has reached us, was already available. 

The Arabic version of Agat'angelos demonstrates that the list 
of 16 princes was already found in the anti-Nestorian recension of 
this work. The story of -Andok in the History of Caspian Albania 
goes back to the period 628-630, thus making it possible for its author 
to use the specifically anti-Nestorian version of the list. At the 
same time the absence of any mention of the Gahnamah of Sahak I, 



232 



CHAPTER X 



in this story in spite of the appropriate context, proves that this 
Gahnamah was not yet known to literature. Consequently the 
Gahnamah which has come down to us must have been composed in 
a period later than the middle of the seventh century, especially 
in view of its unofficial origin. 

The Military List is still later in date and likewise unofficial in 
origin. The sy metrical grouping into 21 and 22 families is sufficient 
evidence of its artificial creation. The document reproduces the 
conditions of the Arsacid period in its calculation of cavalry contin- 
gents, while the influence of sixth century Sasanian regulations is 
reflected in its subdivision into four groups or wings. The literary 
character of the List is the only explanation possible for such ana- 
chronisms. The relation of the List to Zorenaci must also be taken 
into consideration in determining the period to which it belongs. 
A link connecting the List with the Armenian historian can be seen 
in his account of the marcher princes. According to Zorenaci, the 
defense of the realm was one of the concerns of the Arsacid king 
Valarsak, and frontier commanders were appointed for this purpose. 
In the east, " ... at the hmits of the Armenian language, these were 
the Princes of Sisakan and Kadme " 101 ; in a careful definition, the 
historian places Sisakan to the east and Kadme to the south, near 
Assyria 102 , In the north, " „. on the side of the Caucasian moun- 
tains ", the defense was entrusted to the mighty house of the Bdes-% of 
Gugark* 103 . The commander in the west was the Prince of Angeltun 
though at certain times this office was taken over by the Bagratid 
princes 104 , The disagreement found here can be explained by the 
influence of the Anonymous History which identifies Angel and Ba- 
garat. Finally the historian also mentions the mighty bdesx who is 
marcher lord of the south-west, i.e. Aljnik* 105 . 

Similarly in the Military List the commanders of the four gates 
are respectively the Princes of Siwnik', Kadme, Gugark', and Aljnik* 
— Angel — Bagarat. The similarity is beyond question, but it is 
not clear whether Xorenaci is dependent on the List or, vice-versa, 
the List reflects the influence of the historian. The words of Zorenaci 
regarding the ranks present in the army (Ijwpqg ^hmnpiiLpkiug) 
indicate that he was familiar with some document similar to our 
List, but the relation of this work to the List is unknown. It is 
possible that a catalogue of the naxarar cavalry was in existence, 
and that this catalogue was subsequently re- worked into the Military 



A QUANTITATIVE ANALYSIS OF THE NAXABABVOMS 



233 



List which has come down to us. Whatever the circumstances, 
a military catalogue was known to -Xorenaci in one form or another. 

The calculation of na^arar contingents, was apparently intended 
to serve as a guide in the drawing up of the Gahnamak, The criterion 
of power is the link between the Gahnamak and the Military List 
This criterion makes it possible to interpret the Preface of the Gahna- 
mak, but what is the sense given in the Preface to the Ramakan nama 
which Sahak had seen in the royal diwan, and what is the relation 
between the Ramakan nama and the Gahnamak of which this is the 
Preface ? 

There is no basis for the hypothetical correction, vramakan for 
ramakan, suggested by Emin, accepted by Alisan, and rightly rejected 
by Professor Khalateants 105a , He, in turn, derives ramakan [nwSwliwh) 
from the Persian word ram, Arm. eram (kpujd), meaning " rabble, 
crowd ", whence the Armenian ramik (mz/z?/^) " common people, 
peasants " and the Pehlevi famak, having the same meaning. Ac- 
cording to this interpretation, the Ramakan nama was, " the list or 
book of the taxable classes ", as against the Gahnamak, which was 
the catalogue of the nobility. If, on^ the basis of the opening words 
of the Preface, we assume that the Ramakan nama is related to the 
Gahnamak, the explanation of Professor Khalatiants becomes un- 
acceptable, and we are more inclined to connect the Ramakan nama 
with the Military List Etymological meaning of ramakan {nwSw^m'ii) 
in no way hinders this interpretation. Even without deciding a 'priori 
whether or not eram, eram-ak (hpwiS, hpwS—wli) is of the same origin 
as ram, ramik (awd, nwdfilj), we believe that the meaning of famak 
is determined with sufficient precision on the basis of the Pehlevi 
ramak and the Armenian eram-ak {kpwS—wli), <( herd ". The word 
ramak which has the general sense of li crowd, collection (of people 
or animals) ", takes on the sense of a " group of soldiers, detachment 
of cavalry ", in the expression Ramakan nama. According to this 
interpretation, the term with which we are concerned meant " The 
Book of Contingents ", with ramakan as the genitive plural of ramak 
rather than an adjective derived from ram; consequently, it was 
the exact synonym of Xorenaci's vast (^w^in) ws. The evolution 
of meaning from famak to ramik, from " crowd " to " people ", is 
admirably illustrated by the analogous relation of the Russian ftW 
to the German volk. 

According to our interpretation, the Preface of the Gahnamak 



234 



CHAPTER X 



now becomes perfectly understandable: Sahak knew of the existence 
of a Mamakan nama, i.e. of a catalogue of nayarar contingents, and 
lie petitioned for the regularization of the hierarchy of Armenian 
gah% that is to say for the drawing up of a Gahnamak or Throne List 
which would presumably correspond to the data in the earlier document 
which was the Military List 



XI 



TEKBITORIAL ANALYSIS OF THE NAXABAM SYSTEM 

The naxarar&om as a territorial unit — Naxarar lands in general — Patronymic 
and toponymie nomenclature in the survey of the principalities — ■ The border princi- 
palities — The distribution of naxarar houses in the central provinces — Ayrarat 
in the Armenian Geography- — The origin of territorial divisions and the development 
of naxarar&oms — Basean, Sirak, Ayrarat par excellence, and Bagrewand as the posses- 
sions of the houses of Basean, Kamsarakan, Arsakuni, and Bagratuni — Minor princi- 
palities within the sphere of influence of the Kamsarakans, and on the territory of the 
Arsacids: their locations — Bagrewand and Elogovit — Taruberan in the Armenian 
Geography — Its principal divisions; Taron and Tayk 4 — The transfer of Taron to 
the Mamikonean — The adjacent principalities: Arsamunik* and the Arsamids, Aspa- 
kuneae jor and Zoyt s — Hark', the possession of the Manawazean — The princi- 
palities of Apahunik', Xor^oruniV, and Bznunik', and their relation to Hark* — Other 
minor principalities — Vaspurakan in the Armenian Geography — Critical note on 
its composition — ■ Position of the 18 original gawars — Additional subdivisions — 
Chief princely houses: Mfciini in Tosp, Arcruni in the Albak, Mardpets in Mardastan — 
The connexion of the Mardpets and the Arcruni — The Amatuni in Artaz — The 
grouping of the remaining principalities around the leading houses. 

Wherever the naxarar system had spread in it, Armenia consisted 
of a network of independent territorial units or separate principalities. 
After our study of naxarar houses, an investigation of the corresponding 
territorial units is equally imperative, since land was the main basis 
of the naxarar system. The setting of boundaries of the naxarar 
holdings and the determination of their locations and interrelations 
should be considered an indispensable step toward the understanding 
of the nature and complicated genis of the naxarar system. In the 
strict sense, Naxarar Armenia corresponded to Arsacid Armenia. 
Although naxarar customs survived in both parts of Armenia even 
after the partition of the Arsacid realm, they were preserved better 
and longer in the Persian portion as a result of the prevailing political 
conditions ; at least the information which has reached us is fuller for 
this part of Armenia. Even Xorenaci, the first historian of the 
naxarars, studied primarily the naxarar&oms within the limits of 
Eastern or Marzypan Armenia, i.e. within the teritory bounded by 
GugarkV Aljnik', SiwniF, and Kadme, Our quantitative study of 
naxarar clans was also limited in the main to Marzpan Armenia, 



236 



CHAPTEE XI 



A comparison of the toponymy with na%arar patronymics stows a 
close connexion between the two. The princely houses either bore 
regional names, or, on the contrary, gave the family name to their 
district, A parallel study of geographic and najarar names supplies 
important data for the classification of nayp/rar clans according to 
their hereditary evolution a . 

The border provinces of Marzjpan Armenia : Siwnik', Mokk', and 
Tayk\ were in the hands respectively of the pricely houses of Siwnik', 
Mokk' and Mamikonean, The remaining clans were scattered over 
the territory bound by these provinces. According to the Armenian 
Geography, this territory was split into three large units : Ayrarat, 
Taruberan, and Vaspurakan. Although this division was estabished 
after the period which is of immediate interest to us, we will make 
use of it for convenience in presenting the material. 

In Ayrarat, the Armenian Geography lists sixteen districts or gawars : 

1 Basean [Bagsen] 

Gabeleank* 

Abeleank' 

Hawunik' 
5 Arsarunik* 

Bagrewand 

Calkunik' [Catkotn] 

Vanand 

Sirak 
10 Aragacotn (the foothills of Aragac) 

Nig 

Kogovit (the valley of Kog) 

The Armenian Ostan [Ostan Hayoc] 

Urcajor (the cleft of Urc) 
15 Arac 

Sarur 1 , 

Ancient authors were not acquainted with Ayrarat in the same sense 
as the Armenian Geography. It is very important to clarify the 
genesis of the geographical divisions, since these divisions were deter- 
mined by the development of the naxarar&oms and consequently 
shed light on the rise and fragmentation of the naxarar houses. 

Classical authors are familiar with Phasiane-Basean, Sirakene-Sirak, 
Bagravandene-Bagrewand, and the Plain of the Araxes (*Apa£r)vdv 
TTtzSiov), by which name we should understand the Plain of the Araxes 
in the narrower sense of the word 2 , Armenian documents older than 



■i ffl-B^v^a . i .1 1 i ' ■ * ■ ■ i 



TERRITORIAL ANALYSIS OE THE NAXABAB SYSTEM 



237 



the Armenian Geography, usually give the name of Ayrarat to the 
eastern half of the later Ayrarat, from the valley of the A^urean 
[Arpa jayi] to the Plain of Sarar, The y Apa^rjvov 7re8(ov corresponds 
literally to the Armenian Eras^ajor, which is the given to the part of 
the Araxes valley bordering on Sirak and later re-named Arsarnnik'. 

Faustus, the historian of the royal period, mentions Vanand and 
Arsarunik' in addition to the four districts already named 2a . Yanand 
was formerly part of Basean and was called Upper Basean, as it still 
is by Xorenaci, but in the Arsacid period it separated from Basean 
to form a separate principality. In the same way, Arsarunlk' was 
formed from the border strip of the Araxes which was formerly included 
in Ayrarat and Sirak. All these division were not only geographical 
but also political units in the royal period, and were under the rule 
of autonomous princely families, Basean, Vanand, Sirak, and Arsa- 
runik' were ruled respectively by princes bearing the same name, 
Ayrarat, within the limits indicated, was the domain of the Arsacids^ar 
excellence, while Bagrewand, we are told, belonged to the Bagratids 3l3 , 
At a later date, Sirak passed to the Kamsarakan princes. The house 
of Kamsarakan was descended from the Arsacids, and probably 
established itself first in Eras^ajor, which was called Arsarunik', 
presumably after its first ruler Arsawir, from that time on; subse- 
quently, the family extended its power over neighbouring Sirak as 
well 26c , Part of Sirak belonged to a branch of the princely house of 
the Dimak'sean known as the Dimak'sean of Sirak to distinguish it 
from the other branch of the family, the DimaFsean of Bu^a, who 
came from the district of Bu^a near Yanand 2a . 

Sirak occupied the valley of the A^nrean between Vanand and 
AyraTat, and is now called Soragyal [Suregel] from the Armenian 
Sirak-gawar or §irak-awan, Basean, now Pasen, lay on the upper 
courses of the Araxes as far as the ancient Salk'ora, i.e. almost 
to the present [1908] B-usso-TurHsh frontier 3 . According to the 
description of the Armenian Geography, the principalities of AbeleanF, 
HawnuniF, and Gabeleank' lay along the Araxes between Basean 
and Arsarunik', of these, the first two lay on the left bank of the 
river, and the third on the right 3a , According to the geographi- 
cal location, these principalities are to be identified primarily with 
Arsarunik'. The three districts with their center Kalzwan [Kagiz- 
man], the ancient Kalzwan, still form a separate district called Qaldiran, 
similarly the former Basean, Vanand and Sirak now make up the separate 
districts of Pasinfler], Ta^tin, and Siiregel 3 *>, 



238 



CHAPTEE XI 



Paustus, the historian of the Arsaeids, never mentions the prin- 
cipalities of Abeleank*, Gabeleank*, and Hawnunik', which are well 
known from subsequent sources. Should this not mean that in 
the royal period these small principalities were unknown, and that 
their lands entered into the composition of Arsarunik' politically as 
well as territorially? In such a case, they must be considered as 
branches of the mighty house of Kamsarakan, and their emergence 
and political emancipation must date from the troubled period of 
the disintegration of royal power in Armenia, We can see that the 
influence of the Kamsarakans grew toward the Araxes in the direction 
of Basean from the fact that the castle of Bol in Basean belonged to 
them in the sixth century 313 . The principality of Asock* also adjoined 
the possessions of the Kamsarakans ; it occupied the northern corner 
of Sirak in which the source of the A^urean is located 4 , 

Ayrarat was the domain or Ostan of the Arsacids [Arsakuni], It 
was composed of the following gawars [districts] : Aragacotn, Nig, 
Varalnunik', Mazaz, Kotayk', the Ostan of Dwin, Urc, Axac, Sarur, 
Cakatk', and Maseacotn 4a . Of these, the last two lay on the southern 
bank of the Araxes between Kolb (now Kulp) and the summit of 
Mt, Ararat. The other gawars were located on the northern bank, 
and their position was determined by the course of the tributaries of 
the Araxes. The upper course of the K'asal river formed Nig, now 
Abaran [Aparan], the rest of the river and the foothills of the Aragac 
[Alagoz] belonged to Aragacotn (literally the foot of Aragac). Varaz- 
nunik' (now Darachichak) lay at the source of the Hrazdan-Zanga ; 
lower along the same river was Kotayk' (now Zangi-bazar). The upper 
course of the A^at- Garni river was occupied by Maza2 (now Gafni- 
bazar) ; lower down lay the ostan of Dwin, Urc and Arac were found 
along the Vedi river, and beyond them the Plain of Sarur stretched 
all the way to the limits of Na^ijewan 41 \ Many of these subdivisions 
of Ayrarat are not mentioned in the ancient sources before the Ar- 
menian Geography; it would be rash, however, to see in this an argument 
for their late origin. Kotayk', for example, was already known to 
Ptolemy, but the first reference to it in Armenia is found in Sebeos, 
in the seventh century 4c . There can hardly be any doubt that the 
other names also go back to a period earlier than that of Sebeos, 
The reason for their absence in the older documents must be sought 
in the fact that these divisions, devoid of any political significance, 
were purely topographical in character, whereas the Armenian writers 



- -^agu j u^uju i ^- i i 



TERRITORIAL ANALYSIS OE THE NAXAMAJR SYSTEM 



239 



of the Marzpan period speak primarily of those geographical divisions 
of Armenia which were simultaneously political and %ayp/rm units, 

The province of Ayrarat, within the limits given, was the domain 
of the Arsacids, and was therefore a single unit from the political 
point of view. The appearance of small na^mm houses in Arsacid 
Ayrarat must he assigned to the period following the fall of the Arsacids, 
Among these, Urc and Arac already emeTge as separate principalities 
in Ehse 4d , Prince Varaznuni is mentioned at the Council of 555, 
According to -Xbrenaei the possessions of the princely house of Varaz- 
nuni were to be found along the Hrasdan river, the present Zanga, 
consequently the historian derives the Vara2nuni family from a 
descendent of Garnik, the son of Gelam, who was the ancestral founder 
of the adjacent province of Gelark'unik' 5 , According to these 
indications, Varaznunik' coincides with the plain of the upper branch 
of the Zanga river, and corresponds to the modern Darachichak (the 
Flowering Plain), which is a translation of its middle- Armenian name, 
Calkunik 5 ^ 

According to the same author, the Princes Gnt'uni lived near the 
Varaimuni, not far from the Lake [Sevan] 7 , The inscription of a 
certain prince Gregory Gnt'uni has been discovered in the ruins of 
an ancient church on the banks of the Kasal river in Abaran. On 
the basis of this inscription, we must admit that Abaran, the ancient 
Nig, was the patrimony of the Gnt'uni nayamrs 8 , 

South of Abaran, in the valley of the same Kasal river still stands 
the historic village of Osakan, which was ruled by the princes Amatuni. 
Lazar P'arpeci notes that this village belonged to the Amatuni 9 , 
Xorenaci asserts that Osakan had passed to this family under king 
Xosrov II Kotak, who gave it to Yahan Amatuni as a reward for his 
valiant repulse of the mountaineer bands of King Sanesan, but Paustus 
does not mention the presentation of Osakan to prince Amatuni in 
his account of these events 9a , Jforenaci considers the Amatuni to 
have been immigrants from the East, i,e, from Persia, and places 
their appearance in Armenia under King Artases, but he leaves us 
in ignorance as to the home of the Amatuni family in the period 
between Axtases and Xosrov Kotak 10 , We shall see that the Amatuni 
first lived in the district of Arbaz and subsequently moved from it to 
Osakan, By Xorenaei's time Osakan was already in the possession 
of the Amatuni, and the historian, guessing a posteriori, set the origin 
of their ownership in the period of Sanesan's Taid, which he knew 
from the History of Paustus 10 *, 



-^a^v*. 



240 



GHAPTEK XI 



An inscription has been preserved on an ancient church of the 
village of Mastara, at the foot of the Aragac, on the border of Sirak. 
This inscription states that the church was built in the days of Bishop 
Theodore Gnt'uni ", Its existence clarifies a reference found in the 
Life of St Nersss, according to which the Gnuui were to be numbered 
among the naxcirars of Sirak 12 , and proves beyond doubt the connexion 
of the Gnuni house with the localities situated at the foot of the 
Aragac. We know from Thomas Areruni that in his time the Gnuni 
were living in Ali-ovit, on the shores of Lake Van. Thomas calls 
Arees, the chief city of the district, « ,,. the city of the gawaf of the 
Gnunik' 13 », .Zorenaci, on the other hand, points out that Sahapivan 
had formerly belonged to the Gnuni, but later passed to the Arsacids 14 . 
Sahapivan is located within the boundaries of Kogovit and seems to 
have been a temporary stopping place in the move of the Gnuni from 
the shores of Lake Van to Ayrarat, or possibly in the opposite di- 
rection. According to Zorenaci, the Gnuni served as butlers at 
court, and had to live close to the royal court, namely in Ayrarat, 
to perform their duties. Consequently, from Xorenaci's point of view, 
the problem of the priority of Gnuni establishments is resolved in 
favour of Ayrarat 14a . 

According to Xorenaei, court functions were also incumbent upon 
the nax^rardoms of the Spanduni and the Jiwnakan, and these should 
be sought in the vineinity of the royal ostan. In his attribution of 
hereditary offices at the Arsacid court to naxarar families, the historian 
seems to derive these offices from the etymology of the family names. 
The opposite conclusion seems more valid, however, namely that 
evidence of the geographical closeness of a given naxamr&om to the 
court had an unquestionable influence on etymological associations 
and derivations. If, for example, the Gnuni had lived very far from 
the Arsacid osian, Zorenaci would hardly have derived the name 
Gnuni from gini «wine», in order to bestow upon that house the 
honorific office of royal butler, or supplier of choice wines. Zorenaci's 
etymologies are unacceptable in most cases, but they are none the 
less interesting since they reflect to some degree the territorial relations 
between the naxcvrardoms and the ostein or royal domain 15 , 

Sebeos also mentions- the Afawelean and the Afawenean together 
with the Varaznuni, Gnt'uni, and Spanduni among the naxarars 
living within the borders of Ayrarat. According the Zorenaci, the 
Afawenean were descended from the ancient kings, and the Afawelean 



TERRITORIAL ANALYSIS OF THE NAXABAR SYSTEM 



241 



from the Alan relatives of the Princess Sat*enik, who had come to 
Armenia with hex, and had been listed among the nayamrs* Whatever 
its accuracy, this explanation proves that the two families lived in 
the royal ostan of Ayxaxat, though we lack exact information as to 
the precise locations 16 , 

The possessions of Prince Sahafuni were also found in Ayxaxat, 
on the border of Sixak, The village of Sahafunik' is mentioned in 
an inscription of the Bagratid king Smbat Sahinsah [Yovhannes- 
Smbat, 1030-1040], dating from 1037. The inscriptions proclaims 
that the king gave this village to the monastery of Hoxomos, the 
ancestral burial place of the Bagratids 17 . In addition to this, we 
know that David Sahaxuni, Curopalates of the Imperial portion of 
Armenia at the time of the Emperor Heraclius (f 641), built a church 
at Mxen 18 , It is evident that at that time Mren belonged to the 
Sahaxuni house. The village of Sahafunik', the ostan of the princes 
of the same name, was probably located between Mren and Hoxomos, 

Adjoining Ayrarat from the south, lay Bagrewand, Calkotn, and, 
or Kogovit 18a . Bagrewand occupied the basin of the two main origi- 
nal branches of the Axsanias. The source of the right branch, that is to 
say the foothills of the Ala dagi range (ancient Calke) which was called 
Calke-otn, the foot of Calke 19 . Next to it lay Kogovit, on the Maku 
river, which is a tributary of the Axaxes on the xight side 19a , One 
of the important sites in Bagrewand was the city of Bagawan, whexe 
the Monastexy of St John the Baptist [the Precuxsox], Swb Karapet 
now stands 30 . In Kogovit stood the famous fortress of Dariwnk', now 
Bayazet' [Dogubayazit] 20a . In Calkotn wexe to be found the foxtress 
of Angel and the minexal springs of Yarsak 21 ; the springs still exist 
near Diyadin, the ancient Tateonk', but Angl should apparently be 
sought nearer to the Ala-dagi 22 . Paustus is not familiar with 
Calkotn; he puts the city of Zaxehawan into Bagrewand, together 
with Bagawan, while in all the other sources Zaxehawan is a city of 
Calkotn 23 . The latter district probably was not a political unit 
and was paxt of Bagxewand. 

In the xoyal pexiod, Kogovit belonged to the Arsacids, and the 
royal treasure was kept there in the fortress of Dariwnk' 24 , In the 
seventh centuxy and thexeaftex, we find Kogovit in the hands of the 
Bagratid [Bagxatuni] pxinces ; the famous stxonghold of Dariwnk' was 
held to be theix ostan, and the buxial place of the family 35 , "We do notknow 
exactly when the Axsacid possessions passed to the Bagxatids, In 



242 



CHAPTER XI 



Paustus the Bagratid ancestral domain is the district of Sper 26 . 
Since the Bagratids in the person of Prince Tiroc took part in the 
rebellion of the fifth century, albeit on the side of Yasak of Siwzdk', 
and since all the participants in this movement were from the Persian 
pait of Armenia, we must presume that, in addition to Sper which 
lay in Roman Amenia, the Bagratids had another principality in 
Persian Armenia/ and that Tiroc was the repressentative of the Pers- 
armenian Bagratids. It is possible that the family was already m pos- 
session of the district of Kogovit, with the fortress of Dariwnk', which 
had passed to them after the fall of the Arsacids. There is even 
some basis for supposing that the Bagratids had lived from ancient 
times in these districts, and, if we accept a connexion between Bagrata 
and Bagra-i>cm<?a, Bagrewand will have to be acknowledged as the 
original patrimony of the Bagratids 27 , According to some theories 
which we shall present when we discuss the origin of the na^arar 
houses, the Bagratid dynasty sprang from the borderland of Armenia 
adjoining Iran or Atrpatakan. The recognition of Bagrewand as 
the original home of the Bagratids agrees with this thesis, since this 
district is not far removed from Atrpatakan, 

With the spread of Christianity, and the triumph of the cross at 
Bagawan, Bagrewand with its shrines passed to the Church, or 
rather to the patriarchal house of St, Gregory the IUuminator. 
Lazar P'arpeci never refers to Bagawan in other terms than as the 
house (ie, the estate of St, Gregory, Although the Kat'ohkos Sahak I 
died at Valarsapat, he was buried in the village of Blur in Bagrewand, 
which evidently belonged to his house 27a . In any case, Sper is not 
acceptable as the original province of the Bagratids, it is rather an 
intermediate point in their move from Armenia to Iberia, 

The second province, Taruberan, also consisted of 16 gawafs ac- 
cording to the Armenian Geography : 



1 Xoyt* 
Aspakunik* [Aspakuneac jor] 
Tardn 
A(r)smunik* 
5 Mardali 
Dasnawork* [G'astavor] 
Towaraeatap' 
Dalaf 
Hark' 
10 Valnunik* 



"T"1%I 



TERRITORIAL ANALYSIS OE THE NAXABAB SYSTEM 

BznuniF 
Erewark' 
Aliovit 
Apahxonik' 
15 (Kor) 

(Xorxofunik*) a8 . 



243 



Of these, the most outstanding fox its antiquity and importance was 
Taron, which was already known to classical writers as regio Taurau- 
nitium 29 ; the extent of its fame is attested by the fact that even in 
the Armenian Geography Tarnlberan is identified by a reference to it 30 . 

In the period with which we are concerned, Taron was ruled by 
the powerful house of the Mamikonean, whose real province was 
Tayk', as is evident from Eaustus, who repeatedly calls it the land of 
the Mamikonean 31 , The latter then was the ancestral home of the 
Mamikonean, but we will speak of this later. We do not know exactly 
when the Mamikonean came to Taron, but we are certain that they 
were already there in the fourth century. According to Eaustus, the 
impregnable fortress of Olakan, on the banks of the Euphrates, 
belonged to them, and was the seat of Musel, the famous Mamikonean 
general of King Pap 32 . The same historian says that during the 
reign of Arsak II, the Mamikonean princes, angry at the king for his 
brutal massacre of the princely houses of the Arcruni and Rstuni, 
left him, and, adds the historian, « ... abandoning their other province 
[house], » withdrew to their district of Tayk' 33 , ; Here, «the other 
province », must refer to Taron, 

Taron also contained ecclesiastical estates. After he had visited the 
holy places at Astisat, the Prince Mardjpet was enchanted by the 
picturesque setting of the residence of Kat'ohkos Nerses I, and out of 
envy, he began to rebuke King Trdat III bitterly because, in Eaustus' 
own words, « he had given such lands to people wearing female garb 
{ix* to the clergy) 34 », Consequently, one part of Taron with Olakan 
belonged to the Mamikonean, and the other with Astisat to the 
Church, or rather to the family of the kat'oSkos, 

According to the evidence of Xorenaci, Astisat had formerly be- 
longed to the priestly family of the Vahnuni, but later it was taken 
from the pagan clergy, so that Astisat passed to the fisc 35 , The 
truth of the matter is that after Christian shrines had been erected 
on the sites of pagan temples, the ancient temple estates, Astisat 
among them, became the property of the Church, The accuracy of 



244 



CHAPTER XI 



this conclusion is also supported by tlie words of the Mardpet we 
have just cited, since lie places the transfer of Astisat to the Church 
in the reign of Trdat III, who was the first Christian Arsacid on the 
Armenian throne, ^orenaci also asserts that Olakan was originally 
the possession of the princes Slkuni, This family was disgraced as 
the result of their conspiracy with the Persian king against Trdat III, 
The Mamikonean overcame the Slkuni at the order of the Armenian 
King, and received their ancestral lands as a reward 36 , 

It is surprising that Zorenaci does not know that the Mamikonean 
possessed Tayk\ In his account of the legendary migration of the 
family to Armenia under Bang Trdat, he does not indicate the place 
where they settled. According to his information, Trdat III did not 
aasign any particular lands to them, but, having received them with 
benevolence, « ,,. he gave them a place and the means of life, and he 
moved them from place to place for many years, » i.e. until the episode 
with the Slkuni when they inherited the possessions of the latter 37 . 
In view of the obvious ignorance of the historian concerning the native 
province of the Mamikonean, his information as to the date of their 
consolidation in Taron does not inspire particular confidence. It is 
possible that the Mamikonean acquired Taron under Trdat's son 
Xosrov II as part of the estates bestowed by this king on the Spampet 
Vace after his valiant defeat of the invasion of King Sanesan 38 . 
After the death of the Kat'olikos Sahak I, the Mamikonean also 
inherited the patriarchal holding in Taron in accordance with his 
will 39 . The western border of Taron where it adjoined Hasteank* 
was occupied by the princes Paluni 40 . 

Adjoining Taron were the gawafs of Arsamunik' and Aspakunik', 
which had originally been part of Taron, Paustus still considered 
Arsamunik' a part of Taron, while in the History of Lazar P'arpeci 
it has emerged as a separate district 41 . The name of the district is 
derived from an Arsam, who is probably to be connected with the 
founder of the city of Axsamosata. An inscription found quite 
recently on a column in the Mmrud-dag [Nemrut dagi] mountains 
in eastern Taron, likewise mentions a certain king Axsam 42 , The 
princely dynasty of the Arsamids apparently ruled in antiquity in 
the valley of the Euphrates, and their descendents maintained them- 
selves relatively late in the foothills of the Bingol, i.e. in the region 
which has preserved their name. In ancient Armenian literature 
Arsamunik' appears exlusively as a geographical term; the house of 



TERRITORIAL ANALYSIS OE THE NAXABAR SYSTEM 



245 



the Arsamids had already died out, and by the fifth century its pos- 
sessions were in the hands of the princely house of the Mandakuni 43 , 

The sources have likewise failed to preserve any knowledge of the 
princes of Aspakunik', Like Ars^munik', Aspakunik is merely a 
geographical term and is in fact called « Aspakuneao jor (the valley of 
Aspakxmik')» in the Armenian Geography 43 a . It is one of the mountain 
valleys in the vicinity of Xoyt', and lies on the left bank of the Kara-sn 
(the ancient Mel) in the Khandosh [Ha£re§ daglari] mountains opposite 
Sasun, Xoytf was distinguished from Taron and from the other regions 
of Armenia because of its notoriously barbarous customs and speeches, 
and it formed an autonomous district 44 . 

Another district of Taruberan as important as Taron was Hark', 
with its center at Manazkert 45 ; this region was the hereditary pos- 
session of the Manawazean house 45a , The adjacent districts of Apa- 
hunik', Xorxorunik and Bznunik' had originally also belonged to the 
province of Hark', but had later separated from it as the possession 
of princes bearing the same names. This evolution is evident from 
the National Epic which gives Manawaz, -Xor, and Baz, the putative 
founders of these houses, as related among themselves : the first two 
being the brothers of Hayk, while Baz was the son of Manawaz, and 
ruling house of Apahunik' likewise claiming a HayMd origin 46 , In 
addition to these principalities, Apahunik' also contained the domain 
of the Abrahamean princes 46a , 

The Manawazean disappeared without trace as a result of an inter- 
necine war against the Ordunis in the reign of King Zbsrov II, Both 
houses were exterminated at the order of the king, the lands of the 
Manawazean were granted to bishop Albianos of Manazkert, and those 
of the Orduni to the bishop of Basean 47 , The house of Bznuni came 
to an end at the same time with the death pf prince Databey, who 
was executed for treason. His possessions were confiscated by the 
fisc, and seem to have passed subsequently to the bishop of Bznunik' 48 , 

According to Faustus, Apahunik' lay at the foot of a large mountain 
named Masik', and the village of Aiiorsk' was located in this dis- 
trict 49 , The reference here is not to Mount Ararat [Masis], but to 
the mountain called Ne^ Masik' in the Armenian' Geography, the 
present Sip'an [Siiphan dagi] 50 , BznuniF stretched to the west of 
this mountain along the shore of the lake [Van], with Apahunik' to 
the noTth, and Ahovit to the east. According to the Armenian 
Geography, « ,,, the Aracani [Arsanias] passed through all of Apahunik' 



246 



CHAPTER XI 



to the border of Bznunik' 50a ». The two gawafs were therefore con- 
tiguous, Apahunik' occupied the course of the Arsanias as far as 
its westward tend at Manazkert, while Bznunik' lay south of the river 
all the way to Lake Van. Since Manazkert (the present Malazgirt) 
stood on the border of Apahunik 5 and Hark', it was consequently 
placed by historians now in Apahunik' and now in Hark' 51 . 

The district of Kor was, strictly speaking, a division of Bznunik 5 , 
just as AHovit was a part of Apahunik', Below Manazkert the 
Kor su empties into the Arsanias on the left out of the small lake of 
Bulam; ancient Kor was probably to be found here, Zorxofunik' 
lay in the same district now called Bulane^ [Bulanik], between Bznunik' 
and Kor, in the vicinity of Lake Nazik, The positions of Kor and 
Xor^ofunik' remain doubtful, since there is no information about them 
in the Armenian Geography 52 , 

Of the remaining districts, Erewark' lay along the southern shore 
of the Lake [Van] in the neighbourhood of Aljnik 5 . Mardahk' took 
in the sources of the Araxes on the northern slopes of the Bingol. 
East of it stretched in succession Dasnawork 5 , Towraca-tap', and 
Dalaf, above Varaznunik' and Hark', which stretched as far as the 
Arsanias. According to this description, Dasnawork', in the valley of 
the Araxes, was next to MardaHk', Towaraca-tap 5 corresponded to the 
present Karayazi, Dalaf — to the valley of the Elmali, Varaznunik 5 — 
to -Xhus [Hinis] on the upper course of the tributary of the Arsanias 
which flows out of the Bingol, and Hark' next to it stretched along 
the Arsanias 53 . The first three contiguous districts apparently 
formed the domain of the bishop of Mardalik', historical literature 
knows of no princes from these districts, whereas Varaznunik' was 
a principality 54 , 

The land of Vaspurakan, the third of the central provinces, consisted 
of 35 gawafs according to the Armenian Geography, though only 32 
names are listed 54a , There are more gawafs in it alone than in Ayrarat 
and Taruberan taken together. The author of the Geography is not 
equally informed as to the position of the districts; the first 18 are 
more familiar to him than the rest. It must be noted that the descrip- 
tion of the Armenian provinces found in the Georgaphy is generally 
uneven. Of the 15 provinces mentioned, only Ayrarat, Taruberan, 
Armenia IV, and Tayk 5 are described in detail, the other provinces 
are merely listed together with their subdivisions, without any indi- 
cation of the position of the latter. This very unevenness can be 



\. 



TEKRITOHIAL ANALYSIS OF THE NAXAMAR SYSTEM 



247 



seen in the description of Vaspurakan. The double treatment is 
evidently to be explained by the fact that the author of the Geography 
was not equally familiar with all the places he described. The central 
provinces are presented incomparably better than the borderlands, 
In the short version of the Geography, the uneveness of the description 
is not evident because only the names of the districts in each province 
are given without the explanatory comment, 

In the description of the first 18 gawafs, the Armenian Geography 
keeps to a definite pattern. The listing is from south to north in 
three parallel territorial strips 54 b . The first begins with K stunik' in the 
south and ends with Garni, adjacent to Kogovit at the base of Mt, 
Ararat, in the north : 

Kstunik* 

Tosp 

BodordF [Bodunik'j 

Arcisahovit 

Afberani 

Garni up to Kogovit 54c , 

The second strip parallels the first on the eastern side : 

Buzunik' 

Af noy-otn 55 

Anjewaeik' 

Trpatunik' 

Erwandunik' 

Mardastan 

Artaz up to Kogovit, 

The third strip still further to the east but keeping the same pattern : 

Ake 

Great Albak 

T'onfawan 

Cowars-rot up to the Araxes. 



The position of these districts is known for the most part, Estunik' 
is the south-eastern shore of Lake 'Van, with its center at Ostan. 
Tosp is the region around the city of Van, Arcisahovit, the valley of 
the Arcisak, lay north of Van in the region of the small Lake Arcak 



248 



CHAPTER XI 



[Ercek golu] and of the Marmet xiver which flows past it 56 , Aberan, 
more correctly Afberani lay in the northern corner of the lake, along 
the Bendimahi cayi, and was later called after its main city Berkri 
[Bargiri = Muradiye], The upper courses of the same river extended 
to Garni at the foot of the T'ondrak [Tendiirek dagi] and Gure moun- 
tains, which separated Garni from Kogovit, i.e. from the district of 
Bayazet' 57 . 

The first districts of the second row ; Buzunik' and Anjewacik' lay 
on the upper course of the Bohtan cayi or Eastern Tigris, where the 
present district of Norduz is to be found; furthermore, Anjewacik' 
occupied the vicinity of the city of Kangowar, now Kangever and 
of the monastery of Hogeac vank\ South of it lay Buzunik' and 
Afnoyotn, while to the north lay Trpatunik' and Erwandunik' on 
the upper courses of the Xosab [Ho§ap] river whose valley was known 
as the Armenia Gorge, Hay 09 Jor 58 , Further to the north stretched 
Mardastan, from Lake Ercek to Artaz, According to Xorenaci, Artaz 
lay to the south-east of Masis-Ararat, and was called Sawarsan or 
Sawarsakan in ancient times. The village of Ehnd, where Vahan 
Mamikonean's camp had stood at the beginning of his negotiations 
with Ni^or [Vsnaspdat], was in Artaz; it still exists, a few miles 
from Maku and is now called Erind [Bint], Haysun (the ancient 
Haciwn) is also found in that district, The position of Artaz, with 
Maku as its center, along the Ak-cay (the ancient Tlmut), can be 
determined from these indications 59 . 

The third territorial strip runs from the sources of the Zab to the 
Araxes. Albak is the present [1908] haza of Elbak [Baskale], with 
■the town of Baskale. South of it lay Ake, and to the north, Anja^i- 
jor, on the Kotur cayi (the ancient Kotor), with the city bearing the 
same name as the river. Next to this district and west of Kotor, 
lay T'onfawan, with the fortress Nkan, now called Nagan. Sewan, 
another historical fortress, which stood in the gorge of the Limb 
[Lumb], was also found here, Sewan is to be identified with the 
present Seyvan on the Mehmetik river, which flows into Lake Ercek ; 
the village of Lim, which is unquestionably to be identified with the 
ancient Limb, is still found in this district 60 . The plain of Xei [Her], 
Xerakan dast, stretched beyond the Anja^ijor, and still further lay 
Cowaf s-rot, where stood Marakan, the winter residence of the princes 
of Vaspurakan, Xerakan is to be identified with the district 
of the city of Xer, now Xoy, and the lower course of the Kizil cayi 



TERRITORIAL ANALYSIS OF THE NAXABAJR SYSTEM 



249 



(the ancient Karmir get) is now called Xar da St, The village of Mara- 
kend, with the same position as Marakan is found on the riverbank 61 , 

If they were really part of the original version of the Armenian 
Geography, the questionable 13 districts would have formed a fourth 
territorial strip, with the end-points Kicnrnk' and Marand. The 
settlement of Kurucan, north of Kotor, may be identified with Krcu- 
nik', and Marand is a well known city in Azerbaijan 62 , The remaining 
districts should consequently have been found between Xoy and the 
Araxes, But, of the IS districts given, Gokank' and Artasesean are 
found in the basin of Lake Van ; the first one is to be connected with 
the ancient Gokank' in ftstunik', and the second with the settlement 
of Artasesean, not far from Van 63 , It is evident, therefore, that 
not all 13 districts lay in the area of -Xby, and consequently the hypo- 
thesis that they formed the fourth territorial strip must be abandoned, 
As a result, our doubt as to the authenticity of their inclusion in the 
list is all the greater 64 , In fact the questionable districts consist for 
the most part of small sections or perhaps subdivisions of familiar 
districts, and, consequently, are of little importance for our study. 

Thomas Arcruni, the historian of Vaspurakan par excellence, who 
was very well informed about the topography of this province, gives 
a list of the districts of Vaspurakan when he speaks of the division of 
lands between the two brothers Gagik and Gurgen [Arcruni], and at 
the same time divides these districts into two groups : north-western 
and south-eastern. Concerning the first group, the share of the older 
brother Gagik, the historian says that, « .,. these famous districts once 
made up the possessions of the so-called Prince Mardpet 65 », The 
indication found here that part of Vaspurakan bore in antiquity the 
name of the Mardpet, whether or not this part really coincides with 
the share of Gagik, which is merely the result of an accidental par- 
tition in the ninth century, is very valuable. Here, then, is the land of 
Mardpetakan, This name is at times alternately used with the term 
Sep'akan, as geographically synonymous. Thus one and the same 
person is called bishop of either Mardpetakan or Sep'akan 66 , Sep'- 
akan, from sepuhr - akan is a relatively newer form of the ancient 
va-$puhr~akan. Their simultaneous use in documents must be 
explained by the fact that they are not eqxrivalent geographical units ; 
one of them is related to the other as a part to the whole 68a . 

As we shall see, Mard-39^ originally meant the head of the Mardians, 
that is to say, of the people who inhabited Mard-asfcm, Later, 



250 



CHAPTER XI 



Mardpet became a family title similar to Aspet, Ma-^a%, and Sparapet 
With the suppression of the princely house of the Maxds, this title 
passed to the neighboring house of the Axexuni, while Maxdpetakan 
became merely a geographical term 66lD . The Mardpet princes had been 
great enemies of the Axexuni their neighbours, Under King Tiran 
[Tigran III], one of the Mardpets attempted to exterminate the entire 
Arcruni house ; only one infant, Sawasp, who later inherited his father's 
possessions, was saved from the slaughter. The Arcruni were slow 
to recover from this blow, and, as Paustus remarks, they took no 
part in the affairs of the country for a long time 67 , It is true that 
Savasp avenged himself on the man guilty of his family's catastrophe 
by. killing him 68 , but the feuds and mutual hatred continued and 
resulted in the destruction of the Mardpefs house, so that it had 
ceased to exist by the fifth century. The victors inherited the Mard- 
pefs title together with his possessions. In fact, the historian Lazar 
P'arpeci calls Prince Mihr-Sapuh, the Areruni's representative in the 
rebellion of the fifth century, Prince Arcruru and Mardpet; at the 
same point in his History, Ehse merely gives him the name of Prince 
Arcruni 69 , According to Lazar, a « ,,, detachment of military ca- 
valry from Mardpetakan », among others, went together with Vardan 
Mamikonean, to meet the Persians in Albania, while in the corres- 
ponding place in Ehse, mutatis mutandis, appears Prince Nexsapuh, i.e. 
the same Mihr-Sapuh Arcruni, presumably at the head of the cavalry 
from Mardpetakan 70 , Prom this equivalence, and from the definite 
indication of Laaar, it is evident that Mardpet had become the title 
of the Arcruni house. 

Prom ancient times, Albak was held to be the native district of the 
Arcruni, Their residence or ostan was Hadamakert 71 , probably on 
the site of the modern town of Baskale, Lazar and Eh§e distinguish 
two branches of the Arcruni family, Ner- or Mihr-Sapuh represented 
one of these in the fifth century rebellipn, while the representative 
of the other was Aprsam, It is reasonable to suppose that one of the 
branches ruled Albak, and the other Mardpetakan, the inheritance of 
the Mardpet 72 . 

The vast province of Vaspurakan belonged primarily to three or 
four noble families : the. Arcruni in Albak, the Mardpets {sc, Arcruni) 
in Mardpetakan, the Estuni in the district of the same name and 
probably in Tosp 73 , and the Amatuni in Axtaz 74 . Around these 
were grouped the smaller principalities ; Anjewacik', Trpatunik', 



TERRITORIAL ANALYSIS OF THE NAXARAR SYSTEM 



251 



Erwandunik', probably Varjawunik', GokanF, Artaseseank', and 
possibly also Palunik', and TruniF 75 , The princes Eikeuni and 
Anja^eei (= Anja^) were in the sphere of influence of the Mardpet, 
South of Albak were the districts of Ake and Buzunik, The Boduni 
were along the shore of Lake Van, and the Gnuni in Berkri 75a , The 
little known principalities of Bak'an, Patsparunik', Gazrikan, Vizanunik' 
or Vamunik', Cwarsean, Taygrean and others probably lay for the 
most part along the border of Atrpatakan 76 , These families could 
not lay claim to great antiquity; most of them belong to the period 
of the kingdom of Vaspurakan, We must also note that we meet 
families definitely known to have lived in portions of Armenia lying 
outside Vaspurakan, among the princes of Vaspurakan listed by 
Thomas Axeruni 77 . These cases should be explained in terms of 
the growing political importance of Vaspurakan which drew to it 
many princely families which sought the protection of its rulers, the 
Arcruni princes. 



XII 



THE NAXABAB SYSTEM AND THE CHUKCH 

I. The Church in the naxarar system — National traditions on the numher of bishops 
and ecclesiastical provinces in the period of the Illuminator — Ahsenee of foundation 
for these traditions — The evidence of TJxtanes TJfhaeei and of the Arabian version 
of Agat'angelos — Reliable data found in the Conciliar Acts of 451, 505, and 555 — 
Their collation and analysis — Composition of the clergy in the mid- VI century — - 
The Conciliar Lists of 607 and 644 — Analysis of the list of bishops in "Cutanea, compa- 
rison of this list -with the signatures of the Council of 726 — Analysis of the list of 
bishops in the Arabian Agat'angelos — Its confessional background, the list cannot 
be earlier than the beginning of the VII century. 

II. Factual information on Armenian ecclesiastical hierarchy: the Aibianids and the 
Gregorids — The two families as representatives of different Christian traditions: 
the southern Edessene, and the northern Caesarean — Historical elements in the rele- 
vant legends — Rivalry of the two families — Policy of the Church — Adhesion 
of the Gregorids to the Imperial tradition — Structure of the Imperial Church — Its 
relation to the administrative system — Metropolitan and patriarchal authority and 
their interrelation — The hierarchical tie between the Armenian and Imperial Churches 
— The breaking of this tie — Nationalization of the Armenian Church — Its naxarar 
features. 



An examination of the nayarw system is impossible without a study 
of ecclesiastical organization, since the Armenian ChuTeh developing 
in a nax^rar setting inevitably reflected its influence. Even though 
this fact is obvious and fully understandable a priori, it would still 
seem worthy of further investigation, yet scholars interested in the 
history of the Armenian Church either ignore this aspect ox dismiss 
it with generalizations thus taking a position totally divorced from 
Armenian reality in their attempt to clarify the history of the Church's 
origin and nationalization. Consequently, many fundamental problems 
remain unsolved, despite the numerous studies dedicated to Armenian 
ecclesiastical history. Even such a seemingly simple feature as the 
national character of the Church is not yet beyond doubt, Eor one 
scholar, 



254 



CHAPTER XII 



,., the Armenian Church from its very origin was an inde- 
pendent national church in a completely different political 
and cultural setting from that of the great Orthodox Imperial 
Church 1 . 



Another claims, on the contrary, that the Armenian Church in its 
« ...self-determination, over-estimated and over-emphasized its own 
national character x a . Such diametrically opposed views also exist on 
other aspects of Armenian ecclesiastical life. They are for the most 
part to be attributed to a disregard of proper historical method in 
the study of problems connected with place and time. Before at- 
tempting any synthesis, it will be indispensable to determine with 
precision the historical period and the section of Armenia to which 
the evidence discussed is relevant. 

Historically, Armenia was neither a cultural nor a political unit, 
and the fate of each section must be considered in its own surroundings. 
The Armenian Church in one part of Armenia is not at all the same as 
in another part, as a consequence of the political setting in which it 
found itself. Whenever the political aspect of the country was 
altered, the Church changed accordingly to adapt to the new con- 
ditions ai \ Nor did literary documents escape the common lot in this 
changing pattern of life. Historical literature underwent repeated 
revisions and re-workings to bring it into agreement with the spirit 
and demands of the period. Consequently, the literary heritage, too, 
is by no means free from tendentious colouring both in the presentation 
of material and in the fusion of the true with the false. In order to 
uncover the true development of the Church, therefore, it will be 
necessary first of all to demolish these tendentious structures, some 
of them ancient and long-standing, and to present the actual facts, 
free from alterations and accretions. At present, we are interested 
exclusively in the external structure of the Church, and in its hierarchi- 
cal system. Leaving aside the internal life of the Church, and all 
that relates to its dogmatic aspect, we shall examine it only externally ; 
we select, so to speak the architecture of its organization. That is to 
say the characteristic of this organization, as against those of the 
Church in general, which turned it into a national institution. 

Agat'angelos, the historian par excellence of the life and preaching 
of St, Gregory the Illuminator, affirms that the founder of the Armenian 
Church ordained more than four hundred bishops for the various 
localities of Armenia 2 , In another place, however, the same historian 



THE NAXABAB SYSTEM AND THE CHURCH 



255 



says that there were only twelve persons found worthy of the episcopal 
dignity and ordained by the IUuminator ; these he lists by name 3 , 
The discrepancy here is due to the compilatory nature of Agat'angelos' 
work and must be attributed to the sources from which our version 
of the Life of the Illuminator has been drawn, At the end of the 
listing of the 12 bishops by name, there is a rather unclear note, 
whose general sense is that even with the best intentions it is im- 
possible to give the names of the remaining bishops, and whose purpose 
is to hnk the two statements in some way and reconcile their dis- 
crepancy 4 . 

The words of Agat'angelos about the 400 bishops are repeated by 
the historian "Octanes of Urha, However, since U^tanes possessed 
from another source the information that the Illuminator had esta- 
tabhshed only 30 episcopal sees, he stated, when giving the list of 
30 bishoprics that St. Gregory had also ordained 370 additional 
bishops. Moreover, again with the intention of smoothing out contra- 
dictions, he seems to have accepted a difference between bishops 
with and without sees 5 . According to yet another historian, the 
Illuminator had established not 30 but 36 sees. Stephen Orbelean 
relates that 

... Gregory established the ecclesiastical hierarchy (i.e. the 
gahs and jpatiw of the bishops) ; he allowed 36 bishops having 
thrones and gold embroidered cushions to sit with him, eighteen 
on the right and eighteen on the left. [The first on the right 
was the bishop of Hark'] ; the first place on the left was alloted 
to the bishop of Basean, whereas he alloted the seventh place 
on the right to the bishop of Siwnik\ And all this is attested 
by Samuel Kamrjajoreci 6 . 



The author to whom Orbelean refers, was the Abbot of the monastery 
of Kamurjajor in Arsarunik', the philosopher Samuel of whom Asohk 
also speaks, calling him « ...well versed in the Holy Scriptures and in 
psalmody 7 ». Samuel lived at the end of the tenth century, and 
was a contemporary of U^anes and of the Kat'ohkos Xacik I at whose 
order U^tanes wrote his History 8 . Since -Xacik, before his elevation 
to the patriarchate, had been bishop of Arsarunik', that is to say of 
the province in which the monastery of Kamurjajor was located, it 
is reasonable to suppose that "Octanes and Samuel derived their 
information as to the number of bishops from the same source, namely 
from the tradition of the community of Kamurjajor. As we shall 



256 



CHAPTEE. XII 



see presently, the seventh place in "Octanes' list is assigned to the 
bishop of Siwnik', and this agrees with what Orbelean has to say 
about Samuel's list. However, the latter gives the bishop of Basean 
in first place, rather than the bishop of Hark* listed in "Octanes : 
furthermore, one list has 30 bishops and the other 36. Should we 
not attribute these variants from U^tanes' work simply to the particu- 
lar manuscript of Samuel's work available to Orbelean? Samuel's 
list probably did not differ from that of "Octanes, As for the division 
into right and left, it should be considered Orbelean's own invention, 
since he says the same thing about the princes, 

The story of the existence of 400 bishops must naturally be rejected 
in Mo. It is found together with the account of 400 princely gaits 
which we have already shown to be without any historical value. 
It is noteworthy that the sentence concerning the 400 bishops is missing 
from the Arabic version of Agat'angelos. Evidently it was put into 
the Armenian version after the story of the 400 princes, based on an 
incorrect interpretation of Zenob G-lak, had been elaborated 9 . 
This circumstance brings the last version of the Armenian Agat'angelos 
down to a period later than that of the hypothetical Zenob 10 . 

Not only is the figure 400 fictitious, but the far more modest number 
given by "Octanes is also unsuported by evidence n , Among literary 
sources, the most valuable evidence is to be found in "Octanes and 
in the data of the Arabic version of Agat'angelos, since it is not com- 
posed of mere assertions but is supported by the relevant 
material, "Octanes lists the 30 episcopal sees supposedly established 
by the Illuminator. Not counting the Kat'oKkos, there are 30 bishops 
in all : 



1 The bishop of Hark' 


16 The 


bishop of Apahunik s 


» » » Ostan 




» 


» > 


> Arsarunik* 


» » » Tayk' 




» 


» > 


> Gnunik' 


» » » Marda£k< 




)> 


» > 


> Goltn 


5 » » » Arsamunik' 


20 


» 


» > 


> Gardman 


» » » Arerunik' 




» 


» > 


> Ake 


» » » Siwnik' 




» 


» > 


> Bazunik' 


» » » EstuniF 




» 


» ) 


> Kotak [Erutak] 


» » » Mokk' 




» 


» > 


> Syria 


» » » AmatuniF 


25 


» 


» > 


> Anjewaeik* 


» » » Basean 




» 


» > 


> Palunik' 


» » » Mamikoneank' 




» 


» > 


> Mehnunik* 


» » » Bagrewand 




» 


» > 


> El 


» )> » Xor^ofunik' 




» 


» > 


> Zarehawan 


5 » » » Vanand 


30 


» 


» > 


> the Other Syria 12 



THE NAXABAB SYSTEM AND THE CHUKCH 



257 



The Arabic version of Agat'angeios brings fundamental additions 
to the Armenian text, giving more precise limits for the spread of 
Christianity under the Illuminator, The Armenian Agat'angeios is 
satisfied with the general indication that Christianity, 

.,, spread out over all of Armenia from end to end. From 
the city of Satala to Xaltk, to Kalar], to the far limits 
of Moschia, to the gate of the Alans, to the border of 
Kaspe in Paytakaran, a city of the kingdom of Armenia, 
And from Amida, past the city of Mcbin, alongside the border 
of Syria, to the countries of Norsirakan and Korduk', to the 
grim country of Media, to the domain of the prince of 
Mahk'ert, and all the way to Atrpatakan 13 . 

The additions brought here by the Arabic version consist in the listing 
by name of the countries and districts where bishoprics had been 
established. St. Gregory sent bishops not only to the Armenian 
lands, but also to Georgia, to the country of the Durzuks, and of the 
Alans, Ibir-b-z-^ua was sent to Georgia, Sophrordos to Abkhazia, 
and Thomas to Albania. Furthermore, 

... he sent bishops to the countries of Angeltun, to Atjnik' 14 
to Greater Cop f k', to Lesser Cop'k', to Hasteank', to Siwnik' 
to Mokk', and to Mardpetakan ; and thus to eveij place where 
the ruler was well disposed 14a . 

Gregory also hastened to send bishops to the remaining provinces 
of Armenia 14b : 



Albianos — to Bagrewand, and to all the inhabitants of the 

the banks of the Euphrates 
Euthalios — to Basean 
Bassos — to Kol 15 
Movses — to Ekeleae and Derjan 
Eusebios — to Daranalik' 
John — to Karin 
Habib — to Sper 

Albios [Agapios] — to Taron and Bznunik' 
Artites — to the country of the Mafyaz 
Arsikios — to Sirak 
Antiochos — to Korduk* 
Tyrichios — to Atrpatakan 
Kyriakos — to Arsamunik', 



d^ 



258 



CHAPTER XII 



In spite of its value, this evidence cannot be used without a preliminary 
discussion because of the controversial state of the problem of the 
origin and versions of Agat'angelos, We have, however, reliable 
sources which describe the state of the Armenian hierarchy in the 
fifth and sixth centuries, fairly clearly, Starting from the situation 
discribed in them, we may go on to draw conclusions about the fourth 
century, which was the first century of the Armenian Church ; it will 
then be possible to determine the credibility and reliability of Agat'- 
angelos' and "Octanes' statements on the composition of the Church 
hierarchy in the time of the Illuminator. 

The surviving Conciliar Acts are our main source for the description 
of the hierarchy in the fifth and sixth centuries. We have chosen 
here three important Councils : that of Artasat, on the eve of the 
Armenian rebellion in 451, that of Dwin, under the Kat'ohkos Babgen I 
in 505, and the Second Council of Dwin, under the Kat'ohkosNersesII 
in 555, The representatives of the following sees or ecclesiastical 
provinces were present at these councils ; 







Council of 450 17 


Council of 505 18 


Council of 555 19 








Kat'oJikos 


Kat'olikos 


Kat 


'oKkos 




f 




Joseph [Yovsep'] 


Bat 


>gen 


Nerses II 






1 


Bp, of Ayrarat 
» » Siwnik* 


Bp, 


of Ayrarat 

» Mamikoneank' 


Bp. 


of Ayrarat 
» Taron 


1 






» » Arerunik' 




» Mardpetakan 




» Sep'akan 




Ir- 




» » Taron 




» Hark' 




» Hark 4 






5 


» » Estunik ( 




» Bznunik* 




» Bagrewand 


5 






» » Manjkert 
» » Bagrewand 
» » BznuniF 
» » Basean 




» Basean 
» Arsarunik 4 
» Zorxorunik* 
» Tayk' 




» Basean 
» MardaKF 
» Arsamunik' 
» Siwnik* 






10 


» » Mardastan 
» » Vanand 




» Armmuni¥ 
» Estunik' 




» Arcrunik 5 
» (BmuniF) 


10 














[Bagrewand] 






» » MokF 




» Mokk< ^ 




» Xor^ofunik e 








» » Anjewacik* 




» Arcrunik* 




» ApahuniF ■ 




- / : _ -: 




» » TayF 




» Amatunik* 




» Syria 






15 


» » Tar(u)beran 




» Paluni¥ 




» Vanand 


15 


V y - " ~ . 




» » Mananali 




» GnuniF 




» ArsaruniF 




'"■'.* ■• 




» » [MardaKF] 




» Zarehawan 




» Palunik' 




£'■%- 




» » Amatunik' 




» Tmori¥ 




» Goltn 




f l '"' ' * 


19 


» » Apahunik' 




» Anjewacik* 




» Mehenunik* 












» MehnuniF 




» Amatunik* 


20 






s 




» (Siwnik 4 ) 




» Rstunik e 





1 V 



THE NAXARAR SYSTEM AND THE CHURCH 



259 



(MardaliF) 

(Vanand) 

(Apahunik 4 ) 



» Mokk< 

» Anjewacik 4 

» Ake 

» Zarehawan 

» Bznuni¥ 

» (Tayk') 

» (Gnunik') 



25 



28 



These lists give us an understanding of Armenian hierarchy and of 
its evolution. The most complete of them is the list of 555. The 
increase in ecclesiastical provinces manifests itself partly through 
the division of already existing sees, and partly through the appearance 
of new ones. In the mid-fifth century there were 18 episcopal sees; 
these axe all found in the list of 505, with a few purely superficial 
variations. The Mardastan of 450 is called Mardpetakan in the list 
of 505, and Sep'akan and Mardpetakan in that of 555 ; we have already 
seen that these expressions are equivalent 19 a . The see of Manazkert is 
also called that of Hark', according to the name of the province in 
which Manazkert was located. Bagrewand and Arsarumk' taken 
together formed one eparchy whose head bore the title of bishop of 
Bagrewand in tazar [450], but of bishop of Arsarunik' in the list of 
505 ; by 555 Arsarunik' had been removed to as a separate see. The 
adjacent provinces of Apahunik' and Gnunik' were probably in the 
same situation, but if we admit that the bishop of Apahunik' was 
absent from the Council of 505, we must assign the appearance of 
the bishop of Gnunik' to that date. 

The bishop of Arberan, or more correctly Taruberan, as he is called 
by EHse, cannot be found in the later lists, but we have several extTa 
bishops in them as against those found in tazar. One of these un- 
questionably corresponds to the bishop of Taruberan, and in our 
opinion, it is the bishop of Xorxofunik'. The princes Zorxofuni,* 
the noble Mafyazs, whose representative at the Council of 505 bore 
the imposing title of Mafyaz of Armenia, were among the naxarars of 
oldest lineage, and it is unlikely that they lacked a bishop of their 
own in the fifth century. Although we cannot accept as correct the 
statement found in the Arabic version of Agat'angelos, that the creation 
of the bishopric of the Xor^ofuni dated back to the time of the Il- 
luminator, it testifies nevertheless to the antiquity of this see 19fe . We 
believe that Taruberan was the geographical term used for the province 
which was otherwise called Xor^orunik', after the name of its ruling 



260 



CHAPTER XII 



house, Later, the term Taruberan acquired a wider connotation as 
the name of one of the fifteen provinces of Armenia ; at that time it 
was composed of Xor^oihinik' and of the districts surrounding it, 
Taron and Bagrewand 19c , 

In the provinces of Armenia bordering on the Empire, ecclesiastical 
authority was apportioned in various ways. In 450 there were bishops 
in Vanand, Basean, and MardaUk', while in 505 two bishops from 
Basean and one from Arsamunik' make their appearance, without 
any mention of the bishops of Yanand and Mardalik\ Basean had 
formerly included Vanand, and this is the sense in which it is used 
here; apparently one of the two bishops was meant for Vanand, 
As for the bishops of Aisamunik' and Mardahk*, since they alternate 
(i.e, where one is listed the other is absent), we might conclude that 
the two provinces were under the jurisdiction of one bishop who 
correspondingly bore two names, but their geographical positions 
precludes this hypothesis. Divided from each other by the moun- 
tainous mass of the Bingol-Srmanc, one of them adjoined Basean and 
the other Taron, According to "Faustus, Arsamunik' was part of 
Taron, and it unquestionably was subordinated to TaTon in ecclesi- 
astical matters 194 . Under Kat'oKkos John II Gabelean the successor 
of Nerses II, MaTdahk' was listed as part of the bishopric of Basean, 
Under NeTses II, the bishop of Basean had been called Gregory, and 
that of Mardahk 5 , Nerses, but after Nerses II both provinces are found 
under the authority of a bishop Ners(es). This latter is unquestionably 
the bishop of MardaHk' to whom Basean had passed after the death 
of its bishop, Gregory 20 , The reverse occurred in 505. The absence 
of the bishop of MardaHF from the Council held that year is to be 
explained by the fact that his district, in its turn, was temporarily 
subject to the bishop of Basean Just as Mardahk' emerged out of 
^Basean, so Arsamunik' separated itself from Taron, In 555 both 
districts are represented at the Council together with the* sees of 
Basean and- Taron. 

Thus by 505 five new bishoprics had been added to the 18 found 
in 450, namely the bishops of Arsamunik 3 , Gnunik', Palunik', Zare- 
hawan, Mehenunik', and Tmorik', the last of whom corresponds to 
the bishop of the Orthodox Syrians in the list of 555, By 555, the 
circle of bishops had been increased by four more eparchies, Gottn, 
Ake, Binunik', and Axsarunik', which had replaced Bagrewand in 
505, Consequently, by the middle of the sixth century, the Armenian 
Church included up to 27 episcopal sees. 



Q^- 



***+tw&&^ 



THE NAXARAB SYSTEM AND THE CHURCH 



261 



It is rather difficult to trace the development of the ecclesiastical 
hierarchy any further, for with the end of the sixth century we enter 
the period of dogmatic unrest and dissentions, the country was split 
into factions, and as a result, the Councils of the period do not represent 
the entire ecclesiastical hierarchy because many of the Church re- 
presentatives refrained from attended them, for one reason or another. 
Ecclesiastical quarrels persisted right up to the beginning of the eighth 
century, and several Councils were held for a variety of reasons during 
this period : in 607, 644, and 726 20a , John II Gabelean ascended the 
patriarchal throne after Nerses II, and was succeeded in turn by 
Movses II [574-604]. After Movses' death, a Council was summoned 
for the purpose of electing his successor as well as dealing with other 
matters. According to one testimony, 50 bishops and 390 priests 
were presumably present at this Council, but the names of only 12 
bishops are given 21 , The election of the kat'ohkos did not take 
place, and the bishops consequently assembled once more at the 
request of Smbat Bagratuni, Marzjpan of Yrkan [Hyrcania], and elected 
Abraham I bishop of Kstunik'. On this occasion, the same bishops 
who had attended the first Council were again present, except fore 
Theodore, bishop of Mardpetakan, and John, bishop of Arcrunik' aa . 
Having ascended the patriarchal throne, the newly elected kat'oHkos 
again called a new Council in 607, primarily to assure himself of ths 
orthodoxy of the bishops from the provinces recently transferred to 
the Empire according to the terms of the treaty of 591 23 , The names 
of only 15 bishops can be found in the Acts of this Council a4 ; among 
them, the names of Yohannik, bishop of EH, and Thaddeus, bishop 
of Aini, the representatives of Armenian sees on the border of Atr- 
patakan occur for the first time 25 , 

The Council of Dwin of 644, under the Kat'olikos Nerses III and 
the Emperor Constantine [Constans II], likewise failed to attract many 
bishops. Only 16 bishops appeared at the summons of the Emperor, 
most of them from the provinces won back under Maurice, which 
had remained definitely in the Empire after 628, All of them repre- 
sented sees already familiar to us 26 , 

The study of Conciliar material leads us to a conclusion which may 
serve as a basis for evaluating the information relating to earlier 
times. By now it should be altogether clear that the account from 
the monastery of Kamurfajor as to the foundation of 30 or 36 bishoprics 
by the Illuminator is very far from the truth. If we look more closely 



262 



CHAPTER XII 



at this List of sees, which, we have already cited as given by U^anes, 
and compare it with the results obtained from our analysis, of the 
Conoiliar Lists, we find that the only two unf amUiar bishops in "Octanes' 
list are, strictly speaking, those of Eotak 27 and Syria II, It is true 
that of the bishoprics already studied by us, the one from Mardpetakan 
seems to be missing, but it is, in fact included in the see which is 
called Ostan by "Octanes, At the Council of 644, as we have seen, 
both provinces had been represented by a single bishop, who was 
called the bishop of Ostan and Mardpetakan 37a , Similaxly the see 
of Afni found in the 607 list unquestionably belongs in the neigh- 
bouring see of Eli. Of the two unfamiEar sees, that of Eotak already 
existed in the second half of the seventh century, since we know that 
Sahak III of Jorap'or was bishop of Eotak before his elevation to 
the patriarchal throne in 680 38 , As for the bishop of Syria II, he is 
mentioned fox the first time in the Kamurjajox List. Consequently, 
ibis List which was identified with the founder of the Armenian Chxirch 
cannot be assigned to a period eariier than the end of the seventh 
century. 

A few particulars of "Octanes' List support the hypothesis that it 
is in fact the protocol of a Council which took place in Armenia 
after the -date just given. One of the most important and popular 
of the Councils of the subsequent period was that held at Manazkext 
in 726 under the Kat'oKkos John IY of Ojun, In the Armenian 
Acts of this Council, the names of only eight of the more important 
participants axe given. Fortunately, a complete listing of the signature 
has been preserved in Syrian literature, specifically in the Chronicle 
of the Jacobite Patriarch, Michael the Syrian, We give here his Est 
in extenso as well as the paraEel signatures which have survived in 
the Armenian version : 



Syrian 



Armenian 



Iwannes, Kat'olikos of Armenia 
1 Halphai, ep, Arkiws Alp'eos of Hark 1 

Theodoros, ep, Aram 

Sahak, ep, Mamikonean 

(Esqw), ep, Basean 
5 Sargis, ep, Ditpis 

Theodorios ep. Beznunis 

Theodoros of Asamunis 

Grigorios of Asarunis 



T'adeos of Ostan 

Sahak Mamikonean 

Yesu of Basen 

Sargis of Tayk c 

Teodoros of BznuniF 

Grigoris, chorep, of Ajsarnnik* 



TEE NAXARAH SYSTEM AND THE CHURCH 

Nwzwn of Asibw[gn] 
10 Habel of Amatunis 

David of Erestunis 

Iowsep of Artsranis 

Grigor of Wanand 

Narkisos of Khorkhorunis 
15 Esayi of Golt'n 

Iwanes of Gnunis 

Gorgi of Kotakay 

Iowsep of Bakratunis 

Mik c ayel of Bagrewand 
20 Eremia of Apahunis 

Solomon of Mrina 

Gabriel of Arzon 

Khosrov, doctor 

David of Suphrin 
25 Solomon, monk of Mak'enis 

Kaphael, monk 

Simeon, doctor 

Iwanes, chorepiscopos 

Grigor of Taraun 
30 Sahak, chorepiscopos of Matnis 

Sargis, ep. Sanasnaye, and other priests and monks 29 . 



263 



In plus of the kat'ohkos, 31 representatives of various eparchies 
attended the Council, whereas in "Octanes' List we count 30 eparchies, 
the one additional representative at the Council of 726 being that of 
the bishop of Asibw(r)gn, i.e, Sep'ukan or Sep'akan, otherwise known 
as Mardpetakan. At the Council of 644 this province had shared 
one representative with the Ostan ; evidently "Octanes likewise included 
it in the Ostan 30 . In exactly the same fashion, Bznunik c and Taron 
also formed one eparchy, which was the see of the Mamikonean as 
given by "Octanes, But in 726, the bishops of Bznunik* and of the 
Mamikonean are listed separately. It is very likely that Bznunik' was 
given instead of Bznunik', or more correctly Bazunik c , and that the 
representative of this see, bishop Theodore, was the Theodore called 
Bazen by Stephen the Philosopher bishop of Siwnik' 31 , 

If we compare the list of bishops present at the Council of 726 
with the list of eparchies in U^tanes, we find that of 30 names, 20 
coincide completely, while 10 differ. There are numerous errors in 
the 726 list ; some of which have been corrected satisfactorily by its 
editor, while others can easily be given a proper reading through 
comparison with "Octanes' Thus Mrina or Mrdin, Kazwn [sic], Suphrin, 



264 



CHAPTER XII 



and Mainis are distorted and incorrect forms of Mardalik', Zarehawan, 
Palunik', and Mehnunik\ The learned monk Khosrov, called the 
doctor (i.e. vardajpei) of Armenia, attended the Council as the repre- 
sentative of Mokk' (or Anjewacik'), while Sargis, bishop of Sanasnaye, 
corresponds to the bishop of Syria in Uxtanes' List* 1 *. 

In the Conciliar List of 726 the bishops of Ake, Anjewacik', and 
Syria = Tmorik' are missing. The priests Solomon, Simeon, and 
I wanes, who are listed among the participants without indication of 
place, probably represented these three eparchies, The representatives 
of Gardman and Eli = Arni are also missing, but bishops of Bhrtunis 
(var. Sriunis) and Taron are listed. Since the bishop of the Bagratuni 
was also bishop of Bagrewand, but the latter had already been listed 
and similarly the bishop of Taron was bishop of the Mamikonean, 
and had likewise been listed earlier, should we not see in the names of 
the bishops of Bkrtunis and Taron distorted forms of b- Gardman and 
t-Aran ? 

The corrections offered here are based in the main on a comparison 
of the two lists. The Armenian names have been so distorted in the 
Syrian version that a variety of readings is possible, and our inter- 
pretations should not be seen as unduly strained. A demonstration 
of complete agreement between the two lists is not actually demanded 
by the fundamental aspect of our problem, nor is there any need to 
insist upon it. It is sufficient for us that the two lists coincide in 
the main. The Council of Manazkert was one of fundamental im- 
portance, and it inaugurated a new era in the history of the Armenian 
Church. The belief (whether correct or incorrect is another matter), 
that Armenian ecclesiastical tradition, presumably stemming from the 
Illuminator but interrupted and forgotten in the Chalcedonian trend 
of the seventh century, had been restored, was identified with this 
Council, This is the reason for which the list of bishops participating 
in the Council of Manazkert was subsequently archaized and attributed 
to the time of the Illuminator 31 *\ 

The list of eparchies found in Agat'angelos' History is equally 
unpromising from the point of view of historical authenticity and 
arouses our skepticism. But the Arabic version of the work provides 
a key for the discovery of the nature and real origin of the list it 
attributes to the Illuminator 310 . The' first interesting point is the 
classification of countries and provinces. Three groups are distin- 
guished here, Iberia, Abkhazia, and Al(v)ania are mentioned in 



THE NAXAMAB SYSTEM AND THE CHUECH 



265 



the fixst ; the eight Armenian provinces given in the Acts of the Council 
of 726 quoted. above, in the second; and in the third, thirteen other 
Armenian provinces, or more exactly sees are given, The separation 
of the first group from the body of Armenia is understandable, but how 
can one explain the grouping of the Armenian provinces ? Why are 
the autonomous Satrapies together with Siwnik', Mokk' and Mard- 
petakan separated from the other parts of Armenia to form a special 
group ? The key to the date of composition of the Arabic version 
of Agat'angelos is hidden in the answer to these questions. 

There can of course be no question here of an association on political 
grounds. The only hnk capable of tying together such scattered 
districts as Angeltun, Anjit, Cop'F and Hasteank', on the one hand, 
and Siwnik', Mardpetakan, and Mokk' on the other, was confessional 
unity. The Arabic version evidently goes back to a period of quarrels 
between these provinces and the other central districts of Armenia 
based on confessional disagreements. Such a moment came in the 
second half of the sixth century when the sources show Siwnik', 
Mardpetakan, and Mokk' leaning toward Nestorianism, The general 
success of the Nestorian doctrine in Armenia can be seen from its 
penetration to the very heart of the country and from its establishment 
at Dwin, residence of the patriarchate, The Kat'olikos Nerses II 
was forced to use very harsh measures in his struggle against it, and 
the Council of 555 was summoned for this specific purpose, The 
kat'olikos, in a special letter, to certain bishops, summoned them in 
particular to the Council, threatening them with excommunication if 
they failed to attend. These are the bishops of Arcrunik', Rstunik', 
Mokk', and other nearby provinces, whose sees lay along the Syrian 
border and had apparently been the ones most affected by Nestorian 
influence 32 , The Acts of the Council bear the signatures of the bishops 
of Arcrunik', Palunik', and Mehnunik', but the others apparently did 
not answer the summons of the kat'olikos, because of their lack of 
sympathy with the Council 32a . AfteT the Council, Nerses II found 
it necessary to address himself once again to the bishops of Arcrumk' 
and Mardpetakan; although they had been present at the Council, 
he reminded them once more of its decrees and urged them to take 
special measures against the followers of Nestorius 33 . Nerses' 
successor, John II, wrote to Vrtanes bishop of Siwnik' informing him 
of the presence of numerous Nestorians, who must be repressed, in 
his see, and reminding him of the precepts of the Illuminator as well 



266 



CHAPTEK Xn 



as of his labours for the land of Siwnik' 34 , The same points were 
repeated in his Letter to the Church of Albania 35 . 

This evidence shows that the grouping of ecclesiastical provinces 
reflects the dogmatic pattern of Armenia in the sixth century. Hence 
the list of bishoprics is of the same origin as the list of princes studied 
above and belongs to the turn of the sixth to the seventh century. 
If the ecclesiastical isolation of Siwnik' is in any way related to the 
political separation of this province from Armenia in 591, the List 
and the composition of Agat'angelos must belong to the period between 
591 and the year 612 noted above 35a , In the Armenian Agat'angelos 
not only Iberia, Albania and Abkhazia, but even the eight Armenian 
districts leaning toward Nestorianism have been passed over in silence, 
In this version only twelve bishops are listed by name, without the 
corresponding mention of their provinces, though the fact that « . . . there 
also were other bishops whose names, with the best intentions, could not 
be written down » 35 *>, is not concealed. The sense of this comment can 
be explained through a comparison with the Arabic version; it was 
evidently a reference to the bishops of the eight provinces whose 
names had not been given. The Armenian compiler was more inter- 
ested in the bishops than in their sees, and since his source, namely 
the Armenian original of the present Arabic version, had not given 
the names of the bishops of the first group (i.e, of those who adhered 
to Nestorianism), he was satisfied with the passing comment that 
for all his good intentions he was unable to provide their names. 

The Armenian original of the Arabic version was used during the 
composition of the surviving text of Agat'angelos, but even this 
original does not seem to have been the first version of Agat'angelos. 35e 
The unevenness of the composition noted in the episcopal List is great 
enough to warrant the conclusion that the work was a compilation, 
and a re-working of a still older Life of St Gregory, The repetition 
of the name of bishop Albianos in the List points in the same direction. 
According to the Arabic Agat'angelos, St. Gregory had sent Albianos 
as bishop to Bagrewand and to all the inhabitants of the banks of the 
Euphrates, while he sent bishop Albios to the country of the Sparapet 
- Mamikonean, that is to say to Taron and Bznunik\ But the figure 
of Albios cannot be found in the Armenian version, and only Albianos 
is mentioned. Unquestionably the two names refer to one and the 
same person, namely the well known bishop Albianos 35d , The division 
of the person of Albianos results from the fact that the compiler used 



THE NAXAMAB SYSTEM AND THE CHURCH 



267 



different sources, with tlie name Albianos given by one and that of 
Albios, by another. Under the influence of the contemporary situ- 
ation, in which Taron with Bznunik', and Bagrewand with its neigh- 
bours, formed different eparchies, the author mistook Albianos and 
Albios for two different persons. All of this evidence suggests that 
the Arabic version went back to an earlier text as does the present 
Armenian version which is a re- working of this same text 36 , 

It is very difficult to trace the pattern of accretions and to separate 
the relatively ancient elements in the text from the more recent ones. 
One thing is certain, that the episcopal List which interests us was 
not a part of the Armenian version, and likewise could not have been 
a part of the original version. There can be no doubt about the eight 
Nestorianizing provinces, Of them the see of Siwnik', for example, 
was instituted by St. Mesrop according to the testimony of his disciple 
and consequently could not have figured in the original version among 
the sees established by St. Gregory 36 a , A study of the composition of 
the next 12 bishoprics leads to a similar conclusion. These were 
bishops intended for Bagrewand and the lands bordering on the 
Euphrates : Basean (Vanand), Ek c eieac-Derfan, Daranahk', Karin, 
Sper, Taron, Bznunik', Xor^orunik', Arsamunik', Sirak, and also 
Korduk c and Atrpatakan. With the exception of the last two, these 
districts lay in the Imperial part of Armenia according to the division 
of 591 . As we have already said -Xusr 5 II Abharvez ceded an important 
part of Persian Armenia to the Empire after having regained his 
father's throne with the help of the Emperor Maurice. As a result, 
out of the central provinces of former Eastern Armenia only part of 
Ayrarat and all of Vaspurakan remained Persian. The provinces 
which were still Persian are precisely the ones not represented in 
Agat'angelos' List of bishops. 

The political break of 591 was accompanied by religious dissentions 
in the history of Armenia. Already in 572, at the time of the disorders 
arising from the murder by the Armenians of the Suren who was the 
Persian Marzpan, the Kat'ohkos John III and Vardan II Mamikonean, 
the instigator of the rebellion, had gone to Byzantium to ask for 
help against Persian reprisals. While they were there, they accepted, 
in one form or another, the Chalcedonian doctrine, According to 
contemporary testimony, this betrayal of the national tradition pro- 
voked discontent among the population of Armenia 37 . Soon after, 
Persia also found itself in a period of difficulties over the succession 



268 



CHAPTER XII 



to the throne. The Armenians -under the leadership of Musel, likewise 
a member of the Mamikonean house, supported -Xusro II against the 
usurper Vahram Choben, nevertheless, the valiant Musel did not 
escape the persecutions of Xusro II and was forced to take refuge in 
the Byzantine capital, following the example of Vardan 38 , According 
to one account, the Emperor Maurice also proposed to this prince to 
abandon his native Church and adhere to the Imperial one. Musel 
excused himself on the grounds that certain disputed points first 
had to be cleared, For this purpose he advised the summoning of 
the kat'olikos and of other representatives of the Church to a Council 
in the capital. The Kat'oHkos Movses II categorically rejected the 
Emperor's invitation, The bishops of the country of Aspurakan, 
which was under the authority of the Persians, similarly refused point- 
blank. But the bishop of Taron, and others who lived under the 
authority of Rome, set out for Constantinople and there agreed with 
the Emperor in all things 39 , 

The bishop of Vaspurakan is precisely the one missing from Agat'an- 
gelos' List together with his neighbour from Mokk* : and only Imperial 
Armenia is represented in it. The eontent of the List is clearly 
Chalcedonian in its colouration and corresponds to the point of view 
of the last decade of the sixth century, The grouping of the eight 
bishoprics belongs to the same period. It seems as though these 
provinces were put into the list of countries christianized by St, 
Gregory in order to fill gaps in this list. Although Atrpatakan, by 
which Persian Armenia in general was probably meant, had been 
included in the list, the absence from it of the provinces of Mardpetakan, 
Mokk ? , Siwnik', and the Satrapies, must still have struck the eye. 
Consequently they were the ones added to the list. It is also very 
likely that confessional ties were taken into consideration, so that 
the list of Christianized lands carried a partly Chalcedonian and partly 
Nestorian stamp. Up to the seventh century two conflicting currents 
existed in the Armenian Church. The national party fighting against 
Nestorianism unintentionally fell under the influence of Chalcedo- 
nianism, which also condemned Nestorian doctrine 39a , By the mid- 
seventh century Chalcedonianism took the upper hand in Armenia 
and became the dominant doctrine. It was only at the beginning of 
the eighth century that the Armenians succeeded in renouncing 
Chalcedonianism and in working out a national religious policy; 
Agat'angelos was then re-worked once again to conform with the 
change. 



THE NAXABAE SYSTEM AND THE CHURCH 



269 



II 



The present analysis lias demonstrated the total uselessness of the 
evidence of Agat'angelos fox our task of clarifying the ecclesiastical 
hierarchy under the Illuminator. IPor the early history of the Arme- 
nian Church, the History of Ifaustus must still remain the most ancient 
and the most reliable source 39l \ In the period described by this histo- 
rian, there were very few bishops in the Armenian Church, Two houses 
are very familiar to Faustus, that of St, Gregory the Illuminator, and 
that of bishop Albianos, The bishop of Basean is mentioned once, 
and the name of bishop Paustus, who ordained St, Nerses I as deacon, 
is also given 40 . At the end of his work, Faustus presents a survey of 
bishoprics in the reign of King Xosrov III of Armenia, but here too, 
besides the representative of the family of Albianos, he mentions 
only the same, now aged, bishop Faustus, the bishops of Basean (Zorf, 
Artit\ Tirik, and Movses), Kirakos of Tayk', Zoxt'ual of Vanand, 
and a certain bishop Aaron 41 , If at the end of the century only a 
few bishops made up the entire hierarchy, it evidently must have 
been faT smaller still under St, Gregory the Illuminator at .the beginning 
of the century, With his great concern for the clergy, the historian 
would not have forgotten to mention other bishops had they existed, 

In the first century of its existence the hierarchy of the Armenian 
Church consisted in fact of two families, that of bishop Albianos 
and that of the Illuminator, and these were rivals for the first place 
in the land, Their rivalry played a certain part in the establishment 
of the forms of ecclesiastical organization, and we think it necessary 
to pause over the origin and nature of this competition. 

Historical scholarship does not deny the existence of Christian 
communities in Armenia earlier than the mission of the Illnminator, 
Here as everywhere, the beginnings of Christianity are lost in the 
dimness of legendary accounts. National traditions distinguish three 
branches in the spread of Christianity to Armenia ; these axe identified 
with the names of apostles Thaddeus, Bartholomew, and with St, 
Gregory, No matter how legendary, the tales of the apostolic mission 
to Armenia still contain some elements of truth and prove that Ar- 
menian Christianity went back to the Apostolic period. We may 
question whether any of the apostles ox of theix disciples came to 
Armenia, but there are no grounds for denying that Armenia belongs 



270 



CHAPTER XII 



among the earliest countries illuminated by the new doctrine. Next 
to the Armenian lands, Syria with its capital Antioch was one of the 
centers of apostolic activity ; from there waves of the new movement 
could easily have reached Armenia, At first, Christianity spread 
rapidly and won important victories in the East in the period of 
Parthian rule. But with the rise in Persia of the Sasanian dynasty 
with its anti- Christian policies, the results of two centuries weTe 
reduced to naught. All that remained from that period were dim 
memories which, mixed with later ones, formed the mass of varied 
legends preserved in ancient sources. Historical criticism has not 
yet mastered the legends of Thaddeus and Bartholomew and cleared 
them of various accretions, nor has it clarified their relation to the 
analogous Syrian legends of Addai and his disciple Mari, the apostle 
of the provinces of pre-Atropatene, What has been done already in 
this domain gives us grounds for regarding the legends of Thaddeus 
and Bartholomew as echoes of Christian worship in Armenia in the 
pre-Sasanian period 41a . 

The three cycles of legends grouped around the names of Thaddeus, 
Bartholomew, and Gregory, mark three periods in the penetration of 
Christianity into Armenia. The traits common to all of them indicate 
that they were successive periods. The new proselytism revived dim 
memories and put on them the seal of identity. The mission of Thad- 
deus and Bartholomew came from the south. The intermediary 
country between apostolic Antioch and Armenia was Osrhoene, with 
its capital Edessa. Edessa played a leading role in the diffusion of 
Christianity in the Persian East, and Armenian legends are also 
connected with this city. The appearance of an organized Christian 
community in Armenia is consequently inadmissible before the con- 
solidation of the Christian Church in Edessa. 

With the successful analysis of the legend of Abgar scholars have 
accepted the identification of this famous ruler of Edessa who had 
embraced Christianity with Abgar IX, a historical figure who 
lived at the end of the second century ; the legend shifted authentic 
historical events backward to the beginning of our era and attri- 
buted them to another historical personage, Abgar V, a contempo- 
rary of Christ. Hence, there is no reason to admit the existence of a 
Christian Church at Edessa before the end of the second century. 
As for the earliest mention of a Church in Armenia, it must be put in 
the middle of the third century, at which time we have a reference 



THE NAXABAH SYSTEM AOT THE CHURCH 



271 



to a bishop Meruzan [Meruzanes] of the Armenians who was in corre- 
spondence with bishop Dionysios of Alexandria (248-265) 42 . It has 
been supposed that this Meruzan was one of the bishops of Lesser 
Armenia, probably of Sebasteia, but Gelzer raised an objection to 
this hypothesis on the basis of the bishops' characteristic name, and 
sought him within the limits of Greater Armenia, He considered 
him, entirely correctly, to be a scion of the Areruni in whose house 
the name Meruzan was very common, Gelzer was mistaken only in 
placing Meruzan in Vaspurakan 43a . As we shall see, the original 
province of the Areruni was Sophene, so that it is more coxiect to 
identify the seat of bishop Meruzan with Sophene. 

The question of a genetie^Bnk between the church of Meruzan and 
that of Edessa naturally presents itself. The orientation toward an 
Edessan tradition of Armenian legends dealing with early Christianity 
derives from this historical fact. According to the evidence of one 
Syrian document, the apostle Addai had suffered a martyr's death in 
the land of Sophene at the castle of Agel, Armenian Angel, while 
other documents are either silent about his death or place it at Edessa 43 
This legend reflects the fact that Angel was the residence of bishop 
Meruzan, the pastor of an Armenian Church which was a daughter of 
Addai's church in Edessa. The existence of a Christian church in 
the border Armenian province of Sophene in the mid-third century 
should now be accepted as a fact no longer open to doubt, From 
there Christianity was able to penetrate into Armenia. Just as the 
Edessene Church had given birth to the one in Sophene. so after the 
further development of the latter, a bishopric also appeared in the 
province of Taron adjoining Sophene, 

The church ruled by the house of bishop Albianos was created in 
the same way, in our opinion. In the Arabic version of Agat'angelos, 
Bagrewand and the banks of the Euphrates were given to bishop 
Albianos, while Taron with Bznunik' went to Albios. We have 
already seen that only one person was meant. From Eaustus we 
learn that the possessions of bishop Albianos had greatly increased 
under king Xosrov II, for upon the destruction of the house of Mana- 
wazean, the possessions of the dead princes had been transferred to 
the church of Albianos, and in particular, «... Manazkert and the 
neighbouring territory along the Euphrates river 44 », These are the 
very lands to which Agat'angelos refers as being along the banks of 
the Euphrates, but they were given to Albianos only under St. Gre- 



f^ 



272 



CHAPTER Xn 



gory's second successor, Yrt*anes, and not in Gregory's own lifetime, 
as Agat'angelos indicated, Eaustus tells us nothing about the original 
ehurch of Albianos, or about the location of bis church before the 
possessions of the Manawazean passed to him. Nevertheless, Agat'- 
angelos' xefexence to Taron and Bsnunik' as the poxtion of Albianos 
is appaxently not entixely devoid of histoxical foundation. The 
repeated assextions of Eaustus concerning the pxecedence of the elruxeh 
of Taron and its position as the leading church among those in 
Armenia, xest primaxily on his knowledge of its priority in point of 
time, of the fact that the church of Taron txaced its foundation all 
the way back to the pexiod of St. Gregory, 44a 

The legend of the mission of Thaddeus or Addai, beginning in Edessa 
and ending in Artaz, the legend in which the fate of the Edessene 
King Abgar is interwoven with that of the Armenian king Sanatrak, 
also points to the progress of the Christian faith from Edessa, through 
Sophene and Taron, to Axtas. The Sanatrak of the legend is a 
historical figure, he is the king Sanatrak of Armenia, whom the 
Emperor CaracaHa (211-217) summoned, supposedly to reconcile 
him with his rebellious sons, but in fact to deprive him of his crown 
and to turn Armenia into a Roman province as had already, been 
done with Osrhoene, when CaracaHa had treacherously seized king 
Abgar 45 , The actors in the legend, Sanatrak and Abgar, were 
contemporaries in reality, and were connected by their similar fates. 
In the legend both kings are shifted from the beginning of the third 
century to the period of Christ, This parallel in the development of 
the subject, together with the Armenkation of the story of the king 
of Edessa, who is presented as the uncle of Sanatrak, can be explained 
only on a religious basis, as a reflection of the genetic relation of the 
Armenian Church to that of Edessa, These memories of the origin 
of Christianity in Armenia and the tradition of the first Christian 
church connected with the Syrian world, subsequently grew dim and 
all but disappeared, lost in the second and moxe powerful current 
of Christianity under St. Gregory the Illuminator. 

At the end of the third century, great changes took place* in the 
history of Armenia, The wars between the Persians and the Romans 
over Armenia ended in- a Roman victory and, under the terms of 
the treaty of 298, Armenia came under the protectorate of the Empire. 
Royal power was restored in Armenia and Trdat III, the descendent 
of the Arsacids assumed the throne of his fathers, having obtained 



L_ 



/ 



THE NAXAJRAJS SYSTEM AND THE CHURCH 



273 



Ids rights with the help of a Roman army. This political transfor- 
mation was accompanied by the entrance of Christianity into Greater 
Armenia from the neighbouring lands of Lesser Armenia, At that 
time, the Empire was passing through a period of violent straggle 
between the old religious tradition and the Christian teachings, and 
it was on the eve of the victory of the new doctrine. Spreading 
throughout the Empire, Christianity had already had time to gain a 
secure place for itself in many areas, and it won its officially recognized 
victory in the beginning of the fourth century. 

Among the portions of the Empire which were thoroughly Christi- 
anized at an early date were the lands both of Cappadocia and Lesser 
Armenia, The church of Caesarea [of Cappadocia] had enjoyed great 
renown from the third century, since it had been ruled by bishops 
Alexander (ca. 200) and Eirmilianus (f 269), the friends of the great 
fathers of the Alexandrian church, Clement and Origen, Thanks to 
Eirmilianus, Caesarea had become a center of theological instruction 46 , 
and Melitene, the capital of Lesser Armenia, was no less important 
as a center of Christianity, The Christian elements there were so 
powerful that the frequent revolutionary uprisings which occured in 
the city were attributed to them 47 . 

The development of Christianity in the provinces of Lesser Armenia 
and Cappadocia and its influence explain why the new faith was pro- 
claimed as a state religion in neighbotiring Greater Armenia before this 
was done in the Empire 47a , With the subjection of Arsacid Armenia 
to the superior protection of the Empire, the political obstacles to a 
close relationship between the Armenian population of Lesser and 
Greater Armenia, two divided portions of a single ethnic whole, were 
removed, and Christianity poured from, the right bank of the Euphrates 
to the left one. The apostle of this Christianity was St, Gregory the 
Illuminator, To distinguish this branch from the earlier Syrian or 
Edessene Christianity, we will call it Lesser Armenian or Caesarean. 
It is important for our study to distinguish these two paths in the 
spread of Christianity in Armenia. 

The Christianity stemming from the lands of Lesser Armenia first 
occupied the provinces immediately adjoining it, namely Daranahk', 
Ekeleac, Derjan, and others, Eaustus underlines repeatedly the 
allegiance of DaranahF and Ekeleac to the house of the Illuminator. 
Gregory and Vrt'anes weTe buried in T'ordan and Aristakes in T'il. 
Though he died in Sophene, Yusik was brought back and buried in 



274 



CHAPTER XII 



T'ordan in the burial place of his family 48 . In the same way the 
earlier current of Christianity coming from the side of Sophene had 
first permeated the neighbouring districts of Taron and Bznunik'. 
Under the sons of St. Gregory the chorepiscopus Daniel, a Syrian 
by birth, had stood at the head of the church of Taron 49 . To this 
church also belonged P'aren, the successor of Yusik in the kat'ohkosate, 
He in turn was succeeded by Sahak, from the family of bishop Albianos, 
who ressembled P'aren according to Faustus, It is possible that all 
these figures, Daniel included, were of the same origin. The miracles 
and halo of saintlintliness which surround Daniel's name in Faustus' 
History bear witness to the fact that he was an unusual personnage, 
and that he played an important part in the life of the Armenian 
Church 50 . 

In spite of the fact that all of Faustus' sympathies are on the side 
of the descendents of the Illuminator, he cannot hide the fact that 
the representative of the church of Albianos, in the person of Daniel, 
enjoyed a great influence, and that in the struggle for the patriarchal 
throne the representatives of this house undoubtedly had precedence 
over that of St, Gregory 50a . Later, when the cult of St. Gregory 
had triumphed, only dim memories remained of the original Syrian 
church. The more authoritative representatives of this church, such 
as Daniel and Albianos, were represented in the later tradition as 
disciples of St. Gregory. 

The true inter-relationship between the two currents of Christianity 
in Armenia is presented in the admirable legend which relates how 
the parents of St. Gregory, at the time of their flight from Persia 
to Armenia, stopped in Artaz near the grave of the holy apostle 
Thaddeus and, «... here took place the conception of the mother of 
the great and holy Illuminator and, therefore, having received the 
grace of this same apostle, he who was conceived near his tomb com- 
pleted his unfinished spiritual task 51 ». The legend in itself is devoid 
of true foundation, but it is correct as the symbolic expression of the 
belief that the activity of the Illuminator had found a ground already 
made fertile by the martyrdom of St, Thaddeus, and that the work 
of St. Gregory was the continuation of the mission of the apostle. 
An entire century of struggle raged before the total merging of 
the two currents which we have observed and the creation of a single 
Church, In the course of this struggle the principle of nationalism 
in church organization was worked out. 



THE NAXABAR SYSTEM AND THE CHURCH 



275 



If we follow carefully the alternation of the descendents of Gregory 
and Albianos on the patriarchal throne, we can observe a certain 
coincidence between it and the successive political superiority of the 
two states on which Armenia depended. The rule of the Gregorids 
coincides with the ascendant influence of the Koman Empire, and 
that of the Albianids with that of Persia. The protectorate of the 
Empire over Armenia, consolidated by the treaty of 298, remained 
officially undisturbed until 363. During this period, the Persians 
attempted several times to intrude into Armenian affairs, and they 
were finally successful in 338 when dissentions arose among the heirs 
of Constantine after the emperor's death. The king of Armenia at 
this time was Tiran [Tigranes VII], and it is precisely in his reign 
that the Albianids, P'aren and Sahak, ascend the patriarchal throne 
after Aristakes, Vrt'anes, and Yusik, The son and successor of 
Tiran, Arsak II was considered to be an imperial apointee. Nerses I, 
the descendant of St. Gregory, was summoned to the kat'ohkosate 
together with Arsak. As soon as the friendly relations between 
Arsak and the Emperor turned to enmity 52 , and the Armenian king 
concluded an alliance with the Persian king Sahpuhr II, Nerses I found 
himself removed from his see and replaced by a bishop Cunak, probably 
from the house of Albianos 5S . Arsak soon quarreled with the Persians 
and ended his life tragically after he had been emprisoned by them. 
His son, Pap, came to power with the help of a Roman army; the 
deposed kat'oHkos, Nerses, found himself once again at the head of 
the Church 54 . The friendship with the Emperor did not last long, 
however, and Pap turned away from the Empire to join the Persians 55 . 
This is the point at which the murder of Nerses occured, and the 
appointment in his place of Yusik, one of the descendents of Albianos 56 , 
Finally after the division of Armenia, the patriarchal see remained in 
the Persian portion and was occupied successively by the Albianids, 
Zawen, Sahak and Aspurakes 57 . 

In this manner, the heads of the Church changed in accordance 
with political influence. The Gregorids came forward as supporters 
of the imperial policy, while the Albianids sympathized rather with 
the Persians, and their alternation reflects the genetic difference 
between the two currents of Christianity, the Greek and the Syrian, 
which they represented. The policy of the Albianids was basically 
that of the Arsacids, and it became national as it came to serve as the 
foundation for the organization of an Armenian Church 57a . The 



276 



CHAPTEB XII 



process of development and gradual nationalization in the Armenian 
Church becomes cleaxer and moie understandable if we compare it 
with the evolution of ecclesiastical organization in the Empire in 
general, 

From the very first period of its appearance, Christianity displayed 
a natural tendency toward organisation. Straggling against a hostile 
environment, it was obliged to marshall its forces, both for the mainte- 
nance of positions already occupied and for farther advances, as well 
as for the formation of communities, and the establishment of common 
ties among them in order to keep up with the increase and widening 
of the circle of the faithfol. The clergy : bishops and priests standing 
at the head of separate communities, formed the first connecting link. 
Several bishops next joined together and formed an administrative 
unit under the leadership of one among their number, The unification 
of communities and the grouping together of their bishops followed 
the lines elaborated in the political structure. The evolution of the 
ecclesiastical hierarchy followed in the steps of administrative or- 
ganization. Imperceptibly, the Church adapted itself to the adminis- 
trative structure of the Empire, its own divisions paralleling those 
of the state. The work of organization in the Church had already 
reached a measure of completion by the time of the Council of Nicaea 
in 325, which consecrated the then existing order, This order was a 
faithful reflection of the imperial machinery which had taken a clearly 
defined form of its own at an earlier date 57t) , 

The Church responded rapidly to the new administrative divisions 
created by Diocletian at the end of the third century, and by 325 
it had already found the time to make the corresponding alterations 
in its own hierarchy, so that it coincided fully with the new system, 
"We have already seen the main aspects of the reform of Diocletian. 
On the one hand, existing provinces were broken up into smaller, 
administratively autonomous, units, on the other, several of the 
new provinces were joined together to form larger administrative 
units, the dioceses, in a hierarchical pattern of authority. The 
Asiatic possessions of the Empire formed three dioceses : Asiana, 
Pontica, and Oriens, to which Egypt also belonged, At their head 
stood vicars, is. substitutes fox the Praetorian Prefect of the East, 
while at the head of the provinces were found governors or praesides 
subordinated to their vicars. 

Ecclesiastical authority was distributed in exactly the same way 



THE NAXAMAR SYSTEM AST) THE CHURCH 



277 



and in the same framework. A province was simultaneously an 
ecclesiastical unit — an eparchy — and in each, eparchy a metropolitan 
bishop, equal in rank to the praeses of the civilian administration, 
likewise had his seat. Similarly, the dioceses simultaneously took on 
the sense of ecclesiastical units in which bishop-patriarch corresponded 
to the vicars. Just as in each province one of the bishops, namely 
the bishop of the provincial capital who ranked as metropolitan, 
stood out from the rest, so in each diocese one of the metropolitans 
occupied the leading position of patriarch or head metropolitan. 
The metropolitan of Ephesus held this rank in the diocese of Asiana, 
the metropolitan of Antioch, in that of Oriens, and the metropolitan 
of Caesarea [of Cappadocia] in that of Pontica 57c , 

Certain discrepancies raise the question of the date at which these 
relationships were established. Some scholars believe that the system 
of metropolitans was first established at the Council of Mcaea and 
that of patriarchates at the Council of Constantinople I. Others 
concede that the formation not only of the metropolitanates, but even 
of the patriarchates antedated the Council of Nicaea and that this 
Council merely sanctioned a situation already in existence at the time ; 
for this they rely on Canons IV and YI of the Council, The first of 
these deals with provincial bishops and their relations with their 
metropolitans. This canon marks a perfectly clear stage in the 
development of the metropolitan sees, A similar stage in patriarchal 
organization was reached in 381, and this point is marked by the second 
Canon of the Council of Constantinople, which states, 



V 



Bishops of a diocese should not extend their authority over 
churches outsides its boundaries, nor should they mingle 
churches. But according to the canons, the bishop of Alexan- 
dria should concern himself only with what is in Egypt, the 
bishops of Oriens should care for the East, with due respect 
for the seniority of the Church of Antioch, according to the 
canons of the Council of Nicaea, As for the bishops of Asiana, 
they should concern themselves only with the affairs of Asia, 
the bishop of Pontica with the affairs of Pontus, and the bishop 
of Thrace with those of Thrace 58 . 



According to the opinion of some scholars, diocesan divisions patter- 
ned on the political system were first established in the Church as a 
result of this Canon, Others, on the contrary, suppose that the Council 
merely clarified a system which had existed before 381 and which 



278 



CHAPTER XII 



was already accepted in principle at the Council of Nieaea, as evidenced 
by its sixth Canon, which prescribes that, 

.,, . To preserve the ancient custom whereby the bishop 
of Alexandria enjoys the highest authority in Egypt, Libya, 
and the Pentapolis .... Similarly with regard to Antioch, 
and the other eparchies, the right of seniority must be observed 
for these churches 59 , 



According to one interpretation, the words i^ovuia and ret -n-pecr/JeZa 
refer to metropolitan prerogatives, while according to another, the 
matter under discussion is the supra- metropolitan or patriarchal 
authority. In our opinion both hypotheses are incorrect ; the question 
will appear in its proper light if we remember the fundamental principle 
according to which ecclesiastical hierarchy was patterned, and if we 
use our knowledge to clarify these problematic stages in its develop- 
ment 59a . 

As a result of the reform of Diocletian, the number of provinces 
was increased, while their size was reduced. Thus, for example, before 
Diocletian, Cappadocia had included Lesser Armenia, Galatia, and 
Pontus Polemoniacus ; Mesopotamia had consisted of the later Meso- 
potamia, and Osrhoene; Syria of Syria, Phoenicia, Palestine, etc.. 
After Diocletian the fragmentation went even further : Cappadocia 
was split into Cappadocia I and II, Armenia I and II, Syria was 
subdivided into seven provinces (Euphratensis, Syria I and II, 
Phoenicia I and II, Palestinia I and II), Egypt into five provinces, 
Asia into seven 60 . How was ecclesiastical power to be altered as a 
result of these changes ? 

The eparchies were subdivided together with the provinces and new 
metropoleis made their appearance. As a result of Diocletian's reform, 
a whole series of new metropolitan sees arose around Antioch, Caesarea, 
and Alexandria, Did the new sees maintain some sort of subordination 
to their former metropolitans, or were they legally considered to be 
their equals? For instance, Caesarea had been the metropolis and 
the center of political life in the province of Cappadocia before Dio- 
cletian ; for this reason the bishop of Caesarea had necessarily occupied 
the position of a metropolitan. When Lesser Armenia was separated 
from Cappadocia, the bishop of Sebasteia, its capital, became the 
metropolitan of the newly created province. Were the relations with 
the former metropolis at Caesarea consequently broken? In short, 



THE NAXARAB SYSTEM AND THE CHURCH 



279 



when a bishop was elevated to metropolitan rank, what determined 
his relations to his former metropolitan ? 

We believe that the answer to this question is given by Canon VI of 
the Council of Nicaea cited above. The fathers of this Council assumed 
that existing traditions should not be abbrogated and that the centers 
indicated should preserve their rights of seniority. According to this 
classification, Antioch, Ephesus, Caesarea, and Alexandria, which had 
already reached a superior position before Diocletian, should continue 
to enjoy the same status after him, whether or not the Church adapted 
itself to the diocesan pattern. When the Church subsequently did 
accept the diocesan divisions, nothing was altered thereby in the 
position of these centers, since they became diocesan capitals. The 
Council of 381 consolidated and gave a clear formulation to a situation 
which had been established earlier. The metropolitans of diocesan 
capitals became superior metropolitans, each within the boundaries 
of his diocese. The patriarchal authority of Antioch, Caesarea [sic], 
and Ephesus, which were diocesan capitals began and developed in 
this fashion, Egypt, although a part of the diocese of Oriens, remained 
a separate unit because of its location ; consequently Alexandria, as 
the capital of an autonomous country, was assimilated to a diocesan 
capital from the ecclesiastical point of view, and its metropolitan was 
considered to be a patriarch. Hence, the diocesan division merely 
limited the sheres of influence of the senior metropolitans to the 
boundaries of their particular territory, i.e. the diocese, and conse- 
crated the patriarchal authority which had developed before and 
outside the diocesan system 60 \ 

With the appearance of the parallel between the ecclesiastical and 
civilian organizations, it becomes possible to discuss the rivalries of 
the patriarchs for a position of primacy in terms of their desire to 
create for themselves a hierarchical rank equivalent to that of praetori- 
an prefect. One of the patriarchs had to become the first, so that 
the others should be subordinated to him as the vicars or, governors 
of dioceses, were subordinated to the prefect. The Praetorian Prefect 
of Oriens lived for the most part at Antioch, especially at first ; later 
he ruled the East while residing normally in the Imperial capital. 
As a result it is understandable that the see of Antioch was the most 
successful in the struggle for primacy, until the contest was settled in 
favour of the patriarch of the imperial capital 60 *\ The later evolution 
of the hierarchy followed a path of centralization, to the detriment 



280 



CHAPTER XII 



in the first place of the patriarchs of Ephesus and Caesarea, and 
the infringement of their rights, In this sense, the Council of Chal- 
cedon marked the victory of the Church of Constantinople in 451. 
At the Council of 451, the holy fathers reviewed the decisions of the 
earlier oecumenical council as to the pre-eminence of the Church of 
Constantinople, approved it, and went still further in Canon XXVIII, 
which proclaims, * 

„, the metropolitans of the dioceses of Pontica, Asiana, and 
Thrace, and also the bishops of the barbarians who reside in 
the aforesaid dioceses, shall be ordained from the aforesaid 
most holy see of the most holy Church of Constantinople, The 
metropolitans of the aforesaid dioceses, together with the bishops 
of their eparchy, shall ordain bishops for the eparchy, in accor- 
dance with the holy canons, but the metropolitans of the 
dioceses [as has been said] shall be ordained by the archbishop 
of Constantinople 61 , 

As a consequence of this canon the patriarchal prerogatives of the 
sees of Ephesus and Caesarea were lost and they lost their importance 
forever, The negative attitude of the Armenians toward the Council 
of Chalcedon derives from this fact. The schism came about on a 
basis of hierarchy and not over dogma, as implied by later sources. 
The patriarch of Constantinople did not halt at this stage, but also 
strove to subordinate the other patriarchates, and to advance the pri- 
macy of his see. In 558 [sic] John the Easter* assumed the pompous 
title of Oecumenical Patriarch, Thus the Church did not lag behind 
the secular powers, impregnated with the same autocratic tendencies, 
it achieved the concentration of its powers in the person of the patriarch 
of Constantinople in the veTy moment of the triumph of absolutism, 
the period of Justinian 61a . 

This outline of the development of the Imperial Church allows us 
to draw two conclusions and to set down two prerequisites for a survey 
of the evolution of the Armenian Church, Eirst its external relations, 
i.e. its relations with the Imperial Church, must be clarified in relation 
to the political situation ; in other words, the subordination of the 
Armenian Church had to be equivalent to the dependence of the country 
on the Empire, Second, the Armenian Church had to correspond to 
the local secular pattern in its internal development just as the Imperial 
Church had adapted itself to the administrative structure of the 
Empire, 



THE NAXABAM SYSTEM AND THE CHUKCH 



281 



We have already seen that the Armenian lands did not present a 
politically unified whole but were divided into several portions with 
different political aspects, Consequently, we cannot speak of*arsingle 
ecclesiastical organization for the whole of the Armenian territory, 
Lesser Armenia, organically bound to the Empire, was also subject 
to the general pattern in ecclesiastical matters, First, as a part of 
Cappadocia, it had belonged to the eparchy of the metropolitan of 
Caesarea ; then, having been made into a separate province, it had a 
metropolitan of its own at Melitene, With the division of Lesser 
Armenia into two provinces, Melitene became the metropolis of 
Armenia II, and Sebasteia that of Armenia 1 62 , Both these metro- 
polis recognized the seniority of their former metropolis, the see of 
Caesarea; finally at the time of the Council of Chalcedon of 451, they 
were put under the direct jurisdiction of the patriarch of Constantinople, 

Greater Armenia vacilated in ecclesiastical matters because of its 
political instability, now leaning toward the Empire and now away 
from it, according to the political situation, In the first century of 
the existence of the Armenia Church, from St, Gregory to the partition 
of the country at the end of the fourth century, Armenia was in the 
main under the protectorate of the Empire and therefore, their eccle- 
siastical relations were likewise very close. Consecrated at Caesarea, 
St, Gregory, seemed an ordinary bishop, like the bishops of Melitene, 
Sebasteia, and other cities, in his relations with Caesarea, with the 
only difference that the subordination of Gregory, who was the bishop 
of an independent nation, under imperial protection, was purely 
formal in character 63 , 

Gregory might have been ordained as bishop by the nearest metro- 
politan, for instance that of Melitene, since the ancient tradition 
clarified by the fourth canon of the Council of Nicaea gave to each 
metropolitan the right to ordain bishops in his eparchy. Since Gregory 
addressed himself to Caesarea, his ordination must go back to the 
period preceding the reform of Diocletian in 296, which turned Lesser 
Armenia into a new province with a separate metropolitan, To be 
sure even after this reform the successors of Gregory went to Caesarea, 
but this can be explained by the natural desire to preserve the tra- 
ditional link with the former, leading center of ecclesiastical life and 
to avoid becoming dependent on a metropolitan who had formerly 
been subordinate to the see of Caesarea on a par with the bishop of 
Armenia. The relationship of the Illuminator to Caesarea was that 



A. 



282 



CHAPTEE XII 



of a bishop to his metropolitan, while his successors, or those of them 
who received their consecration at Caesarea, behaved with regard to 
it as metropolitans to a patriarch. 

This situation continued with certain interruptions until the partition 
of Armenia, These interruptions were the results of Persian interference, 
and with the increase of their influence, the Armenians showed a 
tendency to end their ecclesiastical dependence on the Empire and 
to create an autonomous church. We have seen that the spokesmen 
of the two trends in ecclesiastical policy were the representatives of 
the houses of Gregory and Albianos. The division of Armenia into 
two kingdoms provided a solution for its dual ecclesiastical policy; 
the two currents withdrew into their respective boundaries : the 
Imperial one into western Armenia, and the national one into eastern 
Armenia, From the time of the partition, eastern Armenia entered 
resolutely upon a policy of nationalization by means of autonomous 
hierarchical institutions, while western Armenia remained in a position 
of dependence on the Imperial Church. 

The nationalization of the Church meant its ^a^ararization or 
feudahzation, Since the political regime in Armenia was feudal in 
character and the power was divided among many princely houses, 
ecclesiastical authority likewise had to assimilate itself to the existing 
society and adopt the forms which were ready to hand. The process 
of waxararization in the Church began with the break in its relations 
with the Imperial Church during the reign of king Pap. After the 
murder of bishop Nerses I, Pap appointed bishop Yusik, from the 
house of Albianos in his place. On hearing of the king's wilful decision, 
taken without his knowledge or permission, the patriarch of Caesarea 
was greatly displeased, A synod of bishops from the eparchy of 
Caesarea met on this matter, and sent a wrathful letter to King Pap 
in which according to the words of the historian, 



... they took away the authority of the kat'olikosate and 
(decreed) that the persons chosen as patriarchs [of the Ar- 
menians] should have only the right to bless the royal table, 
but should not presume hereafter to consecrate bishops for 
the Armenians as had been the custom before. Thereafter, 
" continues the historian ", the right of ordaining bishops 
was taken from the Armenians, and those designated as bishops 
for the various provinces and lands of Armenia journeyed 
to Caesarea and were consecrated as bishops there, since from 
that time, " the historian repeats ", the Armenians did not 



THE NAXABAR SYSTEM AND THE CHURCH 



283 



dare to ordain bishops, but whoever was the senior bishop of 
all sat above the others at the royal table and blessed the 
king's bread 64 , 



The reliability of Faustus' account is supported by the fact that Basil 
the Great, who was the bishop of Caesarea refered to, also mentions a 
clash between himself and Pap in his Letters. Basil rejected Pap's 
request to consecrate Faustus, the bishop sent by the king, and 
proposed in his stead his own candidate, a certain Cyril. The Arme- 
nians then by-passed Basil and having addressed themselves with the 
same request to Basil's rival Anthemios of Tyana, Metropolitan of 
Caappadocia II, obtained their wish 65 . The Faustus mentioned here 
should be identified with the bishop of the same name, who is 
said to have ordained Nerses I and to have lived until the period 
of King Xosrov III, according to Armenian historian 65a , The account 
of Basil the Great does not quite agree with the Armenian evidence ; 
they can be reconciled only if we admit that first Ki n g Pap supported 
the candidacy of Faustus, and only afterward broke with tradition 
and appointed his own candidate because of Basil's refusal to acceed 
to his request. 

The concept of the Armenian historian we have just cited as to 
the nature of the kat'olikos' authority can be seen from his account, 
This authority consisted primarily in the right of episcopal ordination, 
and with its loss, the position of kat'olikos was reduced to that of 
court bishop. Whether the bishops of Armenia in this period bore the 
title of kato&kos, and whether they had the right of ordination is still 
a debatable question 65l) . The influence of the somewhat later period, 
when the position of the kat'olikos had been established, can be felt 
in the historian's words. As for his assertion that after the incident 
with Pap the Armenian bishops were consecrated at Caesarea, it 
should be taken as correct, but only for the Imperial or western 
portion of Armenia. We know that after the partition there was no 
kat'olikos in the Imperial part of the country, and that the bishops of 
this portion were no longer subject to the kat'olikos who resided in 
Persian Armenia. None of the bishops from western Armenia 
participated in the rebellion of the fifth century. 

We do not know the state of ecclesiastical affairs in Imperial Armenia 
after the partition. Judging from certain sources, its Church con- 
formed to the political regime found in each of its component parts. 
We see from the signatures of the Council of Chaleedon in 451 that the 



284 



CHAPTER XII 



representatives of the autonomous Satrapies : the bishops of Sophene, 
Anzitene, Angelene, Sophanene, and Martyropolis, were present at the 
Council and made up the eparchy of Mesopotamia under the metro- 
politan authority of the bishop of Amida 66 . After the separation 
of Osrhoene, Nisibis had become the metropolis of Mesopotamia, and 
after the transfer of this city to the Persians in 363 it was replaced 
by Amida, The bishops of the Satrapies were as autonomous as 
their ruling princes. Just as these princes received the insignia of 
their rank from the Imperial authorities, so the bishops received their 
consecration at the hands of the nearest Imperial metropolitan ; they 
attended the Council of 451 in this position of ecclesiastical subordi- 
nation. 

The status of the Church in Imperial Interior Armenia is less definite ; 
no representative came from it to the Council of Chalcedon. Gregory, 
bishop of Justinianopolis, is mentioned at the Fifth Council of 553 6? . 
At the time of the schism in the kat'ohkosate at the end of the sixth 
century, bishop Theodore, who was considered the instigator of the 
troubles, had his seat at Theodosiopohs 68 . Koriwn mentions a 
bishop of Derjan, while Theodore, bishop of Ekeleac or Justinianopolis, 
and George of Darana3ik c or Kemah, are mentioned among the parti- 
cipants at the Council of 680, This same George was also present 
at the Council of 692, together with bishop Marianos of Kitharizon 69 . 
Consequently in the sixth and seventh centuries bishoprics existed in 
the main provinces of Interior Armenia ; Karin, Derjan, Ekeleac, and 
Daranahk', The date and order of appearance of these bishoprics 
is not as clear. Interior Armenia did not become a Eoman province 
in the strict sense of the term after its reunion with the Empire, but 
kept its internal wiya rar structure to the time of Justinian. Eor this 
reason, in the ecclesiastical sphere as well it was not a separate me- 
tropolitan see, but each province, as an autonomous principality had 
its own bishop, and all the bishops were the subordinates of the 
Imperial Church since they were ordained by it. At the Council of 
Chalcedon it was decreed that, «... heareafter the bishops of the 
barbarians shall be consecrated by the most holy see of the most holy 
Church of Constantinople » 69a , on a par with the metropolitans of the 
dioceses of Pontica, Asiana, and Thrace. Before that time Armenian 
bishops had been ordained by the nearest higher see, namely that of the 
church of Caesarea, but thenceforth, under the terms of the canon 
just cited, they were put under the direct jurisdiction of the patriarch 
of Constantinople. 



THE NAXAMAB SYSTEM iJSTD THE CHURCH 



285 



The real sense of Paustus' statement that the Armenian bishops 
were consecrated exclusively at Caesarea after the clash between the 
patiiaTch of Caesaxea and King Pap becomes clear from what has 
just been said. We see in his assertion a reflection of the state of 
affairs found in the Imperial portion of .Armenia, whose bishops were 
indeed the subordinates of the see of Caesarea, According to Armenian 
tradition, an attempt was made to create a separate kat'ohkosate for 
Imperial Armenia in the period of troubles which followed the partition 
of the country 70 . It is possible that the patriarchal authority was 
also divided at the time of the division of the realm between the two 
brothers who had inherited the throne, one of whom ruled in Persian 
and the other in Roman Armenia* In that case, the reference of the 
historian to the kat'olikos' loss of his power of ordination and to 
his degradation to the rank of court bishop should be applied specifi- 
cally to the holder of the rank of kat'olikos in Imperial Armenia 70a . 

The same statement may in fact also be applied to the Armenian 
kat'ohkos before the partition of the country. The solution to the 
problem of his right of ordination will be found in the actual number 
of bishops found in the Armenian Church at that time. According 
to the decision of the Council of Nicaea, ordination to the episcopate 
was to be carried out with the agreement of all the bishops of a given 
eparchy. Should it prove difficult for all of them to assemble, the 
actual presence of at least three bishops and the written approval of 
those absent was indispensable, and the decision was then to be sent 
for approval to the metropolitan 71 , We know that all the Armenian 
kat'ohkoi from St, Gregory to the time of Pap were consecrated at 
Caesarea; only Aristakes was ordained by his father, If this is 
the case, the consecration of Aristakes probably occurred before the 
Nicaean Council, at a time when such a right had perhaps been conceded 
to individual bishops. Had even three bishops existed in Armenia, 
the ordination of the kat'olikos would have taken place there, and 
only then have been sent to Caesarea for confirmation, Paustus tells 
us that after the deposition of Nerses I, King Arsak II summoned the 
Armenian bishops to consecrate a certain Cunak, but all refused, 
with the exception of the bishops of AljniF and Korduk', who came 
and carried out the king's wish 72 . The true facts here are that there 
were no Armenian bishops, and that Arsak was consequently forced 
to address himself to the representatives of neighbouring provinces 
who were in fact considered to be members of the Church of Syria, 



286 



CHAPTER XII 



The Armenians could not ordain bishops or a kat'olikos for themselves 
without asking for the co-operation of the nearby churches. When 
Paustus says that the bishop of Caesaxea had deprived the Armenians 
of their right of ordination, he is incorrectly attributing to a single 
period the conditions prevailing in the Armenian Church up to that 
time 72a . 

It would not be incorrect to say that the Armenian ecclesiastical 
hierarchy, which had consisted only of the houses of St. Gregory and 
Albianos, began its development in the period of the kat'ohkosate of 
Nerses I. In his description of the prosperity of the Church under 
Nerses, Paustus asserts, that, «... he [Nerses] increased the number of 
clerics [in all the localities of his diocese] . . . and appointed bishops in 
all the districts », The historian's further comment that, «... the 
honours of the father-bishops grew in all the provinces correspondingly 
to their merits, 73 » has the same meaning. The time of troubles after 
Pap must have benefitted the development and increase of clerical 
powers. The gradual decline of royal authority weakened the po- 
sition of the king as a restraining element and created conditions 
which favoured the spread of feudal power throughout the nation. 
The brief reign of Pap's successor Varazdat, who was the imperial 
candidate, was followed by the regency of Manuel Mamikonean, who 
devoted the seven years of his rule primarily to the regularization of 
the na^arwr system 74 . This is the starting point in the adaptation 
of the ecclesiastical organization to the na%arar pattern. Ecclesiasti- 
cal authority began to be parcelled out among the more powerful 
princely houses through the creation of separate bishoprics for each 
one of them. The system of metropolitans was not found here; 
there was no territorial basis for the dehmitation of ecclesiastical 
provinces 75 , The ecclesiastical administrative unit coincided exactly 
with the district of the princely house to which it belonged irrespective 
of its size. In the period preceding the fall of the Axsacids, the 
ecclesiastical hierarchy expanded to such a degree that by 451 there 
were 18 bishops representing the leading princely families of the 
country. 

Among these bishops, one bore the title of kat'olikos (KadoAiKos) 76 , 
and occupied with regard to the others the same position as the king 
to the princes, i.e. he was frimus inter fares. The bishops were 
called ter {wtp) [lord] like the princes, and just as they were the 
ecclesiastical representatives of the principalities, so the kat'olikos 



THE NAXABAM SYSTEM AND THE CHUECH 



287 



was seen primarly as the bishop of the royal province, specifically 
of the Axsacid domain of Ayrarat. Hence the statement of the 
historian already cited, that the kat'ohkos was the bishop of the 
royal court and blessed the royal table, is absolutely correct 77 . 

Spiritual rank and authority, just like that of the secular prices, 
was the hereditary prerogative of certain families. This was a strange 
and incomprehensible phenomenon to the Imperial Church, and was 
a subject discussed at the Council « in Trullo » of 692. Canon XXXIII 
of this Council decreed that, 

Having heard that in the land of Armenia only persons of 
clerical descent are accepted into the church, in imitation of 
Jewish custom .., we decree that henceforth, if anyone desires 
to enter the clergy, family shall not be taken into consideration, 
his worthiness for the spiritual vocation shall be proven accord- 
ing to the requirements laid down by the holy canons, and 
he shall then be consecrated, whether or not he be the descen- 
dent of servants of the church 78 , 

Several Armenian bishops were present at this Council, among them 
Marianos, bishop of Kitharizon. The above information had un- 
doubtedly been provided by them, and the problem must have been 
brought up for discussion at the Council at their suggestion. Conse- 
quently, there is no reason to doubt the accuracy of the evidence 
furnished to the Council, There was no question in this case of 
imitating Old Testament practices as the fathers of the Council 
supposed. If any similarity between the two existed, it must be 
explained by the fact that the Armenian hierarchy, just like the Old 
Testament priesthood, had developed in a society where family ties 
prevailed. Hereditary succession and other points of similarity 
between them were the results not of imitation, but only of the identical 
social conditions under which the Armenian Church and the biblical 
priestly caste had developed 79 . 

The patriarchal authority was likewise hereditary, and it stayed 
in the house of the Illuminator as long as the corresponding political 
power remained in the hands of the Arsacids. At the moment of 
the fall of the Arsacids at the beginning of the fifth century, the 
question of the deposition of the patriarch Sahak I was simultaneously 
raised. After this plan had been carried out, the patriarchal throne 
remained in an equivocal position. Just as the bishops were the 
ecclesiastical representatives of the various principalities, so the 
patriarch depended on the king and was appointed by the Arsacids 
to his position as court bishop. After them, this prerogative passed 



288 



CHAPTER Xn 



to the Persian Mug, However, neither the Persian appointees : 
bishops Surmak, Brkisoy, and Smovel, nor the close disciples of 
Sahak I, Mesrop and Yovsep', all of whom ruled over the Church, 
were ever considered to be kat'ohkoi, but only substitutes. The 
reason for this is unquestionably to be found in the fact that ecclesi- 
astical society was dominated by hereditary succession, and that the 
question to whom the throne should pass in the absence of legitimate 
heirs was still open and awaiting a solution. We can now see why 
the forgotten but honourable house of Albianos came once more to 
the fore at this point, and why its bishops, Melite and Movses ascended 
the patriarchal throne 30 . This attempt to solve the problem his- 
torically was not successful, and bishop Giwt' from the Mamikonean 
province of Tayk' took over the patriarchal dignity from the Albianids, 
probably as a result of the patronage of Vahan Mamikonean. The 
position of the Church was consolidated by the favourable outcome 
of the rebellion at the end of the fifth century and the importance of 
the clergy began to grow. The problem of patriarchal authority 
was resolved in favour of an elective basis as a result of rivalries among 
bishops, and possibly also as a consequence of the influence of neigh- 
bouring Churches with whom relations were tightened in this period. 

The bishop of the Mamikonean held the leading position in the 
Church after the patriarch, or kat'olikos ; the influence of the nax^rar 
system is obvious in this case. Just as the hereditary Mamikonean 
sparajpets stood at the head of the naxarars under the Arsacids and 
even later, so in ecclesiastical affairs, the chief administrator found 
at the side of the kat'otikos was the representative of the same house. 
The leading role played in the ecclesiastical affairs of the sixth century 
by Nersapuh, bishop of the Mamikonean, is well known. 

If our hypothesis of a nayarar foundation for the ecclesiastical 
organization is correct, some relationship between princely houses and 
ecclesiastical representation must have existed ; we should expect the 
more important princely houses to have their own bishop, and vice- 
versa. And indeed, on comparing the list of bishops with the list 
of principalities, we find that all the powerful houses have ecclesiastical 
representatives. The best index of the relative might of the principa- 
lities we have investigated is found in their territorial relationships* 
Unimportant principalities are those whose possessions were included 
in large princely territories. With few exceptions, the former are 
the very principalities which lack representatives in the ecclesiastical 
hierarchy. 



THE OEIGIN OF THE NAXARAR SYSTEM 



XIII 



PEELIMINAKY EXCUBSUS 

Definition of the problem and methods employed for its solution — Positive and 
constructive elements in the study of the problem — The basis of the state in the East 
— Tribal organization as the basis of the state — £var7)fj,a juet£ov — The Iranian 
tribal system and its features: tmima, vi9, zaniu, dahyu zana — Clan life and the abi- 
caris-bazar-iAevdepa dyopd — The disappearance of tribal institutions in the Parthian 
period — The structure of the Arsacid Empire. 



In order to determine the nature of the na-^arar system as a historical 
institution, the formulation of the problem is of the greatest im- 
portance. Since our immediate purpose here is to outline the position 
of the naymwrs in the century of Justinian, which is our concern, 
a study of the qualitative aspects of the naymar system must likewise 
be made within these limits, i.e. within this span of time, but is it 
possible to carry out such an undertaking from a strictly scholarly 
point of view a ? 

In the period under consideration, the nayamr system was passing 
through certain alteration as it entered a new phase in its development. 
This evolution of the na-^arar system, like every other aspect of Ar- 
menian life, did not follow a straight line in either time or space. 
The evolution of life was far from uniform throughout the country ; 
at a time when a particular situation was barely making its appearance 
in one section, it was already dying out and obsolete in another. 
Varying topographical conditions and uneven degrees of contact with 
the general political life of the country created socially and culturally 
dissimilar areas. In one section ancient forms of life were still alive 
and the traditions of a very distant past were occasionally still preser- 
ved, while in the same span of time, other sections altered their social 
structure several times. 

Historical literature is unable to reproduce this social scene in all 
its variety. Writing made its appearance in the period of disinte- 
gration of the Arsacid system, and the sources relating both to this 



290 



CHAPTER XIII 



r 



and to a slightly later period axe not political in nature. It is true 
that these sources contain very valuable information on this or that 
aspect of nax^Tar customs, but in general they are insufficient for a 
definition of the entire pattern. This is all the more true because 
we are often unable to grasp their real value and consequently lose 
ourselves in conjectures. A correct interpretation of the technical 
terms found in the sources should uncover the fundamental aspects 
of the naxarar past, but any attempt to establish their exact meaning 
by means of textual comparison is doomed to failure, One and the 
same term can be used by a single author in completely different 
senses, and the assumption of stylistic defects or textual errors in 
such cases is not always warranted. It is true that the sources were 
reworked, and that the replacing of one term by another is altogether 
possible, but the lack of uniformity in technical terminology and its 
variations in meaning cannot always be explained in this manneT. 
Many terms were inherited from a distant past, had taken on new 
shades of meaning corresponding to new relationships as they passed 
from century to century, and had finally emerged as complicated 
concepts with an elaborate content in the period of literacy. To 
uncover the historical layers accumulated in each term, it will be 
necessary to trace once again the genetic evolution of the naxo/rar 
system, 

The surviving literary sources on the history of the naxarar system 
do not give us a complete understanding of it. Using the simile of a 
famous investigator of feudalism in Western Europe, we can say that 
the naxo/rar system seems like an ancient tree. At best, the sources 
display before our eyes the leaves and the top of the tree, but its 
roots remain hidden from our sight, and to reach them we must dig 
far into the ground 1 . For such a purpose, a purely philological 
method based on the study and comparison of texts is not sufficient, 
and it will not produce the results desired. Some guiding concept 
leading to the hidden roots is indispensable, To be scientifically 
tenable, such a concept must be drawn from the historical context in 
which the Armenians were living. The nax^ar system is a socio- 
political institution, hence its roots must be sought in the common 
political and social conditions of the Iranian states. These were the 
conditions under which Armenian life developed, and it was necessarily 
formed under the influence of political and legal concepts and norms 
similar to those of Iran, Once we have characterized the bases of 



THE NAXABAE SYSTEM — PRELIMINARY EXCURSUS 291 

the Iranian state, we shall have a foundation allowing us to tie together 
our scattered material on the nax^rars and to weld it into a whole. 

The state as a legal system was understood and formed in a different 
manner in the East and in the West. In the West urban organization 
played the leading part in the development of the concept of the 
state; in the East, the same role was assigned to family associations. 
Experts have characterized the Roman Empire as an association of 
city states, whereas they see the Persian Empire as the highest mani- 
festation of the tribe as a political unit 2 . In the subsequent develop* 
ment of political life in Iran, family relations remained the basic 
generative principle of political life, regardless of the supreme power 
which was passed from one dynasty to another. 2 a 

At the time of their appearance on the historical scene, the Persians 
were composed of numerous tribes, as were their neighbours the Medes. 
Deioces the Mede, having gained the leadership of the Median tribes, 
destroyed the power of the Assyrians, while Cyrus, having united 
the scattered Persian tribes, led them against the Medes and laid the 
foundation of the Achaemenid Empire. Alexander of, Macedon in 
turn destroyed the power of the Achaemenids, and his successors 
ruled until they were forced to give way before the greater Parthian 
threat. Through all these dynastic changes, the Iranian Empire 
remained a loose tribal confederation (owny/za /zei£oj>), united 
voluntarily or often also, by forcible means. At the head of the 
confederation stood the closest knit or best organized tribe such as 
that of the Pasargadae, whence sprang the Achaemenids, who looked 
upon the kingship as the possession of their clan or iauma, Darius 
never speaks otherwise than in the name of his clan and it is evident 
from his many statements that family or clan interests motivated 
his policy 3 . The struggle culminating in the suppremacy of the 
Achaemenid house was carried on first, against the Median Magians 
and subsequently, against other tribes which sought to break away. 
As a result of this constant struggle, the Achaemenids succeeded in 
uniting the tribes and in creating a single powerful confederation 
(cjvarrjfjia juei£oi>), according to the formula of ancient authors 4 , 

The Persian authorities recognized several dozen countries with 
different peoples. These countries were called dahiu [dahyu], and 
their heads, dahyu-paiii. Dahyu is the Iranian form of the Sanscrit 
dasi, « an enemy » or « a slave », and designated the countries of enemies 
or slaves 4 a . This characteristic term renders correctly the attitude of 



292 



CHAPTER xm 



the ruling clan toward the other tribes. The dahyu formed a more 
or less uniform ethnic group, a so-called zana. On his inscriptions 
the Persian king is called nysayaBiya dahyunam vispazananam», or 
« puravzananam, » i.e. the dahyu with various or numerous zana, 
« peoples » 5 , Zana is the ethnic form of the concept whose territorial 
equivalent is the dahyu. 

The subjection of the various countries or dahyu was demonstrated 
by their payment of a given tribute, bajim abarantd. In its function as 
a provincial unit or tax district, the dahyu was known as a saira [^saBra] 
and its head as a x§a8rapa or satra-pati [ ?]. In other words, the unit 
called dahyu from the point of view of tribal possession became a 
xsadra according to politico - administrative divisions. The two 
units did not always coincide in size : occasionally several dahyus 
made up one -gsaBra, Under the Achaemenids there were 23 to 28 dahyus 
while, according to the division of Darius, the whole Empire was 
split up into 20 satrapies. The dahyus varied in importance and size ; 
some of them were even autonomous kingdoms whose rulers were 
entitled ysdyaBia = sah, and in relation to them the king of Persia 
became ysayaBianam — xsaydBia, i.e. sahanmh. Persia, Media, 
Parbhia, Armenia, Babylonia, and others were dahyus of this type 5a . 

The dahyu or zana was the largest autonomous unit included in 
the Empire, The zana had its own interior organization : it was 
subdivided into separate tribes and the latter were split into larger or 
smaller clans, Thus, for example, the Persians and the Medes were 
dahyus, the former having 10 tribes, and the latter 6 6 . One of then 
Persian tribes was that of the Pasargadae, and one of the subdivisions 
of the Pasargadae was the clan of the Achaemenids, Herodotus calls 
the Pasargadae, y4vo$> and the Achaemenids, <f>prjTp7]. In the in- 
scriptions of Darius, the clan of the Achaemenids is designated by the 
term tauma; consequently, as a social eoneept, tauma is equivalent to 
^prjTpT},. The zantu \*dantu] was the next step in the social grouping, 
corresponding to the Greek yivos. One of the Median tribes is called 
dpildvroi, i.e. «the noble zantu or tribe)); hence we may conclude 
that zantu is the name of each of the 8 [sic] units or yevza. Since 
the Persian tribes were social units of the same rank as the Median 
ones, the term zantu is also applicable to them. Thus, the whole 
Persian nation formed one zana, the Pasargadae as one of the tribes of 
that zana were a zantu, and the Achaemenids as one of the clans 
composing the zantu were a tauma 6a , 



THE NAXABAM SYSTEM — PREUMINAIIY EXCURSUS 



293 



In the inscriptions we also find the terms vid and Mm. Side by- 
side with the expression mana taumaya or amayam taumaya we find 
vidam tyam amayam, i.e. « our iauma, our vi9)>. Kara refers to the 
entire nation Persian or Median, «Mra Parsa Jiya vidapatiy 
Jcara Mada hya vidajpatiy {i.e. « the people of Persia living 
according to wfi-s, the people of the Medes living according to w0-s»). 
Prom these expressions it is evident that the Jcara was split into vids. 
The relations of these terms to the tauma and the zana is not clear, 
They might be taken as intermediary steps in the social grouping, 
but it seems more likely that vid was the territorial expression of the 
ethnic concept tauma, just as dahyu was the equivalent of zana. As 
for Mra, « powerful », it is rather a military term and designated the 
active element of the zana, the element fit for war, whence comes its 
secondary meaning of « army » 7 . 

.. In the literature of the Avesta, the terms nemana [nmana], vis, 
zantu, dafahu [dahyu] are often found for different levels of social 
groupings made up of men related by blood. According to relatively 
late scholia, the nmana was composed of five couples of men and women, 
the vis of 15 couples, the zantu of 30 couples, and the dahyu of 50 
couples, The important part here is found not in the numerical fig- 
ures, which should be taken with reservations, but in the hierarchy of 
the groups indicated by these figures. The nmana is given as the first 
step of association, designating the sum of individuals living under 
one roof. Several similar nmanas formed a vis, several vis a zantu, 
and several zantu made up one dahyu or people, i.e. a zana. The 
members of these groups were called nmanya, visya, zantuma, dahyuma, 
and their heads, nmano-paiii, vis-paiti, zaniu-faiti, and dahyu-jpaiii 8 . 
The term tauma, which is already familiar to us, is not found here ; 
it seems to be replaced by nmana, which is given in the inscriptions 
of Darius in the form maniya. But we have already said that iauma 
is the ethnic counterpart to the territorial vis; it is, therefore, more 
likely that nmana, as a social term, belongs to a late period and 
corresponds to the family derived from the tauma. 

Among tribal terms, some indicate blood relationships (termini 
geniis), namely tauma, zantu, zana ; others point to a common territory 
(termini loci), namely nmana, 'vid, dahyu. It is difficult to establish 
correspondences between the terms of these two categories ; with the 
passage of time certain terms supplanted others, a territorial term 
takes on an ethnic connotation, etc... 9 It is important to determine 



294 



CHAPTER XIII 



the lints in the tribal organization. The last of these is the dahyu, 
above which comes only the association into a political group and an 
administrative unit under the leadership of the supreme ruler, 

From the point of view of the internal life of the tribe, the viB 
should be considered the most active link. The starting point for 
the history and law of the Aryan peoples was not the family but the 
clan, gens. This was the original form of human association, and 
from it developed the subsequent, more complicated units ; similarly, 
the patriarchal family was derived from it by way of the disintegration 
of the clan into smaller groups. The vid apparently was such a 
starting point in Iran. Darius I, in his inscriptions, often speaks of 
the vid and stresses its interests, The Medes and the Persians were 
divided into vi8s, viBd'paiiy, The king proclaimed that he had rebuilt 
the dyadana destroyed by the usurper Gaumata and had taken from 
him the abicaris, gaiddmcd mdniyamcd vidbis-cd* It is evident from 
this that the institutions listed were also divided according to viBs. 
Finally, even the gods were called by the name of viBs, bagaibis 
viBaibis 10 , This point is very important, since vid, as well as other 
terms, was preserved, with very interesting alterations, in Armenian 
nax^rar terminology, as we shall see later. 

Interesting information on Persian tribal hfe is found in Xenophon, 
It is true that the Greek writer's account is somewhat tendentious, 
but it is essentially truthful and throws a good deal of light on certain 
aspects of Persian life, The historian relates that, 



They have their so-called " Free Square ", (iAzvdipa dyopd) 
where the royal palace and other government buildings are 
located .,, . This square, enclosing the government buildings, 
is divided into four parts; one of these belongs to the boys, 
one to the youths, another to the men of mature years, and 
another to those who are past the age for military service. 
And the law requires them to come daily to their several 
quarters — the boys and the full grown men at daybreak ; 
but the elders may come at whatever time it suits each one's 
convenience, except that they must present themselves on 
certain specified days. But the youths pass the night also 
in light armour about the government buildings — all except 
those who are married; no inquiry is made for such, unless 
they be especially ordered in advance to be there, but it is not 
proper for them to be absent too often. 

5. Over each of these divisions there are twelve officers, 
for the Persians are divided into twelve tribes. To have 



THE NAXARAR SYSTEM — PRELIMINARY EXCURSUS 



295 



charge of the boys, such are chosen from the ranks of the elders 
as seem likely to make out of the boys the best men ; to have 
charge of the youths, such are chosen from the ranks of the 
mature men as seem most likely on their part to develop the 
youths best ; to preside over the mature men, those are selected 
who seem most likely to fit them best to execute the orders 
and requirements of the highest authorities; and of the elders 
also chiefs are selected who act as overseers to see that those 
of this class also do their duty. 

The historian then pauses to consider the Persian system of education, 

... they learn to shoot and to throw the spear, 

This, then, is what the boys do until they are sixteen or 
seventeen years of age, and after this they are promoted from 
the class of boys and enrolled among the young men. 

9. Now the young men in their turn live as follows: for ten 
years after they are promoted from the class of boys they 
pass the nights, as we said before, about the government 
buildings. This they do for the sake of guarding the city and 
of developing their powers of self-control ; for this time of life, 
it seems, demands the most watchful care. And during the 
day, too, they put themselves at the disposal of the authorities, 
if they are needed for any service to the state. Whenever it 
is necessary, they all remain about the public buildings. But 
when the king goes out hunting, he takes out half the garrison ; 
... [12] ... . And of the youths who remain behind, the autho- 
rities employ any that they may need, whether for garrison 
duty or for arresting criminals or for hunting down robbers, 
or for any other service that demands strength or dispatch. 

Such then, is the occupation of the youths. And when they 
have completed their ten years, they are promoted and enrolled 
in the class of mature men. 13. And these, in turn, for twenty- 
five years after the time they are there enrolled, are occupied 
as follows. In the first place, like the youths, they are at the 
disposal of the authorities, if they are needed in the interest 
of the commonwealth in any service that requires men who 
have already attained discretion and are still strong in body. 
But if it is necessary to make a military expedition anywhere, 
those who have been thus educated take the field, no longer 
with bow and arrows, nor yet with spears, but with what are 
termed " weapons for close conflict " — a corselet about 
their breast, a round shield upon their left arm (such as the 
Persians are represented with in art), and in their right hands 
a sabre or bill. From this division also all the magistrates 
are selected, except the teachers of the boys, 

And when they have completed the five-and-twenty years, 
they are, as one would expect, somewhat more than fifty 



296 



CHAPTER Sin 



years of age; and then they come out and take their places 
among those who really are, as they are called, the " elders ". 

14, Now these elders, in theiT turn, no longer perform military 
service outside their own country, but they remain at home 
and try all sorts of cases, both public and private. They 
try people indicted for capital offenses also, and they elect 
all the officers, And if any one, either among the youths or 
among the mature men, fail in any one of the duties prescribed 
by law, the respective officers of that division, or any one else 
who will, may enter complaint, and the elders, when they 
have heard the case, expel the guilty party ; and the one who 
has been expelled spends the rest of his life degraded and 
disfranchised. 

15. Now, that the whole constitutional policy of the Persians 
may be more clearly set forth, I will go back a little ; for now, 
in the light of what has already been said, it can be given in a 
very few words. It is said that the Persians number about 
one hundred and twenty thousand tool^jx; and no one of these 
is by law excluded from holding offices and positions of honour, 
but all the Persians may send their children to the common 
schools of justices Still, only those do send them who are 
in a position to maintain their children without work; and 
those who are not so situated do not. And only to such as 
are educated by the public teachers is it permitted to pass 
their young manhood in the class of the youths, while to those 
who have not completed this course of training it is not so 
permitted. And only to such among the youths as complete 
the course required by law is it permitted to join the class of 
mature men and to fill offices and places of distinction, while 
those who do not finish their course among the young men 
are not promoted to the class of the mature men. And again, 
those who finish their course among the mature men without 
blame become members of the class of elders n . 



According to Xenophon, his hero Cyrus passed through all the 
classes. What the historian is apparently attempting to present here, 
is an ideal Persian system of education so that the details are probably 
exaggerated, but in the main his account corresponds to reality. 
One of the ancient institutions of tribal life, the free square {iXsvddpa 
dyopd) is described ; this is the common gathermg place, the institution 
which is called abicaris on the inscriptions, and the one over which 
Darius struggled with the impostor Gaumata. Abicaris coincides 
phonetically and semanticaHy with the Sanscrit sabMcam (from 
sabha, « gathering » and cam, «to go») which means « an assembly », 
Among the Indians, the assembly of communities living in villages, 



THE NAXABAM SYSTEM — PRELIMINARY EXCURSUS 



297 



grama, under the rule of an elder, gramani, was called sabha. The 
sabM was sximmoned together for the discussion of public affairs 
as well as for festivals and common entertainment. These assemblies 
had so much importance that men were sometimes evalxied according 
to their usefulness in assembly ; a man was prized as sablieya 13 , 

The representatives of the twelve Persian tribes met in the common 
square. Up to ten thousand men were found in each tribe so that 
there were altogether 120 thousand Persians (incidentally, this is a 
staggeringly low figure for a nation wielding such fearful might). 
This institution supports our hypothesis that the Persian Empire 
was founded on a basis of confederation. The explanation of Xeno- 
phon that, only the rich assembled in the common square, even though 
all the members of the tribe were entitled to do so, proves that the 
ancient tribal institutions were losing their true nature, and shows 
the fashion in which an institution based on the equality of its members 
was becoming a source of class differentiation and inequality. The 
former simple assembly was transmuted into a sort of military school 
intended in reality for the wealthy classes of the society; the poor 
had no means of access to such an institution. Cyrus in his speech 
to the soldiers comments in passing, « ,.in our own country you did 
not enjoy equal privileges with us, not because you were excluded 
from them by us, but because you weTe obliged to earn your own 
living 13 ». In this fashion, the free square or abicaris served as a 
means for the subdivision of society into strata ; it created a class of 
privileged individuals and secured for them the leading positions and 
the exlusive right to govern the country. Among the thirty thousand 
warriors of. Cyrus, only one thousand belonged to the Tank of the 
privileged, or, as he calls them, of the omotimoi 14 . 

Xenophon asserts that all commerce was forbidden in the free 
square I4a , Nevertheless, both the New-Persian bazar, « market » 
and the Armenian vacar {i[w£wn) are derived from abicaris 14h ; conse- 
quently the earlier institution obviously degenerated at a later date 
into a market for the exchange of goods. As a term, abicaris lived 
through three stages of development, passing from a tribal institution 
to a rnilitary-aristocratic and eventually to a middle class one. The 
abicaris led to the formation of a class of omotimoi within which lay 
concealed the seeds of the future noble estate. The square once called 
free, probably in the sense of its common accessibility, subsequently 
became the exclusive property of the free class. Differentiation 



298 



CHAPTEK XIII 



within the homogeneous mass points already to the disintegration of 
the tribal bases of life, and is attended by the inauguration of the 
process of feudalization 14c , Careful investigation of the phenomenon 
known as feudalism has revealed all the complication of its nature, 
and we now speak not only of medieval feudalism, but also of primitive 
feudalism developing from the dissolution of tribal society. In early 
periods of history, one of the means of maintaining social equilibrium 
was communal land tenure. With the destruction of this form of 
possession, economic inequality developed, and brought with it the 
uneven distribution of political power. This is the fashion in which 
one of the main components of feudalism, namely the association of 
land tenure and political power makes its appearance. 

It is impossible to determine the exact point at which tribal forms 
turned into feudal ones. The seeds of feudalism as well as those of a 
tight political structure in the Roman sense were already visible in 
the Achaemenid period 14d . By-passing the ethnic groups, of which 
there were as many as 70, the Achaemenids first divided the whole 
of the Empire into 20 administrative units, the so-called satrapies, 
and separated military and civilian powers,, entrusting the former to 
jpJirourarc'hs and chiliarchs and the latter to satraps 15 . If any periodi- 
sation is permissible here, the period of Parthian domination is the 
one to be taken as the turning point in the history of socio-political 
relations in Iran. Under the Achaemenids, the political structure 
rested on tribal forms of association. In the subsequent Macedonian 
— Seleucid period, political bases began to gain strength under the 
influence of the western conquerors. At the time of the accession of 
the Parthians who were emerging from the tribal order, like other 
Iranian peoples, the developing state clashed with tribal foundations. 
Parthian tribal traditions had given way in the Seleucid political 
milieu and acquired the characteristics of feudalism. The process of 
feudalization was expressed by the fact that the Parthians, having 
turned into a military class, became omoiimoi with regard to the rest 
of Iran. The evolution in the internal life of the Parthians, as a 
separate ethnic group, inevitably led to results similar to those found 
among the Persians, i.e., their society was fragmented as had been 
that of the Persians 15a , 

Parthia had been settled by various tribes belonging to the Iranian 
family. Although called Parthian, the royal house of the Arsacids 
was not descended from the Parthians themselves, but from the 



THE NAXABAR SYSTEM — PRELIMINARY EXCURSUS 



299 



neighbouring tribe of the Da5n. The Dahi were a nomad Iranian 
group pushed "beyond the border of the country- yparthau- and may 
perhaps be related to nomads with the same name mentioned by 
Herodotus 16 , The former lived next to the Mardians on the Iranian 
plateau; then, moving to the west, they found themselves on the 
border of Iran, in the plains of present-day Turkestan. The Dahi 
consisted of various tribes 17 and the new conquerors, as personified 
by Arsak [Arsaces], were descended from the tribe of the Aparnoi. 
In the middle of the third century [B.C.], as the result of internal 
tribal dissentions, the Aparnoi were compelled to abandon their lands 
and attacked neighbouring Parthia, which they conquered. Thus, in 
the strict sense, the Arsacids were conquerors with regard to Parthia. 
A century later, having consolidated their position, they emerged on 
the historical stage in the mid-second century under the guise of a 
Parthian dynasty. The fact that Dahi as well as Parthian tribes 
were included in the royal family is no longer questioned 17a , 

In Achaemenid inscriptions, Parthia is mentioned as an autonomous 
dahyu. Nothing is known of its internal tribal organization, but some 
information concerning its political structure in the royal period has 
survived 18 . A type of collegiate institution, a council called ordo 
jprobulorum, existed under the king, and individuals were chosen 
from among its members for military and civilian offices 19 . In its 
functions, this institutions is reminiscent of the class of omotimoi, 
or more exactly of mature men among the Persians, since they too 
had been a ruling class with exclusive rights to public offices. Accor- 
ding to one account, the Parthians had two such institutions. One 
was composed of the relatives of the ruling dynasty, the other of the 
wisemen and Magians, and the kings were selected jointly by the two 
councils 20 . Historians often mention a Parthian senate on whom 
depended the fate of the kings 21 ; the above institutions were probably 
the ones refered to. If the first of these corresponds to the class of 
mature men among the Persians, then the second should be equated 
with the class of elders among the same Persians. Both institutions 
were unquestionably descended from the tribal period and were the 
ultimate development of primitive institutions. The existence of 
such institutions is the best indication of the process of social strati- 
fication taking place among the Parthians. 

We know that Parthian society was divided into freemen and slaves. 
The number of slaves grew constantly, since the custom of manumission 



300 



CHAPTER XIH 



did not exist. Because of the small number of freemen, the main 
contingents of the army, in contrast to the practices of other nations, 
were composed of slaves who were taught to ride and to shoot like 
free men 21a . The free men, according to their wealth, furnished a 
number of knights to the ruler. Thus, at the time of the war with 
Marc Antony, 400 freemen led against him a cavalry of fifty thousand 
men. One of the characteristic customs of the Parthians was their 
attachement to horseback riding ; They almost never dismounted, and 
not only fought, but also ate. and rested on their horses ; public business, 
trade, and assemblies likewise took place on horseback, The outward 
distinction of freemen from slaves was that the former rode, while 
the slaves went on foot 22 , Life on horseback, the distinguishing 
custom of nomads, serves here as an indication of the indebtedness 
of the Parthians to their nomad past, In the new political context, 
the Parthians' natural inclination toward horsemanship turned them 
into a superior warrior class, and this circumstance was reflected in 
their administrative policy. The sigmfieance of Parthian military 
customs must be taken into consideration in any explanation of the 
causes which led their state toward feudalization 22a . 

According to Pliny, the empire of the Parthians, bound by the Red 
and Hyrcanian Seas, consisted of eighteen kingdoms 23 , Of these, 
eleven were called swperiora and stretched from the border of Armenia 
to Scythia, while the other seven made up the regna inferior®, The 
historian does not give the names of these kingdoms, but they pre- 
sumably were the provinces listed in the Itinerary of Isidore of Charax 24 , 
The provinces described by Isidore include Parthia, Media, Assyria, 
and Mesopotamia, together with their subdivisions, "Whether or not 
eighteen kingdoms were to be found within the boundaries of these 
provinces, they must be taken as the Parthian possessions par excel- 
lence. . Along their periphery lay Atropatene and Armenia in the north, 
Persia-Karamania in the south, Adiabene and Osrhoene in the west, 
and Bactria with Sogdiana in the east 24 *v In the period of uprisings 
against the Seleucids, some of these lands broke away, at the same 
time as the Parthians, and formed independent countries, while 
others had accomplished this even earlier. This is the form in which 
they were subsequently incorporated into the Parthian Empire so 
that the border territories were differentiated from the central provinces 
through their political status 24 V 

The Arsacids looked upon their conquests as the possessions of their 



^^^^7?^ 



THE NAXABAJR SYSTEM — PRELIMINARY EXCURSUS 



301 



house, and consequently sought to seize the thrones of the subjected 
kingdoms to apportion them among their kinsmen. They succeeded in 
replacing the native dynasties by ATsaeids in the kingdoms listed 
above, but the junior Arsaeid lines soon identified themselves with 
the countries in which they were installed and took the same hostile 
attitude toward the central power as the former native dynasties. 
Family claims to power were one of the obstructions which prevented 
the creation of a single political organism. The Parthian Empire 
was not a single state, but rather a conglomeration of small units 
with sovereign functions. 

The main Arsaeid line enjoyed a position of seniority; the heads of 
the component kingdoms were entitled kings, while the head of the 
central power was the king of kings. This pompous title was an 
inheritance from the Achaemenids, while the Achaemenids in their 
turn had borrowed it from the Assyrians. Since the first Arsacids 
had simply been called kings and they first took the title of king of 
kings only after the great extension of their territory under Mithra- 
dates I [ca. 171-138 B.C.] 24c , we must suppose that this action was 
not merely an imitation of the Achaemenids, but that the new title 
corresponded to the existing political situation. According to Muslim 
sources which go back to Sasanian accounts, the power of the Askuni 
was acknowledged by 90 kings, who were the rulers of 90 countries 25 . 
According to the same source, the empire of the Parthians was a 
union of kings (muluk-ai-TawdSf), i.e. a federated empire composed 
of kings from various nations and tribes. The formula uvor^jxa 
psilov is also applicable to the Parthian Empire, since it was as great 
a confederation as that of the Achaemenids, with only the components 
changed. The confederation was joined together on the same basis 
as before : the actual dependence of the moluks or reguli was expressed 
by their fulfilment of the basic duties which had characterized Iranian 
kingdoms even before the Parthians, namely the paymentofj^tribute 
and the performance of military service 36 . In such a system, the 
problems and functions of the state are reduced to the rninimxim, and 
this type of relationship of the parts to the whole is the very essence 
of vassalage, one of the essential attributes of feudalism. 

Faced with such a mechanism, the state obviously could not have 
a profound influence on life. According to ancient concepts, power 
sprang from two sources, either from the right of j&oafuest, or from the 
right of birth S7 , In the first case, the power was limited in its ac- 



302 



CHAPTEK XIII 



tivities by the narrow bounds of the existing political structure. 
Power based on the pre-eminence of blood, on the other hand, included 
all other ramifications and aspects of society and functioned through 
the forms of tribal institutions. The rights of blood were the bases 
of society, and the legal relationships derived from them were charac- 
terized by a remarkable stability. This should serve as a guiding 
line in the achievement of our goal : the clarification of the bases of 
social life in Armenia, 



XIV 



THE TEIBAL BASES OE THE NAXARAR SYSTEM 



Two periods in the history of Armenia: the Tigranids and the Arsacids, i.e, the eras 
of the Icomarchs and the strategoi — Tribal life in the earlier period — The ethnic 
structure of Armenia — The decisive moment in the creation of the Armenian nation — 
Its essential characteristics — The transformation of tribal features: malxaz, aspet, 
mamah, ter, through an a posteriori reconstruction — ■ The etymology of these terms 
as synonyms designating tribal leaders in different ethnic groups — The process of 
social stratification Iberia and Albania — The aristocratization of Armenia in the 
time of Tigran the Great — The leading representatives of the nobility in this period: 
the four bde$xs f Bagadates, Mithrobarzanes, and Mankaios as the ancestors of the 
princely houses of the Bagratids, Sophene < Arcruni, and Mamikonean — The Xor- 
Xofuni house — The appearance of classes in other tribal groups: the Mardpet as the 
leader of the M avians, the prince of Moh¥ or Moxene, the prince of the Kurtisms or 
Korcek, hypotheses concerning the origin of the Amatuni and Muracan — The tribal 
origin of Siwnik'-Sisakan, §irak, and others — The completion of the social stratifi- 
cation in the epoch preceding the appearance of the Arsacids in Armenia. 



Armenia, as one of the component parts of the politico-cultural 
Iranian world, was destined to pass through the same evolution. 
The essential phases of development, the outstanding moments in 
the history of Iran, necessarily affected Armenian Hfe. In this 
sense, the historical periodization of Iranian life is also applicable 
to that of Armenia, Parthian domination is the turning point intro- 
ducing a new era in the history of Armenia as well as in that of the 
Persians a . Since the Parthian dynasty was consolidated in Armenia 
somewhat later than in Persia, specifically at the beginning of out 
era, we should begin our reckoning of the Arsacid period of Armenian 
history from this date. The period preceding the appearance of the 
Arsacids may be called Achaemenid, from the external point of view, 
and from the internal one, Tigranid, after the dynasty to which be- 
longed Artaxias [Artases] and Zariadris [Zareh], as well as their 
famous descendent Tigran II the Great \ Socio-politically, this 
epoch corresponds to the pre-Parthian period of Persian history, 
and it is primarily an era of tribal relationships. 



304 



CHAPTER XIV 



From the point of view of social evolution, these two periods in 
Armenian history : the Tigranid and the Arsacid, may be called the 
eras of the Jcomarchs in the former case, and the generals [strategoi] 
in the latter la . Xenophon's description of the Armenian countryside 
at the time of the retreat of the Ten Thousand (ca, 400 B.C.), gives 
an adequate characterization of the social life of the country. Armenia 
was composed exclusively of villages headed by chieftains, Kwiidpxrjs 
or ap^cov rrjs kcojjltjs. These Jcomarchs were members of the ad- 
ministration, as were the satraps, who were the representatives of 
the Persian king. From the patriarchal picture of the villages given 
by Xenophon, it is easy to guess that the Jcomarch was the head of 
the particular clan settled in a given village. The characteristic 
fortifications around the villages, the primitive households with 
subterranean dwellings in which the animals, sheep, goats, cows, and 
the domestic stock in general were housed together with human beings, 
all these are part and parcel of a tribal form of community life. The 
elder who guided the Greeks refused all that the strangers offered him, 
but, « ...whenever he caught sight of one of his hinsmen, he would 
always take the man to his side 2 ». It is impossible to miss in this 
fatherly concern the characteristic function of the father-leader of 
the clan, 

The picture of the country found a few centuries later, specifically 
at the beginning of the Arsacid establishment in Armenia, is quite 
different. By this time, Armenia was divided into 120 districts or 
prefectures called strategies, among which were several former king- 
doms, « ...dividitur in praefecturas, quas strategias vocant, quasdam ex 
Ms vel singula regna quondam, barbaribus nominibus CXX 3 ». To be 
sure, the nature of these divisions is not very clear, and we do not 
know who gave them the name of strategies, or to what local terms 
this name corresponded. We will return to this problem later, but 
in any case, there can be no doubt that strategia is a military term and 
that as such it indicates that m ilitary activity was the first duty of 
its head or strategos. We know that the Armenian princes were 
liable to military service, and that they were obliged to present 
themselves before the king, as soon as summoned, at the head of their 
cavalry contingent. Consequently, we have grounds for admitting 
that the strategies were ruling principalities, even if only in the initial 
stage of their development. 

The strategoi and the Jcomarchs stood on different steps of the social 



THE TRIBAL BASES OE THE NAXAMAB SYSTEM 



305 



scale and were separated by four centuries. They mark two very 
important points of demarcation in tlie Hstory of the growth of 
social forces in the country. Our problem consists in discovering the 
nature of the historical process which led from the Jcomarclis to the 
stmiegoi, from the tribal stage to the feudal one, 

A tribal pattern is acknowledged to be the indispensable stage in 
the life of every people. Armenia offered a particularly favourable 
environment for such a pattern because of its ethnic and geographical 
setting, Set at the crossroads of great popular migrations, she retained 
and absorbed into her own soil numerous racial strains and national 
currents, The natural conditions of the country were not favourable 
to fusion : tall mountain ranges blocking off the land in all directions 
created a network of valleys and gorges, each of which formed an 
isolated community peculiarly suited for the preservation jof^tribal 
characteristics. A multitude of such communities led "to a tribal 
pattern of life and foreshadowed its conservatism 3a , 

Little factual data on the tribal period of Armenian history is 
admissible a priori. What there is consists primarily of those terms 
giving us some understanding of the social structure in the days of the 
JcomarcM which have survived in early Armenian literature, and of 
their transformations. The most notable among these are the Iranian 
terms : mn£S [tohm] tawma, wqli [azri] zcma]%\ the survival of vi$ in the 
form * t[jju [vis] * i[wu [vas], uktymi [sefjyuK], i[w£mn [vacaf] abiearis, 
w^fvwpi [asxwrft] xs a @ m > J"^ 2 "^ [$®hwp] SaBrapa [? ], et al. Some of these 
are borrowings from the Sasanian period, others go back to antiquity, 
possibly to the period of the Achaemenids 3b . The borrowing of ad- 
ministrative terms such as a§xwh or sakap is understandable and 
explicable as the result of the political influence of the conquerors. 
But we should expect tribal terminology to have a purely Armenian 
character, since it developed on the basis of native tribal relationships. 
Here too, however, Iranian words are found, and with nuances which 
make it impossible to consider them simple borrowings. These phe 
nomena also occur in lexical material of different origin which links 
the Armenians to their neighbours on other sides, Consequently, a 
familiarity with the ethnic composition of the Armenian people is 
indispensable for a classification and clarification of certain terms 
which characterize and are important for our problem, 

At present it is still impossible to trace the ethnic strata all the 
way back to the first settlers of Armenia, but if we limit ourselves to 



306 



CHAPTER XIV 



the portion of history accessible to us, we find that the Armenian 
nation was composed of different ethnic as well as tribal components. 
Under the Achaemenids the territory subsequently called Armenia 
was inhabited by various peoples and made up two satrapies. The 
Paktyians and the Armenians, together with their neighbours as far as 
the Black Sea, formed the Thirteenth satrapy, while the Matienians, 
Saspirians and Alarodians formed the Eighteenth 4 . On the western 
border of Armenia lived the Cilicians, and on the eastern one the 
Matienians, Moreover, the Armenians were separated from the former 
by the Euphrates, and from the latter by one of the tributaries of the 
Tigris, probably the Zab 5 . Since they formed one satrapy with 
the Matienians, the Alarodians and Saspirians evidently adjoined 
them on the north-western side. In the expedition of Xerxes, the 
Armenians fought under one standard with the Phrygians, while the 
Alarodians and Saspirians were under another, Among the adjacent 
peoples, the better known — the Moschians and Tibarenians, the 
Makronians and Mossynoechians, the Mares and Colchidians, the Utians 
and MyMans — were coupled under separate leaders, while the Medes 
were under a special command 6 , At the time of the retreat of the 
Ten Thousand, the land of the Phasians and the Hesperites, i.e. 
Saspirians, as well as of the Taochians was already considered to be 
part of Armenia and was called Western to distingush it from the 
other Armenia. The ruler or satrap of one part was Tiribazos, and 
of the other Orontes 7 . In this period, the neighbouring peoples, 
namely the Karduchians, Chalybians, Chaldaeans, Makronians, Col- 
chidians, Mossynoechians, Koetians, and Tibarenians were independent 8 . 
One century later, at the time of the battle of Gaugamela, there were 
again two satraps in Armenia, Orontes and Mithraustes; as for the 
neighbouring Medes, they were under the leadership of Atropates, 
and with them were found the Kadusians, Albanians, and Saka- 
senians 9 . 

All of these nations, in whole or in part, were settled on territorry 
later occupied by Armenia. Though numbering more than twenty, 
they can be reduced to a few groups through blood ties. Of the 
Armenians it is said that, « ,,, The Armenians, who are settlers from 
Phrygia, were armed like the Phrygians » 9a , hence similarity of weapons 
may be taken as an index of relationship. Since the Utians, Mykians, 
and Parikanians were all armed like the Paktyians, the Tibarenians, 
Makronians, and Mossynoechians like the Moschians, the Alarodians and 



THE TEIBAL BASES OE THE NAXARAR SYSTEM 307 

Saspirians like the Colchidians, and the Medes had their own arms 9l5 , 
we can deduce that, except for the Medes and the Armenians, all the 
other nations were grouped around the Paktyians, Moschians and 
Colchidians, Some of these peoples have vanished, bequeathing their 
names to their lands. Thus, for instance, the Alarodians, Saspirians, 
Taochians, and Phasians have disappeared as peoples, but their mem- 
ory has been preserved in the toponyms, Ayrarat, Sper, TayF [Tao], 
and Basean. The name of the Paktyians is preserved in modern 
Bohtan, along the eastern Tigris ; there is also a Kurdish tribe named 
Bohtii, The Mykians, the Armenian?//^ [Mokk'], who were related 
to the Paktyians, moved in part to the north and settled between the 
Kura and the Araxes, Prom them are derived ITnLfuwh - g [Mu^ank 5 ], a 
district in Area^, and Mughan, the name of the famous steppe. The 
Col-chidians, Mos-chians, and other peoples were also pushed northward. 
These disturbances attended the formation of the Armenian nation- 
ality 9e , 

One of the decisive moments in this formation has been recorded 
by history; it came in the period of Zariadris and Artaxias, and 
manifested itself in the formation of a single language accessible to 
all 10 . According to reliable testimony, it was under these leaders 
that Armenia, formerly a small country, grew at the expense of the 
neighbouring lands, She conquered Kaspiane, Saunitis [Phaunitis], 
and Basoropeda from the Medes, the foothills of Mount Paryadres 
(TayF and Sper), Chorzane and Gogarene from the Iberians, Karerdtis 
and Derxene [Xerxene] from the Chalybians and Mossynoechians, 
AMlisene from the Kataonians, and Tamoritis [Taronitis] from the 
Syrians 11 , The process of unification sprang from two centers, the 
realms of Artaxias and Zariadris, which had originally made up two 
of the earlier satrapies ; the movement then spread and took in the 
territories of the neighbouring nations in the directions of Media, 
Iberia, Cappadocia, and Syria. 

The Kataonians were part of the Cappadocian population ; according 
to Strabo, ancient writers had considered them a people distinct from 
the Cappadocians, but in his own time, they had lost their tribal 
characteristics and were similar in language and customs to the 
rest of the population of Cappadocia 12 . As for the inhabitants of 
Cappadocia, despite their dissimilarity to it, they had originally 
belonged to the Syrian world. The Greeks called them Syrians or 
White Syrians [Leucosyrians] to distinguish them from the other ones 13 . 



308 



CHAPTER XIV 



It seems then, that Cappadocia was ethnically related to the Seirdtic 
family, and, it has now been reasonably well demonstrated that the 
Iberians too were related to the Semites 14 , Thus, an ethnically 
related territory once stretched from the plain of Mesopotamia to 
the Black Sea; this agrees with the testimony of Stxabo that the 
name of the Syrians in antiquity reached from Babylonia as far as 
the Euxine Sea 15 . Following Strabo, let us call the Syrian or Semitic 
world Aramaean, On the basis of ancient sources, the great geographer 
also included the Armenians in this world, thus contradicting Herodo- 
tus, according to whom the Armenians were natives of Phrygia, 
A controversy had arisen among the ancients on the subject of the 
nation of the epe/x/Jot mentioned by Homer, and Strabo supported 
the opinion of Poseidonius, saying, 

But it would seem that the view of Poseidonius is best, for 
he derives an etymology of the words from the kinship of 
the peoples and their common characteristics, Por the nation 
of the Armenians and that of the Syrians and Arabians betray 
a close affinity, not only in their language, but in their mode 
of life and in their bodily build, and particularly wherever 
they live as close neighbours, Mesopotamia, which is inhabited 
by these three nations, gives pToof of this, for in the case of 
these nations the similarity is particularly noticeable. And 
if, comparing the differences of latitude, there does exist a 
greater difference between the northern and southern people 
of Mesopotamia than between these two peoples and the Syrians 
in the centre, still the common characteristics prevail. And 
too, the Assyrians, the Arians, and the Arammaeans display 
a certain likeness both to those just mentioned and to each 
other, Indeed, Poseidonius conjectures that the names of 
these nations also are akin; for, says he, the people whom we 
call Syrians are by the Syrians themselves called Armenians 
[Arimaeans] and Arammaeans; and there is a ressemblance 
between this name and those of the Armenians, the Arabians 
and the Erembians, ,„ 16 , 



Returning to the same problem at the end of his work, Strabo repeats 
once more the opinion that the three peoples living next to each 
other show inter-relationship and are called by similar names : some 
Armenians, others Arammaeans, and the third Arabians, as though 
one nation had split into three, following climatic conditions 17 , 

It is evident from this valuable indication that the ancient world 
saw elements common to the Armenians, the Syrians, and the Arabians. 



THE TKIBAL EASES OE THE NAXABAJR SYSTEM 



309 



These were reflected not only in their appearenee and mode of life, 
but also in their language, and these similarities were particularly 
striking in those places where they lived together. The etymological 
connexions of Poseidonius are based on actual facts, on the unity 
and similarity of the peoples compared, and not vice versa; i.e. the 
author is not distorting reality for the sake of etymology. The cate- 
goric statement that Poseidonius « derives an etymology of the words 
from the kinship of the peoples and their common characteristics », 
is very important 18 . 

Side by side with the testimony on the descent of the Armenians 
from the Phrygians, we have a similar testimony on the kinship of 
the Armenians with the Aramaeans, The truth of either point of 
view can be demonstrated only through an analysis of the admirable 
but complicated structure, which is the Armenian language. If by 
Phrygian we are to understand Indo-European, then the words of 
Herodotus are justified. Studies of Armenian from the point of view 
of Indo-European phonetics have shown ties of kinship between the 
Armenians and the great Indo-European family and have placed 
Armenian in the western group of Indo-European languages 18a -, 
There is no doubt that research along historical lines will uncover 
the contact of the Armenians with the Aramean family as well. The 
success of such a task may be taken as certain of the basis of the 
lexical material already isolated. 

To show the relationship of Aryan and Semitic elements in the 
Armenian language, we must admit that Armenia was originally part 
of an area settled by various tribal groups : ifarcfc-chians, Mos-chians, 
Sa-spr-ians, with primarily Aramean components. The well known 
invasion of the Cimmerians in the seventh century B,C, caused a 
shift in the population of Asia Minor, The Phrygian Armenians, 
driven out of their territory, crossed the Euphrates and drove a wedge 
into the local Aramean population. Some of the latter remained in 
the north, and some in the south, while the conquered middle group 
mingled with the conquerors 19 , The core of the Armenian nation 
was formed from this fusion of invading Phrygians and of natives 
who were in the main of Aramean descent, The best proof of the 
double composition of this core is the double name of the Armenians : 
Hay to themselves, Armenios to their neighbours, one derived from 
the invaders, the other from the aborigenes, The process of formation 
of the Armenian nation is presented with remarkable accuracy in the 






310 



CHAPTEB, XIV 



ancient Armenian legends preserved by Movses Xorenaei and the 
Anonymous History. Two eponymous figures, Hay-k (Zujj—1[) and 
Araman~eak {JXpwSwh—hwli) or Armen-ak (UpM—wlf) personify the two 
components Indo-European and Aramaean. The name itself, armen 
(wpiilh), Armni-k ? {JXptfu— jig) is derived from Aram, according to 
.Xbrenaei, and at this point the Armenian account supports Posei- 
donius, who compared appivioi and apap,€loi„ but the attempts to 
explain the origin of hay still remain unsuccessful 20 . 

Other peoples joined the original nucleus, new ethnic currents 
continually flowed into it from the frontiers, from the lands of Atro- 
patene, Iberia, Cappadocia and Syria. The assimilation of peoples 
continued until the time of Zariadris and Artaxias, when an important 
period for the Armenians came to an end, Xteutha-&YfiJQf ..th§.jDxeatipn 
of the empire of Tigran the Great, tribal differentiations had been 
obliterated to such a degree that the entire population spoke a single 
language. The moment of synthesis of a common national language 
should be taken as the end point for the tribal stage in the life of a 
people 20a . 

The tribal pattern is the necessary and natural framework of life in 
a country so rich in tribes and peoples. A man's position in a multi- 
tribal environment had to be established primarily through blood 
ties to some clan. The reference of Herodotus to the Armenians as a 
pastoral people is an eye-witness testimony to the stage of development 
reached by the Armenians at that time 21 , Our knowledge of the 
life of the neighbouring people who formed a single cultural world 
with the Armenians gives us some insight into this phase. Thus, 
common tribal property ruled over by the oldest member of the tribe 
among the Iberians and the Arabs, precedence of brothers over sons 
and community of wives among the Arabs, characteristic administrative 
institutions among the Assyrians, absence of individual inheritance 
linked to the burial of their possessions with the dead among the 
Albanians, the maintenance and general position of the king among 
some of the peoples belonging to the same world — all these are 
characteristic phenomena of the tribal stage of society 22 , 

Certain traits observable in the hereditary nobility of Armenia at 
a later date, such as the undivided tenure of lands by an entire clan 
under the leadership of its oldest member point toward the roots of 
such customs. We do not know the precise forms taken by tribal 
life in Armenia, In such a diversified tribal composition of the popu- 



THE TRIBAL BASES OE THE N AX ARAB SYSTEM 



311 



lation, the forms of life could be of the utmost variety, depending on 
the specific stage of development reached by a particular ethnic group. 
Nevertheless, a general classification according to its major ethnic 
components is possible. 

Among linguistic remains, particular attention should be given to 
certain terms of naxamr terminology, which were an inheritance from 
a tribal way of life. Correctly interpreted, these can give us an 
insight into the grouping of the main ethnic strata in the tribal period, 
and into the manner of their integration of national forces at the 
time of the dissolution of tribal forms. These terms are : maixciz 
[jjwyfuwi[\, aspet [uuii^m], *mamak [SmSmli\, and ter [wtp] in their original 
sense, which we will attempt to uncover. All four terms originally 
designated the holder of power in various parts of the contry. 

Mafyaz [tywtiluwqj, the hereditary title of the princes Xor^ofuni, 
meant in general, « ruler, prince, king », from the Assyrian malxazu = 
Arabian malih. The unit of rule corresponding to this title on a 
smaller scale was alx [""z/" <{ olausura »], from *haxl, which has the 
same origin as the Georgian sa^U, « house ». From the same root also 
comes alaxin (a/jp/u/ji)), which originally meant a « person belonging 
to an «/^/"»> and subsequently a « female attendant » 22 a . 

The second term, aspet (wuuikm), was also a hereditary title and 
belonged to the Bagratid princes 23 , The word has been interpreted 
in various ways which we consider unjustifiable. It should to be 
divided into as and pet, of which the second part means « leader, 
head », so that all the difficulty lies in the first syllable alone. There 
are grounds for comparing this syllable with the initial syllable of 
another equally important term in naxarar life, namely se-puh (uh— 
uinU), of which the etymology has already been traced 24 . Sepuli, 
from the earlier form se-puhr, is the Iranian vida-pudra, where pudra 
(—ujiiU) means « son », while vid, as we have seen, is one of the steps 
in the tribal organization; consequently, se is a contracted form of 
m# 24a , 

The term ukuintC has also been preserved in Armenian literature 
in the form Aspurak (tXuuim.pwlj) 9 the name of a famous bishop in Eaus- 
tus' History 241 \ The presence of a pre-tonic hl [u] proves that this 
m. stood before two consonants, i.e. the word was pronounced Aspuhr- 
ah. SpuJiTy the element left after the removal of the affixes, should 
be linked with sepuh in its archaic form sepuhr. The identity of the 
initial syllables as- and se- is beyond doubt. It is interesting that 



312 



CHAPTER XIV 



JXuuinipmli [Asjpuralc] can also be found with an initial v, ^urni^/u^a/^ 
[Vasjpuralc], but that the name of the province of ^iHi/^/upu/p£ (Vas- 
purakan), which, is the same word, is given in non- Armenian documents 
without the initial phoneme, *A cnrovp < a/c > dv 35 , The existence of the 
form vasjpur next to asjpur shows even more clearly the derivation 
from viBajpuBra* Indeed, merely on a phonetic basis of the absence 
of an initial v in Armenian, the Iranian viBajpuBra or vaiBajpuBra bad 
to pass into Armenian in the form isapuhr, asapuhr, or, with the 
dropping of the pre-tonic a, ispuhr, aspuhr 26 , 

With the discovery of the identity of as-puhr and se-puhr, or of 
as- and se- 3 the etymology of the term aspet becomes clear. Aspet is 
to be derived from viBa-pati in the same way as asjpuhr was derived 
from viBa-puBra, In other words, aspet means « head of a viB, or 
clan », and sepuh, « son of a clan », The word i[ukii [vsern], iluwS [vsam], 
in its long form i[uhwS \yseam\ is undoubtedly the Iranian viBya-ma 
and is related to viB as zantu-ma is to zantu. Its original meaning 
was, « member of a clan, clansman » ; later it came to mean « noble, 
outstanding [excelms] 27 », 

*Mamak {fiwiiwli) is similar in character to the preceding term. It is 
found in historical literature as a proper name in the Mamikonean 
family, but it had originally served as a hereditary title for one of 
the princely houses, and subsequently become a family name, as 
had the two titles already analyzed, ^SwSmli [mamalc] or *&*^ 
[mamik] gives 1Timf[ilj[nh~kwh [Mamikon-ean], as Uu^£w— ju£/t [Aspet-urd] 
derives from uiuu^km [asjpet], 0Tlfinj]puj^-nLh[i [Mal^az-uni] from Smqjuwti 
[malx^z']. We assume that mama-k is nothing more than the 
Armenized form of the Iberian mama = « father », Consequently 
this term is also derived from the tribal period and meant, « father 
of the clan, clan-leader », thus being synonymous with the Georgian 
mama-sa-glisi a8 . 

It is interesting to note that the Armenian tradition likewise traces 
the Mamikon-ean back to a certain Mamik, Our hypothetical ety- 
mology is therefore supported by the tradition and throws new light 
on the traditional origin of this great princely house. The Armenian ~ 
tales derive the Mamikonean from the Cenk' [z?£%], In Faustus 4 
History, the Mamikonean themselves proudly proclaim the descent 
of their family, « ,,. from the kings of the nation of the Cenk' 29 ». 
Faustus does not indicate what is meant by the country of the Oenk', 
but the Anonymous History implies that the name refers to one of 



THE TRIBAL BASES OE THE NAXASAS SYSTEM 



313 



i 



the countries east of Persia, beyond Balkh, i.e. China, According to 
the legend, the hypothetical original chieftains of the Mamikoneans, 
Mamik and Konak, were the sons of a famous nobleman, who ranked 
first after the king in e Cenastan - China, After the death of this 
nobleman, the king married his widow and had by her a son named 
Cenbakur, who was the heir to the throne. His half-brothers on the 
mother's side, Mamik and Konak, plotted to keep him from the throne, 
but the plot failed and they saved themselves by fleeing to the Arsacid 
ruler of Balkh. Cenbakur demanded the return of the fugitives, but 
the Arsacid king refused, calming him with the assurance that, « he 
would send them far to the west, to the end of the earth where sets 
sun»; and indeed, to save them from the pursuits of Cenbakur, the 
king sent them to his Arsacid kinsman in Armenia, The purpose of 
the last part of the story is to explain the manner in which the Mami- 
konean reached their hereditary district of Tayk' from China : the 
king had intentionally settled them on the edge of the world. The 
legend is mistaken in its identification of the nation of the Cenk\ 
In the original tradition, Cenk' did not mean the distant Chinese 
but the neighbouring Tzans, who lived not far from the Mamikonean 
hereditary district of Tayk\ The Tzans are familiar to the Armenians 
whether as &«&— rfwb— ftlj [can-, ean-ik] in Movses JXbrenaei, or nowadays 
as Canik, The family traits of the Mamikonean, hot temper and 
remarkable bravery, link them rather to the Tzans, who were still 
famous in the sixth century for their warlike character, than to the 
peace-loving Chinese, In such an interpretation, the legend con- 
cerning the Mamikonean acquires a historical foundation 29a . 

Let us note in passing that the Orbeliani princes in Georgia also 
sought their ancestors in the country of the CenF, and that Cenbakur 
likewise figured among them. The coincidence is not fortuitous ; it is 
obviously a survival of the Mamikonean family tradition, and can 
be explained by the fact that the Orbeliani considered themselves 
descendents of the Mamikonean, There is nothing surprising in this ; 
just as the Bagratids moved from Sper to Georgia, so the Mamikonean 
may have moved to Georgia from the neighbouring Tayk\ The 
hostile relations of the Orbeliani to the Georgian Bagratids seems an . 
echo of the family enmity of the Mamikonean. and the Armenian 
Bagratids 29to , 

Ter {wip), the last term under consideration, is a contraction of 
mjt [ti-] and mjp [m/r]. The second half of the compound is a familiar 



314 



CHAPTER XIV 



Armenian word meaning « person, man ». St*—l{fi [ti-hin], « lady » is 
composed on the same pattern as uifi—injp [ti-ayr]. The sense of ti- 
is nearly beyond our reach; even its relations to wjp [ayr] and Itfib 
[kin] is not clear; does it modify them or is it modified by them. 
The construction of these words recalls the Georgian mama [« father »]- 
up'ali and deda [« mother »]-wp'aH, of which the former is familiar in 
its contracted form mep'e, from ma(m) / p > a{^i). Like the Armenian ter, 
the Georgian up'ali corresponds semantically and etymologic ally to 
the Semitic baal, « lord », the name given to the gods protecting the 
tribes among the Semites. This word, which had the sense of « god, 
lord of heaven », among the Semites, was transferred by the Georgians 
to the king, the lord of the earth. Something similar apparently 
took place in connexion with the Armenian mfi [ti]. In phonetic and 
semantic content, ti may be derived from the Indo-European root di-, 
which is familiar in many languages and in various forms. The words 
derived from it can be reduced to two basic forms, di and div- ; both 
of which include the concept « day - light », and « heaven - god ». 
Both roots exist in Armenian in the forms w/fi/ (tiv), « day », and m^— g 
[ti-k ']plur. tant, « age », literally, « days,». The form ti, which concerns 
us here, goes back to the root di. Originally m[i [ti], like the Latin 
deus, designated the concept of a superior being or « god » ; 30 subse- 
quently it began to be used as a designation for the «king», and in 
general for a « lord on earth, i.e. it underwent the same fate as the 
Georgian up'ali, In order to set off this last meaning of mji [ti], the 
element — wjp [ayr], « man » was added to it thus producing the 
compound wji—wjp[ti-ayr] = w£p[ter]. This mental process was interes- 
tingly to be repeated once again. After the form wfj—wjp [ti-ayr] had 
had completely replaced mfj [ti], and had come to designate a lord 
both o$ earth and in heaven, the word wjp [ayr] was suffixed a second 
time to differentiate these two concepts so that the word mjjp—wjp 
[tir-ayr] was formed as a result. In literature, this last form occurs as 
a proper name, but it was unquestionably used originally in the 
general sense of « lord » or « master 31 ». 

The terms drawn from naxarar society, which we have just discussed, 
are an inheritance from an ancient period of tribal organization, 
according to our interpretation. They belong to the class of local 
titles born by the IcomarcJis. These homarchs needed different titles 
because of the variety of their ethnic environment. The four terms 
discussed should be taken as the most common ones among these 



THE TRIBAL BASES OE THE NAXABAB SYSTEM 315 

titles', and they correspond to the four major groups to which the 
ethnic pattern of Armenia could he reduced grosso modo. In regions 
of dominant Aramaean influence, the heads of clans were called malyaz* 
On the border of Atropatene or Iran, the Iranian terminology, aspet, 
sepuh, vsem was current. In the Iberian area, mamah was the 
most current term, Finally, in localities having a Phrygian population, 
ter, tun, and so forth were in use. Let us stress again that this analysis 
is not exhaustive; not only are the areas selected by us far from 
homogeneous, and contain certain variations necessarily reflected in 
their terminology, but other ethnic groups also existed. Such, for 
instance, were small nations like the Mards, whose head was known 
as the mard-jpet {SwpTj.uikw), and a number of others. 

The period of tribal relations ended in the political ferment which 
occured in Armenia in the time of Zariadris and Artaxias. The dis- 
solution of tribal organization and the appearance of new social re- 
lationships began in this period. The nature of the change manifested 
itself in the stratification of the social mass and the emergence of 
classes. The heads or leaders of clans were gradually transformed 
into noblemen or lords, thus laying the foundations for future heredi- 
tary principalities. Such a transformation begins at the moment when 
the title of the heads of clan becomes hereditary. His descendents are 
singled out thereby and occupy a special position among the other 
members of society, The right of inheritance carries with it the right 
of ownership over all that had formerly come under the authority of 
the clan leader. The social organism once homogeneous and equali- 
tarian is split in this fashion into two socially unequal halves. On one 
side are the few, the rulers, descended from the elan chieftains, on the 
other the remaining majority of the population, the subjects. Once 
he has become a lord, the head of a clan finds it easy to extend his 
power over other clans, and, having gradually extended his sphere of 
influence, to achieve the rank of king. 

This process is admirably illustrated by the pattern of social strati- 
fication found in Iberia. Here, according to Strabo, four clans or 
castes of men were distinguished. The kings were chosen from the 
first group, the priests from the second. The third was made up of 
those concerned with agriculture and war, and the fourth of the 
common people who were the slaves of the king, and supplied him 
with the means of existence. Finally, « their possessions are held in 
common by them according to families, although the eldest is ruler 



816 



CHAPTEK XIV 



and stewart of each estate 31a », The royal and the priestly castes, 
that is to say the artificial ones, were descended from the man who, 
as head of his clan, had consolidated behind himself the hereditary 
right to this calling, and such a man came originally from the same 
clan as those who later fell into slavery, In other words, the king and 
his slaves had once formed a single clan unit of which he was the head, 
but with the concession of hereditary power to its head, his family 
began to be singled out socially from the rest of the clan ; the higher 
it rose, the lower the rest descended, Finally the riding family reached 
royal estate at the price of the enslavement of its own clan. Later it 
subordinated to itself the regaining population of Iberia, which was 
divided into clans partly concerned with agriculture, and partly with 
military affairs 33 , 

We might think that the priestly clan had the same origin as that 
of the king, i.e. that it arose on the basis of the destruction of its 
own clan ; but in such a case the priests too should have had slaves, 
which is not the case, Strabo asserts that one of the members of the 
royal clan, namely the oldest, inherited the throne, while the second 
after him was entrusted with the administration of justice and with 
the command of the army 33a . This comment suggests the manner in 
which the priestly caste developed. The king, having evolved out of 
his earlier function of clan leader, first united all functions in his own 
hands; he was simultaneously ruler, war leader, judge, priest and 
general overseer. The time then came when a single individual found 
it difficult to perform all these duties, and one of the men close to 
him, the heir to the throne, began to assist him 33 . Subsequently 
one of the branches of government, religious duties and international 
politics, were entrusted to him. The union of two such apparently 
dissimilar functions can be explained by the fact that a religious 
concept of the world joined to the great authority of the priesthood 
and the existence of common cults provided the common ground on 
which people of different cultures could meet. As a result of existing 
conditions, priestly duties became hereditary in the family of the 
second man after the king, then, another branch of the government, 
namely justice and the command of the army, was also entrusted to 
him, This is precisely the pattern found in Iberia 34 , These additional 
functions, like the priestly ones, were undoubtedly destined to become 
the hereditary prerogative of a certain family and, in fact, this very 
development took place under the Arsacids, Thus, the four groups 



THE TRIBAL BASES OE THE NAXARAB SYSTEM 



818 T 



of men into which tlie population of Iberia was divided were not 
only socially but also genetically different units. The first and last, 
i.e. the royal and slave elans, were derived from a single group, while 
the sacerdotal clan had developed out of the royal one. These two 
clans had furthermore evolved from family units and were castes 
rather than clans 35 . 

The social process just described also took place in nearby Albania, 
but since this country was composed of twenty-six tribes speaking 
different languages, the process occurred at twenty-six points 36 . At 
first, each of these had had a king, but subsequently all came to be 
united under the authority of a single ruler. The royal power de- 
veloped among these tribes in the same fashion as among the Iberians ; 
the Albanian king was a former clan head who had reached the rank 
of ruler or prince. The need for a common defense against the raids 
of mountain peoples forced the tribes to unite and form a single 
political body. Among the Albanians, as among the Iberians, the 
sacerdotal office was filled by the most important figure after the 
king, and he ruled over vast and densly populated sacerdotal estates 
and temple slaves 37 , The administration of justice and the command 
of the army were undoubtedly similar to those in Iberia, i.e. they too 
were entrsuted to the man who stood next after the king 38 . In Iberia, 
however, the supreme power was supported by an alliance of clans or 
of their leaders, while in Albania, it depended on the federation of 
minor kings or princes. In other words, some twenty-six small 
kingdoms had first been formed, as a result of the ethnic pattern, and 
only then were they united together into one. 

Armenia was no less diversified than Albania, from an ethnic point 
of view. Hence, we should expect on a basis of analogy that the 
evolution of society here should have taken place in the same manner 
as in Albania, i.e. that unification inside the country should manifest 
itself at first along the lines of ethnic differentiation, Unification 
should be expected initially in the major ethnic groups : around the 
nialxaz, the aspet, the mamaJc, the ier> etc. An investigation of the 
scant evidence provided by the historical events supports this hypo- 
thesis. 

The dissolution of tribal relationships, accompanied by the rise of 
classes ruled by kings who were to be the future hereditary princes, 
took place in the stormy period of Tigran the Great, The unrest of 
a period of wars creates an atmosphere favourable for the trans- 



318 



CHAPTER XIV 



formation of tribal relationships into those of classes 38a . The political 
might of Armenia under Tigran the Great, his vast conquests from 
Mazaka of Cappadocia to Ekbatana of Media, and from Iberia and 
Albania to Arabia and Palestine, show first of all that the unification 
of the country had already been achieved. We lack direct historical 
evidence on the organization of social groups and on the territorial 
and social forms which contained them, though historians are generous 
in their colorful descriptions of external events, Nevertheless, we 
fortunately can find some specific indications of great importance for 
our purpose in the midst of generalities about the military strength 
of Armenia under Tigran, 

Armenian historians depict the king surrounded by numerous 
vassal - kings. In addition to Armenians and Qordyenians, the army 
of Tigran included contingents of Medes and Adiabenians under the 
leadership of their own kings, of Arabian tribes from the shores of 
the Persian Sea, of numerous Albanians from the shores of the Caspian, 
and of their Iberian neighbours, as well as of many nomads from the 
banks of the Araxes, Plutarch comments, 



Many were the kings who waited upon him [Tigranes], and 
four, whom he always had about him like attendants or body- 
guards, would run on foot by their master's side when he rode 
out, clad in short blouses, and when he sat transacting business, 
would stand by with their arms crossed 39 , 



The Armenian princes are given here the pompous name of kings, and 
the four kings in perpetual attendance upon Tigran are none other 
than the four bde§x$, the four marcher lords who were considered 
first in rank or cushion at the Arsacid court 40 , Sycophantic pane- 
gyrists of Lucullus, drawing their information from his own boastful 
accounts, stopped at no exaggeration in the glorification of their hero, 
who had presumably overthrown so great a king. Plutarch's raising 
of the Armenian princes to royal rank in this passage stems from a 
similar intention. 

In addition to the bdesxs, some of the figures among the Armenian 
princes are mentioned by name, One of these is a certain Bagadates, 
When Tigran became master of Syria, Cilicia, and all the lands as 
far as Egypt, after his defeat of Antiochus, he appointed this Bagadates 
stmtegos of the newly conquered territories, Bagadates ruled this 
area for fourteen years, until the arrival of Lucullus in 70 B.C., when 



THE TRIBAL BASES OE THE NAXABAR SYSTEM 319 

he was forced to hasten to the assistance of Tigran, and Syria reverted 
once again to the Seleueids 41 , We likewise have mentions of Mithro- 
barzanes and Mankaios, The former, with two thousand horsemen, 
opposed Lucullus as he crossed the Euphrates and invaded Sophene, 
while Mankaios was entrusted with the defense of Tigranokerta. 
Mithrobarzanes fought Lucullus and was put to flight, while one of 
Lucullus' lieutenants, Sextilius, surrounded Tigranokerta where he 
besieged Mankaios 42 , According to another highly embelished 
account, Mithrobarzanes was a member of Tigran's entourage, and 
the one who first told him the truth when no one dared to inform 
the king of Lucullus' advance. As a punishment for his boldness, 
Tigran ordered Mithrobarzanes to move against the enemy and 
bring their general back alive while exterminating the rest of his 
forces. But Lucullus sent an army under the command of Sextilius 
against Mithrobarzanes, who died a hero's death in the encounter 43 . 
Finally, we have references to a certain Guras, who is identified 
as Tigran's own brother. He defended the city of Nisibis against 
Lucullus when the latter crossed the Taurus and besieged this im- 
portant strategic point after his repulse from Artaxata 44 . 

The name BayaSdrrjs is the original form of the Armenian pwqw — pmm 
(Bagarat), from Bagadata, « god-given » derived from baga and data — 
Greek ©eoSoros, before the Armenian form was affected by the law 
of phonetic mutation, Tigran's general bearing this name was unques- 
tionably the ancestor of the famous dynasty of the Bag<a>ratuni 
or Bag<a>ratids, and he provides the reconciliation and justification 
for the two divergent Armenian accounts of the Bagratids' origin. 
According to the Anonymous History, the Bagratids were descended 
from h-Ayk, the ancestor of the Armenians, but another tradition, 
preserved by -Xbrenaei, speaks of their Jewish blood. The very name 
of the dynasty, and especially its hereditary title of aspet, which we 
have discussed earlier, point to an Armeno-Iranian origin. The original 
home of the family apparently lay on the border of Atropatene, as 
we have already noted, and we have already postulated an etymolo- 
gical connexion between Bagarat and Bagrewand 45 . Unless it is an 
invention of Zbrenaei, as has been assumed by his critics, the second 
account, which traces the descent of the Bagratids from the south and 
from Jewish stock, must be a distorted memory of the fact that 
Bagarat had been sent south under Tigran, and had been appointed 
governor of Syria, an area with a Semitic population 46 . Indeed, the 



320 



CHAPTER XIV 



Anonymous History is also familiar with. Bagratids on the southern 
border of Armenia, more exactly in the province of Angeltun, which 
it considers to be their hereditary domain 47 . Whatever the case, 
this tradition is connected with the sojourn of Bagarat in the south, 
and it is even possible that after Tigran's loss of Syria the defense of 
Armenia's southern frontier passed to Bagarat, and that the princely 
house of Angeltun was a branch of the Bagratid family 48 . 

A more accurate reading of the name of the general who first met 
the Eoman army is MiBpo^ov^dvrjs instead of Mi9poj3ap£,dvr}$ = Ar- 
menian JThCpntdwh (Mehruzan) 49 , and this personage is none other 
than the prince of Sophene, The first Armenian district threatened by 
the Romans after their crossing of the Euphrates was Sophene invaded 
by the legions of Lueullus, Mithrobuaanes, as ruler of the district, 
set out to defend his country, but was compelled to retreat before the 
onslaught of the Romans because of the insignificance of his forces. 
Everything related by Plutarch about the advance of Mrthrobuzanes 
is nonsense intended for effect. Mithrobuzanes was forced to advance 
against Lueullus because of his position as ruler of the territory, and 
not because Tigran wished to punish him, as we are told by PlutaTch 50 . 
Mithrobuzanes undoubtedly belonged to the family of Zariadris, the 
king of Sophene contemporary with Artaxias. The name Mithro- 
buzanes was a common one in this house : Zariadris' son, who fled 
from the persecutions of Artaxias to king Ariarathes IV of Cappadoeia, 
and later returned with his help to his father's throne, likewise bore 
this name 51 . The king of Sophene under Tigran II was Artanes, who 
perished at the hands of Tigran; 51a Mithrobuzanes must have been a 
personage close to Artanes, possibly a son or a brother. The descen- 
dents of Zariadris and Artanes preserved their sovereign rights in 
Sophene, although they had lost their royal rank. 

The third name with which we are concerned, ikfay/ccuos, should 
read M<xp,KoZos or Ma(p,a)KaiQs, to correspond to the Armenian Ifwitfjli 
[Mamik], A verification of the manuscripts on this Teading is of the 
utmost importance since a demonstration of this hypothesis will 
necessitate the shift of the appearance of the famous Mamikonean 
house to the period of Tigran the Great 511 \ 

All the figures we have just discussed and whom Tigran raised to 
outstanding positions because of the imminent war with the Romans, 
namely Bagadates, Mithrobuzanes, and Mankaios, not to mention 
Guras, were numbered among the kings who crowded at Tigran's 



THE TKIBAL BASES OE THE NAXABAM SYSTEM 



321 



court, Thus crar fundamental premise that principalities based on a 
class structure had already appeared in tlie time of Tigran is proved 
in essence, whatever the soundness and acceptability of our further 
conclusions about the individuals themselves. Furthermore, there is 
no serious basis for disputing the conclusion that these officials were 
the ancestors of such subsequently distinguished princely families as 
the Bagratids, the Mamikonean, and the princes of Sophene, The 
princes Arcruni must be the descendents of this last principality, as- 
evidenced by the names current in their family 52 . Thus, three of the- 
most important princely houses were already in existence under 
Tigran the Great, and.were probably making their first appearance at 
that time. The house of the Mal^az, which judging by its name was 
even older than these families, must also have unquestionably existed 
in this period 52a 

Despite the social stratification and the aristocratization of the 
country, its ethnic diversity continued to exist. However much the 
different groups may have been assimilated, their components in the 
last analysis still preserved their peculiar colouration. The power of 
the malx&z among the Xor^ofuni, of the aspei among the Bagratuni, 
of the mafnak among the Mamikonean, and finally the supreme power 
of the ter indicate the preponderance of the corresponding ethnic 
groups during the period of general unrest. The surrounding areas 
which were in direct contact with neighbouring peoples were less 
liable to assimilation. Because their mixed population, continually 
replenished by the influx from nearby peoples, they were to some 
degree isolated from the central districts. The result of this partial 
isolation was the formation of four small political units on the borders 
of Armenia ; these were the domains of the bdes-gs : G-ugark', Aljnik', 
Nosirakan, and a fourth whose name is unknown, The genesis of 
these units was presented altogether incorrectly in the Armenian 
legends. It is true that the bdefys guarded the frontiers of Armenia, 
but they performed this duty not at the order of the Armenian king, 
but as a result of their position as rulers of border districts who simul- 
taneously acknowledged the suzerainty of the Armenian king 521 \ 

The ethnic diversity of Armenia was not limited to the elements or 
groups already discussed. Other tribal units were scattered throughout 
the country, and principalities subsequently developed out of them in 
the manner already described. Some of these groups could trace 
their origin back to very early periods, while others were relatively 



««a 



322 



CHAPTER XIV 



latecomers. These newcomers were drawn in the main from nomadic 
peoples and often brought back into the country social forms which 
it had long since outgrown. Consequently, more or less important 
groups still leading a tribal life under their tribal chieftains could be 
found side by side with more evolved principalities. Ethnic diversity 
naturally brought about variations in social conditions. Hence our 
periodization of Armenian history, like all such divisions, is correct 
and acceptable only for the general aspects of its evolution ; anachro- 
nistic elements have always found a way to survive outside the main- 
stream. 

Some of the newcomers, who were to be the ancestors of great 
princely houses, had already arrived on the scene under Tigran II. 
According to Plutarch, Tigran sent a large contingent of cavalry 
against Lucullus at the time of his assault on Artaxata. Mardian 
mounted archers and Iberian spearmen, who were the best hope of 
the king, stood in the forefront of this army 53 . The Mardians were 
the descendents of Iranian tribes scattered along the Zagros, who 
lived by sheep herding and brigandage 54 . When Xenophon crossed 
the Kentrites or Eastern Tigris, he was attacked by Mardians as well 
as Armenians. The former were undoubtedly Armenian Mardians who 
had moved to Armenia from Iranian territory 55 . In 68 A.D., the 
Roman general Corbulo, a successor of Lucullus, had to cross the land 
of the Mardians during his retreat from Artaxata to Tar5n, At that 
time the Mardians were still living by brigandage and attacked the 
Romans, but hid in the mountains when they met with resistance 56 . 
It is evident from this account that the Mardians were already occu- 
pying then the area which the Armenians later called Mardastan or 
Mardoc-k'. i.e. the land of the Mardians 57 . The princely house whose 
representative bore the title of Swpq.mhm (mardpet), or leader of the 
Mardians, was also descended from this tribe. Subsequently this 
name became a hereditary title similar to that of malxaz or aspet, and 
just as the bearers of the latter titles began to call themselves Mal^azuni 
and Aspetuni, so the house of the mardpets as well their possessions 
came to be known as Mardpet-akan, i.e. Mardpet-ian 57a . 

According to Pliny, the Menobardi and Moscheni lived in the vicinity 
of the Armenians, The first name should apparently be read [ar]- 
menomardi, whereas the second, *mochs-eni i corresponds not to the 
Moschians, as might seem likely, but to Armenian Moxene [Mokk'], 
Henefe, the principality of Mokk' also had an ethnic background 58 , 



THE TRIBAL BASES OE THE NAXARAR SYSTEM 323 

The Kvprtoi (Kurtians) lived together with the Mardians between the 
Zagros and the Niphates mountains. Like the Mardians, they settled 
on the Armenian border, but south of the Mardians, in the region 
which later bore their name : ^np££g (Korcek'), These Kurtians 
were the ancestors of the modern Kurds and should not be confused 
with the Karduchians who are a people of different origin, The 
country of the latter was called ynprpLg [KorduF] by the Armenians to 
differentiate it from ^nptyg [Korcek'], the country of the Kurtians. 
In the History of Faustus, Korcek' still designates the region of 
Salamas, but later it also included the neighbouring Korduenian lands, 
evidently as the result of the growing influence of the Kurds. Both 
the Kurtians and the Karduchians were less influenced by the Arme- 
nians than other tribes, but they nevertheless filtered imperceptibly 
into the Armenian territory, and in time their lands came to be con- 
sidered Armenian provinces 59 . The princely house of Kadme, i.e. of 
the Kadmeans, seems to have had very close connexions with Kor- 
duene 60 . 

The Matiani lived at first in the land of the Kurtians and subse- 
quently in their vicinity. The data of ancient writers whereby the 
maii were related not only by blood, but also by name to the Madians 
[Medes] (mada) has been corroborated by the findings of modern 
scholars. Manda, another form of the same name, is the one given 
by Assyrian inscriptions to the people ruled by Astyages. Mada has 
the same relation to manda as maii to manii (the form found in Mav~ 
Tiavrj) a name for Lake Urmia which was undoubtedly derived from 
people who once lived in the area 61 . The existence of these parallel 
forms is to be attributed to dialectal variations 62 . The names of 
the Armenian princely houses of A-mat-um and Mand-ak-xmi are 
closely related to these peoples. The home of the Amatuni was in 
Artaz, and Armenian legends refer to their Iranian origin 63 ; both 
these facts support the connexion we have postulated, 

The Madians [Medes] were usually known to the Armenians under 
the name of Mars, where mar is also to be derived from mada. Hence, 
the Armenian tradition that the principality of Muracan was descended 
from the Mars does not seem altogether devoid of foundation. The 
district settled by Mars must have been called* Maroc( Jfwpnij) m 
Armenian, just as the home of the Mardians was called Mard-oc. 
[The former must have given Wuipnij—hwh [Maroc-ean] or Ifwpng — mh 
Maroc-an] with the Persian suffix an, which later became ITnLpwgwh 
[Muracan] 84 . 



324 



CHAPTER XIV 



Separate tribal groups penetrated into Armenia from the lands of 
Atropatene which had been settled from ancient times, together with 
the shores of the Caspian Sea, by a multitude of tribes leading a 
pastoral ox nomadic form of life. In the period of Tigran the Great, 
many of these tribes, which were moving along the Araxes, supported 
the Armenian king against the Romans 65 . Side by side with Gelans, 
Kadusians, and AmaTdians. we find Ovroi (Utians or Uitians) to 
whom the Armenian Utians are related. The Utians first lived south 
of the Araxes, but later settled north of the river; and the Axaxes 
was considered to be the boundary between Otene and Atropatene. 
In Ptolemy, Otene is already located along the Kura and consequently 
corresponds to the Armenian Uti 6S . The Gelans, from whom the 
Armenian 9*% fykqwlinLbfi [gel, gelakuni] are probably derived 67 , and 
the Anianoi, who provide the link between the Zwbfi [Hani] of P'ayta- 
karan and Xkbp [Am], the city in Sirak 68 , moved into the area between 
the Araxes and Kura rivers together with the Utians, 

Saka sen, one of the districts of Otene, is usually identified with 
the Sakasene of ancient sources, which was held to be the land of the 
Saka. "We are told that the Saka moved from the steppes of Central 
Asia to the neighbourhood of Bactria, and occupied the district called 
Sakastan after them, now Seistan. Some of them moved westward and 
reached the Pontic shore, after having crossed Armenia, The Saka 
who remained in Armenia established themselves in the very fertile 
district known thereafter as Sakasene 69 , The connexion of Sakasene 
to the Saka has been disputed in recent times, but without any ap- 
parent justification 70 , Under Darius Codomanus, the 27a/cecrtW, are 
mentioned as living in the vicinity of the Albanians and of the Kadu- 
sians ; all three peoples were found in the Median camp under the com- 
mand of Atropates 71 , Irrespective of the explanation given for the 
origin of the Sakesinai or for their relation to the Saka, the existence 
of such a people is not in question. Siwnik' is the best known of the 
Armenian provinces bordering on Albania. It formed a single princi- 
pality like Mokk' or Tayk', and although part of Armenia, it was 
characterized by separatist tendencies. This fact must undoubtedly 
be attributed primarily to the ethnic particularism of the area, and it 
is understandable that scholars have sought traces of the Saka not 
only in Sakasene, but also in nearby Siwnik' or Sisakan, as it was 
called by the Persians, Attempts were even made to read HiuaKavq 
for 2aKao7}vrj„ though this runs counter to the textual evidence 72 , 



THE TRIBAL BASES OF THE NAXAMAB SYSTEM 



325 



This hypothesis is untenable also because the form Sisakan, whieh is 
the equivalent of Siwnik' in the opinion of -Xbrenaci, cannot be found 
in early documents. Morphologically it is of the same period and 
origin as the terms Swhnunhpwlpuh, yk^mlimh^ ^inuuinLpwlpuIi [Tanute- 
rakan, Sephakan, Vaspurakan], The contraction U{wwl{wb qm.hq [Sisakan 
gund] is similar to these names. The origin of the double name is 
still obscure, and it is not possible to say anything definite about the 
relation of the two forms to each other or to the Saka, The only 
certain fact is that Siwnik', the Albanian borderland of Armenia, 
had a population which differed somewhat from that of the central 
districts of Armenia, 

The tribal peculiarity of Siwnik' was supported and reinforced by 
a stream of migrations from the adjacent mountains, which have left 
traces in the toponymy : Thus, the ethnic origin of pmqf [Balk'], 
U"P4> [Sot'k'], U^witfy ]Alaheck'[, ^hmp^mp—m^li^ [Gargar-acik'], 
and ITriLJuwtig [Mu^ank'] are beyond question, 

The first of these names is to be found in connexion with Balasakan 
(Pwqwuwliwh), and is derived from the prnqwuty—g [Balasci-k'], a moun- 
tain people already known to Faustus 73 . 

Sot'k' is a district of Siwnik' which owes its name to the Sodi, Ar- 
menian tymlqt — wgfi [Covde-aei], who lived in southern Albania, This 
district corresponds to the Sodukene mentioned by ancient writers as 
being next to Kolthene (Armenian *injiP [Kolt']) in Otene 74 . 

The G-argarians are recorded by Faustus among the mountaineers 
who poured into Armenia in the reign of Xosrov II Kotak, together 
with the Piuqmu^lj^ [Balascik'], For Xorenaci, these same Gargarians 
are the neighbours of the inhabitants of Uti and Gardman, as well 
as of the Sodi [Covde], Later, ancient Uti or Otene, far excellence, 
with the city of Partaw, came to be known as the Plain of Gargar 75 . 
One of the tributaries of the Kura near Partaw is still called ^mplpup 
[Karkar, Gargar] in our time, The Armenian Gargarians were evi- 
dently descended from mountain Gargarians who lived in Albania in 
the foothills of the Keraunian of" Albanian branch of the Caucasus 76 , 

Muxank' (ITnLfu— u/D— ^), a subdivision of the district of Arca^, is derived 
from the Mvkoi„ one of the peoples living along the Caspian Sea. 
Their memory has been preserved in the name of the Mu#an[Mughan] 
— dast [IfnL^iuIi—quj^uj] Steppe, i.e. the Steppe of the MyMans 77 . 

In the same manner, the Pazkan-k' [^u/i^ii/fr— ^], or more correctly 
Parsakan, also in Arca^, were descended from the Ildpmoi,, The 



326 



CHAPTER XIV 



Kust-i-p'afens \yinmw~~fi—tjiumlhu] is perhaps to be related to the 
Ilappduioi 78 . 

All the ethnic groups listed, as well as others, provide the foundations 
on which the corresponding princely houses later developed. The 
immigration of new groups, the infiltration of separate tribes still 
continued. New groups often came in from neighbouring lands and 
were transformed into principalities in a similar fashion. Under these 
circumstances, the aristocratization of the country cannot have been 
carried out at once, and if we point to the period of Tigran the Great 
as the. time when tribal, institutions disintegrated and an aristocracy 
of classes was created, this must be taken as no more than & gener- 
alization 788 . 

The result of this ferment expressed itself in the transformation of 
the multiple homarchs or clan leaders into princelings. They brought 
forth the one hundred and twenty strategies found in Armenia at 
the accession of the Arsacids, and the formation of these strategies 
may be taken as the starting point of a new legal pattern in the history 
of Armenian socio-political evolution 7 81 \ 



XV 



THE FEUDAL BASES OF THE NAXARAE SYSTEM 



I. The replacement of the Tigranid dynasty by the Arsacids — Trdat I, the founder 
of the new dynasty — His personality, his historical aspect reflected in Armenian 
legends — The relation of Trdat to the 120 strategies — The situation of Armenia 
at the time of the appearance of the Arsacids — The feudal organization of the Parthian 
Empire and its origin — The social structure of Persia under the Arsacids; sahrdaran, 
vos r jmliran > vuzurgan, and azadhan, and their interrelations — The Arsaeid organization 
of Armenia — The distribution of lands and offices: naxarar, nahang, azat — The 
coalescence of tribal and administrative institutions — Reciprocal terms: tanuter and 
nahaj>et, 

II. The feudal aspect of Armenia — The nature of feudalism — Koman and German 
antecedents in the development of west-European feudalism — The naxarar system 
as part of feudalism: ^wjpkbjig and u^wptj.hLwl^mh^ as types of land tenure similar 
to allods and beneficia — Vassality: bumwjnLpjiLh servitium — The characteristic 
features of feudalism — The seigneurie as the fundamental institution of feudalism — 
The Armenian principalities as fiefs-seigneuries — Seigneurial legal relationships: 
[wuthriL-'] mlpnLppLU - seigneurie, the evidence — Eeudal relations in the narrower 
sense: hommage, auxilia, consilium and their equivalents in Armenia — Sub-infeudation 
and seputfium, its origin — Ostan freedom, 

III. Servile tenures — Conditions in the West — The tax system of the Parthians 
Basic obligations: Jcharaj and jizya\ or sale and oaz — cens and chevage, bahrah ~ cham- 
part, hor-behar = corvees — The size of the taxes — ■ Our conclusions compared with 
the account of Movses Xorenaci — Xorenaci as the historian of the naxarar system. 



I» 



In the beginning of our era, more precisely in A.D. 36, the dynasty 
of the Tigranids came to an end in Armenia with the death of Tigran IV, 
and a bitter straggle to seize the vacant throne began among the 
surrounding powers l . The Romans acted through the Iberians and 
supported the candidacy of these princes against the pretensions of 
the Arsacids, Vologaesus I, who had ascended the Parthian throne 
in A.D, 50, drove the Koman candidate, Rhadamistes of Iberia from 
the Armenian throne and installed his own brother Trdat I as king 
in approximately 53 A.D. The Roman legions under the command 
of Corbulo immediately made their appearance and forced the Parthians 



328 



CHAPTER XV 



to abandon Armenia ; Tigran V, whom Nero had appointed to assume 
the royal power in Armenia, arrived at the same time 2 , While 
Vologaesus, hampered by a war in Hyreania, was unable to drive out 
the Romans, Tigran began to act in a provocative fashion and to 
devastate neighbouring Adiabene, This action was taken as an insult 
by the Parthian nobles, among whom the most indignant were Mono- 
bazus of Adiabene and Trdat I, who had been driven from the Armenian 
throne. We are told by Tacitus that, 

Tiridates, too, dethroned and exiled, carried a weight in- 
creased by his silence and his restrained protests: — " Great 
empires were not conserved by inaction — they needed the 
conflict of men and arms. With princes might was the -only 
right [Id... aequius quod validius]. To retain its own possessions 
was the virtue of a private family: in contending for those 
of others lay the glory of a king " 3 , 

Influenced by these words, Vologaesus summoned a council and spoke 
as follows : 

This prince, the issue of the same father as myself, having 
renounced to me the supreme title upon the ground of age, 
I placed him in possession of Armenia, the recognized third 
degree of power; for Media had already fallen to Pacorus, 
And it seemed to me that, in contrast with the old brotherly 
hatreds and jealousies, I had by fair means brought order to 
our domestic hearth. The Romans forbid; and the peace 
which they have never themselves challenged with success, 
they are now breaking to their destruction, I shall not deny 
it: equity and not bloodshed, reason and not arms, were the 
means by which I should have preferred to retain the acqui- 
sitions of my fathers, If I have erred by hesitancy, I shall 
make amends by valour 4 . 



After these words, he once again tied the diadem around Trdat's 
head, and sent him to Armenia accompanied by a contingent of cavalry 
commanded by a nobleman named Monoaeses with the order to expel 
Tigran V and seize the throne. 

Trdat and Monoaeses manoeuvered so brilliantly that Corbulo was 
put into a position where, to " ... risk no further the laurels earned 
in the course of so rhany years ", says Tacitus, he asked for the appoint- 
ment of another general to defend Armenia 5 , His successor, Caesen- 
nius Paetus, a man of no ability, suffered a total defeat and began 



THE FEUDAL EASES OE THE NAXAMAE SYSTEM 



329 



negociations with. Vasak, tlie commander of the cavalry, who was the 
Parthian plenipotentiary. The terms accepted by Paetus : to remove 
the legions from Armenia, and to surrender all fortifications and sup- 
plies to the enemy, were so shameful, that Corbulo returned once 
more to the rescue. The Romans agreed to leave Armenia to Trdat I, 
on the condition that, « .,, Tiridates should lay the emblem of his 
royalty before the statue of the emperor to resume it only from the 
hand of Nero 6 », 

To fulfill these terms, Trdat set out on a long journey accompanied 
by his own family, his wife and children, the sons of Vologaesus, 
Pacorus of Media, and Monobazus of Adiabene, as well as by three 
thousand horsemen, After nine months, Trdat arrived in Italy, and 
since Nero was in Naples at the time, made his way to that city. 
The reception of the Armenian Hng was attended by pomp such as 
the Romans had granted to no other foreign ruler, Naples and Rome 
took on a festive appearance, The latter was illuminated and deco- 
rated with garlands 6a ; the entire city poured forth to receive the king 
and see him in person, so that the crowd filled the streets and even 
the roofs of the houses. In the middle of the square where the coro- 
nation was to take place, the citizens stood in order of rank clad in 
white robes and bearing laurel wreaths, while the soldiers were drawn 
up all around. The ceremony was unprecedented for solemnity and 
splendour. After the coronation, theatrical performances were given 
in honour of the king, and on this occasion the entire theatre was 
decorated with gold 7 , All this pomp demonstrates the high value 
set by the Romans on the form, flattering to their vanity, in which 
their defeat in Armenia was clothed, The Parthians had obtained 
their goal and consolidated their position in Armenia to the accom- 
paniement of Roman jubilation, 

The accession of Trdat I marks the beginning of the Armenian 
Arsacid line, notwithstanding national traditions which single out 
Valarsak as the founder of the dynasty 7a , The real ancestor of 
the Armenian Arsacids, Trdat I, seems to have been one of the 
most interesting figures not only among the Armenian, but also 
among the Parthian representatives of this famous dynasty. He 
united in his person both military valour and political perspicacity. 
The statement that " Great empires ,,. needed the conflict of men and 
arms ", which Tacitus puts into his mouth, characterizes him admi- 
rably as a military and political figure. His elder brother held to 



330 



CHAPTER XV 



the more humane view that, " equity and not bloodshed, reason and 
not aims ", were superior, but subsequent events showed the side 
of the truth, and the accuracy with which Trdat had perceived the 
true state of affairs, His boldly explicit belief, " id ... aequius quod 
validius, with its teaching that " .,. might was the only right", 
is equally apt for our times. It is an admission characteristic of a 
national leader who had experienced in his own life all the bitterness 
and tribulations of the powerless. 

The proud Arsacid prince bowed his head before the statue of the 
man he hated in order to receive a crown which already belonged to 
him, Naturally, he did not undergo this humiliating procedure 
because he lacked pride. According to the observation of ancient 
writers, a realization of their own worth bordering on haughtiness, 
and fearlessness were the characteristic features of the Arsacid princes. 
These qualities are also visible in Trdat, as evidenced by his desperate 
struggle against the Komans, His barely restrainable native pride 
was ready to burst forth at any moment. Thus, when he was required 
to remove his sword to be presented to Nero at Naples, the proud 
prince categorically refused to obey, and agreed only to put the 
sword in a scabbard, Cassius Dio comments that Tiridates " ... quel- 
ling his pride made himself subservient to the occasion and to his 
need, ... in view of the prize he hoped to obtain » 8 . 

The figure of Trdat I, the founder of the Arsacid dynasty, cut itself 
deep into the national memory and left a vivid imprint in popular 
legends. Armenian popular traditions, even in the form in which 
they have been preserved by Xorenaci, occasionally reflect with 
remarkable accuracy specific historical moments and events, though 
these are often distorted by the false pragmatism imposed on the 
national traditions by Xorenaci or the writer who first reworked them, 
The characteristic traits of king Trdat were preserved piecemeal in 
these traditions. They are clearly present in the personalities of 
Valarsak, of Artases, the darling of national bards, and even in that 
of Trdat III, the contemporary of the Illuminator. According to 
the Armenian tradition, the founder of the Armenian Arsacid line 
was called Valarsak and not Trdat. Valars-ak [*lui%wp£ — ml[\ is the 
popular form of Vologaesus (Persian, Valas), the name of Trdat Ps 
older brother who was ruling in Persia 9 . Vologaesus was deeply 
concerned with the fate of Trdat, and probably for this reason his 
name was interchanged with his brother's in legendary accounts. 



THE FEUDAL BASES OE THE NAXARAR SYSTEM 



331 



In appointing his brother king of Armenia the Parthian king showed 
him the boundaries of his domain, adding the comment, « ... the boun- 
dary of the brave is his sword, as much as he cuts off, so much he 
possesses". This statement which is strikingly reminiscent of the 
words of Trdat I quoted above, " ... id ... aequius quod validius", 
render admirably their fundamental concept 10 . Morp'iwhk 
[lfnpipjiLqJjli\ 9 against whom Valarsak was compelled to wage war, is 
none other than the Roman general Corbulo, the enemy of Trdat 10a . 

According to trustworthy historical evidence, Trdat, the first Arsa- 
cid ruler, repelled an attack of the Alans and drove them from the 
territory of Armenia in 75 A.D. National tradition, however, attri 
butes this action to Artases II in the romantic tale of the princess 
Sat'enik 10 *\ Some of the actions of Trdat I have also been transferred 
by Xorenaci to the king of the same name, who was the contemporary 
of the Illuminator. Thus the famous lariat episode, which in reality 
involved Trdat I and occurred during the Alan campaign, of 75 A.D., 
has been attributed to Artases II n , whereas the exploits related by 
Xorenaci about king Trdat III are echoes of events which occurred 
during the gladiatorial games offered by Nero in honour of Trdat I, 
where the Armenian king was said to have brought down two bulls 
with a single arrow 12 . These revivals and echoings of the period 
of Trdat I in relatively late accounts prove that memories of the first 
Arsacid ruler were very vivid indeed, and that his reign had aroused 
great interest. 

The fame of the king presumably rested not only on external events, 
but also on his internal accomplishments. The national traditions 
depict the first Arsacid king as the renovator of the country, and 
they see his period as the beginning of its rebirth. There is a certain 
degree of truth in these views. The reforming figure of Valarsak 
sketched by Xorenaei is a reflection of the historical Trdat I. Trdat 
ascended the Armenian throne ca. 50 A.D,, and ruled with only a 
brief interruption until the eighties of the first century A,D. After 66, 
conditions became favourable for development, and were not affected 
by the brief Alan invasion of 75 A.D. 

The first possibility which comes to mind in this connexion is to 
attribute to the period of Trdat I the division of Armenia into one 
hundred and twenty prefectures or strategies. The earliest author 
to speak of these is Pliny the Elder, Trdat's contemporary, who 
perished a victim of the erruption of Mt. Vesuvius in 79 A.D. Strabo, 



m 



332 



CHAPTER XV 



who died in 23 AJ),, and who was well informed concerning Armenia, 
is not acquainted with these strategies, though he mentions divisions 
bearing this name in Cappadocia 12a , If the strategies developed as 
the result of anyone's initiative in the interval of time between Strabo 
and Pliny, this initiative can only be that of Trdat I, but the divisions 
represented, and the reason for which they were called strategies are 
not altogether clear. The terminology of classical authors with 
reference to foreign administrative units is not known for its accuracy, 
in contrast to irrapxia or provincia, arpar^yia seems to be an unusual 
expression designating any district which did not fit into the Roman 
institutional pattern. Whatever their date, however, these divisions 
were unquestionably native in origin, and this fact is not altered even 
by their Greek name, just as Parthian coins remain Parthian despite 
their Greek inscriptions, 

Pliny gives the name of one of the prefectures or strategies, namely 
Carenitis or Karin, and thus clarifies the nature of the one hundred 
and twenty divisions from a territorial point of view 131) . They are the 
Armenian districts familiar to us from classical and Armenian literature, 
Pliny describes Armenia as being within the following limits : from 
the city of Daseusa on the banks of the Euphrates to the Caspian Sea, 
and from Tigranokerta to Iberia 13 , i.e. within the same boundaries 
as those given in the Armenian Geography, According to Pliny, the 
entire country was divided into one hundred and twenty strategies, 
" barbaris nominibus ", although the Armenian Geography gives up 
to one hundred and eighty five small districts. Nevertheless it is 
not difficult to isolate the one hundred and twenty older districts 
from the larger number ; the names of more than fifty of them occur 
in Strabo, Pliny, and Ptolemy, The divisions indicated by Pliny are 
not units of an administrative nature ; they developed independently 
through private initiative, Hence, all that we can attribute to the 
first of the Arsacids is to have put them in order and to have brought 
them into a definite system. The new dynasty merely sanctioned a 
situation which it had found in the country. 

There are no reliable indications concerning the socio-political aspect 
of these divisions or of the country in general. The historians of the 
wars of Trdat distinguish two strata within the population : the 
proceres and the plebs, i.e. the aristocracy and the common people; 
the nobles are also called megistanes, primores, or simply nobiles 14 , 
The social content of these terms can be postulated on the basis of 



/pO^_ 



THE FEUDAL EASES OF THE NAXABAM SYSTEM 



333 



our previous analysis. Since we have seen that tribal forms of life 
disintegrated in the period of the Tigranids, and that the tribe split 
into two parts with the aristocracy devoloping from one of these, the 
proceres and megistanes cannot be anything but the former variously 
named clan heads, who by this time had reached the level of rulers 14a . 
The nobiliias which had arisen on the ground of the disintegrated tribal 
life formed a closed circle in which many of the customs of the earlier 
tribal stage continued to operate, and tribal terminology was also 
transferred to this smaller circle. Thus, the heads of noble families 
were called, ier, aspet, mal-gaz, mamaJc, etc, as before, in spite of the 
altered nature of the clan, and the terms manuk, sepuh, vor-ear, and 
aiaxin, which had formerly designated the children of tribal communi- 
ties, began to be used exclusively for the children or members of 
noble families, though manuk and alaxin also kept their more general 
meaning 15 , Analogous phenomena are observable in Latin where 
liberi means simultaneously, « children » and « free men », Another 
such example is found in the forms patres and pairieii, which have 
the same relationship to each other as aspei and wpuh* Here pairicius 
is the adjective from pater and indicates membership in a family, while 
the pairicii are the liberi of fathers of senatorial rank 15a , A number 
of terms meaning « clan» — a^ &$, mn£j), ^mp^[azg, azn, iohm y zarrn] — ■ 
also acquire a connotation of nobility, and similarly, mnJi [tun = 
" house "] takes on the sense of a princely clan or domain 151 '. The 
remainder of the population which did not belong to the nobility, 
i.e. the pUbs, was called according to the district — ■ qb^m^ ^fjhwlimi, 
nrndfili^ qnk£[ili [gelyulc, smahcm, famih, greMJc], perhaps also ^«/# 
[msalc], etc, — JRamih and msah seem to be derivatives from apa^ioi 
and poo-xoi, the names of the native peoples who were conquered 
and subjected by the newcomers, and whose names consequently took 
on a connotation of inferiority. The Aramean wymfulifi [aiaxiri] acquired 
the sense of a serving woman for the same reason, As for the term 
gfeMk, it corresponds to the Georgian glexh and has the same origin 15 °, 
Such, then, were the conditions present in Axmenia at the accession 
of the new dynasty. The process of aristoeratkation was Teaching 
its term and had consequently brought forth one hundred and twenty 
rulers, both great and small. The princely houses which were to 
dominate the country in later times were already in existence, 

According to national traditions, social relationships under the 
Axsacids entered the phase of development characterized by the term 



334 



CHAPTER XV 



na-^arar. "We have already seen that the period of Arsacid rule 
marks the stage of Iranian history characterized by a socio-political 
system reminiscent of western feudalism. The nayarar system is its 
counterpart in Armenia 15d . 

Scholars have given a great deal of weight to the factor of conquest 
in the origin of Arsacid feudalism, but such an interpretation is not 
altogether correct 16 . As a result of its development a tribal society 
evolves naturally into a feudal pattern, In Persia, feudalism had to 
follow the tribal stage as an indispensable phase ; the real problem is 
to determine whether the presence of secondary factors accelerated 
the process of change. The Persian Empire arose in an area fertilized 
by the cultural achievements of the vanished empires of Assyria and 
Babylonia, and it was also the heir of much of their political wisdom, 
The earlier empires had possessed fairly elaborate administrative 
systems, which necessarily influenced that of the Persians to some 
degree. Among the Persians themselves, administrative institutions 
had also existed from ancient times side by side with the tribal 
organization ; next to the dahyu we find the x^dra, ruled by a satrap. 
These all-powerful servants of the king of kings were not merely 
officials or governors. Their appointments to the countries they 
ruled were made on an individual basis, but in practice they 
consolidated their hold over their area through the hereditary 
inheritance of the office itself. In this sense the satrapy was an 
important social as well as political factor. 16a 

Men who had been granted lands, though they did not govern them, 
and who were present at the king's court as his attendants were also 
ranked as satraps. Lands were given out for administrative purposes 
but also merely as a source of revenue. When King Cyrus of Persia 
distributed satrapal positions to his friends, he said to those whom 
he kept by his side, 

" I have further decided that any of you who remain here, 
and to whom I may occasionally give the trouble of going 
on business for me to those nations, shall have lands and houses 
there; so that they may have tribute paid to them here and, 
whenever they go there, they may lodge in residences of their 
own ". 

Xenophon goes on to comment after this statement, 

he gave houses and servants in the various states which he 
had subdued. And even to this day those properties, some 



THE FEUDAL BASES OF THE NAXARAB SYSTEM 



335 



in one land, some in another, continue in the possession of the 
deseendents of those who then received them, while the owners 
themselves reside at court 17 . 



Not only courtiers but government officials in general should be 
included among such men. Under the Achaemenids officials such as 
the collector of taxes, the dispenser of the revenue, the supervisor of 
public works, the supervisor of the treasury, the supervisor of supplies, 
the overseer of the stables, the master of the hunt, the commander of 
the infantry, the commander of the cavalry, the generals, et ah were 
already known in addition to the provincial satraps (idv&v aaTpairai), 
The king likewise had a door keeper, a baker, a cook, a butler, a bather, 
table servants, a chamberlain, an adorner, etc., 18 All these public 
or court officials were distinguished from the provincial governors 
and two types of land tenure were distinguished accordingly. In the 
first case land tenure meant a territorial estate, a possession, while 
in the other, namely for the satraps, it meant territorial rule, ix, 
the union of a territorial holding with political rights. Such a fusion 
is the general and the main foundation of feudalism. 

When Cyrus sent the satraps to their appointed posts, he admonished 
them to follow his example in all things, among otheT things, he also 
demanded of « , . , as many as received lands and authority », that they 
present themselves at the satrap's court, that they bring their children 
to be educated there, and in general, that they serve the satrap 19 , 
According to this indication, the landed aristocracy in a given satrapy 
stood in the same relation to their satrap as the satraps to the King 
of Kings. The satrap is shown here as a sort of king or ruleT in the 
territory entrusted to him. He is the one endowed with " yr\v koX 
apx^a ", i.e. territorial and legal sovreignty, the two basic aspects 
whose fusion characterise feudal land tenure, In this sense, the 
Achaemenid empire already showed the rudimentary forms of a feudal 
pattern 19a , 

The same administrative system with its distribution of lands 
functioned also in the Axsacid period. Like the Achaemenids who 
had usually distributed the satrapies among their kinsmen, the Arsa- 
cids maintained a tribal point of view and entrusted the government 
of the conquered lands to members of their own family. "When the 
Parthian king Vologaesus^I appointed one of his brothers in Media and 
the other in Armenia, he believed that he was thereby settling the 
affairs of his family " ... familiae nostras penaies rite composuisse 20 ". 



336 



CHAPTER XV 



A council of notables (ordo probulorum) seemingly composed for the 
most part of the representatives of the noble families, existed under 
the Arsacid kings. According to historical testimonies, commanders 
in wartime and governors in time of peace were chosen exclusively 
from this group, " ,,, ex hoc duces in bello, ex hoc in face rector es Jia- 
bent 21 ". The term rector es designates not only provincial governors, 
but administrative officials in general, and such offices were numerous, 

A Byzantine author has preserved for us the information that seven 
famous families had hereditary control of the more responsible and 
honourable branches of the government, according to the ancient 
laws of the Medes (i.e. the Parthians), The first of these was the 
Arsacid dynasty which held the royal power, and to which the right 
of crowning the king also belonged, After it came the families which 
performed the following duties : the command of the army, the civilian 
administration, the supervision of justice, the command of the cavalry, 
the levying and collection of taxes, the supervision of arms and military 
dress 22 , This information must refer to the Parthian period as is 
obvious from the words referring to the ruling family. But even if 
we admit that seven noble families existed among the Parthians, these 
can have had no connexion whatsoever with the seven satraps found 
under Darius I. Such a division of power would have been alto- 
gether fatal in a country which had repeatedly suffered dynastic and 
other overturns. The incontestable and essential part of Simocatta's 
description lies in his indication that certain duties were considered 
particularly honourable in the Parthian realm, and that these functions 
had been assigned to certain families as their hereditary right.Ad- 
ministrative offices were obviously becoming hereditary just like the 
satrapies, and this development brings us close to the concept of fief- 
office in western European feudalism 22 a . 

Hence, from ancient times, a purely administrative structure based 
on service had operated side by side with a tribal organization, and 
had brought about the creation of feudal classes under the influence of 
this tribal society. We have already seen that classes had also de- 
veloped from the disintegration of the tribal mode of life. We do not 
know how the results of this double process manifested themselves, 
nor the form of class structure found under the Arsacids, but some 
understanding of this may be obtained from an analysis of the Had- 
jiabad inscription of Sahpuhr I (240-270) 231 >. 

This inscription was set up during one of the king's hunting trips, 



THE FEUDAL BASES OF THE NAXABAB SYSTEM 



337 



< -• 



and the following sentence is found in it : « I shot my arrow in the 
presence of the satrdamn va barb%ian va raban va azaian " [$ic] ? or in the 
Pehlevi version, "xmdrddran u vaspuOrakan u vacarMn u azaian" [sic] 2 *. 
These terms are a list of the various ranks of free classes 34 . It is 
evident that we are not dealing here with a haphazard collection of 
names but with a rigorously defined class stratification, since Arabian 
authors are also familiar with four noble classes among the Persians, 
although they mistakenly attribude their creation to Ardasir I, the 
founder of the Sasanian dynasty 35 . While it is true that the Had] la- 
bad inscription belongs to the reign of Sahpuhr I, its evidence on class 
structure belongs to the preceding Arsaeid period. At the time of this 
inscription the new dynasty had not had the time to make fundamental 
changes in the organization of the country, especially since it had 
been forced to spend a great deal of time in the consolidation of its 
political position. 

The first term in the list consists of satra, " country ", and dar, 
" to hold ", and corresponds to the mtrapa (Armenian iwCmui [sahap]) 
of the Achaemenid period, which is derived from the same satra and^a, 
" to keep '\ The fact that the sairdaran accompanied the king on 
the hunt shows that they were not ordinary satraps or provincial 
governors, but that this term has broader implications, since it is 
clearly not possible to believe that the satraps were summoned by the 
king from their provinces merely to go hunting. The persons bearing 
this title seemingly stayed constantly in the capital, and were probably 
connected with the court, Hence, sairdar designates a social or 
class rank rather than an administrative one. These were the titled 
magnates, the representatives of the great territorial estates, and 
they can be traced back to the king's friends whom he had endowed 
with lands, as we have seen above. They were equal in social rank 
to the satraps but had no official or administrative functions 25a . 

The next rank consisted of the vaspuhran. This term which is 
taken from tribal society, is already familiar to us ; it designated the 
son of a clan, and, after the disintegration of the tribal pattern, the 
son of a ruler. In tribal society, the sons of the clan ranked high, 
since they were the chief associates of the clan leader both in battle 
and in the pursuit of booty 26 , The high rank of the vaspuhrs is noted 
in Avestic literature. Here, the price given for a nmanopaiti is one 
bull of lowest quality, that of a vispaiti is one bull of medium quality, 
that of a zaniupaiii is one bull of highest quality, and that of a dahyupaiti 



338 



CHAPTEK XV 



is a chariot with four horses. The visoypuhra is left out of this list, 
but his value is one bull of highest quality, i.e. the price of a zaniwpaiii. 
Consequently, the visoypuhra was ranked second only to the dahywpaiii, 
and the latter, as the governor of a province, was equal in rank to a 
satrajpa or sahrdar (dahyu = sahr). Thus the Avestic tradition on 
the rank of the vdsjpuhrs coincides with the evidence of the Hadjiabad 
inscription, since the vdsfuhrdn also follow the sahrddr an in this 
inscription 27 . The vdspuhrdn were the sons of the sahrddrs 27a , We 
should understand by this not only sons in the strict sense of the word, 
but also the members of the family in general, i.e. all the male represen- 
tatives of that family except for its head, the sahrdar himself. During 
the lifetime of the sahrddr, his sons had potential authority as his 
heirs, for this reason they ranked below him but above everyone 
else a7 *\ 

Third in rank stood the vacarkdn or vuzurgdn, that is to say the 
elders, from vazraJca (Persian buzurg), " great ". This title is best 
suited to those noblemen who were descended from ancient tribal 
leaders, The institution of the abicaris or iAevdzpa ayopa, with 
which we are already familiar, had provided the crucible in which 
the heads of clans had been transformed into nobles and court officials. 
The word abicaris may have been pronounced vacar in Pehlevi as 
is suggested by the Armenian form t[ui£um [vacar]. The name of the 
third rank may, therefore, be read vazrakdn or vacarkdn, In the 
the latter case it can be derived from vacar, with the sense of a parti- 
cipant or member of the abicaris. Only the wealthy came to the 
abicaris and concerned themselves with military affairs. This 
was the circumstance which made possible their transformation into 
a privileged class. Eegardless of its etymology, it seems to us that 
vuzurgdn must be understood as the general term for that part 
of the nobility which had grown out of a tribal background. 

The last in rank were the dzddh-dn, who were related to the vuzurgdn 
in the same way as the vdspuhrdn were related to the sahrdar an. 
Azddh-an is the plural form of azddhd, the past participle of d-zan, dza, 
" to give birth ", The dzddhs were the sons and heirs of the vuzurgdn ; 
in a legal sense they shared the rank of the vdsjpuhrs, i.e. they had power 
as future vuzurgdn 27C , Such then, is our interpretation of the inter- 
relation of class ranks ; the evolution of classes in the Arsacid period 
ended with them, and we must bear this circumstance in mind when 
we study the Arsacid legislation in Armenia. 



THE FEUDAL BASES OP THE NAXABAB SYSTEM 



339 



A two-fold development of the class nobility, based on tribal as 
well as purely administrative institutions, must be acknowledged 
as having taken place in Armenia as well as in Parthia. The custom 
of distributing land and offices existed also among the Armenian 
Arsaeids. Faustus, the historian of the Christian Arsaeids, testifies 
that the Armenian princes owed much to the Arsacid house, First 
among these benefits were territorial possesions: the rulers gave to 
the nobles great provinces, countries, villages, and other estates; 
they granted them benefits by bestowing upon them honourific 
or administrative and service offices 27d . As we have seen, fefvwiwL- 
PfiLU [isxanutHwTi] was distinguished from qnpbinIiwinLp[jdi [gor- 
cakalutHwrb], the princely rank from the service or official position. 
Members of the former were identified according to cushion, those 
of the latter according to rank: " ... ifhbwilkbgh jfiLpwgwh $Lp qwCn^ 
zf-nphwl^iu^h jfiLpuigwh^fiLp ^wipiiL : [He re-established ... the great 
each to his cushion, the officials each to his rank] " 38 . 
Princes had existed in Armenia before the appearance of the Arsaeids, 
while the administrative machinery was introduced by them. Accord- 
ing to Xorenaci, the achievement of the first Arsacid king consisted 
in his distribution of diverse public and court functions among the 
native princely families. The official duties noted in the Parthian 
Empire were also present in Arsacid Armenia 28a . 

The honourific duty of crowning the king was considered to be the 
prerogative of the Surens in Persia and of the Bagratids in Armenia 29 . 

The supreme command of the army, uuiwpiuujkmnLpfiLh [spara- 
peiufiwn], was hereditary in the family of the Asparapets [Aspahbadh], 
or Kosmids in Persia, and in that of the Mamikonean in Armenia 39a , 

The office of hazarapet was of great importance ; according to all 
the evidence it corresponded to the third office listed by Theophylact 
Simocatta, i.e. to the one concerned with " rets ttoXitik&s <j>povriha$" , 
as opposed to the one dealing with "ttqXzixikjjs crwTa|<=tos , " 29b . In 
Faustus' History, the hazarapetut'iwn, or office of the hazarapet, 
was in a certain sense contrasted with that of the sparapet The 
former official was appointed for " public welfare [supervision] 
w^fuuiptimnkub fuhiudwliwini.pkwhb ", and concerned himself with 
qhtywhriLpkujIi, ix, with the peasants working the land. The Gnuni 
house, which held the office of hazarapet, was called ^fihwlimhm^ih (sina- 
Jcanasen [peasant ruling or supervising]). On the other hand, the 
Mamikonean, as spampets, were said to stand, " ft *[hpwj fe/uujIwLphuibh 



340 



CHAPTER XV 



p.ni^Luhrj.iuli fi i[kpwj mdkhwjh ynpmy tjnpwihnpnLpbwhh, above all the 

princes, above all the zoravarh* or military commanders] " 29c , The 
Armenian army was made up of many contingents furnished by 
the princely houses, Each of these detachments was commanded 
by its own prince, but the supreme command belonged to the 
hereditary sparapets, the Mamikonean house, who, in this sense, 
stood " above all the princes ", and their armies. 

The entire peasant population was under the authority of the 
Jiazarapet Hazara-pati = %iMapxo$ ; " cMliarch " or " ihousander ", 
was a military term which had been transferred to the civilian sphere 
as early as the Achaemenid period. According to Xenophon, Cyrus, 
concerned with the regularization of finances because of the enormous 
expenses resulting from such a vast realm, decided to adopt the 
system of military divisions for this purpose, and consequently set 
up dehadarchs, hcliagoi, chiliarchs, and myriarchs in the financial 
branch to simplify its administration 30 . According to this system, 
Jiazarapet must be the title of the highest ranking official in the ad- 
ministration of the state revenue, and his relatione to the taxeable 
population, i.e. to the peasants becomes understandable, The ham- 
rapet might be thought to have supervised taxation as well, but we 
know from the list cited earlier that the levy and collection of taxes 
formed a separate department, It is possible that this duty was at 
first included among those of the hazarapet, but that it was later 
removed to a separate division. The all-powerful Sasanian magnate 
Mhr-Nerseh was called great Jiazarapet (dhb £wqwpwu{hw) by the 
Armenians, and vuzwrg-framadhar by Tabari, a title which, 
according to his own translation, meant grand vizir 31 , Nevertheless, 
Mhr-Nerseh had under him a certain Mahgusnasp, who supervised 
the department of taxation and was called vastryosan-salar, i.e. 
" overseer of the tillers of the soil " 32 . This title is similar to that 
of the Jiazarapet in the sense in which Faustus used it to characterize 
the Gnuni family 33 , Perhaps the former office of hazdrhadh was 
split in Persia into two departments: one for^as* wohriKas <j>povTi$as 
was supervised by the vuzurg-framadhar, and the other by the vasiry- 
osdn - solar, whereas in Armenia this office remained unaltered. 

The judicial power or the administration of justice lay in the hands 
of the Magians or of the clergy in general. With the conversion of 
the country to Christianity, this passed to the Church, and^bejpnged 
to the bishop in each province, The authority of bishop Daniel of 



THE FEUDAL BASES OE THE NAXARAR SYSTEM 



341 



Taron is defined by Paustus as that of " qnpbwlfiuLriLpkwhh 
ilhbft tiwmwLnpnLpbwhh : [the office of great justiciar ot high 
justice] " 34 . Both the interference of the clergy in political affairs, 
and its great importance in both Armenia and Persia is partly explained 
by the fact that it served as a judicial institution 35 . 

The command of the cavalry ttjs Ittttov) was also a hereditary 
right, but we do not know to what family it belonged, The Armenian 
army consisted primarily of cavalry, but it was under the over-all 
command of the sparapet According to some indications, the Persian 
cavalry was commanded by the Spandiyadh house 36 , 

The supervision " rwv ottXcov kcli ttjs ttoXgijuktjs icrdjjros " corres- 
ponded to the modern quartermaster's duties and was apparently 
connected with the house of the Mardpet 36a , The exalted rank of this 
house at the court of the Christian Armenian Arsacids was noted 
in historical sources, but its duties remain altogether unclear, and 
we have nothing but contradictory evidence on this subject. The 
origin of the title, mardpet, which we clarified earlier, had been forgotten 
at the time of the advent of literacy in Armenia, On morphological 
grounds, this term seemed to contain the Armenian word Siupq [mard], 
" man ", and this mistaken etymology interfered with the under- 
standing of the true nature of the office, Xorenaci did not even 
mention the title of mardpet, and admits truthfully that he knew 
nothing about the house of prince Z&jp [Hayr] 37 , Mardpet was a 
tribal princely title, and had nothing to do with any office, The 
mardpet " was called, according to his duties, father of the king 
[Smpquihm^ np lufwLwhhw^ ^ n >tp £«^ *l n P^ n fi * ^ m JP PwqwLnpfih) " 38 

For this reason, Suipquikmntpjidi [mard-petufiwn] and ^mjpnipfiLh, 
[hayrufiwn} are used as synonyms 39 . 

We do not know the true meaning of this imposing title ; we know 
only that the supervision of the royal treasure and of the fortresses 
in which it was kept was one of the duties of the mardpet, and that 
" this duty and the office of mardpet called Jiayr [father], had been 
the prerogative of eunuchs from ancient times " 40 , The pompous 
title of the mardpet indicates a very close connexion with the Toyal 
court. Because they were eunuchs, the mardpets served as the guar- 
dians of the queen 41 , and the supervision of the royal wardrobe 
probably also belonged to them. When King Pap decided to kill 
the mardpet Glak, he ordered first that the mardpet be dressed in 
inordinately wide garments, then that he be taken to the <c house of 



342 



CHAPTEK XV 



the wardrobe {wndi luiumSn ^wiling) ", and executed there 4S . If the 
wardrobe was under Glak's control, the execution of its supervisor 
within its very walls must have seemed particularly humiliating, 
as the king had intended 43 . 

The administrative system of the Arsacid state was not limited to 
the offices listed. Our purpose in investigating them was to demon- 
strate the similarity of the Arsacid administrative systems in Persia 
and in Armenia. The custom of granting lands and offices necessarily 
produced the same results in Armenia as in Persia, namely the creation 
of social entities similar to the native princes both in rank and terri- 
torial possessions, but distinguished from them through the origin 
of their status, The Persian Arsacids created such positions for 
their relatives, but the Armenian kings, at least in the Christian 
period, for the most part promoted influential representatives of the 
native hereditary nobility, though it is possible that they too, at 
first, primarily favoured their kinsmen. In either case, a new element 
was added to the native legal system, either through giving a conno- 
tation of officialdom to the aristocracy, or through the creation of 
new noble houses, whose privileges derived from sources other than 
those of the native princes. 

This innovation was reflected in the relevant title, which is that 
of nax&rar. Unfortunately for us, this term does not yet lend itself 
to etymological analysis 44 , Originally it probably designated the 
provincial governors, rectores. As a social concept, nayarar was the 
equivalent of the Persian sahrdar, and the corresponding territorial 
unit, the nahang, was equal to the sahr. It is also possible that the 
term azat designated the son of a naxarar. The new terminology 
did not oust the old, since the new concepts it expressed had not 
replaced but only paralleled the earlier ones. But the older tribal 
terminology tended toward greater uniformity as the result of the 
wear and tear of new relationships. Ter became the predominant 
title among all the terms by which clan leaders had once been known ; 
the others either vanished or became localized and frozen in a particular 
clan as its hereditary title. Ter became the technical term used to 
designate the head or senior member of a princely clan. The terms 
used for the sons or children of a clan underwent a similar standardi- 
zation, Sejpuh was assimilated exclusively to princely families, 
while the other terms kept their original general meaning. In the 
same way, among territorial terms, tun, li house ", became the tech- 
nical term used to designate the corresponding territorial concept. 



THE FEUDAL BASES OE THE NAXABAR SYSTEM 



343 



Tims, side by side with ter, sepuh, tun, which had been the concepts 
of the tribal society, appeared nax^rar, azat, nahang, which were the 
concepts of the administrative order, but the tribal traditions proved 
so powerful that whatever new relations arose had to be fitted into 
the earlier framework. For this reason, the sources alternate ier 
and naxarar, sepuh and azat, tun and nahang, as equivalent terms. 
With the consolidation of the administrative pattern, the term na%amr 
became the dominant one with the meaning of " prince " or " noble- 
man ", while ter turned into an uncharacteristic expression for " lord " 
or " owner ". Secondary expressions, such as nahapet and tanuter 
which replaced ier and naxcirar, became necessary for the identification 
of the head or senior member of a princely house. The origin of the 
first term is not clear, but it seems to have had the same relation to 
nahang as tanuter to tun. These newer expressions must have taken 
over the original meaning of ter and naxarar. They did not, however, 
replace ter and naxarar altogether, but were used in preference to the 
earlier terms whenever the head of a clan had to be singled out 
from it 44a , 

This is the manner in which the more characteristic terms of Ar- 
menian social life made their successive historical appearance. Up 
to now, we have pursued our investigations according to general 
historical principles, while maintaining a contact with the historical 
reality reflected in the sources. Eeturning to our earlier simile, 
we may say that starting from the top of the tree under investigation, 
we have tried to uncover its roots according to the guidelines found 
in its visible or accessible parts. < The hypothesis we have elaborated 
on this basis can only have the validity of all such inductive work. 

We shall now turn to an analysis of that phase of the phenomenon 
we are investigating which is known to Armenian literature, and 
which has served us ex silentio as a foundation for our earlier excursions 
into antiquity. 



II 



Arsacid Armenia is depicted in historical literature as a country 
divided into several dozen principalities. Although legally equal, 
these principalities were differentiated in fact by their political im- 
portance and their degree of dependence on the crown. The most 
independent were the principalities which lay along the borders. 



344 



CHAPTER XV 



First among these were the domains of the so-called frdel^sr&nd next 
to them the principalities of Siwnik 4 , Mokk e , Tayk', and Sophene 
[Cop'k'], Among the central principalities those of the Mamikoneans, 
Bagratids, Mardpets, Arcrunis, Xor^orunis, and Kamsarakans, were 
especially outstanding. The remaining majority consisted of small 
principalities whose rulers were greatly inferior by birth and ancestral 
possessions, and who mostly took refuge within the sphere of influence 
of the more powerful princes, 

All territorial units were considered equal in rank. Each unit 
was a tun with a ter or tanuter at its head, and these units were called 
fcm^erdoms [tanuiemfiwnF] or w^arardoms \nax^rmutiwn¥\ The 
tanuter&oms were closed circles with sovereign prerogatives. They 
had judicial and administrative autonomy as well as the other attri- 
butes of sovereignty. They were a type of small kingdom, and their 
ruling princes were little kings or minor sovereigns. The limits of 
each one's jurisdiction depended de facto on the power he possessed 
at a given moment. The historians often speak of great nobles 
(ilhbutifkb im.wifaihfi [mecamec awagani]) and distinguish the senior 
{mum} [awag]) tanuters or naxarars from their juniors (l(pumhp 
[krster]), evidently on a basis of relative power. The general position 
of a tanuter&om was determined by the actual relationship between 
the power of its prince and that of the king, and by the predominance 
of centripetal or centrifugal tendencies. Whenever circumstances 
were favourable, the kings did not hesitate to confiscate the territorial 
possessions of guilty princes 44 K 

This general picture outlining the political aspect of Armenia, 
should prove sufficient to compare its internal structure to the system 
known to us under the name of feudalism 44 °, A many-headed political 
body with little administrative unity, is particularly characteristic 
of a feudal state, Iii the period of the flowering of feudal institutions, 
France, the locus classicus of feudalism presented the same aspect 
as Arsacid Armenia 44d , The country was split into a multitude of 
large and small principalities which enjoyed sovereign rights to various 
degrees. Some exhibited a highly developed feudal organism and 
enjoyed full sovereign powers; others were great lordships [seign&unes] 
with the same powers ; finally there were small holdings with limited 
rights, From the point of view of sovereignty, all principalities were 
divided into two categories: great and small lordship (grandes seigneu- 
ries, petites seigneuries). The former, up to forty in number, enjoyed 



THE FEUDAL BASES OE THE NAXABAM SYSTEM 



345 



all the rights inherent in a state (droits regaliens) and made up the 
various duchies, countries, and vis-counties (duches, comtes, vicomies). 
The small lordships were very numerous and consisted of cMiellenies, 
secondary vicomtes, vidamies, and avoueries; these did not have the 
same political importance as the grandes seigneuries, and their lords 
were often indistinguishable from vassals, because of the limitation 
of their rights 45 . The feudal character of Armenia was not expressed 
exclusively in its external aspect, however ; the similarity went deeper 
and also included the legal system operating within the country. 

Feudalism is a well known politico-social pattern which is not 
exclusively characteristic of the Middle Ages. In a certain sense 
it appears to be an inevitable stage in the development of human 
societies, and can be traced among various peoples and in different 
periods of the history of culture 45a . It has been shown by historical 
science that feudalism develops in both ascending and descending 
orders, that is to say, either when the cultural level is rising or when 
it is falling and disintegrating. In this sense the controverse between 
the Eomanist and Germanist schools over the origins of feudalism 
ceases to be a true conflict. Both sides are right albeit one-sided: 
The dissolution of the Roman Empire already disclosed processes 
of feudalkation, but the barbarians, who had vanquished the Romans, 
were also set on the path of feudaliaation insofar as their own tribal 
pattern was disintegrating in the fifth century, The two independent 
processes were leading to one and the same system. Their meeting 
and clash after the Germanic invasions merely complicated the nature 
of the institutions developing at that time. 

One of the most characteristic traits of a feudal system observed 
by scholars is fragmentation, or the dispersal of supreme political 
power. According to this definition, feudalism begins to develop 
when the territory, the supreme power, or the political and legal ties 
in general are divided among many regional rulers and when groups 
subject to private rather than public authority make their appearance. 
Large-scale tenure of land simultaneously marks the point of departure 
of feudalization ; it is the foundation of the entire system, all of whose 
other features then follow from it. 

Two categories of holdings are recognized as the bases of feudal 
land tenure: the allod, ancestral or royal, and the beneficium, or grant, 
The institution of the fief [feodum], from which the entire system took 
its name, is supposed to have developed from the second category, 



346 



CHAPTER XV 



The word feodum - fief does not appear in documents before the 
tenth century, It was first used side by side with benefice, and in 
the same sense, but subsequently, from the end of the eleventh century, 
it replaced the earlier term. From that time on, fief began to designate 
land whose use was contingent upon military service. Scholars have 
also shown that a fief was further distinguished from a benefice by 
the fact that the fief was granted as a hereditary tenure, whereas 
the benefice was only granted for life 46 . 

The origin of the fief and its relationship to the benefice have not 
been interpreted altogether correctly by scholars in my opinion. 
Although the term feodum = fief appears in document only from the 
tenth century on it is unquestionably of ancient origin. Moreover 
the Roman and Germanic antecedents of feudal concepts and termin- 
ology must also be distinguished on analysis. The fief like the allod, 
is an inheritance from Germanic society. The equivalent Roman 
social concepts were the latifundium and the beneficium. Allods, 
as ancestral holdings, i.e. hereditary lands free from all obligations. 
They went back genetically to lands once held in joint ownership by a 
community, but which had passed after the dissolution of the com- 
munity to its chieftain or to its more powerful members who sub- 
sequently turned into an aristocracy. The feodum, in opposition to 
the allod, in all likelihood originally designated land liable to taxation 
or subject to certain obligations. Whereas the allod, because of its 
lexical composition (from all. (i all " and od, " possession ") indicates 
total possession, or more accurately the possession of the whole clan 
or community; feodum -fief derived from feih and od (where fieh = 
German vieh, from the same root as Latin ypec-us and with the same 
double sense of pecus and jpecunia), characterized a particular category 
of land according to the revenue furnished by it. In this sense, 
feodum could designate both a censive tenure [tenure a cens] or a benefice, 
depending on the nature of the obligations laid upon it. This fact 
would explain why in feudal times, when the fief had acquired the 
sense of hereditary benefice, the term fief also designated censive 
tenures (terres roturieres), or peasant holdings in certain regions of 
France 47 , From the beginning, hereditary tenure was inherent 
in the fief, and from the tenth century we also meet the expression 
perpetuum beneficium, which indicates that the evolution of the benefi- 
cia also tended toward hereditary tenure. It is very likely that this 
circumstance assisted the fusion of the benefice and the fief and the 
disappearance of the former. 



THE FEUDAL BASES OF THE NAXARAB SYSTEM 347 

The characteristic feature of feudalism is the fact that territorial 
relationships are closely tied to relationships between individuals, 
Land is not only a source of wealth, but also a means of ruling over 
men. The owner of free lands becomes a lord [seigneur], while the 
holder of a fief becomes a feodalis, or vassal, depending on a lord. 
The roots of this type of subordination are also traceable back to 
Germanic and Koman antecedents. Various types of patronage 
and clientele had existed among the Romans, particularly in the 
period of decline. Among the Germans, we find lords and his gassindi 
or vassi, i,e, a group of companions or a bodygard composed of men 
loyal to their lord. In exchange for their services to the lord these 
men enjoyed his protection, but their type of service differed from 
the one found in the feudal period. Service, even military, may 
also exist without territorial compensation, and such service can 
probably be called vassalage, but the basis of feudal vassalage was 
service, primarily military, in exchange for land. This service deter- 
mined by land tenure is the crucial element in the feudalization of 
personal relationships. Thus the allods and laiifundia provided the 
foundation for free and seigneurial land tenure, while the feodum and 
beneficium were the basis for feudal or conditional tenures ; sociologic- 
ally the suzeT&hx-seigneur was descended from the German lord and 
the Roman magnate, while the vassal was descended from the vassi 
and clientes. 

These basic elements of a feudal regime were clearly present in 
Armenia. The indispensible foundation of land tenure showed great 
similarities to the feudal pattern, not only in type but in origin 47a * 
The Armenians also distinguished two categories of lands: ^wjpkhfig 
[hayreniF] and ujujpqliwliwhg [pargeivakanF], i.e. ancestral posses- 
sions and benefices. In addition, we also find mentions of guwlpu- 
qjihg [k'sakaginF], or purchased estates, a type of lands also found 
in western feudal society after the twelfth century, but the main 
categories of land holdings both in the West and in Armenia were 
ancestral holdings and conditionally granted benefices 48 . 

The Merovingian allod corresponds exactly to the Armenian 
Cwjpkhfig [hayreniF], as a type of land tenure. Both are inherited 
lands transferred from one house to another on terms of outright 
possession. Zwjpkbfy [Hayrenik'] was the general term designating 
the whole of a princes hereditary possessions, each unit of which was 
called a mni.h [tun], " house ". These houses are depicted as lordships 



348 



CHAPTEK XV 



in the sources, and they make up a mipniPfiLU (terufiwn) or lord- 
ship; in the same manner the feudal dominium, or unit of allodial 
tenure lay at the foundation of the lordship or seigneurie 483 ; 

In the case of ujiiipqliwliiuhg [jpargewalcanF] lands, the term was 
perhaps used by historians to designate exclusively small holdings, 
such as hamlets or villages, bestowed on a prince as reward for some 
service, but in earlier times, more extensive territories had also been 
subject to grant. According to the Arsacid historian whom we 
have already cited, many of the princes became the rulers of great 
provinces, countries and important villages thanks to the favour of 
the Arsacid kings. These words are corroborated by the concrete 
examples cited in historical literature 48b , 

We have already presented the hypothesis that a nahang had the 
same relation to granted lands as a tun to clan or patrimonial lands, 
but that it was at the same time an administrative unit, A naxarar 
held a nahang just as a ter, or tanuter, ruled over a tun, and the tenure 
itself was called naxararutiwn* There were no essential differences 
between naxarar&oms and tanuter&oms; both weTe 4erriW£al-pohtieal 
xinits similar to the seigneurie. Moreover, the naxaramtfiwm can be 
equated with western comtes and duches whieh had grown out of 
former Roman provinces, while the tanuterutfiwns were the equivalents 
of allodml lordships (alleux souverains) 48 *. 

The evolution of the allods took place in two directions. Some 
gradually extended their boundaries and became such important 
territories powers that their lords easily obtained political rights 
through so-called immunities. The crown conceded to them judiciary 
power, taxation, and other branches of public jurisdiction within 
the limits of their allod ; at the same time, they were often invested 
with administrative functions such as the office of count, The rights 
and privileges so acquired generally became hereditary; once it had 
entered upon a political course, the allod tended to become a seigneurie, 
and indeed all important allods eventually became lordships. Other, 
and very numerous, allods were unable to expend either territorially 
or politically ; on the contrary, they were forced through circumstances 
to give up their privileged character of free inherited lands, and came 
closer to the category of benefices, or of the conditional service- 
holdings known as fiefs. The owners of such former allods became 
the vassals oifeodales of more powerful lords, 

A similar process can also be traced in Armenian land holding 



THE FEUDAL BASES OE THE NAXABAB SYSTEM 



349 



relationships ; but first, how should the relationship of the Armenian 
princely estates to the crown be characterized from the feudal point 
of view ? According to the indications given by Armenian sources, 
the princes — nayarar^ or tanuters, — were the bwnwjg [cafayF] 
or " servants " of the king; their relation to him was called bumwjin.- 
Pftth [cafayufiwn u service "] to them the king was a mtp 
per], or lord. The terms servant, and service correspond to servitium 
and obsequium, the terms expressing the obligation of a vassal to 
serve his lord, and the vassal himself was nothing more than a servant, 
famulus 49 , The obligation to serve may have been founded on fides 
or on the fief; the Armenian sources do not present this matter very 
clearly, On the one hand, the princes are shown as the absolute 
owners of their lands, with the same unalienable rights as the king. 
On the other, there are relatively common cases where the king either 
confiscates the lands of a prince for the benefit of the treasury, or on the 
contrary, presents a prince with new lands. The right of confiscation 
and grant affected the dominium utile which characterized lord - vassal 
relationships, The confiscation of princely estates under the Arsacids 
may be equated to the right of forfaiture claimed by the feudal lord 
against his guilty vassals, On the whole, conditions in Armenia 
were marked by instability, which is a feature characteristic of a 
feudal society. 

The Armenian form of princely land tenure is perhaps best defined 
as the type known to historians of feudalism as fief- seignewie. 
Historians distinguish true fiefs from 1 this type, which may be classified 
as a lordship from the point of view of its legal status, although it 
is called a fief 50 . These two categories of fiefs were founded on two 
opposite principles of feudal land tenure; on the one hand, the fief 
was seen as a conditional land-hold, subject to confiscation in case 
of a breach of the oath of service by the holder ; on the other, the feudal 
tenant who no longer wished to serve a certain lord, moved with his 
land into the service of another lord by means of the so-called commen- 
datio. Such transfers indicate that the individual was acknowledged 
to be the absolute master of the land, and could serve whomsoever 
he wished. In such cases, then, the personal dependence of the vasals 
on his lord prevailed over the dependence of the land. This type of 
lordship is especially noticeable in the initial period of feudalism. 
In the opinion of a recent historian, the territorial sovereignty of 
counts or princes extended only as far as their own domain. Outside 



350 



CHAPTER XV 



this domain, their authority expressed itself not so much with regard 
to land, as over the persons bound to them by the vassal oath of 
allegiance 51 . The relations of the king to the Armenian princes may 
be defined in similar fashion. They too were characterized by personal 
rather than territorial dependence. The king might deprive a prince 
of his possessions for a particular offense, but this punishment did 
not extend to his descendents ; the nearest heir of the guilty prince 
assumed his rights, and the confiscated lands returned to him as the 
unalienable possession of his house. In reality, in Armenia as in all 
of feudal society, social relations were governed not so much by legal 
regulations as by concrete force, and the relationships entered upon 
varied a great deal as a consequence. In the West, some lordships 
were true states in which the king was the head of allied princes, other 
lordships enjoyed the same rights to a lesser degree but were still 
considered sovereign units, and finally there were fiefs of various 
status, some with the prerequisite of service, others with the right of 
commendatio, etc. Similarly in Armenia, the marcher lords or bde$x^ 
were seen as lesser kings, the senior tanuters possessed de facto rights 
which were incomparably greater than those of their juniors, and there 
were princes, who occupied very subordinate positions as well. 

According to the generally accepted pattern, the feudal system 
consisted of three links of personal and territorial dependence: 1) the 
lord and his domain, i.e. the part of his lordship which he held directly, 
2) the group of his vassals and their fiefs, ix* the lands granted to 
them from the lordship, 3) the common people and their lands called 
censive tenures. Formerly, the fief, that is to say the territorial 
dependence, was thought to have been the fundamental feature of 
feudalism; at present, however, historians give greater importance 
to the commendatio of the individual holding the land, and they 
shift the center of gravity of the feudal system to the seigneurie. 
But, in fact, the commendatio did not contradict the principle of 
conditional tenure; it only preserved the fief from enslavement. 
The mark of conditional tenure stayed on the fief, since its holder 
was not freed from the obligation of serving by passing to another 
lord. Every schematization of feudal relationships, which attempts 
to establish distinctions between the important and unimportant 
aspects, is in some sense an abstraction and a hypothetical structure. 
In reality, these relationships presented themselves in a variety of 
forms and combinations. For purposes of historical investigation, 



THE FEUDAL BASES OF THE NAXABAB SYSTEM 351 

it is more fruitful to trace this phenomenon in its concrete forms than 
to dwell on the degree of comparative importance of this or that 
feature. We should be on the right path if we divide those legal 
relations in the feudal system, which correspond to the two classes of 
persons, lords and vassals, found in feudal society, into two categories; 
seigneurial and feudal, in the strict sense of that term; and if we 
then study each category separately. 

Arsacid Armenia as a whole may be taken as a single large lordship 
in which the lord was the Arsacid ruler, the vassals were the princes 
acknowledging his authority, and the remainder of the population 
was made up of the common people or peasants. It is evident from 
the relations of the wa^arardoms to the crown that each of them could 
be equated in some degree to an ordinary lordship. Like the French 
crown, which was above all a lordship of superior rank, the Arsacid 
kingdom formed a single we^arardom, while the wa^arardoms, in 
their turn, tried to turn into miniature kingdoms of the Arsacid type. 
For an investigation of lordship and vassalage in Armenian society 
it is important to note the presence of the characteristic features of 
na^arardom, whether these features appear as attributes of royal 
or of strictly naxarar power. Because Armenian life was subject 
to tradition and custom more than to law, it is vain to seek in it the 
legal codifications of relationships found in western feudalism; the 
forms expressed through legal aspects in the West appear in Armenia 
under the guise of custom. This difference does not prevent us from 
comparing the two systems, since feudal legislation itself was merely 
the sanction of an existing order, which had been worked out in 
practice. 

The institution called seigneurie in France, manor in England, and 
Ghundherrschaft in Germany, was designated in Armenia by the terms 
uitprnPfiLh (terufiwn) and wwhnLin£pnLpfiL.h (tanuterufiwn), or hwfvwpmpnL- 
Pjilu{ na^afarutHwn) and hw£wuikinnLpfiLh (nahapetutfiwn). The lord 
of such a holding was called correspondingly, wip[ ier] and wwhrnmip 
[tanuier], or huifuwpwp [naxamr] and hm( % mu^hm\naha , pei\. In the 
historical sources that have come down to us, these terms alternate 
as absolutely synonymous concepts; the only variation being that 
some authors prefer one of them, and others another 52 . 

Terutiwn and naxararutfiwn were the political formulae correspond- 
ing to the territorial concepts expressed by the terms wnrfj [tun], 
m^fumpi \asx^rh"\, qwnun [gawaf], etc. The central core of such 



352 



CHAPTER XV 



territorial units was the nuwwb \ostan\, or " court " (lit, " threshold ") 
which corresponded to the seignorial curtis dominicalis m the West; 
another name for it was ip£ qwinjhg [gah, gahoynJtf], " place, 
throne", in the sense of residence 53 , The lordship, par excellence 
was held by the Arsacids, Just as the head of the Church was acknow- 
ledged as kat'olikos of all Armenia but also as bishop of Ayrarat, 
so the Arsaeid rulers were simultaneously Mugs of Armenia and 
princes of Ayrarat, To distinguish it from other princely holdings, 
the domain of the Arsacids was called the royal osian ox " house " 
(wpgm&ft mm.h wpgm&fi numwh} 53a , 

We can see from its relationship to the royal power that the tanu- 
ierutHwn was not simply an estate but a lordship, i.e. a territory 
possessing sovereignty. It is well known that the power of tanuter 
was hexeditaxy in a given house and passed from one of its membexs 
to anothex according to seniority, as was the case in feudal lordships 54 , 
Even so, the sovereign interfered in questions of naymwt inheritance; 
first the native kings, in the period of the Arsacids, and, after their 
downfall, the Persians, watched continuously to see who inherited 
the power of the tanuters. Historians even claim that the king ap- 
pointed the tanuters or that they granted the tanuters their position 55 , 
Koyal interference seems unnecessary where a strict pattern of suc- 
cession already exists, and there is no evidence that the kings acted 
against the rights of seniority, Hence, the appointments or grants 
of tanuter positions mentioned by the historians must evidently be 
understood in the sense that the king acknowledged and confirmed 
the rights of the legal heirs. Had the tanuterdoms been essentially 
private holdings or estates, the confirmation of the king would not 
have been needed. The tanuter performed important duties, and 
the purpose of the royal confirmation of the ianuier's authority was 
to insure his obligation to sexve the rulex as his vassal 55 *. 

In the feudal world, a vassal's tenurial right ended with his death ; 
the lord had to revive it in order to transfer it to another vassal, and 
usually, in his own interest, he left it to the dead man's son, In 
Armenia, royal recognition of the rights of the heir of a tanuter was 
not a general custom ; it was stressed only in particularly important 
cases. This grant of recognition had the same significance as the 
installation or investiture customary in the West, The symbols of 
the investiture in the West are well known ; they consisted of a green 
branch or a piece of sod representing the land, together with a hat, 



THE FEUDAL BASES OP THE NAXAMAM SYSTEM 



353 



a ring, a standard, etc, depending on the customs of each locality, 
A similar practice can also be observed in naxarar society. We know 
that each ianuier had a ring which served simultaneously as his seal, 
and each princely clan had its own standard; the hat was replaced 
by a diadem (^iww/u., faiiw), which was part of the headdress. We 
know of cases where the diadem and ring were given at the time of 
entery into the ianuterufiwn, a significant gesture giving to these 
objects the character of an investiture 56 . The standard (fl/mij 
draws) undoubtedly had the same significance 5 7 , We had occasion earlier 
to refer to the branch or wreath with which the Persian king greeted 
the Armenian princes, though we do not know what they signified 58 , 
In the West, the lord gave his vassal a green branch or piece of sod 
as a symbol when bestowing his lands upon him, and in all likelihood, 
the gift of a branch or wreath had the same symbolic meaning among 
the Armenians, Finally, the seignorial character of the Armenian 
ianuierutiwn is shown by the fact that the term mipnLpjjLh [terutf- 
iwn\, the older and more common name for the authority of the 
ianuier, eventually acquired the meaning of state, Moreover, the 
lands of a ianuier were called [asxarh, gawaf], that is to say they were 
designated by terms likewise having connotations of statehood. 

The above discussion gives a basis for considering the ianuterufiwn 
as a politico-social institution in the general spirit of the feudal sei- 
gneurie of the Middle Ages, but there is little evidence on the manner 
in which the rights of the ianuiers were more closely defined. In the 
West, the lord's power included the most "important aspects of life, 
He had the right to legislate, droit de ban, and to issue decrees, consti- 
iutiones or assisiae, which were binding within the limits of the lands 
directly subject to him, If, however, they affected his entire lordship, 
including the fiefs, the agreement of the vassals was required and a 
curia flenaria was summoned to the lord's court for that purpose 59 . 
In Armenia, the right of legislation was inherent in the royal power, 
and the naxarars and ianuiers unquestionably also enjoyed this right 
to a greater or lesser degree. If references to legislative acts are rarely 
found in the documents that have reached us, this can be explained 
in part by the fact that Armenian life was governed primarily by the 
rigid tradition of the past and consequently had no pressing need 
for legislation, In such a society, usages and customs (pwpg k 
vmlnprnpiiLh^ bar¥ ew sovorufiwnF) rule in preference to laws 60 , 

Eaeh naxarar, like every feudal lord, was judge within his own 



354 



CHAPTER XV 



domain 61 , Powerful princes in the West passed judgement over 
their vassals, in their quality of seigneurs justiciers. The Armenian 
king, as the supreme lord, had the same jurisdiction over the princes 
subject to him 6a . A medieval lord supervised administration together 
with the council of his vassals. He also had various minisieria, or 
instruments of government whom he appointed in a manner similar 
to that of the king, and the courts of the various counts and dukes 
likewise reflected the protocol of the royal court; they had their 
senechal, chambrier, connetable, bouteiller, and chancelier, who served 
at court and simultaneously performed the administrative functions 
they had received as fief-offices. The system of administration 
current under the Armenian Arsacids was relatively complicated. 
Offices were known as qnpb [gore], " function, work " ; an individual 
invested with such an office was called qnpbwIjuiL [gorcahal], 
" officer ", whence we find ^npbw^wipLpjiLij [gorcaJcalutHwn] = 
*' ministerium, office ". The corresponding Persian form was ^puifilimp 
[k'rtikar], Jcarta-Mra = ^pb^m^ [gorc-hal]. More than nine hun- 
dred gorcakals were said to have existed at the court of the Arsacid 
kings, but this figure drawn from Faustus' History, is unquestionably 
an exaggeration 63 , 

We have already studied the more important offices which have 
much in common with the leading offices (grands offices) found at the 
court of great lords and kings in the feudal period. In Armenia, 
as in the West, the chief offices were held as hereditary possessions 
by the more powerful houses. Foremost among the leading officials 
were the senechal and the chancelier; the former was considered to 
be the most important personage in the lordship after its lord, and his 
importance was based on the fact that he was the commander of 
the army. The chancelier, on the other hand, had a court title, 
but he simultaneously administered the revenues of the state, and 
derived his influence from this fact 64 . The functions of the senechal 
and of the chancelier were essentially equivalent to those of their 
Armenian counterparts, the sfarafei and the hazarafet. In Armenia, 
as in any feudal state, land subject to the king was held partly by 
princes, who ruled over estates, and partly worked by peasants. 
Two offices were needed to correspond to this division, one " [i ifipwj 
fefvwhnLpkwhh [over the princes] ", and the other " ft ftpiuj 
^huilimhntphmhb [over the peasants] ", The obligation of the 
princes was to render military aid, i.e. to furnish knights; for 



THE FEUDAL BASES OF THE NAXABAB SYSTEM 355 

this reason the official placed over them was the sparapet, who was 
the commander of the army. The hazarajpet, on the other hand, was 
the supervisor of the affairs of the taxable population, that is to say 
of the peasants 64a . We have no information about the internal organi- 
zation of Armenian peasant communities ; we know only that they 
had headmen called qmuwuihm^ [dasapetk'] or qhtigwuuq [gelja- 
wag], who may be equated with the doyens = decani, and maires 
— maiores of France 65 , Similarly we know very little of the internal 
administration of the principalities. Like the king, their rulers 
probably had administrative officials commensurate with the size 
of their possessions 66 , 

We have already distinguished seignorial duties from those of 
feudatories or vassals. Feudal obligations in the latter sense took 
their point of departure from hommage et fidelite — hominium et 
fidelitas ; since the dependence of the holder of the land, or feudatory, 
on the owner of that land, or lord, the so-called engagement de vassalage, 
was created by this means. Hence, the obligations of the vassal, 
made possible in practice by land tenure, were derived from concepts 
of fealty and homage, The form taken by these obligations were 
primarily auxilium, consilium, and participation in the administration 
of the lordship. The basic aspect of feudal obligations can also be 
traced in nayarar society. Hominium (from homo) consisted in the 
act by which a given individual acknowledged himself to be his lord's 
man (homo), and confirmed this by an oath ; the Armenian engagement 
or vow (m-fum, u%t), by which the princes swore to the king that they 
would be his faithful servants, had the same significance. This 
engagement was in turn accompanied by an oath whose symbolic 
expression was the offering of salt 67 ; as a result of this oath, one side 
took upon itself the duties of lordship and protection, and the other 
those of faithful service and obedience 6S . 

The military obligations of a vassal consisted of three types of 
duties: service in wartime (service d'ost et chevauchee), defense of his 
lord's castle (estage), and surrender of his own castle at the demand 
of the lord 69 . Military service was likewise obligatory in nayarar 
society. Just like western-European vassals, the Armenian princes 
had to appear together with their contingent of knights at the summons 
of the king in order to participate in military undertakings, either 
campaigns (hostis, uiuiinkput^ — ■ faierazm), or raids (chevauchees = 
cavalcata, lmiujwinwlj — aspatah). The listing of na^arar cavalry 



356 



CHAPTER XV 



contingents in tlte Military List discussed earlier, gives us some idea 
of the military strength of the naxarar&oms. In the West, barons 
of the first rank furnished a few hundred knights to their suzerain ; 
a senior fief furnished only one knight, and frequently several fiefs 
joined together to furnish a single armed man 70 , According to 
the Military IAsi, the Armenian princes were divided into several 
categories for this purpose, hut according to this List no prince furnish- 
ed fewer then fifty knights 71 , 

The obligation to defend the lord's fortresses was the duty of a 
single Armenian family, and formed its hereditary prerogative on a 
par with other hereditary offices, The defense of the royal fortresses 
in Cop'k 5 and Angeltun was entrusted to the prince Mardpet 72 . 
Among the Persians, who considered this office very honourable, 
this office was held by a member of the royal family, who was called 
the Argabadh (the ancient arJcajpaies) 73 . Each principality contained 
its own castles which were, however, at the disposal of the Arsacid 
kings; the prince's castles were jurabilia et reddibilia to them, as was 
required by feudal custom in the West, After the disappearance of 
the Armenian kingship, the Persians occupied many fortresses in 
various parts of the country and kept permanent garrisons in them 74 , 
We presume that these strongholds came to them not by way of 
conquest, but as an inheritance from the Arsaeids, 

A duty similar to the feudal consilium also existed among Armenian 
princes, and the Armenian nobility unquestionably shared in the 
administration of the country, since a large body of officials was 
recruited exclusively from among of the princes. Moreover, the 
princes frequently assembled at the king's court to discuss with him 
various questions of internal or external policy; in important cases 
the king summoned them himself, while on some occasions they came 
of their own accord 75 , The rule that the mightier princes, those 
having more than a thousand knights, should reside continuously 
at the royal court and ride forth with the king, was attributed to 
Trdat Ill's son king Xosrov II, but in fact, Cyrus had already obliged 
those who possessed lands and power to spend some time at the court 
of the satrap and to have their children educated there 76 , This 
custom was apparently traditional in Armenia as well. 

One of the characteristics of the fief was its tendency toward hier- 
archy. With the passage of time, a fief might grow territorially 
and politically; the holder would then parcel out estates from his 



THE FEUDAL BASES OP THE NAXARAB SYSTEM 



357 



lands and grant them to Ms friends on a basis of vassalage; thus, 
while remaining the vassal of his ownjord, he became simultaneously 
a minor lord in his own right, A subinfeudation of lands and indivi- 
duals had taken place, and the new estates and vassals were considered 
to be sub-fiefs and sub-vassals (ameres-fiefs, ameres-vasscmx) of the 
original higher lord, A similar phenomenon can be observed in the 
Armenian feudal system. Here the division within fiefs took place 
inside the limits of the clan and as a result of the disintegration of 
seniority rights. 

The inalienability and indivisibility of the fief through a system 
of single inheritance was characteristic of the feudal world and served 
directly the interests of the lord, since the obligations of vassalage 
were better guaranteed under such circumstances. If the lands 
were to be divided equally among the heirs, the vassals would soon 
loose the minimum territorial endowment indispensable for the 
performance of their duties. Consequently, the lords stood for unity 
in the fief, and for the rights of seniority in inheritance. Little by 
little, however, equahty of rights fought its way to the fore, At 
first other sons were allowed to inherit as a favour, but the best part 
of the land was preserved for the eldest, and he was considered res- 
ponsible for himself and his brothers ; he alone did homage and received 
the investiture, but in the end, the principle of equal inheritance 
gained the upper hand. Hereditary tenure and seniority were funda- 
mental features of Armenian f eudalism. The so-called rights of sepuhs 
grew only in proportion to the weakening of this basic aspect 76a , 
As we have already seen, sepuh had originally meant a " son of the 
clan ", and the word was used in the na-^arar period as a technical 
term to designate any member of a princely house. With the exception 
of the tanuter, or head of the clan, all its other members were considered 
to be sepuhs. In the sources, the tanuier's brothers, sons, nephews 
and kinsmen in general are designated in this fashion without any 
distinction 77 , As long as the system of single inheritance continued 
to function, the power of the tanuter passed to the senior sefuh, the 
rikh [mec] or wLwq uku[nU [awag sejpuJi], With the passage of 
time, however, the principle of the indivisibility of princely lands 
was undermined, and the sejpuhs demanded their share. A „new 
sysleia.of.divide^^oldings, according to which the lands were parcelled 
out among the sepuh heirs, appeared, with the result that a new rank 
in land tenure as weH as in the nobility came into being, The portion 



358 



CHAPTER XV 



of land alloted to a sepuh was named sephakan, and the sepuh class 
of free men was referred to as the sepuh nobility [uhu^ml^m'u 
mqiumnLpjiJ} y sephahan azaiutHwn) 173 '. 

The process by which the sepuhs were transformed into an 
independent group cannot be traced. Some indications of the in- 
dependence of sepuhs can already be seen in the historians relating 
the events of the fifth century. In all of them the sepuhs appear 
next to their tanuier, as seemingly free agents. At the time of the 
Armenian uprising, the sepuhs were invited for negotiations at the 
Persian court along with the tanuters, During the rebellion, decrees 
were signed not only by the tanuters but also by the sepuhs, and, 
in general, on all important occasions, the sepuhs appeared inseparably 
from the tanuters 78 . The first manifestations of sepuh rights appear 
in a period following the fall of the Arsacids. Xorenaci asserts that 
according to the custom of the Arsacid house, only the heir remained 
at court, while the other children and kinsmen of the king lived else- 
where, in districts especially set aside for them, first in HasteanF, 
and later in Aliovit and Arberani, In time they found themselves 
crowded in these districts, and petitioned the king to have their 
possessions enlarged. But the king refused the request and decided, 
cc not to set aside new estates for them, but to divide equally among 
them the lands in their possession ", A new division of the lands, 
conforming to the number of individuals, and the Arsacid princes 
were evenly distributed in the districts indicated 79 , 

The more credible and valuable part of the story lies in its indication 
that the custom of dividing land according to the number of heirs 
was common among the Arsacids. Since the royal house was internally 
held together on the same basis as the families of the princes, this 
custom of dividing lands, noted by the historian, should hold for 
princely houses in general, This phenomenon does not belong, 
however, to the period of the Arsacids in which it has been set by the 
historian. We have in his account the reflection of a practice con- 
temporary with himself, and which he has set back into earlier times. 
The division of the inherited domain among the members of a family, 
is a manifestation of the new -sepuh system, which apparently made 
its appearence in the sixth century, since it was quite clearly developed 
by the time of the Council of Dwin of 641 80 , Private quarrels over 
the division of property resulted from the sepuh system, and were 
duly recorded by Zorenaci, in whose time a rich documentation on 



THE FEUDAL BASES OE THE NAXABAE SYSTEM 359 

litigations, both in Persian and in Greek, was still available. The 
historian made use of these documents and he confirms the fact these 
acts dealt for the most part with the inheritance rights of sepuhs 81 , 
Subsequently, the terms sepuh and sephalcan, acquired the meaning 
of private property. At the present time, there is no term for private 
property in Armenian other than u^i^iw^ui&iji^riji [sephaha- 
nutHwri]. In ancient documents this word is also used as a synonym 
for dwnwhq, dwnuihqnLpfiLh [zafang, mfangufiwn\ " heir, inheri- 
tance ". The second meaning of the word is easily derived 
from sepuh, as a son and heir, and it apparently also served as an 
intermediary step to the sense of private property 82 . The sepuh 
system of inheritance undermined the foundations of naxarardom, 
through its replacement of the rule of seniority. As a result of the 
division and redistribution of lands, naxarar holdings necessarily gTew 
smaller. Inevitably artificial means were sometimes taken against 
the disintegration of princely houses; thus, for example, marriages 
of near relatives were allowed to preserve the unity of the land 83 . 

The status of a sepuh separated from his clan is unclear. We do 
not know the relation of a sepuh who had left his clan taking his terri- 
torial portion with him to his former ianuier. As the holder of a 
relatively small landed estate, he could not aspire to an independent 
position. According to some indications, such an isolated sepuh 
sought the protection of a more powerful prince and bound himself 
to him through an obligation of service. In the ninth century a 
man of sepuh rank served Prince Asot as his bodygard, and there were 
other sepuhs around Asot performing various other duties 84 ; sepuhs 
in such positions should be classed as vassals. As owners of an 
independent territorial property, they were free to serve their own 
ianuier or to seek another, on the same pattern as holders of allods, 
The development of the sepuh class was halted by political conditions 
which brought about the concentration of feudal forces and the forma- 
tion of several large units which engulfed the smaller proprietors. 

Another distinct category of free men is recorded in the sources; 
these are the so-called men of the ostan, or ostanik\ [nuwwhfig] 85 , 
Ostan, as we have already seen, meant " court " ; the osianiF were 
" court people, courtiers ": " ostan freedom ", corresponded as a 
class to that of the Eussian nobility [dvorianstvo, from dvor, " court "]. 
The sense of osianih\ however, was equivalent to mp^nLiifj [ark'mi], 
" royal ", a term by which it was often replaced or explained. Each 



360 



CHAPTER XV 



princely court was called its ostan, but the term was used primarily 
to designate the royal court as the most important one among them, 
and it sometimes referred to the entire royal domain, Osian freedom, 
as is implied by the name itself, was closely connected with the royal 
court. In my opinion, the ostaniF had the same relation to the 
royal house as the seypuhs to the naxarars 85 a . 

The problem of the origin of the ostan freedom and of its nature 
already attracted the attention of X orenaei, but the historian appar- 
ently had no more information on this matter is available to us, He 
came to the conclusion that the ostaniFs were descended from the 
four regiments supposedly created by king Valarsak out of the descen- 
dents of the ancient Armenian kings of the house of Hayk in order 
to protect the royal court. Subsequently, according to the same 
historian, the Persians formed regiments composed of other men and 
called these ostan, Movses does not know the reason for this trans- 
formation, whether the former clans had died out, or whether they 
had been dismissed for some act of disobedience so that other regiments 
called wpgnibfi [arJc'uni] had to replace them; he merely insists 
upon the fact that the former regiments were descended from the 
house of the first kings, like those which in his time were called seyfecul 
by the Georgians 86 , Xorenaci's conjectures aro undoubtedly based 
on suggestions found in the Histories of Lazar P'arpeci and EUse, 
which we have already cited, Having noted that the terms nummh 
[ostan] and wpgiuhlt [arFuni] alternated in them, Zorenaci postul- 
ated correctly the identity of the two terms and suggested hypothetic- 
ally that the first term was older, $\e, Arsacid, while the second belonged 
to the Sasanian period, That ostan regiments did indeed exist is 
evident from the historians who speak of " ^jp^l 1 " or " qnpu 
jwp^uLhfj mwh£" [the cavalry or forces of the royal house] S6a , 
but that these regiments were intended especially for the defense 
of the court, or that they were four in number, are mere guesses of 
the historian inspired by the information he had about the prince- 
bdesx& who held the first rank at the Arsacid court 87 , The authentic 
part of the historian's explanation is his statement that the ostaniF 
were men of royal origin ; but we must take this in the sense of descen- 
dents of the Arsacids and not of the earlier kings, as Zorenaci believed. 
The ostaniF were Arsacid sepuhs, and the historian is right when he 
equated them with the sep'e&uls in Georgia, There is also no reason 
to doubt the correction of his assertion that the composition of the 



THE FEUDAL BASES OE THE NAXABAE SYSTEM 



361 



ostaniF in his own time was not the same as before. The ranks of 
the ostaniF must have included not only Arsacid descendants but 
also men of other origin who had distinguished themselves in some 
fashion. Among them were the minor princely houses found within 
the Arsacid domain, i.e. the province of Ayrarat. Perhaps this is 
the reason for which this class of free men began to be designated 
by the uncharacteristic name ostaniF (men of the ostan) rather than 
sepuh, the term probably used for them until the change in their 
make up, 

The documents speak of ostaniF only in the royal province (jwpgnLhft 
wwh, [in the royal court or domain]), meaning Ayrarat. In speaking 
of the origin of ^a^arardoms, Jforenaci notes that men were raised 
to the rank of free men for outstanding service to the king 88 , Such 
cases became more common after the fall of the Arsacids, and these 
are the means by which the Persians prepared a circle loyal to their 
interests. Consequently, Zorenaci attributes the change in compo- 
sition of the ostaniF to the Persians, The Armenian Arsacids had 
also rewarded persons who had rendered important services with 
landed estates in their own domain of Ayrarat. The enormous staff 
of officials serving at court consisted of men from the free class, and 
after the fall of the Arsacids, these must have passed into the ranks 
of the ostaniF, Thus, the so-called " ostan freedom " was composed 
of persons of three types: descendents of the Arsacids who had sepuh 
estates in Ayrarat, men having received grants of land there, and 
finally court officials after the abolition of the royal power in Armenia 88a . 



Ill 



There is very little information concerning the Armenian rural 
population and particularly its social status. The sources sharply 
distinguish the peasants from the rest of the population and contrast 
them with free men, but they say almost nothing about their relation 
to each other, or about the dependence of the peasants either on the 
land or on their lords. According to the sources, all those who did 
not belong to the classes of free men, nobility or clergy, formed a 
single mass of inferior population known under the name of ^bui- 
IfwhiiLpiiLh (sinalcanutfiwn), SinaJcan, in the sense of peasant, 
is the term most commonly used, but it alternated frequently with 



S62 



CHAPTER XV 



mm/Jili (ramih), qb^m.!} (geljuk), and the ecclesiastical term 
dinini[nLpq (zolovurd)* 9 . The hazarapei stood at the head of the 
peasant population, just as the sparapet was at the head of the princely 
class. We do not know whether the entire peasant population of the 
country was subject to the hazarapet, or whether he was the official 
supervising exclusively those peasants found in the royal domain. 
Although the first thesis seems to have better support, we are still 
not clear as to what determined the relations of the peasants to the 
princes and to the king, and what was meant by peasant land holds 89a . 

In western feudalism the status of peasants was known as servage, 
Legally the holding of a serf was considered as being merely a life 
tenure conditioned by the performance of certain services. On the 
death of the serf his portion returned to his lord to dispose of at will. 
The serf did not have the right of hereditary transmission, but was 
said to hold in mortemain insofar as the transmission of land to his 
heirs was concerned, though in practice, the lord usually transferred 
the allotment of a serf to his son. The serf was also personally subject 
to his lord ; he could not leave his land without his lord's permission, 
and this is the characteristic trait of serfdom. The legal status of 
the medieval peasantry was also reflected in its financial subjection 
to the lord ; the peasants were subject to taxation for the lord's benefit. 
The basic tax was the chevage = census capitis, a poll tax collected 
from all of the common people, Servile lands paid tenures a cens, 
or rent for the use of the land and tenures a champart = campi pars, 
or the share of the harvest designated for the lord's use. In addition 
to these there were the corvees or obligatory services of various kinds 
such as carrying, building, work on the lord's fields, etc. 89b . 

Among the Armenians, the peasant allotment was known under 
the following names: likwhg [JceanF], m p m p [arar], liwipmib [halo- 
wac], though the same terms were also used for the possessions of 
the princes and of the clergy. ^wifiLwd [Jcalowac] corresponds 
semantically to tenure, and Ijkwhg [keank*] literally means " life ". 
Might we, therefore, conclude on this basis that the legal nature of 
these allotments was a tenure and a life holding ? 90 

Definite information about the territorial obligations of the peasants 
has also failed to reach us, No evidence on this subject is found in 
Faustus 9I . We know only that the Armenians paid a tribute to 
the Persians in the period of the Marzpan&te ; such a tax had undoubt- 
edly also been paid to the Arsacid treasury. The Marzpan Vasak 



THE FEUDAL BASES OE THE NAXABAR SYSTEM 



363 



is quoted as having boasted, " I have before me all the taxes of the 
Armenians, and the Persian officials are in my hands, and much wealth 
I took from them " 92 . After the fall of the Arsacids, the Persians 
took a new census most likely to determine the amount of the taxes 
and to distribute them accurately. According to Elise, one of the 
results of the census was an increase in taxation designed, as he 
assumed, to ruin the rural population. The taxes were enormously 
heavy, 

where one hundred dahekans should have been levied, 
twice as much was taken ; even bishops and priests were taxed ; 
not only were the inhabited lands taxed, but even the ruins. 
Who, " exclaims the historian, 'is able to speak of the burden 
of taxes, of Smfig kt uml^lig^ fuudfig kt £wu]itj At £wplfwij levied from 
mountains, and fields, and forests' ? They did not take as 
becomes the royal dignity but plundered like brigands " 93 . 



The Armenian princes come before Yazdgard, claimed that Armenia 
was paying in this time more Snung [muth 6 ] and uw1{£ [sakfc], and 
other fopfe [harhF] than in his father's days. These statements 
are valuable in that they apparently contains the technical names 
of the various taxes, and it is possible to form some idea of them, 
in the absence of direct evidence, through a comparison with the 
tax system common in Persia, 

Two types of taxes existed in the Sasanian Empire: the hharddj, or 
land tax, and the gezitf [jizya], or head tax. The same taxes were 
also custommary among the Romans: the tributum soli and the tribu- 
tum capitis, or, according to the terminology of the reform of Diocletian, 
capitatio terrena [jugatio] and capitatio humana or plebeia [eajpitatio], 
Under the Arabs, two types of land tax were distinguished: the wazlfa 
and the muqasama* The first corresponded to the European cens, 
or rent for the land, and consisted in the payment of a given sum by 
each unit of land; the second was the European champart, or share 
of the harvest due to the landlord. The tax system of the Arabs was 
not created by them, but went back to the preceding period and was 
their inheritance from the Sasanians 93 a , Among the Arabs men younger 
than twenty and older than fifty years of age, the years unfit for 
military service, were free from taxation, and we know that the Per- 
sians performed military service between the ages of twenty and 
fifty 94 , This fact indicates the antiquity of the regulations inherited 



364 



CHAPTER XV 



by the Axabs from the Persians. The Arab wazlfa in fact corresponded 
to the kharddj, and the muqasama was called bahrah in Persian 95 . 

The types of taxes mentioned above: 5nLw, um^ pmd^, £,wu and 
faplf can be explained within this framework. Of these urnl} [sale] 
and p.wd [baz] are very familiar in Persian literature, and are charac- 
teristically often mentioned together: sak and baz, ox saw and baz 9f \ 
We consider this pairing characteristic because it shows that these 
terms expressed definite tax regulations. It is more than likely 
that sak and baz corresponded in content to kliaradj and jizya, i.e. that 
they designated respectively the land and head tax. An explanation 
for these synonymous expressions may be that some of the terms 
apply to an older period and others to a more recent one, or that some 
were customary in one part of the empire and others in another. 
The etymological content of the terms might have solved the problem 
had it not raised doubts. Baz, the Armenian pwd [baz] means 
" share ", as does the Greek SaajjLos (from Scuta, " to divide ") = 
" a share or a tax ", The tax collected by the Achaemenids from the 
subject nations was called bazi [baji]* In the Cyrojpaedia, SavfjLos is 
used to designate the tax collected by Cyrus, and is a translation 
of the Persian bdji 97 , The term [sak], the Persian sd or sdv, is 
unquestionably related to the Armenian word um\ [sak], " count ", 
uwli—wpliki [sak-arkel] 98 , but its Persian or even Aryan origin is 
doubtful. There are also grounds for relating the khardj (from the 
Persian *hardka) with the Armenian twplf [hark], Georgian bdf^o, 
thus linking them with the root from which is derived the Ar- 
menian &plihi [herk-el], " to plow ". Finally, gezitf corresponds 
letter for letter with the Armenian ^ymfi \ff^\ which means " a 
measure of wool, a fleece ", from the root 3^ [gzeT\, Semitic gizzd, 
<e to comb out waves ", whence comes gizzad. This word acquired 
the general meaning of tax, evidently because in some period gizzad's of 
wool had been paid in lieu of tribute 98a , Originally all taxes were paid 
in kind. The payment in wool is particularly suited to a pastoral 
mode of hfe, There is nothing surprising in this, since even in the 
middle ages the feudal lords sometimes still took several pounds of 
wax for clievage or census capitis. 

If the Iranian origin of sak were indisputable, it would be possible 
to state that sak and baz, and kJiarddj and gezitf were synonyms ex- 
pressing one and the same idea in the Aryan and Semitic cultural 
spheres, It is altogether possible that iim.mg [mutkf] and fau [has], 



THE FEUDAL EASES OE THE NAXAEAB SYSTEM 



365 



which the historians give side by side with sah and baz, served as 
the corresponding expressions on Armenian soil, In the same way, 
aimriLri [<piul] was the equivalent of bahrak. Baz is probably the land 
tax, and sak the head tax, In fact the same terms may have been 
used with different meaning depending on circumstances of time and 
place. So fox instance, kharadj among the Arabs indicated the land 
tax, while in Hebrew documents this is called iasqa, B,jxdkliaradj[keraga'] 
has the meaning of head tax ", For our purpose it is the tax system 
itself that matters and not its terminology. In addition to the taxes 
mentioned, a tax of the same nature as the European corvees = 
corrogata opera, " service ", was known and called p-hl^wp (bekar), 
This word is derived from the Persian expression belcar amadan, 
"to go to work ", and is used to this day among the people in the 
sense of corvee, In practice, the service was called Mr, Armenian 
l\nn [kof], " work ", and occasionally the two forms are given together, 
fan m pkfap [kof u bekar], but the term behar is the one found in the 
Canons of the Council of Dwin of 641 10 °, 

Thus, the main types of obligations were the kharadj, or land tax, 
the gezitf [jizya], or head tax, bahrak, or payments in kind, and bekar, 
services, A few indications exist as to their amounts. According 
to the information preserved in Tabari, the Persian land tax before 
Zusro I Anosarvan was collected in kind and corresponded to 1/3, 
1/4, 1/5, or 1/6 of the crop, depending on the fertility of the land or 
the quality of the harvest, In order to regulate the land taxes at 
the end of his reign, king Kavadh had intended to take a land-census 
of both plains and mountains, but he died soon after and left the 
completion of this task to his son JCusro I, Xusro carried out 
his father's intention and took an additional census of the taxable 
population, and at the end of this undertaking established regulations 
for the collection of taxes. The unit of land called ganb, and equal 
to 3,600 square ells, was taxed differently according to the nature 
of the crop: a garlb of wheat or barley land owed one dirrhem in taxes, 
a ganb of vines owed eight, a ganb of lucern, seven, etc 100a , 

All men between the ages of twenty and fifty were subject to the 
head tax except princes, courtiers, soldiers, priests and officials, 
From each man 12, 8, 6, or 4 dirrhems were collected according to 
his income. The last figure was apparently the one generally accepted, 
since Chinese sources mention that in Persia each individual paid 4 silver 
coins as a tax, It was also laid down that payments were to be made 



366 



CHAPTER XV 



in thirds, on three occasions or, as this was called in Persian, in three 
simarmks 101 . 

We know that in the Koman Empire a measure called jugum served 
similarly as the basis for the land tax. Each jugum was subject 
to a tax whose amount was regulated by imperial decree. The jugum 
varied in amount according to the nature of the land ; for instance, 
in Syria a jugum consisted of 5 jugera of vines, or 20 jugera of cultivated 
land of first quality, 40 jugera of medium quality, or 60 of poorest 
quality, etc. It is interesting that in the Empire taxes were collected 
on three dates: 1 September, 1 January, and 1 May 101a . 

All the taxes mentioned were also current in Armenian feudal 
society; the technical terms which have survived in the literature 
are our best evidence for this, but we have no specific information 
about the actual amounts of these taxes unless we take into consider- 
ation some of the evidence drawn from ecclesiastical life. The Ar- 
menian Church was feudal in structure ; it reproduced the social and 
economic regulations customary in nayamr society, and preserved 
thejn in part after the disappearance of the secular feudal nobility 102 . 
Consequently information taken from ecclesiastical life is also relevant 
for nayarar society. 

In the Canons of the Council of Dwin it is noted that certain princes 
oppress almshouses, demand from them taxes equal to those of living 
[lay] men, and send their men there to levy taxes even on items of 
food and drink, while on the contrary, " they should watch over them 
with care, granting them a share of all the produces of the threshing 
[£iwj, of the pressing [#}<W&], and of all produce " 103 . It is evident 
from this that a certain share of the rural produce went for the use of 
the feudal lord, Even more valuable is the decree of the Albanian 
Council under king Vacagan. The clergy decreed here the following 
taxes : 



This shall be the rule concerning the (fruits of the earth to 
be given) to the priests by the people. He who is rich shall 
give 4 bushels (griw [garib]) of wheat, 6 of barley, and 16 jugs 
of sweet (wine) ; the poor man shall give half a loaf of bread 
and as much wine as he can ; and nothing shall be taken from 

him who possesses no field or wineyard He who has 

sheep in his household shall give one sheep, three fleeces ( = 
gzafs) and one cheese; (he who has horses — one foal, and 
he who has cattle — • one calf) 104 . 



THE FEUDAL BASES OF THE NAXA8AJR SYSTEM 



367 



This official document provides a reasonably clear definition of the 
material and financial obligations of the population toward the spiritual 
and, of course, the secular authorities, The information that, in 
addition to the tillers of the soil, there was in Armenia at the time of 
this Council (i,e. no later than the sixth century) a group exclusively 
occupied with stock raising is important for a picture of the social 
structure of the country. Xorenaci was consequently not speakingly 
idlely when he complained that " not all part of Armenia were given 
over to agriculture, but here and there lived men who ate raw food 
and the like, as in northern lands " 105 , 

According to the information incorporated into the Law Code of 
M^iVar Gos, 

... from the plowed lands a fifth to be taken ,,, purchased 
fields, orchards and gardens shall be exempted from the fifth 
as are mills, houses, and shops. Persons occupied with crafts 
or trade shall pay a tax; the head tax shall not be collected 
from the peasants but only from foreigners conquered by force. 
Irrigated fields shall be subject to a fifth, and non-irrigated 
ones to a tenth. 



Limits were also set for required services and for taxes on household 
cattle 106 . These taxes are often mentioned in the inscriptions from 
Ani and we know that several dozen taxes of various types were 
enumerated and named on them 107 , This evidence, however, 
belongs to a period outside the Hmits of our immediate problem, 
although this needs not mean that the tax system given was a product 
of Arab rule. Muslim law undoubtedly brought about alterations 
in the naxarar system, and single regulations or the evaluation of 
taxable items might be affected, but the system of taxation as such 
was inherited by the Arabs from their predecessors. 

The form taken by nayamr relationships under the Arabs, and the 
degree of their influence are problems which concern scholars of the 
Bagratid period. Our own task has been to trace the nayamr system 
from its origin to the period when Persian rule, which had provided 
the foundation for its development, was replaced by that of the Arabs, 
As such, the period of Xorenaci marks the close of our investigation, 
and our intention has been to repeat the task performed by him in 
his own time. It is, therefore, interesting to compare our conclusions 
on a given problem with those reached by Xorenaci. This comparison 
will serve in part as a verification of our hypothesis. 



368 



CHAPTEK XV 



Although negative criticism has succeeded in undermining the 
authority of Xorenaci as a historian, it has in no way shaken the 
importance of his work as a historieo-literary source reflecting the 
atmosphere and attitude of the author's worid, From this point 
of view, the work of Xorenaci is of the greatest interest still. Indeed, 
what really matters in problems relating to the naxarar system is 
not this or that item of factual information or a particular explanation 
given by a historian, hut rather the value of that author's ideology ; 
the manner in which he understood the nature of naxarar society 
and the methods he used in solving the problems he raised about 
the naxarar system, 

Xorenaci could not free himself entirely from the contemporary 
scene. Every investigator of antiquity suffers to a greater or lesser 
degree from the weakness of looking at the past through the prism 
of the present, and many of the ideas of Zorenaci concerning the past 
of Armenia are echoes of conditions found in the period in which he 
lived, We can doubt the accuracy of his information and whether 
a particular naxarar&om arose in fact in this or that period or under 
these or those conditions, but we cannot doubt that one or another 
quality attributed by the historian to the naxarar system is correct, 
or that the particular method for the development of naxarar houses 
given by him was considered possible, at least in the period when 
he was writing. Our historian undertook a responsible task in the 
bitter knowledge that he possessed no reliable sources for the history 
of his native land, He was forced to work on the basis of disconnected 
and fragmentary data, of occasional references, of incomprehensible 
misstatements, and of similarly unpromising materials. He was forced 
to arrange odd facts and data in the light of the pattern of his own 
times, so that he involuntarily archaized many phenomena and 
situations drawn in reality from the life of a later epoch. JSTorenaci's 
pattern of thought moves from facts to ideas, from actuality to theory, 
Though on the contrary, he may select or even occasionally invent 
a suitable setting to illustrate a preconceived idea, piecing 
together more or less credible facts in the process. Often Xore- 
naci bases his conclusions on an etymological interpretation of geo- 
graphic or naxarar terminology ; this is the method on which he relies 
extensively for his analysis of naxarar genealogies. Irrespective 
of the degree of acceptability of these etymological interpretations 
(indeed the majority of them is not acceptable and many of Xorenaci's 



THE FEUDAL EASES OF THE NAXARAE SYSTEM 



369 



conclusions owe their dubious character primarily to doubtful ety- 
mologies) his method, as such, is unquestionably scholarly 107a , 

Out own investigations resemble those of X orenaci from a metho- 
dological point of view. Taking as a point of departure the existing 
conditions which are reflected in the sources which preceded the 
period of Xorenaci, we have attempted to penetrate into the depths 
of antiquity in order to seek in them the origin of the phenomena 
interesting us. At the same time the etymological analysis of the 
more characteristic terms of naxarar institution and terminology 
have been of substantial assistance to us, The results of our investi- 
gations have also been similar in a certain sense. For Zorenaci 
as for us, nayarm society was closely tied to the Arsacids, Zorenaci 
tells us on one occasion that Valarsak, the first of the Arsacids, found 
naxarar&oms already present in Armenia when he came to power 
and that he consequently asked his older brother Arsak, " whence 
originated the wa^arardoms existing here % " though on another 
occasion, the historian claims that Valarsak " established the naymm- 
doms and nahapets for each one of them " 108 , so that we might accuse 
convict him of inconsistency and contradiction on superficial acquain- 
tance. But when we penetrate to the heart of the matter, it beeomes 
clear that Jforenaei was attributing to the first of the Arsacids only 
the regularkation of the wxcwrdoms and not their creation in the 
strict sense of the word. 

Zorenaci set for himself the task of clarifying, " which of the ntiyaim 
clans had been related to us from ancient times, and which came in 
[from outside] and became related " 109 , Strictly speaking, this 
clarification was really the valuable result reached by the historian 
at the end of his investigation, but he presents it in his account as a 
problem to be examined. Our own analysis justifies his interpretation 
since it too has shown that the nax^Tar system did in fact consist 
first of native and later of foreign elements. The fundamental differ- 
ence lies in the fact that according to Xorenaci the great ^a^arardoms 
were descended from single individuals, while in our analysis they 
were derived from previously independent ethnic groups. According 
to Zbrenaei's version, some nobles, in fact the majority, had risen 
to the rank of %aya mr ^7 waY °^ court a:n( l administrative service. 
Such were: the Bagratids as coronants, the Gnt e uni as keepers of 
the royal wardrobe, the XoxoTftxm as bodyguards, the Varamuni 
as masters of the hunt, the Abelean as supervisors of the granaries, 



370 



CHAPTER XV 



the Gabelean as royal chamberlains, the Arcruni, as eagle [standard] 
bearers, the Gnuni, as butlers, the Spanduni as supervisors of the 
slaughterhouses, the Hawnuni as falconers, the descendants of the 
house of hAyr {i.e. the Mardypets) as eunuchs, etc. ... ; the princes of 
Siwnik', the princes of Kadme, and the princes of Gugark c , Angeltun, 
and Aljnik' guarded the frontiers and the Vahewuni were clerics. 
Some of the nobles were included among the nayarars thanks to the 
nobility of their stock or because of their descent from the progeny 
of Hayk or of the ancient kings; such were the Orduni, Apahuni, 
Manawazean, Bznuni, Afawenean, Zarahewanean, There were also 
noble foreigners such as the Amatuni, Arawalean, Kop'sean, Mami- 
konean, and Kamsarakan. Finally, a few men had obtained nayarar 
status through merit and personal valour, such had been the ancestors 
of the Dimak'sean and of the Truni. 

It is evident from the examples just cited that for Xorenaci two 
qualities were the bases of nayarar status: service and nobility i.e. 
superiority of blood. On this point too we have no disagreement 
with him, since the roots of the nayarar system found by us go back 
either to a tribal stage or to bases of an official administrative cha- 
racter. In Xorenacfs opinion, the aristocracy consisted of the more 
ancient families, primarily those presumably descended from Hayk, 
which were already present in Armenia at the time of the coming 
of the Arsacids. According to us, the aristocracy consisted of the 
houses which had developed through the disintegration of tribal 
relationships. Here, however, our agreement with Zorenaci is only 
one of principle; in actual cases, i.e. on which nayarar&oms belonged 
to which category, we are in disagreement. 

For JTorenaci, as for us, land tenure also provided the material 
basis of the nayarar system. Promotion to the rank of nayarar 
according to him was nothing more than a grant of lands no . Nobility 
and nayarar status were synonymous concepts for him for the very 
reason that all nobles possessed lands, and lands were granted in 
hereditary tenure. The expression mt^. ukpky " to lay the founda- 
tion of a clan ", which he used in speaking of someone's elevation 
to nayarar rank is significant in this connexion. If to grant lands 
was to create a new clan, then it is clear that these lands were given 
in hereditary tenure m . According to Zorenaci, the grant and 
confiscation of land was a prerogative of the sovereign power, which 
is a feature characteristic of feudal land tenure 112 . 



*mm 



THE FEUDAL BASES OF THE NAXABAR SYSTEM 



371 



The clan and its territory were both called naxararufiwn by Xore- 
naci, and this is the term lie commonly uses. Occasionally lie replaces 
it by im^miqkamLpijLh (nahajpetuiHwn), but it is evident from Ms expres- 
sion hwfuiupwpnLpfjLhg h bntjjih huijiwipiupjiLpbuihij biu£iuiqkmnLpfji.hg [" the 

naxarar&oms and the nahapei&oms of their naxarar&oms "], that he 
assumes a difference between the two terms: naJtapet designates the 
head of a naxarar&om. Ter and tanuter are likewise found in his work as 
synonyms for naTiapei, and also m£pm.piuSt [terutfiwn] and wwhm.uitpm.pfirf3 
[ianuierutfiwn] in the sense of naxararAom and nahapei&om 113 . 
Finally Zorenaci is thoroughly familiar with the sepuh system of 
inheritance and with the institution of ostan freedom at the very point 
in their development when they were turning into a separate new 
rank within the ieu&al-naxarar aristocracy, Thus, if we discount 
the factual setting and turn instead to a theoretical study of the 
naxarar system, we cannot deny that X orenaci was correct both in 
his methodological approach and in his setting of the problem. 

The points where our own interpretation of the origin of the naxarar- 
doms and of the nature of the system itself have coincided with X ore- 
naci's should be counted in favour of the accuracy of our conclusions. 
How far these are acceptable and what is their historical value, is 
open to differences of opinion, and we are far from any idea that 
anything beyond challenge can be created on the basis of the insuffi- 
cient and fragmentary material at our disposal. There is unquestion- 
ably much that is hypothetical and biased in our thesis, but these 
are defects which characterize all historical interpretations to a greater 
or lesser degree. What matters here is the general setting of the 
problem and the broad lines of its solution: have we been successful 
in uncovering the roots of the naxarar system, how accurately have 
we observed its subsequent development, and are the points of contact 
between the institution of wx&rardoms and western-European feudal- 
ism suggested by us acceptable? Our clarification of the naxarar 
system should bring a ray of light into the darkness which hangs over 
the Armenian past in the pre-Arsacid period, and should provide a 
starting point for a scholarly analysis of the extensive subsequent 
period of Armenian history which leads all the way to the destruction 
of the naxarar system in the period of the Mongol invasions 113 . 



ABKEYIATIONS 



AASS Ada Sanctorum Bollandiana (Brussels). 

AAWB Ablmndlungen der Ahademie der Wissenschaflen zu Berlin, 

AB Analecta Bollandiana (Brussels). 

ABAWM Ablmndlungen der bayerischen Ahademie der Wissenschaflen zu Munchen, 

AGO Ada Conciliorum Oecumenicorum, Schwartz, E, ed, (Berlin, 1914). 

AEHE Annuaire de VScoh des Hautes Etudes (Paris). 

AIPHO Annuaire de VInstiiiit de philologie el d'histoire orientales et slaves (Brussels). 

AJSLL American Journal of Semitic Languages and Literatures (Chicago). 

AKGWG Ablmndlungen der honiglischen Gesellschaft der Wissenschaflen zu Gbttingen. 

AO Ada Orientalia (Copenhagen). 

AQ Armenian Quarterly (New York). 

AE.BBL Academie Boyale de Belgique, Bulletin Classe des Lettres (Brussels). 

ASGW Ablmndlungen der sachsischen Gesellschaft der Wissenschaflen, 

B Byzantion (Brussels), 

BA Bulletin armenologique, Melanges de VUniversite de Saint-Joseph (Beirut). 

Ber Berytus (Beirut). 

BGA Bibliotheca geographorum arabicorum, de Goeje, M.J, ed, (Leiden), 

BIM Bulletin de VlnstUui Marr (Tbilisi), 

BK Bedi Karthlisa* Revue de Karthvelologie (Paris). 

BM Banber Matenadarani (Erevan). 

BNJ Byzaniinisch-neugriechische Jahrbucher (Berlin). 

BSL Bulletin de la Socieii Linguistigue de Paris, 

BSOAS Bulletin of the School of Oriental and African Studies (London). 

BZ Byzantinische Zeitschrift (Leipzig), 

Ca Caucasica (Leipzig). 

CAH Cambridge Ancient History, 

CHA Collection d'historiens armeniens, Brosset, M.P. ed. (St, Petersburg, 1874- 

1876), 

CHAMA Collection d'Mstoriens anciens et modernes de VArminie, Langlois, Y. ed, 

(Paris, 1967-1869), 

CHK The Catholic Historical Meview (Washington), 

CIG Corpus Inscriptionum Graecorum. 

CIL Corpus Inscriptionum Latinomm. 

CJC Corpus Juris Civilis, Mommsen, T,, Kxuger, P., et al., edd, (Berlin). 

CMH Cambridge Medieval History, 

Cod. Th. Codex Theodosianus, Mommsen, T., et al,, edd. (Berlin). 

CP Classical Philology (Chicago). 

CR Classical Beview (London- Oxford), 

CSCO Corpus Scriptorum Christianorum Orientalium (Louvain), 

CSHB Corpus Scriptorum Historiae Byzantinae (Bonn, 1828-1897). 

DHG Didionnaire d^Histoire et de GiograpMe JScclesiastique (Paris). 



ABBREVIATIONS 



XXIII 



DTC Dictionnaire de TMologie Oaiholique (Paris). 

EHR English Historical Beview (London). 

EI- Encyclopaedia of Islam (Leiden, 1913-1948). New edition (1 954-), 

EO Echos d'Orient (Paris). 

EGH Fragmenta Historicorum Graecorum, Muller, C. ed. (Paris, 1841-1883). 

G Georgica (London). 

G46 Office of Geography, Department of the Interior, Gazetteer No. 46 : Turkey 
(Washington, 1960). 

GGM Geographi Graeci Minores, Mutter, C. ed. (Paris, 1855-1861), 

HA Handes Amsorya (Vienna). 

IAFAN Izvestia Armianshogo Piliala Akademii Nauk SSSB (Erevan). 

IAN A Izvestiia AJcademii Nauk Armianskol SSB (Erevan). 

IANS Izvestiia Akademii Nauk SSSB (Moscow), 

IKIAI Izvestiia Kavkazskogo Istoriko-Arkheologicheskogo Instituta (Tbilisi). 

IZ Istoricheskie Zapiski (Moscow). 

JA Journal Asiatique (Paris). 

JEH The Journal of Ecclesiastical History (London). 

JHS Journal of Hellenic Studies (London), 

JKAS Journal of the Royal Asiatic Society of Great Britain (London), 

JR.GS Journal of the Boyal Geographic Society (London). 

JKS Journal of Boman Studies (London). 

K. Klio. Beitrdge zur alten Geschichie (Leipzig), 

KSINA Kratkie Soobshcheniie Instituta Narodov Azii Akademii Nauk SSSB (Mos- 
cow). 

KV KhristianskU Vostoh 

L Loeb Classical Library (Cambridge, Mass.-London). 

LTK Lexikon fur Theologie und Kirche (Freiburg i/B), 

Mansi Sacrorum Conciliorum Nova et Amplissima Gollectio, Migne, J.B. ed. (Floren- 
ce - Venice, 1759-1798). New edition (Paris, 1901). 

MAIP Memoires de VAcademie Impiriah des Sciences de St, Piiersbourg. 

MBAK Monatsberichte der berlinischen Akademie der Wissenschaften, 

MDGKO Morgenlandische Darstellung aus Geschichie und Kultur des Ostens (Berlin). 

MVG Mitteilungen der vorderasiaiischen Gesellschaft. 

NT Nord Tidsskrifi for Sprogviden (Oslo), 

OC Oriens Chrisiianus (Leipzig). 

OS Orientalia Suecana (Uppsala). 

P Pazmaveb (Venice). 

PBA Proceedings of the British Academy (London). 

PBH Patma-banasirakan Handes (Erevan). 

PG Patrologiae cursus completus. Series graeco-Iatinu, Migne, J.P. ed. (Paris, 
1857-1866). 

PL Patrologiae cursus completus. Series laiina, Migne, J.P. ed. (Paris, 1844- 
1855). 

PO Patrologia Orientalise Graff in, K, and Nau, !F, edd. (Paris, 1903). 

PP La Parola del Passaio, Bivista di Studi Classid (Naples). 

PS Palestinskil Sbornik (Moscow). 

PW BeaVencyclopadie der classischen Aliertumsurissenschaft, Pauly, A., Wisso- 



XXIV 



ABREVIATIONS 



wa, G., and Kroll, W. edd, (Vienna, 1837-1852). New edition (Stuttgart, 

1893). 
REA Bevue des Etudes Arminiennes (Paris, 1920-1932). New series (Paris, 

1964-)- 
REAnc Bevue des Etudes Anciennes (Bordeaux). 
REB Bevue des Etudes Byzantines (Paris), 

REIE Bevue des JJJtudes Indo-JSuropeennes. 

RH Bevue Historique (Paris). 

RHE Bevue d'JBistoire Ecclesiastique (Lou vain). 

RHR Bevue de VHisioire des Beligions (Paris), 

ROC Bevue de VOrient Chretien (Paris), 

RSJB Becueils de la Societi Jean Bodin (Paris). 

S Syria (Paris). 

SAW Sitzungsberichie der philologisch-historische Classe der kaiserlichen Akademie 

der Wissenschaften (Vienna). 
SBAWM Sitzungsberichte der bayerischen Ahademie der Wissenschaften zu Munchen, 
SI A Studia Instituti Anthropos (Vienna). 

SMM SaFarfvelos Muzeume Moambe (Tbilisi), 

SV Sovetshoe Vostokovedenie (Moscow). 

T Traditio (New York), 

USAEM USAF Aeronautical Approach Chart (St. Louis, 1956-1958). 
UZL Uchennye Zapiski Leningradskogo Universiteta. 

VBAG Verhandlungen der berlinischen anthropohgischen Qesellschafi* 

VDI Vestnik Drevnel Istorii (Moscow). 

VI Voprosy Istorii (Moscow). 

VIA Voprosy lazykoznaniia (Moscow), 

W Vizantilskil Vremmenik (St. Petersburg, 1894-1928). N.S, (Leningrad, 

1947). 
WZKM Wiener Zeitschrift filr die Kunde des Morgenlandes, 
ZDMG Zeitschrift der deutschen morgenlandischen Gesellschaft (Leipzig). 

ZE Zeitschrift fur Mthnologie, 

ZKO Zapishi Klassicheskago Otdeleniia Imperatorskago Busskago Arkheologi- 

cheskago Obshchestva (St. Petersburg). 
ZMNP Zhurnal Ministerstva Narodnago Prosveshcheniia (St. Petersburg). 

ZNW Zeitschrift fur neutestamentliche Wissenschaft, 

ZVO Zapiski Vostochnago Otdeleniia Imperatorskago Busskago Arkheologicheskago 

Obshchestva (St. Petersburg). 
ZVS Zeitschrift fur vergleichende Sprachforschung. 



ABKEYIATIONS 



AASS Ada Sanctorum Bollandiana (Brussels). 

AAWB Ablmndlungen der Ahademie der Wissenschaflen zu Berlin, 

AB Analecta Bollandiana (Brussels). 

ABAWM Ablmndlungen der bayerischen Ahademie der Wissenschaflen zu Munchen, 

AGO Ada Conciliorum Oecumenicorum, Schwartz, E, ed, (Berlin, 1914). 

AEHE Annuaire de VScoh des Hautes Etudes (Paris). 

AIPHO Annuaire de VInstiiiit de philologie el d'histoire orientales et slaves (Brussels). 

AJSLL American Journal of Semitic Languages and Literatures (Chicago). 

AKGWG Ablmndlungen der honiglischen Gesellschaft der Wissenschaflen zu Gbttingen. 

AO Ada Orientalia (Copenhagen). 

AQ Armenian Quarterly (New York). 

AE.BBL Academie Boyale de Belgique, Bulletin Classe des Lettres (Brussels). 

ASGW Ablmndlungen der sachsischen Gesellschaft der Wissenschaflen, 

B Byzantion (Brussels), 

BA Bulletin armenologique, Melanges de VUniversite de Saint-Joseph (Beirut). 

Ber Berytus (Beirut). 

BGA Bibliotheca geographorum arabicorum, de Goeje, M.J, ed, (Leiden), 

BIM Bulletin de VlnstUui Marr (Tbilisi), 

BK Bedi Karthlisa* Revue de Karthvelologie (Paris). 

BM Banber Matenadarani (Erevan). 

BNJ Byzaniinisch-neugriechische Jahrbucher (Berlin). 

BSL Bulletin de la Socieii Linguistigue de Paris, 

BSOAS Bulletin of the School of Oriental and African Studies (London). 

BZ Byzantinische Zeitschrift (Leipzig), 

Ca Caucasica (Leipzig). 

CAH Cambridge Ancient History, 

CHA Collection d'historiens armeniens, Brosset, M.P. ed. (St, Petersburg, 1874- 

1876), 

CHAMA Collection d'Mstoriens anciens et modernes de VArminie, Langlois, Y. ed, 

(Paris, 1967-1869), 

CHK The Catholic Historical Meview (Washington), 

CIG Corpus Inscriptionum Graecorum. 

CIL Corpus Inscriptionum Latinomm. 

CJC Corpus Juris Civilis, Mommsen, T,, Kxuger, P., et al., edd, (Berlin). 

CMH Cambridge Medieval History, 

Cod. Th. Codex Theodosianus, Mommsen, T., et al,, edd. (Berlin). 

CP Classical Philology (Chicago). 

CR Classical Beview (London- Oxford), 

CSCO Corpus Scriptorum Christianorum Orientalium (Louvain), 

CSHB Corpus Scriptorum Historiae Byzantinae (Bonn, 1828-1897). 

DHG Didionnaire d^Histoire et de GiograpMe JScclesiastique (Paris). 



ABBREVIATIONS 



XXIII 



DTC Dictionnaire de TMologie Oaiholique (Paris). 

EHR English Historical Beview (London). 

EI- Encyclopaedia of Islam (Leiden, 1913-1948). New edition (1 954-), 

EO Echos d'Orient (Paris). 

EGH Fragmenta Historicorum Graecorum, Muller, C. ed. (Paris, 1841-1883). 

G Georgica (London). 

G46 Office of Geography, Department of the Interior, Gazetteer No. 46 : Turkey 
(Washington, 1960). 

GGM Geographi Graeci Minores, Mutter, C. ed. (Paris, 1855-1861), 

HA Handes Amsorya (Vienna). 

IAFAN Izvestia Armianshogo Piliala Akademii Nauk SSSB (Erevan). 

IAN A Izvestiia AJcademii Nauk Armianskol SSB (Erevan). 

IANS Izvestiia Akademii Nauk SSSB (Moscow), 

IKIAI Izvestiia Kavkazskogo Istoriko-Arkheologicheskogo Instituta (Tbilisi). 

IZ Istoricheskie Zapiski (Moscow). 

JA Journal Asiatique (Paris). 

JEH The Journal of Ecclesiastical History (London). 

JHS Journal of Hellenic Studies (London), 

JKAS Journal of the Royal Asiatic Society of Great Britain (London), 

JR.GS Journal of the Boyal Geographic Society (London). 

JKS Journal of Boman Studies (London). 

K. Klio. Beitrdge zur alten Geschichie (Leipzig), 

KSINA Kratkie Soobshcheniie Instituta Narodov Azii Akademii Nauk SSSB (Mos- 
cow). 

KV KhristianskU Vostoh 

L Loeb Classical Library (Cambridge, Mass.-London). 

LTK Lexikon fur Theologie und Kirche (Freiburg i/B), 

Mansi Sacrorum Conciliorum Nova et Amplissima Gollectio, Migne, J.B. ed. (Floren- 
ce - Venice, 1759-1798). New edition (Paris, 1901). 

MAIP Memoires de VAcademie Impiriah des Sciences de St, Piiersbourg. 

MBAK Monatsberichte der berlinischen Akademie der Wissenschaften, 

MDGKO Morgenlandische Darstellung aus Geschichie und Kultur des Ostens (Berlin). 

MVG Mitteilungen der vorderasiaiischen Gesellschaft. 

NT Nord Tidsskrifi for Sprogviden (Oslo), 

OC Oriens Chrisiianus (Leipzig). 

OS Orientalia Suecana (Uppsala). 

P Pazmaveb (Venice). 

PBA Proceedings of the British Academy (London). 

PBH Patma-banasirakan Handes (Erevan). 

PG Patrologiae cursus completus. Series graeco-Iatinu, Migne, J.P. ed. (Paris, 
1857-1866). 

PL Patrologiae cursus completus. Series laiina, Migne, J.P. ed. (Paris, 1844- 
1855). 

PO Patrologia Orientalise Graff in, K, and Nau, !F, edd. (Paris, 1903). 

PP La Parola del Passaio, Bivista di Studi Classid (Naples). 

PS Palestinskil Sbornik (Moscow). 

PW BeaVencyclopadie der classischen Aliertumsurissenschaft, Pauly, A., Wisso- 



XXIV 



ABREVIATIONS 



wa, G., and Kroll, W. edd, (Vienna, 1837-1852). New edition (Stuttgart, 

1893). 
REA Bevue des Etudes Arminiennes (Paris, 1920-1932). New series (Paris, 

1964-)- 
REAnc Bevue des Etudes Anciennes (Bordeaux). 
REB Bevue des Etudes Byzantines (Paris), 

REIE Bevue des JJJtudes Indo-JSuropeennes. 

RH Bevue Historique (Paris). 

RHE Bevue d'JBistoire Ecclesiastique (Lou vain). 

RHR Bevue de VHisioire des Beligions (Paris), 

ROC Bevue de VOrient Chretien (Paris), 

RSJB Becueils de la Societi Jean Bodin (Paris). 

S Syria (Paris). 

SAW Sitzungsberichie der philologisch-historische Classe der kaiserlichen Akademie 

der Wissenschaften (Vienna). 
SBAWM Sitzungsberichte der bayerischen Ahademie der Wissenschaften zu Munchen, 
SI A Studia Instituti Anthropos (Vienna). 

SMM SaFarfvelos Muzeume Moambe (Tbilisi), 

SV Sovetshoe Vostokovedenie (Moscow). 

T Traditio (New York), 

USAEM USAF Aeronautical Approach Chart (St. Louis, 1956-1958). 
UZL Uchennye Zapiski Leningradskogo Universiteta. 

VBAG Verhandlungen der berlinischen anthropohgischen Qesellschafi* 

VDI Vestnik Drevnel Istorii (Moscow). 

VI Voprosy Istorii (Moscow). 

VIA Voprosy lazykoznaniia (Moscow), 

W Vizantilskil Vremmenik (St. Petersburg, 1894-1928). N.S, (Leningrad, 

1947). 
WZKM Wiener Zeitschrift filr die Kunde des Morgenlandes, 
ZDMG Zeitschrift der deutschen morgenlandischen Gesellschaft (Leipzig). 

ZE Zeitschrift fur Mthnologie, 

ZKO Zapishi Klassicheskago Otdeleniia Imperatorskago Busskago Arkheologi- 

cheskago Obshchestva (St. Petersburg). 
ZMNP Zhurnal Ministerstva Narodnago Prosveshcheniia (St. Petersburg). 

ZNW Zeitschrift fur neutestamentliche Wissenschaft, 

ZVO Zapiski Vostochnago Otdeleniia Imperatorskago Busskago Arkheologicheskago 

Obshchestva (St. Petersburg). 
ZVS Zeitschrift fur vergleichende Sprachforschung. 



NOTES ; CHAPTER IX 



433 



CHAPTER IX 



a {See Introduction n, la for criticism of Adontz's periodization. Eor the naxarar 
system in Armenia see particularly, Toumanoff, Studies, Manandian, Feudalism and 
for a more recent Soviet interpretation, Sukiasian, Armenia, On The Sasanian Empire, 
'the fundamental work is still Christenseri's against which all of Adontz's discussion nmst 
'he checked, See also Erye, Persia.'] 

1 Arm, Geogr., p, 40/53, " Ifnmuj [i hjnpuupwh {pro Iflnpuwwh, short form 
hjnpwjwpwfojg found in the printed text) fynujw fi \,hiirw£, Iptutw jt ]\}npmumh^ 
Ipitum ji ^mu^ljiili np £ hnipfh ^WLlpuunL jkpwhq ", The phrase uupupwiqhmh 
qnp }\}npwLwpwhb ^nihli i.e. " the sparapet of -Xbrwaran " of the western region, 
which the editor Miahan, Ararat, (Eeb,-Apr,, 1906) failed to understand, is found in 
Ananias Sirakaei, \Cf. Markwart, Bran, pp. 16-18, 94 sqq. Also, Christensen, 352 
n. 6, etc], ■ (213, 1) 

2 Kawlinson, The Seventh Monarchy, p. 439 [Of. Christensen,. $. 102 and Toumanoff, 
Studies, p. 158 n. 33], (214, 1) 

3 Noldeke, Talari, pp. 151-153, " Darauf ward Konig Chosrau An6sarwdn, Sohn des 
Kawadh u.s,w, JTachdem er Konig geworden, schrieh er an die 4 Padh6spane, deren 
jeder erne Gegend Persiens Terwaltete, und ihre Gefarten (= Unterbeamten), Eriefe, 
von welchen der an den Padhospan von Adharbaigan gerichtete also lautete ; s ... an 
Zadhoe( ?), den Nachwergan, den Padh6span von Adharbaigan, Armenien und dessen 
Gebiet ... '\ [Of. Christensen, pp. 363-364], (214, 2) 

4 Ibid., p. 155, "' Bevor er Konig wurde, hatte ein Mann die Stelle eines Spahbedh's 
d,h, Oberbefehlshabers der Truppen bekleidet ; dieses sein Amt hatte das ganze Eeich 
umfasst, Als Konig vertheilte aber Chosrau das Amt und die Wnrde unter 4 Manner 
namlieh 1) den Spahbedh des Ostens d,i, Choerasan und dessen ISTachbarsehaffc %) den 
Spahbedh des Westens 3) den Spahbedh von Nimroz d.i. Jemen 4) den Spahbedh von 
Adharbaigan un dessen Nachbarsehaffc d,i, dem Lande der Chazaren *\ [Of. Christensen, 
pp, 130 sqq,, 370, 518, and Erye, Persia, p. 220], (215, 1) 

5 Padhghospdn (which was incidentally the title of the governor of Isphahan at the 
time of the Arab conquest in 641, Talari, HE, xliii), Arm. ^immij.numiiimh, Sebeos, 
xxiii, p. 77 is formed in the same way as marz-pan, and means " protector " or " ruler " 
of a " country " or " region 5 \ Pathos occurs in Pehlevi leterature with the sense of 
" province ", West, Pahlavi Texts, II, p. 297, Hubschmann, Grammatih, I, p, 223, 
The word upuwqrw has a definite meaning for the historian Thomas Arcruni, " umiuglih 
jwpgniSjnuap iqwmqnug u^ujmiiLWLnp^ tppg pwqpog ", Tov. Arc, p, 286, 
it is used in the same sense by Mov. Kal, p. 265, " klft jumtjap lfiuiuiqmt " and 
probably by Lewond, p, 133, " n * Ipnpwtj phquummfunu ujiuintrnimG mjh 
imSpwnhmj_ ^qiptjuh fjLp ". The word underlined being a mistake for u^mmipiu. 
In the examples cited the word ufimnqnu means " ruler, official " in the sense of pathos- 
pan. Unfortunately the etymology of the word is not known. On the basis of the 
Armenian examples it may be broken down into pati~hos, " head of a country or region " 
(of. hos with Arm. fynju). On the other hand hos begs the comparison with host, hostah. 
Perhaps paigos-pan < pat-hos4-pan, while pathos < pathost, We should perhaps 
acknowledge hos as an independent word and derive host from it, (215, 2) 



434 



NOTES : CHAPTER IX 



5a [There must be a misprint in the text at this point probably due to the quoted 
passage of Sebeos which speaks of " Yazdgard the grandson of Xusro ", but the xi* sr 6 
meant here was Xusro II Abharvez, who was indeed the grandfather of Yazdgard III 
the last Sasanian ruler. [Cf. Erye, Persia, p. 283, whereas the divider of Persia into 
four regions is usually held to be .Xusro I.] 

6 Sebeos, xxviii, pp. 99-100, " ♦♦•^ inuiiniunbui^ piuduthktjuih q^uiLp^ 
ui^fuuip/ifiL luwpufiif jkpfiit Smundiu : Quup iffi wju* np ji inwpufirj h. jiupkh^jin 
Ifnqfiwhf; : fci. omLp dp hjnpkdiuj * np jlXunpkumwbbwjn l^nqpwhi : fei nwip 
*fy* n P jlXuipuiwinuifywh l^nqSwhi : pwjfj pwqtULnp hnpw ip ji Sfiiipnb^ L 
uiifkbkglib i/Jiiupwh jqwainiijjh qhw tlfiuipmLmpkui^p inSib ". An echo of 
the Sasanian division of the army into four parts is found in MX, II, liii, where the 
Armenian king Artases divides his army among four commanders. (215, 3) 

6a [See, Chrisiensen, pp. 181 and 371 for the opposing theory,] 

7 West, Pahlavi Texts, II, 7, p, 12 [" As it is said that Tistar is the chieftain of the 
east, Sataves the chieftain of the west, Vanand the chieftain of the south and Haptok- 
ring the chieftain of the north "]. (216, 1) 

7a [See Stein, Bin Kapitel, and Christensen, Excursus II, pp. 518 sqq., for the thesis 
that the organization of the Empire was copied on that of Iran, but in the Vllth century, 
also Altheim, Ein asiatischer Stoat} 

8 Arm. Geogr., p. 40/53, " JXuipuiuiiuinwlpuu, IXpifii np £ Zwjgt ^wp^wb * np I 
^frpg fhwfr np i IXqnuubg) pw[ujiiuil{uiu^ IJpuwl^wb^ Unf, ty*bnwh, fcuih£uih, 
'hjdnLhg, *}*Spu]Lwun.^ Suju^ppumwh^ Phmuby IXfifc npp upvunfb^ wniu^fi 
Ijwj dkq^ ". [Of. Eremyan, Armenia, p. 26]. Which lists 14 districts instead 
of the 13 announced [See Appendix IV B for the context. The Soukry translation 
Ibid., p. 53 omits the district of Are, which is, however, included in the text, Ibid., 
p. 40]. Markwart, Bran, p. 125, takes Sancan (pro er£an) as an interpolation. The 
final comment in the cited passage of the Arm. Geogr., is interesting since it suggests 
either that our version of the Geography is incomplete, or that the section dealing with 
the subdivisions of Persia has been taken from another work in which a detailed descrip- 
tion of the provinces listed had been given. [On the problems of the Arm. Geogr., see 
below, Chapter XI n. 1], (217, 1) 

9 Markwart, Bran, pp. 94-95. Might this indicate that the creation of the four 
toparchies preceded the actual subjection of the districts named to the Persian Empire ? 

(218, 1) 

io Ibn Khordadhbeh, pp. 118/90, Cf. Eremyan, Armenia, p. 66 n. 1. (218, 2) 

n Ibid.,?. 17-8, d U jUUjI ±$y> i oU J^ S Uj ST ( oU ilD^I ( $U dkj^ 

pro jUa^wvw [6^j-»] ^ 0^1 _ ^ C)\$i£ \yj eLi oL^^ 

aU oW^jl ^Jj> J oLi (jfcf.Slj.ST pro d U (jUc^s^ as we believe this 
passage should be read. (218, 3) 

12 Zach. Mityl., II, vii, pp. 327-328. (219, 1) 

23 According to Ibn Khordadhbeh, p. 121 there were 12 farsalchs from Berzend to 
the plain of Balasakan and from there to the frontier of Azerbaijan. According to the 
short version of the Arm. Geogr., the nju^m Ji Pmjmuwljmh is mentioned among the 
districts of Albania, Because of the defective description of Albania in the longer 
version of the Geography, we cannot tell how the subject was treated there. Lewond, 
p. 132 lists the districts devastated by the Khazars in 763 with the comment that 
" wju n,inLwnj> wphjwp^fih JXnnimhfiij " and furthermore that " umfih k 
nnwhhmih nwyuih puiimumbmh ". Balasakan was perhaps moved to Albania 



V 



NOTES : CHAPTER IX 



435 



on the strength of this passage. If the corrections nnm - [i - pinqwu and Pwrpuu 
— nnm " the Balas river " are permissible, Pwqmu^wl^wh might be connected 
with the district of fiwqiuh - nnm or fhnw—fi—pwqiu in P'aytakaran. In Alisan, 
Great Armenia, p. 92, the Bolgaru cay is called Bala-rud and equated with the Balas-fot. 
[Cf. Manandian, Trade, pp. 163-168; Erye, Persia, p. 206 and n. 20; Eremyan, Armenia, 
p. 44, 83, 88, 109; Honigmann and Maricg, Eecherches sur les Res gestae divi Saporis 
(Brussels, 1953), pp. 80-87.] (220, 1) 

*3a [Koriwn, xxi, 5, p. 34 and p, 92 n. 46.] 

13b [Sebeos, i, p. 26, " Qwjhdwii jwniu^ gwh tjwju uiiqumuiSpbiu^ fi pwtj 
bbiurj fi Zwjng whnLwhbw^ ^uj^iub fiyjumh ut^JvwpCfih UpLukwij, k fuhqpbwij 
[t hjnupni^wj jiupguijiti ^ivpufig. t^fi wwpgbh i^jiLwh w^fuwptyh UfiLubiutf 
fj 'ypLUwj ft <pwjuiwl{iupiub ^ujqm^ h lfiupn.buijl; r^gmipngii ft fcwCpdwp 
\Xmpmuiwujliiuhfi. tj^jj fifr Lu ^n^bugjj uibnrfi ZwjniJ J* ^P^J ^nyw ". [Adontz 
translates the last clause, " ... so that in the future it should no longer be considered 
an Armenian city, but hngiu is a plural, as it is rendered by Macler in his translation, 
E. Macler, Hisioire d'Eeraclius par Veveque Sebeos, Paris, 1904, p. 5 and n. 5, " ... de telle 
sorte qu'on ne donnat plus aux Siuniens le nom d'Armeniens ".] 

14 Sebeos, xxxviti, p. 152 who speaks of the princes of Siwnik', " ••♦ npg jpbui^ 
tfiu jumw^wnJJjh jui£fuwp£uii{fipli IXmpmwmmliuihjii iljiii^k pwp&wi. pwq.WLn- 
ptiLpjuhh ^tupufjg^ ••• ". sahr-mar is correctly rendered by the Armenian term 
w^juwp^uiqfip, [Cf. Markwart, Eran., pp. 120-122, Toumanoff, Studies, pp. 131, 214 
and n, 244, 332. Eremyan, Armenia, p. 88.] (220, 2) 

15 Procopius, Pers., I, xv, 1 [L. I, 128/9], " to 8c arpdr ^v/xa tovto JTeporap/xeviW t€ 
/ecu EovvvrCov rjoav, ol 8^ *A}\fi\avQis etaiv ofxopoi ". (221, 1) 

16 Arm, Geogr., p. 40/53 gives £T/r/7H/^i//i/, JTujj^fihpumbC of which the second is 
a distortion of Hp&b miimuli. The Syrian sources distinguish between Arzon and 
Arzon-ostan, see above Chapter II, p. 35 and n. 25. The short version of the Geography 
gives this name as Hp&[ih, Hp£bli„ which Markwart, Eran, p. 25 takes to be a 
dittography, [On Aljnik and Nor-Sirakan, see also Eremyan, Armenia, pp. 33-34, 
72, 77, etc. ; Toumanoff, Studies, pp. 128-129; and below n. 32, 33, 35a]. (221, 2) 

17 The correct form is marzpan from marz, " country, region " and^cm, hence " keeper 
of the country ". Marzban is the modern Persian pronunciation. [Cf. Hiibschmann, 
Grammatih, p. 193, Christensen, pp. 136-137, 519, Toumanoff, Studies, p. 153 etc,]. (222, 1) 

18 Procopius, Pers., II, xxv, 1-5 [L. I, 478/9-480/1], According to which Dwin served 
as a mart for the exchange of Persian and Roman goods, " ... iK re yap 'Ivh&v «rai r&v 
7TXi]aiox(Jop(^v *IfiTJpa>v iravrcov ts <!>s eurztv rwv eV Tlipuais idvwv /ecu 'PcojitcuW nvcov rd 
<j>opria ioKOfiiloiAf-voi ivravda cUAtjAois £vjn/?a/\Aot>cn ". [Cf. Manandian, Trade, pp. 81- 
82], Dwin stood on the bank of the Araxes, presumably at the site of the present village 
of Diurgiun. [The site of Dwin is now identified, cf, Eremian, Armenia, p. 49 et a?.]. 
The name of this city, which was of such interest to Armenists, must, in my opinion, 
have the same origin as Dabana, a town in Armenia Minor and have had the sense of 
6( village " from the pre-Indo-European Armenian *dava-, Georg. q?^6c>. [Cf. Minorsky, 
"Le nom de Dvin ", PEA, X (1930), pp. 117-120, Hiibschmann, Ortsnamen, p. 422,] 

(222, 2) 

19 LP\ Ixxiii, pp. 428-429, " qruhq uwuinfilf h£uw jkplffjpu ^ping, L JiS 
mbubm^ pi ri£ h3 Ifiupnq^ ^wj_ uinm^ fi, jvnju wmJbui^ fi ippfihu Z&jn*J Sow [i 
uwCSmhm^nLpkmhh ^pwg ffuwLl ^^ : ♦** HPPgh £ m ] n 3 ma wppwfi i Lp m 3 
£,wuwhj;fih. k pmhmbbmjD um *[wjp ify ^uimmfifii ywht^wpuig qmqmpljih ". 

(222, 3) 



436 



NOTES : CHAPTER IS 



20 Zewond, p, 168, " iwuwulp [i um^Smuu ^pwif ]} nwuunh Ijnq". [Cf. 
Eremyan, Armenia, pp, 59, 84, 118,* Tonmanoff, Studies, pp, 131-132, 452-456, etc] 

(223, 1) 
20a [ZP\ xxxv, p. 202, " ♦♦♦ ilfiufk [j qkonh np Ijn^i Jtiwqjvwn, jupfypjiu JXqpL- 
wbjiij ". Mlise, p. 75, " ifhph Jj utu^Smhu ^pwg £wunJ;iq Jtfwnjuwq gwnwgfii 
np &ilhpnij lp puiqwLnpwg JXnnuuu[iij *\] 

21 Alishan, Great Armenia, p, 84; Barxudarean, Arca-%, p. 420; Vayu&t gives Xilxil? 
a city in Shamshadilo, on his map. [Gf, Eremyan, Armenia, pp* 75-76, 118; Tonmanoff, 
Studies, p. 132 etc The Zegam or rather Jegam cay is indicated in both the Armenian 
and Azerbaijan Atlas. — AA, AzA,] (223, 2) 

22 FB, HI, vi, " pkpfih jj qwimnlj ptpkmhg ft l^nqSmhu Hqnuithfiij ji 
uw£fiwhu Zwjng j> Zwpwufr j* $oqh ••* JXSwpwq = MX, HI, iii, " phpkpnj 
[i ifingp \Jju.ujw, pwqbjpfj jJX^mpivu mimuji *\ Arm. Geogr,, p, 33/44. (223, 3) 

22a [Idem,, the printed text of the Arm, Geogr,, p, 33 has the following reading, 
" qljjiuwl[wujiu, ij^lnmmli^^ "* *3f* Erem y a31 > Armenia, pp. 70, 117,] 

2 s IsP% xxxvii, p. 216, i% phpkm wuqiu ji uikqm^u ji Zkp A [i QujpwLWun. 
qmuunjiu* ifewpfiu kjj> qnpbnju inwwkpwqifjiu. k n£ ng ifrukp mpnh^ qopnih 
tywpujig £wuwuk^ jm^jvmp^u Zwjn$ uupuhnLpkiuSp k qkpk^pi[ jwp^mljkj_ k 
ifiiwuki". Ibid,, Ixxi, p. 416, " • ♦♦ phpkm ns y tuutp^ ]_]}§]} uihywuk^ pwnfim.- 
pkmuu tfjwpujitj fj rifling fynnpu wyfuwp£fiu ZwjnjJ "• Ibid., \xxxxx, p. 519, 
" fet klikm^ \,jijunp jw^juuip^u Zwjn$ n°£ fejukujtj ilji^uiSjuk^ fi Zwjng mkqjiua, 
mjj_ qiuqmpkiu^ ji qiuuunjiu np tyn^tp Zkp ♦♦♦ ". [Of. Eremyan, Armenia, 
pp. 51-52, 63, 117,] (224, 1) 

23a [Lyneh's Map still indicates Hogeae VanF dne sonth of Kasrik c, SS^^x 
43°25'E. bnt no snch indication appears on TJSAEM 340 B IT.] 

24 Georg. Cyjpr., p. 48, " tj /cAetcrovpa BaX-aA-sfowv Kal apx^irat Kara to apKrQov pipos 
ri MeydXy * Apulia ,J . The kleisnrai are the pass called Qnpwju Jjnju by the Ar- 
menians, FB,, TV, 1, ** luuipjiuiq wb^jth ji Zwjng l[nLu£ n p Qnputjh fynikh^ 
ij.pm.uu qui; fib". Sebeos, xxx, p. 108, " ***£% £wuwiuwp£u Qnpnj jwyfuwp^u 
Swpounj"* ". Nowadays this pass is called /}*iz/£/tfi InciSean, Geography, p, 176; 
Sargisean, Itineraries, p, 274, [Cf. Marlswart, Sildarmenien, pp. 376-378, Eremyan, 
Armenia, pp, 33, 36, 58, 63, 71, 117; Tonmanoff, Studies, pp, 198-199, etc.]. (225, 1) 

. 24a [Arm. Geogr., pp, 28-29/39-40, 33/44; Tonmanoff, Studies, pp. 128-133.] 
2 <n> [FB, TV, 1, cf,, V, yiii-SYi.] 

25 Ibid., Y, viii^XTi, (226, 1) 
25a [Ibid,, V, xiii, " fet q^fULp qkw* npujiu h jumw9u lkmj_ ip y pun. kpljfipu 

HqpLwufiEf h pun. kplffjpu fjLpkwuif uw£Swu mpuiphmj^ IjwtjnL^whijih ",] 

251) [Ibid.,Y,xv, " {?z. qiftu uuiCduiuuh, np jmnm^utu hp ]hwj_ jkpfyfipu Zwjnij 
k pun. hplffipu ^pwtjy np £ fiugh i/kb qkmu ^ULpy wjunhQimhi. jfiugu i£wwpkut^ 
qmnuiujp wumfi ",] 

2* Ibid., VI, i, "Pm jg k J, unywhi pmuniS qwLumg fobbing ^mmkm L ^ 
mjup^ muqp^ h ifuwij umfywL fiwuu jwduwp^wy kpl^n^niuq um hpfjnukmu 
Pmqmtnpuu", - (226,2) 

2 ? Ibid., IV, xxi, » hulj uwtfwuwujw/igu pwqwLnpfi ZwpS, np uumhfi /r 
*}*wuhuili lXmpuimuiuj^uju[jy +*+ ", Ibid., V, it, " ♦♦• ilfiuih. fywu&wlf jHwpiyw— 
jwljuiu fi uw^dwuu Zwjntj ^uijiubijiu*** '% Ibid., V, v, " *»* *} UW L ]} uwCJiwun 
Zwjng p ^mu^mlji fynLu£ jiuinpiuwjwtyiuu wiuul", [Ibid., V, vi, '" ***Pnq[ju 
qim uiSmmimmwH J, ^mUmlf np ip uwtfmb pin <l)uipu It pin Zmju ", 



NOTES : CHAPTER IX 



437 



Ibid,, V, xxxiv, t( ***ilfih^k fi ^wu^wl^ uwCSwu kpfyfiph np fyuipufin fynuit ip-> 
Zwjntj uw£Swh ip ". Of, also, Ibid,, V, i, " fez. iljiu^kL fi pndi Ji uw^Smuuh 
jj ^wh^ui^ J£uipufwwwl}whli+ + * " and Ibid,, III, vii, " • ••qniribhuijfi kpfyfiph 
£wjnn+ •♦• i/jih^L fi tjuigp gwnjugjiljh JJininmnnL^ k iffihik fi ^uihkml^ nmiSwuu 
mmpiummmljmh ". (226, 3) 

28 Hoffman, Auszuge, p. 250; Noldeke, Talari, p. 100 [Of, Ohristensen, pp, 142, 166; 
Erye, Persia, pp. 221 and 259 n, 23 ; see also, Chapter I n. 1 ,] (226, 4) 

29 Pelr. Pair., p. 135, " *Apfj,Gviav 8s ZivBa to Kaorpov ev fXG$optu) rrjs MtjBiktjs K^i^vov 
opi&iv ". Kiepert, Xarte gives a Zindu north of Nineveh near Akra, but it is doubtful 
that it is related to the historical Zintha. [See next note.] (227, 1) 

29a [Qf, Erye, Persia pi, 5, c * Zindan-i Sulaiman, and Manandian, Hist,-geogr. Studies, 
pp. 15-18.] 

30 PB, IV, xvi, " ♦ • • k k plipnpij.il] Irtish nilkb mnthh mmjp hSm jwmpinmmmlimh 
wpfumpifiu". Ibid,, IV, xsr, " • ♦ ♦ k mwij uSm wnih Jkb* n^fi mpdwii nhwn£ 
ji ^mjnu B.iujL um ilkq, wjhin^u mwrjnLg hdw mmu, ^ fi fi Zwjnij iJpnsk ji Spojpnh 
*tyh}k urn ifkt^ ^wSwl} jfiLpnLil mmh w^Ji, tSfih^k um tfkq^ kljkunl ". (227, 2) 

3i kx,II,liii, "-♦ painty ft Sfinpfwy np wjdii Ijn^ ^npnpfig, pjU^^ ". 
[On Alki-Elki, see Eremyan, Armenia, pp, 32, 60, On the various districts, Ibid,, 
pp. 49, 60, 64, 72, 86; also Toumanoff, Studies, pp. 163-166, 180-182, etc.] (227, 3) 

32 ££, p.41, " ^*^pfiuwnukmj^ Xtn^fipwlfwufju ft "bfiunLl; uui^ujun. ", [Of, 
Markwart, Bran., pp. 165 sqq.; Sudarmenien, pp, 378 sqq,, etc; Toumanoff, Studies, 
pp, 161-166, a at,-] . (227, 4) 

33 Syn, Or,, p. 272, [" Ensuite vient le siege d'Arbele, et l'eveque qui l'oectipe est 
metropolitan! de Beit JSTouhadra, de Beit Bagas, de Beit Dasen, de Hamonin, de Beit 
Mahqart, de Dabarinos( ?) et de leurs e>eques — ", Of, Ibid., p, 617, [Of, Markwart, 
Mran, pp. 22-25, and Manandian, Eistor, Geogr, Studies, pp. 18 sqq.], (228, 1) 

34 Gever — Arm, fyauun; the modern name may he Bales gewer. Ghabot, Syn, Or,, 
p, 617 identifies Bales with the modern Baskale, an old theory which Hoffman, Aufzuge, 
pp, 227-230 and n. 1826 considered unacceptable. See also, Ibid., pp. 206-216, Nohadra; 
pp, 202-207, Dasen. The Kurdish tribe of the ♦ V. is found in the Sheref-Nameh, 

cf, CJiarmoy, I, 2, p, 28 and in the JeMn-numa, Ibid,, 1, 1, p. 73. According to Mas'udi, 
II, p. 423, the (jl^^Ui are in Azerbaijan and theirs is the settlement of * a5^53| 
Kinkivar, Markwart, Bran, p. 24. Markwart accepts the recent opinion of Hartman, 
Bohtan, p, 128 that "KinMvar — Arm. ^ujuqnmip a city in the district of Anjewacik', 
[Of, Eremyan, Armenia, p. 58], (228, 2) 

34a [Arm, Geogr,, p, 32/43,] 

3 4i> [Of. Markwart, Mran,, p. 24,] 

35 Arm. Geogr,, p, 40/53, '* \,mnwpmwj pro \,n£wwpwj \\jn\^jipm!jmu ", 
[Of. above n. 16]. (228, 3) 

35a [Of, Markwart, Bran, p, 25, Syn. Or,, p. 272, " Ensuite vient le siege de STisibe; 
Teveque qui Toccupe est metropolitan* d'Arzon, de Qardou, de Beit Zabdai, de Beit 
Kahimai, de Beit Moksaye et des ev^ques qui s'y trouvent ", Ibid., p, 617, II, " Pro- 
vince occidentale : 'Arab, Beit 'Arabaye ",] 

^ [FB, IV, 1.] 

36 Sebeos, iii, p. 37, " fei JdnnnLii &kt^ njmlhuwju kpl^jiph Z w j n g tlp^^k 
n*limufyn£ k nqnuiu JXnnLwhhn, k jlXunpmn l^nnSmhl nJXpmmummh k 



438 



NOTES : CHAPTER IX 



°\r n p &jipujl{uih iljihik yuw£iiuiuu $lu61jlul], qfi wn uuifuubuiLgu fiul{ ^kq 
[kuij i, k puri LupkSnunu iljiuik tj^kuiupfiiu ^wmujqni[ljiugLng ", Cf. also Ibid., 
p, 9, where \,npp[]pujl[ should be read \,np^jipwlj <uiu>. In FB, V, xxxii, King 
Pap says to the Emperor, " bpt ^kuwpfiLUL Qui k mwuu gwnwg ifkp /£«/££, ivpij. 
h piug wnLp. k qf]LiiCuj guiqiug ^jihtm^ £ hwjvukwtjh ifkpnq ". (229, 1) 

3 «a [Arm. Geogr., p. 29/40; Toumanoff, Studies, pp, 132, 163-164 and n, 43.] 
36t> [Arm. Geogr,, pp. 31/41, 33/45; Eremyan, Armenia, pp. 35, 85; Toumanoff, Studies 
pp, 129, 148, etc. Goubert, IS Orient, pp. 167-170, 290-302.] 

37 Sebeos, ii, p, 33, " fei. uimn ^kq ul^nqpwhu aJXunpuirf 1 nlXpnLwumwhh 
wifkuujju llfib^k fj Ifbpfiu gwqiug : h^jkpl^piu Zwjng nw^jumpGi Swununkpujlfuju 
h-sluwhnL&kwh iffiusk gllpiupLuw k *}%pLJ7U gwqwg, k ifjiu^k tjkqp bni[ni.u 
pnunLukwn k tjJXn.kuwwL.uih". Ibid., Hi, p. 45, " • • • km hngw alXprnujumuiu 
qmtlkuLUjh i/fih^k Ejlfbpjiu, k qbpljfipu Zwjnij* np phij. fiLpni^ fr^jvwunLpkwSp 
lp, qwrtLU SwhnLwkpwliLuu ityu^k Ejqkmh ZntpujLiqLuu^ k ljuillutiu ^nmljja 
ifhhik nwLwhu %wnhfi k gkqp bnifnLu PqlmLukiuij, L nJXakumujLUJU k g%nqni[fim 
LjWLwn. djth^SL tjZwgliLLi k gtyuilinL " and not " IXnkuwwLiuu k Lj^nqilfiuj 
as it is given in the printed text. {Cf, Tov. Arc, p. 88, " iljiu^k jlXnkuuiuuiLLUU 
k nytnwiilhm"* ") fei. Ijnnpu ^ujuwnLp ml^mh q^qjih ip fi bumwjnLpjiLu 
wwpufitj uip^utjjiu ". (223, 1) 

37 a [Cf. Chrisiensen, pp. 103 sqq.] 

38 The term i[wuiunLpiulfLuIi is first met in Sebeos in two forms: ijwuwnLpwIimh 
and ukifi{,ujljiuh, xxxv, p. 140. It is used only once in MX, II, lxii. See above, Chapter 
VII n. 55, An interesting composite form, ijwuiunLpwIjuiuiuqnjh is found in the BL, 
p. 170, where Kyrion, kat'ohkos of Iberia wrote to the Marzpan Smbat: ,l \XumnLinb 
qpkq £ mbpwha ijwuwnLpwl^Luhmqnjh uipwutj£ nmLmlpuu^ £,wun.kph. ". Kyrion 
is praying here that Smbat should become not only a mere ier [lord] but a ter among 
ters, so that ilwuLunLpwIjLuuiuqnju is the equivalent of uikpwqnju if such a term is 
possible. This interpretation supports our hypothesis as to the synonymity of 
LJuiuLunLpmljiuu and uimhnLmkpml^wh, (223, 1) 

39 JDiehl, Ch,, Uorigine du regime des ihhmes dans V empire byzantin, Paris, 1905; 
H. Gelzer, Die Genesis der byzantinische Themenverfassung, Leipzig, 1899. [The litera- 
ture on the problem of the Themes and their origin has grown rapidly since the time of 
composition of Adontz's work. See for the more recent views on this disputed problem: 
J. Karayannopulos, Die Bnistehung der byzantinischen Themenordnung, Munich, 1959; 
N. Baynes, " The Emperor Heraelius and the Military Theme System ", BHB, LXVII 
(1952); W. Ensslin, " Der Kaiser Herakleios und die Themenverfassung '% BZ, XLVI 
(1953); E, Dolger, " Zur Ableitung des byzantinischen Verwaltungsterminus 0ej*a ", 
Historia, IV (1955), G. Ostrogorsky, " Sur la date de la composition du Livre des Themes 
et sur Fepoque de la constitution des premiers themes d'Asie Mineure ", Byzantion, 
XXIII (1954), and " Korreferat zu A. Pertusi ", Berichte zum XI. International 
Byzantinisten-Kongress, I (Munich, 1958) ; A. Pertusi, " La formation des themes by- 
zantins ", Ibid., and ** Nuova ipotesi sull'origine dei temi bizantini ", Aevum, XXVIII 
(1954). See also E. Stein, " Zur Entstehung der Themenverfassung ", Studien zur 
Geschichte des byzantinischen Beiches, Stuttgart, 1919, Bin Kapitel, and his " Review" 
of A, Christensen, ** L'Iran sous les Sassanides ", 1st ed. in Le Museon, LIII (1940), 
pp, 123-133, where he discusses the thesis that the Byzantine administrative re-organi- 
zation of the VII was influenced by the Sasanian reforms of the VI ; a thesis adopted 



NOTES : CHAPTER IX 



439 



by Christensen in the 2nd edition of his work, " Excursus II ", pp. 518-526; though 
questioned by most other scholars.] (233, 1) 

40 Diehl, L'Origine, p. 12. (233, 2) 

41 Justi, GescMchte Irans, p. 469 [Cf. Christensen, p. 210 and above n. 39.] (233, 3) 
4ia [The " Leo I " of the Russian text is obviously a misprint for Leo III.] 

42 Georg. Cypr., p. 46, " iitapxlo. MzooTTOTaixias avw tJtoi A 'Apixevias "Afxiha p,7)rpo- 
ttoXis ". Ibid., p. 48, " iirapx^a A 'Apfisvias aXXfjs, Aahtficov vvv {UjTpoTroXis "• [Cf. 
Appendix II ¥ for the context]. Elias, the representative of the latter at the Quinisext 
council of 692, is called, te £ttI(7Kottos Aahlp.a)v J u€T J ocmoAea>S' rrjs A 'IovaTiviavijs "> 
Mansi, XI, p. 992. The Armenian historian JoJi. Katfoh, xvi, p. 88 mentions these 
transformations and renders Iustinianea by QnLuin[fih~]fiwiwLjUliuin, Gelzer, 
Georg. Cypr., pp. xlvi sqq., suggests that the episcopal residence of Armenia IV or Upper 
Mesopotamia was Martyropolis rather than Amida, but this was the situation at an 
earlier date, before the changes just described had taken place. [Cf. Goubert, V Orient, 
Appendices x, xi, pp. 290-302]. (234, 1) 



440 



NOTES : CHAPTEK X 



CHAPTEK X 



a [On the question of Adontz's periodization, see above, Introduction n. la. The "whole 
of this chapter, and indeed, the entire section dealing with the naxarar structure of 
Armenian society must be taken now in conjunction with ToumanofFs major study 
on this problem. Toumanoff 's work, being a continuation and revision of Adontz's 
earlier work, (see. Studies, pp. 149-150), is relevant to all of the aspects treated here, 
but it is far too extensive and detailed to permit a discussion of its contribution at this 
point, since any adequate treatment of this would entail its incorporation nearly in 
extenso, Hence, only separate points from Toumanoff's work can be introduced into 
these notes. See also below, Chapters XIV-XV.] 

15 [The date of composition of the History of Armenia attributed to Movses -Xbrenaci 
has proved to be one of the most controversial problems in Armenian studies for nearly 
a century, and Adontz was one of the leading protagonists in the debate. See, for 
instance, his discussion with H. Levy in Byzantion, XI (1936). The literature on this 
subject is far too extensive for even a meaningful sample to be included here, but see 
on this question, Toumanoff, Date, and Studies, pp. 18, 104-105, 307-308, 330-334, and 
particularly p. 330 nn, 113-114, for some of the bibliography, as well as pp. 108-111, 
for Toumanoff's appreciation of Adontz's thesis,] 

1 MX, I, ix, " • • *funp£m.pn. ji Sw fi hn.fi qfimkj^ ♦ • nLumji umjumpmpni.pjii.ugu 
np mum l{mu : QJi n£ lfmpq.j> Jiu£ ]kmj_ mum jmjmhji^ A n£ ifk^kujiij mm^mmiinLUg* 
A n% u.iJumLnpmg myjump^jiu mnm^Jiuu jmjmuji J; L n£ ijbp^fiuu, A ns mj]_ hh% 
opjiumLnp, mj^ fumnu Ji jumnu mifkumjli A ijmjpkuji " : (237, 1) 

la [JPJB, IV, ii, " ••• rfhbmrfkbgu jfiLpmgiuu^fiLp qm^m.^ A j^npbml^miph 
jJiLpmgmu^JiLp ^mijjnL ".] 

8 Ibid., " fei mil jmjuS m^mg A Ji JunhmpC np qnpbml^mjii muniSi pmphhtp 
mnm^Ji mpgmjjih mmmfiL fi ^jni-Ju pmqpijiu. pnq qhmCmmhmu ilkbmdkbu A 
qmmunLmlpu* npg pjipbrnlim^ ujimju ifih, Jihh Cmpkhp pmpk* np Smmulp [i 
dmS mmfimpjiu nLpmfuni.pkmuu pmqjSml^m^m^u l^mpqkiji^. P nr i nnmhl^mju 
qjipbmljmipLphmuu ummunt ", The punctuation of the 1889 Venice edition is 
incorrect, as a result the words " np ujipbm^m^g rffimju IJih " might seem to be 
related to the preceding phrase, " P nT l uhmimmkmu ", etc., whereas they are in 
fact a repetition of the clause, " np qupbrnfymju mum.h pmp&JiLg ", etc, evoqued 
by the introductory phrase, " P ni l um^mmkmu • •♦ ". Moreover, pmuSml^muunu 
should be read pmuSml^mhmgh which means both " feast " and " seat ", hence 
" pmndmlimumiju l^mpnhip^ " can mean either " the organization of a feast ", or 
"the putting out of seats", Cf., Ibid,, IV, liv, " ***Ji mmjumjiu mpl^mhki 
pmqpml^mh •♦• ". As for the expression " q.npbm^mipipkmh ummuni.", it has 
only one meaning here and refers to the " gorcahal office, or function " and not to 
u service " at the table. (239, 1) 

3a [£ e e below, Chapter XV, pp. 354 and n. 63.] 

3 Nerses, p. 32, " ..♦ A hm l^mpuip fi dmS 6m^njh Ji ukqmh mp^mjjiu 
Uppmlfmj pmp&u snpkg£mpju.p ", [Cf. Appendix III C for the context. Also, 
Toumanoff, Studies, p. 229 and n. 273,] (240, 1) 

3a [Jfterses, pp. 38-39, " qnp unpnq.kmg IXp^ml} Pmqmmp A l}mpn.kmg [i dm 8 
£m^njh [jLpnj Ji ukqmh [up pmpku inpkg^mpJiLpu £pmdmumL ilkbjih \ f kpujiufi*\'\ 



NOTES : CHAPTER X 



441 



4 StepTi. Or &., vii, pp. 63-64, " «•* umpp ^m^whinjwu^hinu ^pjinnpfju^ ^pmSwj^ 
Jhbph Sptpmnwj ty w pi}kl iPm^ininpiiLpjiLli [up pum uwidwuji Ipujuhpwzjb 
Qtiluwij, h mmj^ pum fiLpwgwusfiLp tqmmyiu£ji zp/4 h wwmfiL fofumuwou imPSt 
h. jw^klfl; : flp L q.bnkgl}wiu£u jopjiukm^ fi dwi) £m^nLU uumkj_ pwp&u 400 
fe[uwuwij " ; (240, 2) 

5 Tov. Arc, p, 239, " • • « njumwfu. Smpnu^mhnLphmh qm^mljm^nLpkuju ^mtii/— 
uiiu^bw^ pum tywpqji pwpkmpkp^ntpkmu Prnqmuipiugu £wjntj y Swuwliuuii 
ilkbfiu Spqwinwj pwqiuLnpji ", (240, 3) 

6 ZG, p. 43, " •♦• h wbfib pbq jihghwhu kpfiu fi^juuthu k ^nphg^wpkLp 
mjp puqfip " ; Ibid,, p, 49, " * • • L mmm ilfiiupwu [uwnwzjiiu qhrnzfjih jj quiuuna 
JXu{w£nLukwg^ phtj. fiughwuu mmukjnil q^ft^uwuuu npg £fih fi pkpqh fltjl^wh. 
k n^nphg^wphip mpuu ji qfuwuu wpgtiLuji mwjp mpl^mhk^. [On the accuracy 
of Zenob, see, Abelean, I, pp. 345-362.] (241, 1) 

ea [Cf. Toumanoff, Studies, p, 128 n. 226, and below Chapter XV, n. 63.] 

7 The work of Ya'qubi is known from excerpts and fragments found in the works 
of other authors, The information quoted is found in Ibn al-lTakih who was almost 
his contemporary, Geogr, Arab,, V, p. 281 = Karaulov, Sbornih, XXXI, p. 23, Yakut, 
XIII, century gives 118 instead of 113 principalities, he evidently likewise depends on 
Ya'qubi [Cf. Toumanoff, Studies, p. 229, n, 273], 3JL l&SC!^ S-U J I J>Zj 

*yU ^Jtfj. (242,1) 

8 Sahib as Serir means " lord of the throne ", According to this account, the golden 
throne of Persia fell into the hands of a descendent of Vahram Choben after the downfall 
of the Sasanians, He took it with him and ruled in the country called by the Arabs 
y yd I after the name of the throne. According to Movses Kalankatwaei, the descen- 
dents of Vahram Choben settled in Arran and Juanser [637/8-680/1] was descended 
from them, [Cf Mov. JDas^ H, xvii, pp. 107-109 and p, 107 n, 3 and pp, 109 sqq. 
also Toumanoff, Studies, pp. 392-393 and p, 392 nn. 10-11], In such a case, it may be 
possible to identify Serir with Arran, or rather with its northern portion. (242, 2) 

9 LP\ xxiii, pp, 133 sqq, [3Tor the text of this list, see Appendix III L, i, For the 
lists of princes found in Ijazar and EHse, see also Toumanoff, Studies, pp, 223 and n, 290 
and particularly pp. 246-248 and tables viii-ix]. (242, 3) 

io Ibid., xxv, pp. 143 sqq. = Elise, ii, pp, 42-43 [Cf. LP\ xlii, pp, 236 sqq. [See 
Appendix III P, for these texts.] (242, 4) 

11 LP\ xxxvi, p, 209 = Elise, pp, 74, [See, Appendix III G and n. 14.] (242, 5) 

i 2 LP\ xxxix, pp. 227-228 = Elise, pp. 99-100 [See, Appendix III H, and below 
n. 15, also LP\ xxxiv, pp, 198-199.] (242, 6) 

is Z,P\ xlvii, pp, 272-273 = Elise,y. 193 [Cf. Ibid., pp. 178-179, and Appendix III, L] 

(242, 7) 

14 Elise, p, 92 [See, Appendix III, H, ii,] (243, 1) 

is Elise, p. 119 [See, Appendix III G, ii,] (243, 2) 

i 6 The text has " JXphtjpmu ", instead of " ty*~wpktjkiuh ", as it is correctly 

given in LP\ xli, p, 231 in agreement with the same passage in Elise, p. 119 which 

reads, " Jlpm^h %mphqhwu " ; the representative of the Abeleans was named 

Gazrik. In the list of the partisans of Vasak we find the name of " fyiuwjiynj 

JflnpfannriLufi ", This family is listed earlier under its hereditary title of " Smnjumq ". 

Its representative Xoren is referred to once as " Sw\rj^jumu^ ", LP\ xxxix, p. 225, 

and several times in the same passage as " J\)npjunnniJjji ", [e.g. Ibid., p. 227,] (244, 1) 



442 



NOTES : CHAPTER X 



I? " q&ipi&kwg ", Ibid,, xxxix, pp. 227-228, " Wpmihag ", Ibid,, xlii, p. 237. 

(244, 2) 

is " Sm^pw^<hiu^> " Ibid., xlvii, p. 273, and " Sut^pwj ". Ibid,, xlii, p. 237, 
more correctly, " Swypwtj ", Mlise, p. 194 has " fihwifiiwbhwh '\ Nerses, p. 35 
has " ]Xnmfium.hfij> ", and " ftuiifiukwlig ", p. 36, There were two Arcruni houses, 
one represented by ISTersapuh, as noted above, ZP\ xxiii, p. 135, and the other by 
Aprsam, Ibid,, xlii, p, 237, and xlvii, p. 273. (244, 3) 

19 The prince of Ake is given in the list of the Yasakists, Blise, p. 92, where we also 
find the name of Upturn!} ftynmSip indicating that the &stuni shaxed Yasak's point 
of yiew. The list of Yardan's supporters in Blise, pp. 99-100 has only 26 names, of 
which 20 are known from other referrences. The JJu]£umm.bji are found in L~P\ 
lxix, p. 406, and Ixxviii, p. 470. The jtfnupni^ Qwpkijhiuh of ffiise, p. 100 shoxild 
be corrected into ^wqjipl} Upkqkiufa&s in ZP\ xxiii, p. 135, and xxx, p. 179. Might 
" tywtip— ftlf " — c*? be a dialectal form of " hjnupni[ " ? Cf, " Bmjuhmh ", 
JBUie, p. 194 [n, 1] = ZP\ xlvii, p. 273. The ^mphtjkmh were in Yasak's camp. (244, 4) 

so jw$., Ixx, pp. 412-414; lxxi, p. 419; lxxxi, pp. 483-484; lxxxiv, p. 497; xc, p. 526. 
" ^wpgwjjib". Idem*, should probably be read " Zwpgwjffi"* Among other 
princes already listed earlier, we find the following: 
. 1. Ifw^^nhhiuh^ *l^u)£wh, ^wutu^ IXpww^u, ^mpq., ffm^kq^ 

% l//nJi^ putpp^ Pwpq.lhs ^qpini 11. Uh&kLuigpg, JJIldlIi, Qn^mh 

3+ IXSuimniSsJi, Unhwlf, ^wpwq^w^^ 12* Xfnlpng, Qn^mh 

4. Jfuijjumq^ tjhwwjj^nj, ^ap^njj^ 13. fh^inn diking, mqmmnpkmp 

5. fli-p&wj) ynqphfyi ^wpmijhkpu^ 14. ^ui4kniiji r pw£% 



15. tf}wfndift % ^ptvinml^ 

16. ^whwhqwijfi, *£/*£&, Uqqih 

17. ^wpkrjlupij, JJwmnj, %wpwri 

18. ^fSw^uhhlig, f)pf}p 

19. JJui^mnrnhjiy *luwwl[, -fiw^m^ 



6» pwi^puimndjjjiy JJmCiul} 

7. *lwiiuiijpwl{wh, Zpwiww*, Xfkpuk^ 
JJw£iul[ 

8. ^hndihmg, Umnii, Unwuwnii 

9. ^hpndikwii, ^oman? 
10. JXp^pnthjiy Qm^nLp 

[Cf, Toumanoff, Studies, p. 233 n, 290, who rejects Adontz's additions to the List of 
princes,] (245, 1) 

21 According to the sense of Lazar's words, Atrormizd was of Armenian descent, 
" qjlmpnpifjjqij. nifii wumSi * jw^JuwpQh Zwj n $ Smprjiumh CpmSmjlp pnqnL^***" 
ZP', xl, pp. 229-230, We should perhaps read " jw^uwp^jih Zwjng • + • ", i*e, 9 
c S.e., " to leave Atrormizd as Marzipan, in the land of Armenia ", in which case the origin 
of the M arzjpan becomes doubtful. The same passage in Mlise [p, 128] is very interesting, 
". Qmjhd-md qilh jiuuiiq. huifuwpiupwtjh* JXmpnpiljjtm tuhndi npnj fifhirnhmPfidih 
[julf fuwriit lp phq Zwjng w^fump^fih k ipipbui^pg lp qopwjjwpph *** " 
since it reads as though the historian had the parallel passage of Lazar in mind, and could 
not work out Atrormizd's background. Atrormizd's family name, " Hp^iiil} — inh ", 
is the Persian form of the Armenian " Hp^iufy—nLhp ", and he was probably a 
descendant of the Armenian Arsacids. " Quihqwqkwh ", is the name of an individual 
from " jf]uinwhkwii mmh^h ", LP\ xxxii, p. 188, as is \JI)miqj IXpmiul^nLhlj, wjp 
mimwhjili ", Ibid,, lxxi, p. 419. It is not clear whether 2andalean is a praenomen 
or a nomen gentilicwm; we shall see it given as a naxarar&om in the Military List [See 
Appendix HI B]. The *' ^mfunnwmkinh wpgmJjJi ", is found in , both MUe, 
p, 100, and Nerses, p. 35 and designates the royal equerry, probably because of his 
duties, [Cf, ToumanofT, Studies, p. 233, n, 290]. (245, 2) 



NOTES ; CHAPTER X 



443 



22 BL, p, 42, [The list of those present at the Council of 505, including the names 
given by Adontz in the note though not in the text will he found in Appendix III L, ii]. 
Yard Mamikonean was the hrother of the famous Vahan, Marzipan of Armenia; he is 
also mentioned in £P\ lxiii, p, 357, etc. In the " Letter of the Armenians to the Per- 
sians on Orthodoxy ", BL, p. 48, Vard is referred to as the Marzpan, " p ^mpquij 
JTiui/JiIjnh£fjij mkmnSii L p Zwjnif dwpqujwhi ", We know from Sebeos, i, p. 25 
that Vard was named Marzpcm, " Skm ^m^mhutj Jjwjwl qji^pimhnLppiSt 
^tupq. ^mmppfy kqjiwjp ftnpw". Sahak Kamsarakan is probably the well known 
leader and collaborator of Vahan Mamikonean, £P\ xe, p. 526, " fiw^jvhh ^mihniJip 
is perhaps to be identified with " Pwpgl ^wflwiSip ", a contemporary of Vahan, 
Ibid., lxxi, pp. 420-422. This information shows that our documents really belong 
to the early VTth century. (246, 1) 

23 JBL, p. 74, [The list of those present at the Council of Dwin of 555 including the 
names given by Adontz in the note but not in the text, will be found in Appendix III, 
L, iii]. In the list given here, only eleven names belong to famous princely families, 
the remainder do not seem to be princes, but since they are listed under the heading 
" p^pjiuftg" we must presume that many of those present at the Council were listed 
according to their patronymic and not by their nomen gentiliciiim. Only the first 
signature is given in fall: praenomen, patronymic, nomen geniiliciwm, Judging from 
their names, " JXp^nm p ^mpmqmppnqkmi " " and 2w8wquiuui p JJwCwiiwu " 
are from the Pagratuni family (the form " JXpynm " is interesting as the prototype 
of the later Bagratid praenomen " IX^nm "). " QuiLpwfy p ^qmtnukwti " and 
" JXumnimbmmnip ]Xpywi.pkwh " are from the Kamsarakan family, Anemia Bimlmci, 
i, pp, 514/520, mentions a " Zorak Kamsarakan during the rebellion of Armenia in 571. 
(Concerning Gazawon Kamsarakan, see MX, HI, xlviii), Moreover, " %mmnj p 
Qwp^nLqkwh ", and " *Lpp*l llpmw^pwh " are names common in the Mafyaz 
house. The remaining signatures with the repeated grmnomem ^mpqwh^ Xfni^hr^ 
Xfmhmily ", etc., which are so characteristic of the Mamikonean family, must belong 
to members of this house. The influence of the Mamikonean was strongly felt at the 
Council of 555. They had a powerful representative in the clergy in the person of 
bishop Nersapuh, one of these responsible for the Council, The construction with the 
idafai " Qppqnp p Zfiwjwl} ", is noteworthy. The p becomes a j— before vowels: 
" JJw^tq jlXpmwi-wqifwft, *Lpp4, jlXprnw^pwh ". In the case of the twelfth signa- 
ture, we have either Z^^miimum [A] \j minify p ^mumfywh '% or Hamazasp's 
patronymic has been omitted. (246, 2) 

84 Sebeos, xviii, p. 65 and xxviii, p, 98. Among the other princes, we find references 
to the following; 
1, Jfmiipfynlikiuh^ ^w£mh, p. 24, ^mpqiuiuuippfy^ p, 25, ^wpqwh npqp ^zi/— 

uwfymj, pp, 26, 29, IfnL^ktj^ pp. 37, 52, ZwSwqwiiWy p, 50, JJw£wfy, p, 53, 

^mqpfy npqp ITiuhnd;^, pp. 56, 58, Jfuidmfy^ pp. 48, 56, Ifm^hri npqp 

'hwLpp, pp, 107, Zwdwiiuwui pp, 139 sqq, 

UpLhkuizj) *£///£#£, p, 26, IJmhipwhunu, pp. 48, 56, \JmCmfy £opkqjiwjp \)mh* 

ipnthhnup, p, 58, ^pijuqnu^ p, 29, ^ppqnp, p, 107. 

^unCkmJip) ymfinLJ-j, pp. 48 sqq., JJuipqpu^ ^wpwii—hhput^ ^uinwt), pp. 50 

sqq., h}nupm[, pp, 5% 58, pinqnpmt, p, 108, ^ppqnp, p. 109. 

ttfjippintwiSip, IXwww, p, 69, folnqnity pp. 74 sqq., ^w^mu^ p, 102. 
5, IXdujujnjLhpi Ijnmpuiy p, 56, Cwiqni^ p* 108, 



444 



NOTES ; CHAPTER X 



SpiuwmniJjji, l&inqnu, pp, 50 sqq., Uwpqfiu^ 65 sqq, 

PwqpwmnLhfi, \JSpwm ftpqjj Ifwhrntl^ pp. 53 sqq., ^wiif npqfi IX^nuwij 
wuiukmfi, p. 56, \J Upturn fiwpqinuih ^plpubwj^ pp, 59, 61-63, ^wpwtpnfipm^ 
npqfi hnpw, pp. 63, 68 sqq., JJSpmin npqji ^mpmq — }Jus£wliwj y p, 117, 
IXpbpndi^ tf^ujpqwh, p. 56, ^mprntp—DmiunL^ p. 65, ^mumlj^ npqfj l/ii/^u/^ii/j, 
p. 78. 

^ftfiwgukwu, tyk^pm., p, 63, I7i//^7i^i/, pp. 65 sqq., IXpinwLwqif. 'hjifiuifyukwh, 
p. 149, wpg mmhh ^fiilwguhlifiij, p. 139. 
10, Uufiu£nLh[i^ IXpmmmjtjtj.^ tf^umuifi, ZSuijm^ p. 65. 
IXupu&Llihiutj inip JTmhnLiy p. 65, 

fh^uinLU^ ptnqnpnu^ pp. 97, 103 sqq., ft*}wm.huiliwh, p, 139. 
^hnJj^ JTdkd, p. 101, ^mpmii-^uh^ p. 103. 
JJtu/iiunnLh[i^ ^wlJjP^ p. 103. 
15. IXnuiLhqhwh, hjwskiuh, pp, 108, 139. 

ITnl^wg fejutuh ^wpqjil^ ^n^bgbuj^ p, 109. 

^whwhij.uij fefmuhg, IXnliuLJiuukwuj)^ ^iz/^u/g^/jjiJj^^ p. 20, ^k/?^/^, 

IJiumhijjiihfi^^ ^w^mwlpupwh x J > wpwq--\,kpub£, p, 140. The fryjvwhg k 

qtULpg ^nppnpij. Ifij^hijbiu^ ^mjn^^ Uiukpuigjig, JTuihwqwjgh, Iwpiutiwqw- 

Jgby npg jfclfkqbwij qwLum£^ fyuiphwriljg, Pwubhuirjlig "are likewise found here. 

We have included in our list only the principality of Basean, which from Antiquity is 

referred to only by the name of its territory, as is also the case for Goltn, as is evident 

from ^^111,9, "-.fetuwh Pmubhnj Ifmliim^, Ibid., Ill, xii, ",„&«>«/£ fofuwu,., 

^nqPmh ". The remaining princes came from outside Marzjpcm- Armenia, and may, 

furthermore, have been members of the Arsaeid, Bagratid, and other houses already 

listed. Sebeos, p. 93, mentions a certain " ^ph^ndi^ " = MX, I, xxx, " ^pwh- 

^nihpg " from the name of a vanished naxarar&om. [Cf. Toumanoff, Studies, p. 233 

n, 291, on the inclusion of Tayk 1 and Phasiane [Basean]; 251 and table xii]. (247, 1) 

25 ZG, p. 48, " }Tk{,hm.hkw 3 h wpfy ". BL, p. 70, " frwqimi Ifk&n diking, 

IXuwnLW&wwnLp Pd\&]ni$haiji U^^^jti 8wpk£iuuuh[iij ", Ibid., p, 42, 

" Jfniluiu Hwpp/jtuuhfa fottipujj JTbtfrnrfibwy ". Cf. MX, II, viii, " Qwpb£- 

uiLwhbwhh [j tjmpiffig pmij.muipwg ". (248, 1) 

a5a [Cf. Toumanoff, Studies, pp. 128 n. 226, 227 sqq,] 

25fc [The entire text of the Qahnamah will be found in Appendix III A. See below, 
n. 43 for a discussion of the problems found in its preface,] 

86 The Qahnamah was discovered by J. Akhverdov who realized the full importance 
of this document and transmitted it to J, Emin, who published it as a supplement to 
his Russian translation of the History of Movses Zbrenaei (Moscow, 1858). A photo- 
graphic reproduction of the MS of this document is to be found in Alisan's Ayrarat, 
p. 430. [Cf. Toumanoff, Studies, p. 229 n. 274], The following errata should be corrected: 
" ^w^ngbwh " for " IX^nybwh, ", where the — q has been shifted from the numeral 
Jjj.; " ITiuiipbpiuijfi" for " Swdpbpwgfj" ; and " IX^w^wpbrnh ", which is pro- 
bably a distortion of st JX>fumqmpbmh ". (250, 1) 
a6a [The entire text of the Military List will be found in Appendix III B.] 
a7 This document was first published by bishop Sah^atunean in Storagrufiwn Mjmiacni 
(EJmiacin, 1842), II, p. 59. (252, 1) 
28 TJyianes, I, xxvii, p. 40, " * • -npnlib^ ipqfiLwhh wpgnthji k inrnj^ qipu£— 
hmSwl^ hujfuwpwpwtjh £wjntj . , , fiulf hm qmbmj^ ijjpuihuiSmljh pbpuiL wn^w— 



/ 



NOTES : CHAPTER X 



445 



qwppwlf wpgwjh Zwjng *** A HJ^JU 1 *$ E um rifrn^t Ifwpqt A l^n^ ivbnLiuLu. 
uinm^liV Ufn-hkuitj uikp^ kpfypnpqb* JXuwkin Pwqpuiinndiji^ kppnpqh* JXp&pnihfi, 
wppnpqh* JftviijvinqnLhj}^ ^hqkpnpqh^ IfmSjiI^nhhmh A qjujl ku tyMpitt A £u/— 
urn mint jjitpw^mk^pLp ji£fvwhni.pkwhk »♦ (253, 1) 

29 Siejph. Orb., vi, p. 64, " **• pul^ qppwij A ui^jvwpijik ^npit iffiuijk 
l^m^nL^mhi • •♦ ^pmSmhmmmpu •♦♦ mpkLkjkmli l^nqStuhh qjitfuiuhh JJfuhkmg 
guwhkulfi fi^fuwhog^ £,jiLufiuujjfih l^nrjpmhh qpqkiu^fuh QniMupwij guwhkkpIjnL 
fj^fuwhog, wpkSmjwh l^ntjputhh *JnpqiiLUJijliti guwhkiffi fejvwhog, CwpwLWjJih 
l^nqpiuhli jiyfuiubh XJjiij.bqwwbh guuibbkpfynL fejuw'iiog : Qwju wjuii[£u ji 
q.mCbmSm1jp fj^fumhwgli Zwjng qnp \hnhq qpkwijy A uwIjiul ifjj tjnLtjwhl; 
\Xq-wpujhqknnijJih A \,hpujiujih ". The final remark does not refer to the four 
military leaders which are not mentioned in the Gahnamuha of Agat'angeios and JN"erses, 
but to the princes listed before them who occupied the first gahs. (254, 1) 

30 Lewond, '* tfjwwdni.p]iLi} ^knhqkmj ifhbji i^mpqwmkmji iwjng np jwqmqu 
kpkkjnj U*w£iJkmfi A ^jji//f hnpjihy p£ npiqtu A Ifwd npni^ opfihml^wL injipkgfih, 
ku in nin Lk^ wm^jii] Cwjnij ", Sahnazarean ed, (Paris, 1857), p. 11, Ezov's MS lacked 
the beginning up to chapter xiii. The Paris edition was completed on the basis of the 
Ejmiacm MS which contained the fall title, (254, 2) 

30a [TUhe question of the Anonymous History found in conjunction with the History 
attributed to Sebeos has been the subject of considerable controversy among Armenists. 
For Adontz's contribution to this problem, see, VV, VIII (1901), For more recent 
discussions of the problem, see Abgaryan, Sebeos, and Toumanoff, Studies, pp. 18 and 
306 sqq., who refers to this work as the Primary History of Armenia,'] 

31 Sim, Aparan. The investigators of the Anonymous History have failed to observe 
this fact which was noted by Ali san, Hayapatum, p. 62, It is possible that in the MS, 
the History of tiazar P'arpeci was followed by that of Sebeos, in the version in which 
it has reached us, i.e. with the Anonymous. Simeon Aparaneci then mistakenly connect- 
ed the Anonymous with the work of Lazar P'arpeei which preceded it. [See above, 
n. 30a], (255, 1) 

32 The famous house of the Amatuni is missing from the Military List. We believe 
that the principality of the Amatuni is to be found hidden under the distorted form 
IXSwui^nliji (or also ZwSwuwm-hkiuh) although the size of its cavalry contingent, 
200 knights, is far from representing the true strength of the Amatunis. [Of. below 
Chapter XIV, n, 63, also Toumanoff, Studies, p. 230, n. 282, 237, nn. 305-306,] (256, 1) 

32a [Arm. Qeogr., pp, 32-33/43-45, Of. Eremyan, Armenia, pp, 61, 109. Mazaz is 
the seventeenth district of Ayrarat, see, Ibid., pp. 64, 111, 118.] 

32b [Arm. Qeogr., p. 33/44; Eremyan, Armenia, pp, 82, 107, 109, 111, 116-118,] 

32c [See above n, 21, and below n. 32e.] 

33d [Arm. Qeogr., p. 32/44; Eremyan, Armenia, pp, 80, 117.] 

32e [MX, II, Ixxxiii. Of. Toumanoff, Studies, p, 232 n, 287 who objects to some of 
these identifications as incorrect.] 

32f [Sebeos, xxviii, p, 98 ; vi, p, 48, vii, p. 50, xviii, p. 65, Of. Toumanoff, Studies, 
objections, pp, 221 and n, 266, 233 n, 290 and 235 nn. 299, 301,] 

32 s [Of. Toumanoff, Studies, p. 232 and n. 286.] 

32tl [The test of the Pseudo-Gahnamah of Nerses will be found in Appendix III C, 
Of. Toumanoff, Studies, p. 229 and n. 273,] 

33 [The additional names from the Pseudo-Gahnamak given by Adontz in this note, 
but not in the text, will be found together with the text of this document in Appendix 
HI C] (259, 1) 



446 



NOTES : CHAPTER £ 



33a [Alishan, Hayapatum, pp. 227-229. See, Appendix HIJ,] 

34 Ibid*, p. 227, The princely families mentioned by name are as follows: Upbpm- 
ubmg^ Qwpk^uiLwhji^ IfmSJil^nhkmh^^ ^nLqutpmg^ finh^fuu (sc, J£n&ubwtj), 
UfiLubwtj, fh^uinLhhiuijy J\}np]unnnLubwi], Xtfjkkwtjbwij, Hp^wfymubwij, *?h/i?— 
umpwl^wumg^ JXfiwwnLubwg, ytupumgwij, nnp JJSpimm Pw^pmmnLUph ^jtubm^ £, 
}Jw£umnLubwtj, tflw}nritkwij, QiuqwIfmSihwij? ^mpmd-h\niSi]bmg^ ^utumuqw— 
gbwg, \fnl[wij s ^m^pbpnLubwn^ "^nnwj, $wjnij, JJiumunjii.ubwg^ JXp^mSnthbmg^ 
$W£pwy<ng> and a few more given according to their districts. Several houses 
are given twice: JXjl j/ 1 "^ XkdwmnLubwij^ JXu&kwybwtj^ JJjiiubmg etc.,,, which 
is also similar to the Galmamak, According to the assertion of Anastasius, all of these 
churches, ** ft up J^niuwwpilu ^wumwmbijfiu, pwjg J n l nl l u t 1 V£/w^f " an 
affirmation likewise influenced hy accounts giving Gregory and Nerses as the authors 
or the ones responsible for the creation of the Galinamah [Cf. Toumanoff, Studies* 
p,310n,32], (260,1) 

35 M qv, Kafanfa, II, lii, p, 324, " UJmuuug ijmunpljig*** qnpu fi pnpnj 
bprnubjuiju Huwuwimiwj mn ^m^wu Jfui^fjl^nhbiu^ wnbm^ nmwg *\ Surpri- 
singly, Yahan Mamikonean replaces Hamazasp Kamsarakan in this passage. 
At the end of the List of Churches, Anastasius mentions the kat'olikos John, who from 
the context must he John [II Gabeiean] o&, 573, the successor of Nerses IL This fact 
may perhaps explain the alteration in the name of the prince ; the Albanian historian 
mistook the kat'ohkos mentioned for John I Mandakuni [478-490], the contemporary 
of Vahan Mamikonean. (260,2) 

35 a [The last nine families listed are found in the History of Movses Xorenaei, but not 
in MX, II, vii, viii: ^wuwun., Ibid*, II, vi; ^jiiiwgubwu^ $pnjLufi,U 9 xlvii; U&i/?- 
fjLuji) IXnmUsnbmh, II, lvi ; fynifiubwu, II, bdv ; XfmSpl^nhbuth, II, Ixxxi ; *Jhi i?uju- 
pwlfiuby IX^fuujqwp, II, Ixxviii, See above, nn, 32e and f. Of. Toumanoff, 
Studies, p. 252 and n, 343.] 

35t> ISee Toumanoff, Studies, pp. 232-233 and nn, 286-291 and p, 245 and Table vii, 
for his objections to some of Adontz's hypothesis and his failure to make use of the 
material found in the History of Faustus.] 

36 Xorenaei was probably acquainted with the Artaseseans, In the Gahnamafa the 
Artasesean follow immediately after the npumvubm wpgrnSifi. According to MX, I, 
xii and II, vii, the office of npuunubm^ i.e* of " keeper of the royal hunt " belonged to 
the Varajnuni n&xarars, the descendents of Varja who had been appointed keeper of 
the king's hunt, ***** p fopwj n P un B m pg nLUGUJ B"* According to Xorenaci, 
Ibid., H, vii, this had taken place in the days of Artases, ***** wju nJiujj jwuiLpu 
IXpmm^jiuji ". Might this be a guess of the historian based on his reading of 
" npummbm wpgnLuli JXpww>tubwu " ? 

37 MX, II, lxiv, " *** bt wj£ np jiu£ Ifpuwhpwtpiju 
nJinqfiwSpg ^nptyfin £uiuwuimt wju Sfypwu* npg JfiwhtjmS whwqgfig £pu 
fi i[mpuju k wu&wiip bpkb^jig, k ifwuh hnpm ifipfympbiuu wwwbpwqdbiu^pun. 
($njuu\ npo /t *jnp£lfin k npg h ilbpfit bnnfiwulu, qjwnm^utq.nLh[jg wuwtjbjntj 
Somwunpwq ^Ifibufjg (read n&bmfin) k uwpifhg ^wjl^wnmug^ ndwhg h jbfywg : 
Qnp jm muntmSp lywmifbuijnLg, i jiu% np t[wuu jmjwhp n% j^fjfubjnj ilbq^ h /* 
jihinp i[muu wyJvwmnLpbwh ji i£wutnwl[lijji funju mm^ri^ bppnpn., qji h wu£w— 
umwmnkPfiiSih prnqfiwy' 1 Ifwpbjjm ji ubpgu mb^ pun. puutLU wugiuub^" ; 
Emin's translation of this passage [in his Russian edition, Moscow, 1858] is inaccurate, 

(266,1) 



VVh 



(265, 1) 
bpt wum kbpt 



h* 



NOTES : CHAPTER X 



447 



w*[8ee afove, n, 32.] 

38 -MT, III, li, " • ♦ • /7/7w^ ifuiutjkjpitju wujpqkhwi qlfhmjp^ ipmfimjhwg quim.hu 
bpfywgwhsliLpnqh qlim^kujwh juip^rnhfiu^ qwp&nLgwhk^ jj hnuw* pvjjq iffiwjh 
ft qw£ £wjphbfi n£ Ciuumwwkjj wjj^ ji JvnhuipC gwh qpwqmfiu fimwmqkmj^ 
hwfuwpujpu, fj Ijiupij. lipmuhpuiqm.hfitj upumpwumh^ fiulf qwqqh ZwfiwtpHuiqwj, 
np i mn^fi fiwifyl{nhhwh, ft i[kp fiiuuimqwhk^yy, qfj Ipupqhu^h nthk^ q4jjh — 
qkpnpq qw£ hujfuujpujpwqh Zwjnij^ It [i qfjmihp fiLprnfi qph]_ » ; 

[Of. ZP\ xxvi, p. 111. See below Chapter XV n, 54], • (267, 1) 

39 The number of knights at the disposal of the Kamsarakans and the Amatunis on 
this occasion is not without interest, In the Military List the Kamsarakans have 
600 -knights and the Hamastuneans (sc, Amatuni) haTe 100, or 700 in all. The coinci- 
dence is worth noting, [See above, n. 32], (268, 1) 

40 Khalatiants, Arm. Arsacids, p. 294, [On the problem of a system of precedence 
indicated by the documents, see Toumanoff, Studies, pp, 242 sqq,] (268, 2) 

40a [See, Christensen, pp, 393-394,] 

41 MX, HI, li, " ^muh npnj qnpqpm^hm^q qnpm qZuifiwqwuw Ipuqniuuqhu ft 
*[kpwj qopmqq ummpmtqhm y ifkp ipwfiwhwL, k mqqp qnpm mmqku qqw£ 
£fjhqkpnpq? iqwuimnj hwfuwpwpwijq* k qqhqu k qqmuw^hpmu^ qnp mnihm^ 
i hmfuhhmqh gnij Cwpij qnpm^ fyw^qlih qngw : JJuiw^u It qmmhu q%wuwipupwi} 
wqqmqq^ qnp Ifwnug japgmSijm, flnqnhu h puiq* mhhpl^hupnpmp dumwhqhi 
hngfrh Wjwqnpqmq. pwjfj qw£nuj ^jp^tulpuumg luwmmnj dfi wpwugku 
lupdrnhfj^ qji h ifhg fi ilkpm.fi qftLiuhfj £pwfiwjhqwg qpi>l ". Emin's transla- 
tion is unsatisfactory, (269, 1) 

4ia [Yahram IV was not the immediate successor of Ardasir II, See, Christensen, 
pp, 253 sqq. ; Erye, Persia, p, 283,] 

42 MX, HI, 1st, " •♦* p m J3 ! } flL XwLiuhkgnLuqku qhw* wwj_ Jih<i qkpl^muu 
qmju rifruijh* m{ m £> hmfumpmpw^h Zwjn$y nprn^u k m P 1 $ w L t jV*P mm lpi &■ 
ijmpkqmh ityh^k ywjdfi, ^ n pf iU opfjuwlpuLXpmSujjhugik umjutiqiuj, qji fiwpqwwhg 
iqwpuliljig uji Ipuipwunhh hrnhnLomMt uuiha mjhp uinhkjni^ hhs qhpwqwhtjoplh 
rftnijinjuk^ pum tjwfiwg jizpkwutj » ; fez. dkui, qji qwphnwijj; quimh mqqmlituhjih 
fifinj k gn, qSmhl^uthh ^wquiLnhfj qnpqunji Zpw£,wmm^ jd£ k n£ jjji.pm.fi 
kinkq^ qwp}tul[riLu]ih jwputjkjpi^ uihrnJi^ q£p p fywpq phlikijkiu£ pbq wjj_ Da/— 
fuwpmpiuuh pmku^[ty jjiprnfi mknnui^ fogh Iputfkuij^ npvq^u £wfiwnh hnpw 
IpuSmijpwlpubh L l^iufi qwfiwmniSiJiu Pwiphnj^ fi ^wpifi iqwinmnj h. jumwQw— 
fiwujibb qm4i [t uwnplimfjujjjib^ It fywfi qpkjd} npiq£u qwpgmSjfi qnpbml^w^-- 
m.pjii5ihfiw iwLwmwut}!; khnpnth qutuu^uj^ Ciufimhfimbml^ji puqwhm,pkwfip^*+ » 
Sahak asked for the return of their possessions to " afiwhliwuh ^mqmuihfj qnpqmjh 
Zpiu£wmuij) ", which Emin translated ts Gaaawon son of Hrahat ", Bnt Gazawon 
is none other than the prince mentioned earlier as haying been taken into captivity 
with king Xosror, and Sahak was beginning for Gazawon's son Hrahat in whose favour 
he had already interceded before with Ardasir [MX, III, li] Emin, misunderstanding 
the phrase, " npiq^u £wfiwqji hnpw Ipufimupwlpuub k Ipufi nwfiwwni.ufih, etc... " 
translated it to mean that Sahak, having previously obtained the abbrogation of the 
Kamsarakans' and Amatunis' disgrace, was now pleading for a Gazawon, who was a 
relative of the Kamsarakans, In reality, however, the kat'olikos was merely repeating 
to Varahran in person his previous intercession for the dispossessed houses, Hrahat 
was the son of Gazawon, so that a better translation of this passage would be, " ,„ to 



448 



NOTES : CHAFTEB X 



assign him {i.e. Hrahat) a place according to his [the king's] pleasure as his (Sahak's) 
kinsman or that of Suren Pahlawuni," to whom Sahak's speech was addressed. As for 
the Amatuni, " ,„ having deprived him of his hereditary honourable rank, at least to 
lower him from the first places to the lower ones, or to grant him, the Amatnni heir 
(and not the Kamsarakan one) an official position ", a hint at the position of hazarapet 
held by Vahan Amatuni in Sahak's own time according to Koriwn, [XIX, iv, vi, vii, 
pp. 61-62, 111, 113] and JSUse, [There is no mention of Vahan Amatnni as Jmzarapet 
in the text of ElUe, though his name occurs in several lists of naxarars, e.g. pp. 43, 99, 
193], (269, 2) 

43 [See, Appendix III A for the text of this preface.] The text is far from correct, 
and the opening words are particularly awkward. In our opinion, the phrase, "ft 
gwijnij mSunj d£" is a gloss which has crept into the text, and which originally 
referred to the document in which the ramakan nama was to be found, Prom this 
reference, this document must have resembled the later synaxaries, and the ramakan 
noma was included in it under the date 17 hahc along with other material. The word 
wkufj may also belong with this reference instead of " wku(fi) fi guiqnij mSunj d£". 
In both cases, the phrase " hu qnuiSml^mh hind in " from which the verb is missing, 
must be take as defective, probably as the result of an error of the copyist who in referring 
his reader to another document failed to complete the sentence. If we acknowledge 
that the words " qnp JXpmw^pp " refer to the preceding " qhiuiiui" the grammatical 
error will be removed and the sense of the passage will be that Sahak I saw " the famahan 
nama of Artasir " in the royal diwan, and not " in the diwan of Artasir ", Khalat'iant3, 
[Arm, Arsadds, p. 297 n, 3] suggests that the hi. is superfluous in the phrase " k qfjp 
wpiupfj ". In our opinion, either another epithet referring to the king has been omitted 
here, or, as is more likely, A qfjp should be taken as a lapsus calami for fuhqfip* Kha- 
lat'ians, [Ibid., p. 295] corrects ^w^SuipJih into -\wCidwph. Since the -^r- is clearly 
legible in the MS, and the space for a missing letter at the end is also visible, the reading 
^w^Swpph must be accepted as correct and attributed to the linguistic peculiarities 
of the document. Gf. qij.mihkiSujliph for qgui£tiki>wl[ft. The word hnjhujln in 
front of \,kpuk£ refers to the preceding clause. The qualification pmphpmp for the 
Persian king is likewise to be found in LP\ xlv, p, 261, " fhnpQurjf- ilk^ pwpbpujpnL- 
Pfirfaq Skp". Cf. MX, n, xliii. (270,1) 

44 MX, III, lviii, " puiQiuuip Ipnvjnujwuhpni^ ijJXpwiu^u ••* li ijinjuhw^ 
qmhttii JXpinw^fip ^a^Ajt?^ • * *\ (271, 2) 

44a [See above, n. b,] 

45 FB, HI, ix, "♦•• ijpaihpij pwp<ikp£ij mm£uipjih wpgmhfi+**"* Ibid., IV, 
xvi, " fj dpnui mmjumjj rj.w£m5j Jj dwS m.pwfunLpkwbh prnqplfth*** ", also, 
Ibid., IV, liv, [On the question of precedence, see Toumanoff, Studies, pp, 242 sqq.] 

(272, 1) 

46 FB, IV, liv, " oplhg £fih* qji pmqmuiph tywpujiij h pmqjaiunpi Z w J n g 
ft dftm.fi wmfumfi pmqpijih fi dfinLfi mu£nju x hul( wjh op hmfu %mnJl pmqfiml^iuhm^h 
PwqwLiipinijh) np mhqU £fih y ijpmfkftbijmSiy !fwpqkif[ih. £nuil[ jhmnj tjffif* 
imihhkrjnLhrj ft hkpgnj pn^npfth qjlp^m^wj pwqfiwlpuhh umhifib, • ♦♦ \,wfu 
wdkhkgkwh fippki. piuqdkijuih jfiLpwgwiisfiLp swifim., jbmnj wblfih pwqdkynL— 
gwhlfih i£wpj>iujfih Hp^wlf » ; (272, 2) 

47 LP\ xxvi, p. 148, " pwrniuSg kh • ♦• npg kh ujLiuq. gwh q^fiu k mu^m. k 
wuuppg k jiwro hli* np ft funhwp£ kh ", Mlise, ii, p, 20, " *•• fyn^tp qjuuqiufiu fi 



NOTES : CHAPTER X 



449 



qfihnLnputlpnb ^pfimnhlfitjli : fet fi dmfi qw^mh upuwpiuamnLpbujh qfiLpuigwi^jiLp 
uibqji fhnpQp ftnyw pwqfiwl^mhph ", Ibid., p. 85, " « ♦ « A quipqbjbm^ pwqfim- 
l^mhuh ft hntjmhj? fi mhijjj ^pwfiwjig fiwwsiLijwhb^ ", (273, 1) 

48 Sebeos, xix, p. 68, " • • • h £p htu bppnpij. hwjmupwp fi miu6wpfj pwzpuuipnL- 
pkmhh fymtpmfwj wpgwjfi". (273,2) 

49 Mov. Kalanh., II, i, p. 119, " qnpbbmj_ £my ilbbwiqlu uidbhmjft dbbwilbbwij 
jfihhujilkhfig inn£i/fiij w qui ww in n££fii ffjiupufiij ", [Cf. Dowsett, Mov. Dasx*> PP* 61- 
62 and notes for variants in the translation of this entire passage. St&pTt, Orbel., I, ix, 
p. 69, gives s * uiilbhuijh " for " jfih&iuifbhfjtj " in the corresponding place. The 
subject here is the seven Sasanian noble clans. The original may have had "jlu-P " 
whence came the reading fifth from p [Cf. Dowsett, Mov. Dasx*, p. 61 n. 3]. (273, 3) 

50 Mov, Kalanh., II, i, p. 119, " * • • pwdwlpui. A nuinif wnm^p fiLp * * ♦ ifbbmpbm^ ". 
The king honoured the nobles with a cup and a wreath. The comparison with Elisi, v, 
p. 115," Vff finmnftmjg qkqh h muuwljh h miuwft h qumwmiuhhnii iqiup— 
qkuh y tip fhnp^fi &bq jutpgndinLum " is interesting, (273, 4) 

51 Mov. Kalanh., II, i, p. 119, " qtfjiupufiij II qtywppiiwij qpbwlpiib nfw£[WLfiI[u", 
Here luwipiiLfilj is used as a general term to designate a true or hereditary prince (c/- 
Arm. upuppk = " stately, dignified "), and not a Pahlewi in the strict sense of the 
word, as this has incorrectly been taken by the translators. We have maintained the 
original term in our translation. [Cf, Dowsett, Mov. JDasx., p. 60]. (273, 5) 

5ia [Dowsett, Mov. Dasx*, P* 62 n. 1, observes that although " Andovk ... is often 
mentioned in 3?!B, ... the ... author's account of his feats against Sapuh (iv, 20) bear no 
resemblance to the present passage ",] 

52 ZP\ ban, pp. 360-361, " * • • jwl mjp pwpfiljh ufuhfi ". It is well known 
that one of the sons of the Marzipan Vasak was also named PwpfiJj, and it is difficult 
to determine the relationship between him and " Pnipfil} pwi. ". Perhaps this is in 
fact the Marzjpcm's son. [Cf. Dowsett, Mov. Dasx*, p. 63 n, 3], who questions Adontz's 
identification. (275, 1) 

93 Steph. Orh., I, ix, p. 68, ** -**jwuiv^nLwbng uppnjh fybuipnufj* Ufuhkiuif 
buifjufyniunufi. Ibid., I, x, p. 77, " ♦•• ft ftnjft £umfitjh $bwnb tybuipnufi". Ibid., 
p, 78, " ♦ • ♦ npu^lu qkfynjg tlbq fj hbppnjjfiwltjh fu.pm.fi bputhb^fift tflbmpnu 1 
yfiLtikwrj biufiulfniqiiUy gbppnrpn£opft J/nifufiufi a^mlfhpm : " Ibid., p. 79, 
" • • • jhpmhbjLnjh tyfcmpimf? XJfjihbm^ biqfiuljniunufj, fj hbppnqfimhih umwgy qpp 
wpmpft ft ifb/$ Putpfilj, qpp k *Hwpufil[g jfupbrnhu jbqnth pun. fyn^ifih qftw ". 
J, Emin, the editor, regretably removed this valuable commentary from the text and 
put it as a note at the bottom of the page, in contradiction to the evidence of the MS. 
[The Sahnazareane, Paris edition used in this edition gives the commentary as part of 
the text]. (276, 1) 

53a [See, Appendix III, L, iii, for the list of the participants at the Council on Dwin 
of 555,] 

53*> [IB, IV, xx.] 

54 Elise, i, p. 6, " fei fippb tfiw hu tlbpdbwg Ji pwt}.WLnLpbftf;ft, fj hwfuw— 
pwpuft ZwjnQ mhl^mh^p puJULwuipntPpihii. qfi piu}£m h qiuh&ft jwpgmbfiu 
tyiupulpug bpPwjp, uuil^mjh wjpnjL&fih Zwjng pnifmftrj.aj^ fi &bnb hmfuiupm— 
pwtfli wntu£iwpq.£p fi iqwuibpwtpffi ", For the ceremonial meeting and reception 
of the Armenian cavalry at the Persian court, see, Ibid., ii, p, 44. (278, 1) 

55 Ibid., ii, p. 22, ts •♦• m^fumpCtuqfip mnhbj^ wilbhwjft bpfypfih Zwjng fj 
PnipuPfiLb {jnplpwt} h fi pbpbinLpfu.h bwhpnLpbmh wjpiiL&fiiijft ". Ibid., vi, 



450 



NOTES : CHAPTER X 



/ 



p. 131, " PnqrtL^ £pwiluij£p tj^wplp] w^fuwp^Jih. h ijwjpmAfiu Lu tpupgnLuJi 
pkpkuiijnjcj um dmSmhm^ Jj] ". (278, 2) 

56 Lewond, x, p. 33, " *"wbym.ijwhk^ ji ^wiiwpm. wpgnLhji L wnbnL^ £iwq.'\ 

{278, 3) 

57 Mise, iii, p. 85, " fei. qCmmkwj^ nn£fil^uh l^mpqip wuqpih fiLpwgwh^Lp ", 
Ibid., p. 196, " l^wpqkwg hnrjiv nn£ji^ L qupuuipujmjwLpfiLh uiuumwiiJiunLpkwhh 
^pwSmjkwg jwpgmJmLuin ", also, Ibid., p. 199. (279, 1) 

58 FB, V, xxxiv, " * * • qimfkEiwju tmjwmuh Zwjng l^wjukpuiljmh pnvuiliogh 
qjthninpki" (279,2) 

59 Sebeos, xxxv, p. 138, " ••« mmS kkq WLqhnLpfiLh (nnqfj L fvnpijjLlj) [i 
iJjiwufjiV kpt ^°^l uipdwh Jigi urnhki". (279,3) 

60 Ibid., xi, p. 56, " ••• l^wpqk^ hnijui iw^jil^u jwpgnLunLiiw". (279,4) 

61 Lewond, xvii, pp. 100-101, " • •• w^JvwpCwqJip wn.uk j^ pun. ui^fuuipCu Zwjn$ 
ifwuu buihpwynujwhkipj qmhnLp jbnj buinwjnLpkuiu CuipljwwpnLpkuiu », ", 

(279, 5) 
63 Ibid., xxi, p. 113, " ••• jkpfin aiding L muqp uipq.kikw^ lp njiuifuwpuipuiiju 
Zwjnij h bntjjiu Ckbkjnijh qjvuibfjpuijuh ". 

{280, 1) 

63 Idem., " "+U1WJP Cpwduih ^n.b^fi &nnu unpui wiffi imJfi £n oppjin wSuiyli ". 

(280, 2) 

63a [The reference is not clear, Adontz's text reads " Abdallakh (750-775) " following 

tewond's form Upmuij. This should he a reference to the first Ahbasid caliph Abu'l- 

Abbas as-Saffah, but his rule extended only from 750 to 754, when he was succeeded 

by al-Mansur who reigned until 775, Of. Lane-Poole, Mohammedan Dynasties, p. 12.] 

64 Lewond, xxviii, pp. 128-129, "jiujhj}£kin£ £uiwuil uuilj wpbuipnju, np n.iujp 
mSfi uaJfi jwpgnLUnLum nopuinu Zwjng : fci n^uiSuip £kbk[nnh iqui£uiu£fliu 
ifejuuiuuiiju* k iuipl^ ll*h£p ft inwui] fupkuihn £uiun.kp&k^ nqniSiqu nopuinu ", 

(280, 3) 

65 Sebeos, xxxvii, p. 147. (281, 1) 

66 Lewond, viii, p. 23, " • • . junp^mpn. ijmm Ji 3^ wnkuij^ puinhuij^ nwnw— 
inwfunLiip wn^Su jui^fuwpQu Z^Jjnn £uiuq.hp& unnfiu £kbkjnu[g ". Ibid., x, 
p. 31, " ^iffp*** funp^bnuiL puinuui^ jui^fuwpQu Zwjnn nwn£d hwjvwpwpwtj 
ungfih £kbkimlg". (281,2) 

67 Ibid., x, pp. 31-37, Theoph. Conf., i, p. 372, " ... rovs Se peyivravas rwv MpjueviW 
owpzvaas ivl tqttw £vi ^moKavarovs iirQirjaGv ". This event is also known to Michael 
Syrus, II, p, 474, (281, 3) 

68 Lewond, x, p. 35, "••• n£ pmtkS dfi pum dfin^l inwindh^; QLnuuj qmilhhkuhmh 
pwpkkwj_ fi l^khmg 1 whdinnwhq wnhljih qw^JuwpCu Jj hmfuiuputpiun : 

Swfifi duifiwhwljli pwijjntp kqkwj^ w}fuwp£i] Zwjng ji uin£i)l hwfuwpwpwtf 
dwinu£]ih npiu£u qn^fuwpu ji S^ ^JIPS '* (^81, 4) 

68a [See, Toximanoff, Studies, pp. 234 sqq,] 

69 Indeed, we should note the words of FB, IV, ii, " • * * L £hwiiwbq£p qtJkb^ 
wi/kbub, qjiLpwgwh^fiLp qopu pwdwukiuj^ juii/kLwjIi l^nijpwhg^ uiu^Swhwijh 
Zwjntf uwCfiwuwinuiXu IfuignLijuJutp ". [Cf. Tonmanoff, Studies, p, 235, and 
below Chapter XV, nn. 86-87]. {282, 1) 

70 FB, III, vii, xi, xviii ; IV, ii ; V, i, xxxvii, xliv. [#ee below Chapter XV n. 29a]. 

(283, 1) 



NOTES ; CHAPTER X 



451 



70a [Ob the bdeas%s or wto;^ me below, Chapter XXV, nn, 39-40,] 
7013 [FB, HI, is, " •♦• pqkwyjuh, np ip lift ft wpufttj^ n.w4kp£ij pwphkplij 
mw£wpfiu wpgntSift ". Cf. Toumanoff, Studies, p. 176 n. 115,] 
700 [See above. Chapter IX and n. 6a,] 

71 MX, II, liii, " pmdmhi IXpwm^u k nJr£fuwuni.pfjLU qopniSj phn. gnpu^ 
qwpkkjkwhh PnnnL ft tjkpmj JXpinwLwqifiiijy k nmpkL^mkwhu mwj $fipw—* 
bmjy n^mpmtwjfihu ft XfSpwm £wiwwwj k nifjiufmuijfthh ft QwpH ". (284, 1) 

72 ZG, p. 43, " l^nnpumu^m^ njr^fuwhu JXip&hkwzj inpkg £wqap wpiuSp ... "• 
Ibid., p. 46, " h iftnqnnu pnnkm^ tqwtnkpmni/fth ifprnbu kp![nL (var, ni.p) 

k qnuwiqw^u mknkutgh npwn.pwmni.hkwnu ft^fuuthh l^mnnLnwtiip : ftufy nm^ 
plrij qopmh wivjp ft &knu fefuwhftb JJfttSikwij k ^w^km!} pMj ft &knh Jr^fuwhfiu 
XfSiakmnwh x fiuk nnm^kn^g Jtyjumhuh kiuqnjg ph^hwupn^u [up ". (285, 1) 

73 MX, II, viii, " ♦»♦ k tjwpqv njtunwpm.pkwun k untjm.hzj mnw^fthu k 
kplfpnpiju k kppnptj.il k np ft fywpqftu ", (285, 2) 

73 * [See, Toumanoff, Studies, pp. 236-241.] 

74 Agaf., exxvi, p, 644. Trdat set out for Home, " *** pmnjihS ilkbuufkbog k 
kprnhmumh £wnjupwi. phmftp nopmh £whn.kph. ". This statement is repeated 
in the famous dogmatic letter of the Armenians, Sebeos, xxxiii, p. 129. This is an 
important factor in the critique of the texts of both Agat'angelos and Sebeos. [On 
Agat'angelos' problems see below n, 89a, on those of Sebeos, see above n, 30a] (286, 1) 

7 5 FB, in, viii. (286, 2) 

76 Ibid., IV, xxvii, xxiv, xxx, xxxii, xxxiii, xlviii, xxviii. We also find purely 
legendary figures such as the 400,000 knights of Ibid., IV, xx, or the 600,000 in Ibid,, 
IV, xxi. (287, 1) 

77 Ibid., V, ii, iv, v, vi. (287, 2) 
™ Ibid., V, xxxix, xl. (287, 3) 
79 Sebeos, i, p. 24, " £ £wnwp pbwftp ^wiiA/Jiy ". (287, 4) 
so JoK Mam., p. 13. (287, 5) 
si ZG, pp.40, 43, 46, [On the date of Zenob Glak, see Abelean, I, pp. 345-362], 

v (287,6) 

83 Procopius, Fers., I, xv, 1 [L. I, 128/9], " „. to 8e urpdrevpa tqvto iJepoapftevtW t€ 

kclI Uowit&v ^aav, ot By > AX(v)avots slat opopot". On the sise of the army see, Ibid., 

xr, 11 [L. I, 132/3], (287, 7) 

83 Sebeos, xxxy, p. 138, " Chbbwi ^w^ jiu^fuuipfyn. dt l^mnrnp ". (288, 1) 

84 Zewond, xxt, p. 120," •♦♦ biutu tummpftljh Zwjng ji Pftfymhu onJtwlfwhnLpbuih 
k nifcft phq. fup ppwfrp Akbkjnq dkn wptuhq l^wnkwy ", At the time of the 
rebellion of Vahram Choben, the Armenians supported Xnsto II and sent 15,000 men 
to his assistance, it is evident from Sebeos, however, that this was not the whole of the 
Armenian army, bnt only a part thereof, Ibid., iii, p, 36, " npg ft dwtimh wn&knh 
ifrnkiui". (288,2) 

85 Sebeos, iii, p, 42 " \lfiuiifil{n&kujft] wnSini. phmfip mpu hpfynt imnyjp ", 
Ibid., vi, p*48, " ljnpfunnni.hft ♦♦♦ ^mi/intSifi *** IfmiSfil^nhkmh \k tiijjfi] *** 
Ckbkw^g ftppk kpfyni ^wnwpj)". JM(?., x,p, 5B, u lfft fanjup ijIJuj£wIi JTtnSftl^n-' 
ftkutb, h tlfi imnjnp ft hbnh JJiSpwmmj PrnqprnmnLULnj npn.wj Xfrnhnii^p". 
Ibid., xviii, p. 65, " JXpbpntSift * • * k wjlg ft hwfuwpwpwijb : fei. ^"tLp^ hnpm 
ftppk kplfni. £wqwp Xkbkw^***". Ibid., xxx, p. 107, " Ifm^kn^ tTimffilinukwh 
npn.fi ^utLpft quiLpwifuipft Zuijnii hpho ^wnjup uwwnwnJihwLg ", etc. In FB, 



452 



NOTES : CHAPTER X 



III, xx, the marzpan of Atropatene bad 3,000 men, Of. ^uiuwlfig, with 3,000 men in 
the Military List. (289, 1) 

86 FB, III, viii, " • •• quip opluu* ujp dhbwifhb wLwn.uiujiu, uuifuiupwpgh 
wyfvwp^wlfiu^gh w^juwp^wmkiupgu* np Ijih pfiLpuiLupgli L CwqwpuiLnpgu 
Ifuijahh urn wpgwjjiu ". (289, 2) 

87 Sebeos, p. 36 (34), " wugkw^g ji {ujuqfiufj, [ippki iuttjwpg £ua.hmwuwug^ 
JjLpuiguihiJjLp qm.urj.g hwfuiupwpiuij pum £iupfiLpwLiipiug* pum ^wq^wpiuLnpiuij* 
pum qhqfy* pum qprni^mn fiLpbwun ", The passage is found also on p. 34 as a 
result of a confusion in pagination. (289, 3) 

88 LP\ xxxvi, p, 209 [Of. above, n. 11 and Toumanoff, Studies, pp, 246 sqq, and 
tables viii, ix], (290, 1) 

89 Thus, for instance, the duplication of names some of which are mentioned by the 
historians; Sebeos, xviii, p. 65 knows two JXmm^nLufj, LP, xxxiv, pp, 198-199, and 
Mlise, v, pp. 99-100 distinguish two Dimak'seans, the representative of one branch 
being named P^mpULj^ and of the other ZSmjhmlji Mlise, v, p, 100 also has " iffim hu 
mjj^ %wnjifil^ ^fjjlwgukwu ". We have already noted two branches of the Arcruni 
family. LP\ has two %JiLm ^mOwLuji, one in Vasak's camp, xxxvi, p. 209, the 
other with Yardan in Albania, xxxiv, p. 198. Incidentally, in the same passage of 
Blise, p. 100, we read in one case qm.uqu instead of Qjum [cf. p. 100 n. 23.] (291, 1) 

89a [See above, pp. 215-217 and nn. 49-51, and Appendix III D. Even though he was 
well acquainted with the various redactions (Armenian, Greek, and Arabic) of the work 
attributed to " Agat'angelos ", which he cites repeatedly and discusses in this section 
of his book, Adontz refers here to this work as a single unit rather than as the composite 
source that subsequent scholars have shown it to be. Indeed, many of the problems 
and some of the solutions connected with the enigmatic work commonly known as the 
History of Agafangehs appeared considerably later than Adontz's study. For the 
problems of the " Agat'angelos ", its connexions with other sources, the various compo- 
nent parts of this work, new versions, and the relevance of these questions to Adontz's 
discussion, see Garitte, Agathange, and later studies in AB and Le Museon, as well as 
Toumanoff, Studies, pp. 16, 159-166 with their notes, 243-244, 307, 458-459 n, 98, and 
Ter Lewondyan's, New Arabic Version; also List A, pp. 159-161, List B, pp, 161-162, 
and p. 244 table vi, Eor the sake of convenience, Adontz's terminology will be main- 
tained in this edition, and the documents of the " Gregorian Cycle " will be referred 
to as Agafangehs.) 

90 Mov. KalanTc., p. 182," IXuq. mkumlj qjpmaSngu[hngm] ", Ibid., p. 189, " few *i^Jipnf 
IXq/iLiuhfin ... IfwPnJifjlfnu". [Of. Dowsett, Mov. Dasx>9 pp. xviii, 92-93, 103, 
229], (292, 1) 

"a [S ee Appendix III, E,] 

91 Had we not possessed the Arabic version of the List, this fact might have been 
explained through the influence of the second list of princes in Agafangehs, [cxxvi, 
p. 643], where the four bdesx? are listed separately at the beginning followed by the 
princes of JXuqhq^ tntiLU ei al. In the first list, the prince of Angeltun is immediately 
followed by the bdesx of Aljnik', as a result the author of the story included only three 
bde$x& into the list, so that the prince of Siwnik' might consequently have found himself 
in the 14th place. The fact that the bdesx of Gugark' occurs twice in the list since he 
is also the bdesx of Maskut found in the first list, will have to be explained in terms of 
the author's ignorance of their relationship, [Of. Garitte, Agathange, lxxxviii, p. 72, 
and Toumanoff, Studies, pp. 183 sqq.]. (292, 2) 



NOTES : CHAPTER X 



453 



92 [Armenian] Agaf, cxii, pp. 590-591. Ag., $ip. 68-69. [Cf. Toumanoff, Studies, 
pp. 159-160, also Appendix HID. The order of Ag. has been altered by Adontz to 
reconcile it with the Armenian list.] (293, 1) 

93 Va,, p. 115, [Cf. Garitte, Agathange, lxxxvi, pp. 72-73; also Appendix III D, ii]. 
The MS has ^s^ ^^ tne article J | . Marr Christianization, p. 202, suggests the 

reading ^^J-| in which he, incidentally sees, the Greek yvr/t, a translation of the Arm. 
" uihqq " = " kite ". In my opinion, the first syllable is not an article, and the whole 
word should be taken as a misreading of { yX^\ Anjelin = Angelene, (293, 2) 

94 The MS has 4- *^b* |, which is undoubtedly derived from (^dij I or (J&a1?j I* 

(293, 3) 
9 * The MS has <j bUj I pro Q b ^ [*]. (293, 4) 

96 The MS reads, ss the prince of the Aspeis, entitled Aspet to whom was entrusted 
the guard of the qwsywn and mtznywn mountains, (JU^^^aJ I (U^» JsLa^ j w L* 
iJa-J jJaJUj- [Clf* Appendix III D ii for Garitte's translation]. The qwsywm 
mountains are unknown in Armenian literature, and the word mtznywn is completely 
incomprehensible, This phrase is undoubtedly a translation of the Armenian original, 
" umCSuthmljiui^ jiupht^wjiq I^ntut ". (Cf. MX, II, iii) where the Greek translator 
has mistaken the last word for a proper name, and the Arab translator has confused 
the Gr. opos = uwCdinh with opos = " mountain ". As for the word l)j*1 \JaII 
it is probably a distortion of (J a.,o ytll " west ". (293, 5) 

97 "Prince qmrdl near the qrdytn ". J3 ^J may he a distortion of the Arm. 
^p£hg. ^ (293, 6) 

98 The MS has Qj^A\ pro Oj**£&l (293, 7) 

99 The MS has <jj | >a^o = SeerTroretcSv or ^eairoTwy corresponding to the Arm. 
ZaaiTiavoov of the Greek text according to Marr. But 2^aa7tiav€>v = Arm. ^w^wu^ is 
found in another list which is not found in the Arabian redaction. In context, this 
word corresponds to the Armenian tywuftfiij of which it must be considered a distortion, 

(293, 8) 
99a [Arm. Geogr., pp. 30-35/41-46.] 

100 Sebeos, xxxiii, p, 121, "Ijnuljimiuu Jfmil^ljnhi^ kuifiulfnujnij k Ifmmpinu U&«— 
mndthuiq ; flrfij-fih iqmmpmumuil^wii uihq piq. Jili^bwhu ifflfip uppnjh ^pfiqnpjt " : 
[On the date of this Council, see Garitte, Agathange, pp. 351 and n, 4, 353.] (295, 1) 

101 MX, II, viii, " fct jiupbtk^firj fynqjlwiu£ tjkijkpp £wjl{w l^wh fuoujig 
linrjpjtiml^mjij^ qhplpiL fjkqfiij hm^miqhinnLphmhijh uftuuiljhmh L l^mqifhmjh^-** '". 

(296, 1) 

102 JUd., I, xiv. (296, 2) 

103 Ibid., II, viii, " phqqkii jhpjihh *JujlIiiiwwj*** f^^fa qnLqwpwyuig", (296,3) 

104 jud„ II, viii and iii, (296, 4) 
i° 5 Ibid., II, xxiii and xxx, [Cf. Toumanoff, Studies, pp. 201, 295-296, 318-320]. 

(296, 5) 
iosa [Khalat'iants, Arm. Arsacids, pp. 294-295.] 
106 [MX, II, lx, lxiv]. (297, 1) 



454 



NOTES i CHAPTER XI 



CHAPTER XI 



a [This entire chapter is based primarily on the Armenian Geography, the names of 
the districts discussed should be checked in all cases against the more recent works of 
Eremyan, Armenia (particularly the classified lists on pp. 116-120), Hewsen, Armenia, 
and Toumanoff, Studies, especially, ii, " States and Dynasties of Caucasia in the Eorma- 
tive Centuries ", pp, 147 sqq. See also the Bibliographical Note, and Appendix IV B-C 
for the two versions of the Armenian Geography,'} 

1 [Arm. Geogr., p, 33-4/45], The short version of the Geography has 20 gawafs, the 
additional ones being: ** ^w^wm^ Jfujuhtu^ nwh^ 1X>nggi Ifwqui^ ", Note that 
this version has the better reading ^w^iuuiLUjig for ZuiunuhfiD, [Arm. Geogr,, 11, 
p, 366/7. The Armenian text speaks of 20 gawafs although it actually gives 19, Saint- 
Martin's translation acknowledges this discrepancy " L* Ararat ... contient dix-neuf 
cantons ... ", Ibid,, p. 367. On the two versions of the Geography and their problems, 
see Hewsen, Armenia, and Appendix IV B-C], {300, 1) 

2 Xen., Anab., IV, vi, 5 [L, II, 60/1], " Qamavoi ... "; Ptolemy, V, xii, 4, p. 938, 
" 2ipfucqy$ ... ", V, xii, 9, p. 947, " Baypavavfyvy ... "; Strabo, XI, xiv, 4 [L, V, 320/1], 
" 'ApafyvovTTGhiQv.,." =:bpuiujuui&np; MX,TX,xc. (300,2) 

2* [FB, 3H, xi, xii; IV, xix, etc.] 

«> [Toximanoff, Studies, pp. 132, 137, 201-202, 209-210, 218, 241, 309, 321, 324.] 
20 [Ibid,, pp. 132, 171 n. 90, 202, 206-207, 323-324 n. 81 etc. Of, FB, III, xi.] 
as [Ibid,, p. 204, and 230 n, 278.] 

3 AL, iii, p. 35, " ♦ • ♦ ji uiqwwiimbh* ftwuklmj^ fj uikqLn^h np l(nsh JJwignpuij 
... ", [On Sirak, see Eremyan, Armenia, pp. 73-74. Soragyal cannot be found as 
the name of a district, although it survives in the form Bassuregel, the modern name 
for the village of Sirakawan, Ibid,, p, 74 and G 46, p. 80, In the same district, the 
village of Suregel, G 46, p, 578 is likewise still extant; the Arm. Atlas, p. 107 seems to 
give Soragyal as a district rather than a locality. On §irak in general, see Alisan, 
Siralc, On Basean and the village of Salk'or, see, Eremyan, Armenia, pp. 44, 79, etc.], 

(301, 1) 
3a [Arm. Geogr., pp. 33-34/45, see, Appendix IV B for the text, (7/., Eremyan, Armenia, 

pp. 31, 46, 62, 79, 118, etc.] 

3 *> [See above, n, 3 for §irak, Basean, Vanand and Suregel, Ta^tin is given by Lynch, 

Armenia, map in the form Takhtin and apparently survives as a toponym, though not 

as a district in the village of Tahtakiran NW of Kars, G 46, p, 581.] 
3e [On Bolberd, see above, Chapter I, n, 40 and Eremyan, Armenia, p, 45, Toumanoff, 

Studies, does not give Bol as one of the Kamsarakan possessions,] 

4 Only the short version of the Armenian Geography lists the district of Asoc in 
Ayrarat; it is no longer mentioned in the longer version [Of, above, n, 1]. Georgian 
sources read Aboe as the result of a confusion between y and y. See Alisan, Ayrarat, 
p. 127, [Also Eremyan, Armenia, p. 36 and Toumanoff, Studies, pp, 185-186, 190 and 
n. 199, 323-324 n. 81, etc], (302, 1) 

4a [The list of gawafs given here follows the short and not the long version of the 
Geography. See Arm. Geogr., II, p, 367, Of. Eremyan, Armenia, pp, 64-65, and 
Appendix IV C] 



4^ 



NOTES : CHAPTER XI 



455 



4to [On these districts, seeEremyan, Armenia, pp, 38, 49, 60, 64, 72-74, 76, 82; Tou- 
manoff, Studies, pp. 197-198, 204-205, 222, etc., they are to he found in AA, pp, 7 107.] 
4e [Ptolemy, V, xii, 4, p, 938; Sebeos, xxxii, p. 114,] 
4d [WHse, iii, iv, vii, pp. 74, 92, 179.] 

5 MX, I, xii, and II, xi, " ♦** /i//z bqkpp Zpmq^mhjj^. [BL,j>. 74, see, Appendix 
HI for the text]. (303, 1) 

6 In western sources we find D**"/^*^ = " Zachunuc " Saint-Martin, Mimoires, 
II, p, 287 n, 24 ; Alisan, Aymrai, p. 260. [The district of the upper 2anga is still called 
Daraehichak hy Lynch, Armenia, Map, hut it has now reverted to the Armenian form 
of Calkuni according to the Arm. Atlas, p. 7. Cf. Markwart, Mnistehung, p, 43 and 
Honigmann, Ostgrenze, pp, 193-194.] (302, 3) 

' MX, II, iv, " **+qbmlk^hpkujjuh dwhljm.'hU) npg p Qhqwiluij k [j gmhw- 
hwtji.nij* + * " (so. Gnt'uni), cf, Ibid., I, xix. (303, 3) 

8 Alishan, Aymrai, p, 248, " gu tub jfehw fflpfynp fyhpnilikwij m£p ", [Cf, 
Toumanoff, Studies, pp. 204-205 and n. 233, on the Aparan inscription], (303, 4) 

* LP 3 , xix, p. 113. (303, 5) 

**> [MX, in, ix [Of. PB, III, Tii].] 

10 MX, II, lvii. [On the origin of the Amatuni and their possession of the fortress 
of Osakan, see, Toumanoff, Studies, pp, 197-198 and n, 223, 229-230 n, 110, also below, 
Chapter XIV n, 63]. (304, 1) 

10a [See below, n.74 and preceding note.] 

11 Sah^atunean, Storagrufium Pjmiacni, II, p. 46, "jwilu mkmnh ptn^npnufi 
fybmjbkuiij kn^uf* ^fibhijWL mniSiu * ♦ * " ; Alishan, Ayrarat, p. 135. (304, 2) 

12 Nerses, p. 15, " • ♦♦ iffi uifii Ji tfkb hmfitrnprnpiugiU fj quiLwnlh Cjipmlimj np 
lp iljiwhahwtjkw^ fj jkpfihuy jm^q^h ty&mSjkuiij ", (304, 3) 

is Tov. Arc, HI, xxvii, p, 247, and IV, xii, p, 308, " .„ Jlp&i J gwyutg ^hnihkmg 
quiLmnfih", (304,4) 

14 MX, HI, xxiii, PB, [IV, xxiii] gives pzw^ioiy^ini/ as a city, whereas Xorenaei, 
he. tit., speaks of it as a mountain not far from Kogovit [Of. Eremyan, Armenia, pp, 5^, 
85, also Huhschmann, Ortsnamen, p, 457 and Markwart, Sudarmenien, p. 560], (304, 5) 

i4a [On the Gnuni and their possessions, see Toumanoff, Studies, pp. 205 andn, 234, 295, 
298, 453 n. 63. On Xorenaci's derivation of their name, Ibid,, p. 206, n.234. Of. 
MX, II, vii,] 

!5 MX, [II, vii] asserts that the name of the Qpthwl^mh came from their function 
which was to supply snow to the royal summer residence. The name is more correctly 
derived from the " snowy lands " in which the Jiwnakan dwelt. Their home was 
prohahly on the slopes of the Aragac in the vicinity of the Gnuni, The etymology of 
JJuiujhqni.hjj from uiqwhq * meaning "slaughter house" [supervisor of sacrifices] is 
equally duhious, JJu^ahq may he a contraction of Xjupuhii.mpmp a name favoured 
in the Kamsarakan family, and the Spanduni were a hranch of that house, [Cf. Tou- 
manoff, Studies, pp. 220 and n, 259, 221], (305, 1) 

16 [Sebeos, xxxv, p. 139 ; MX, II, viii, Iviii]. The form HjiuiLhiikuih helongs together 
with JXpkqkwh, Qwphqhuilt, and prince Af awelean may have lived close to them on 
the horder of Sirak in the locality of Osakan (now Asnak near Talis) which is still familiar 
to Seheos. UnuiLwh-biuh is the popular form of JXpmwLiuh-biuh^ and Artawan 
[Artahanos] was a common name in the Arsacid family, The Arawanean may 
have heen descended from it. [Cf. Toumanoff, Studies, pp. 199, 215, 305, etc.], (305, 2) 



1 



456 



NOTES : CHAPTER XI 



17 Sargisean, Itinerary, p, 166, " fi u&k flwC few U3iqww<* + bin ml njiS 
qknu nJJw£umnLhfiu fi tfhp £wbq.uinuipwliu ^mnjuuipm^ fi mpbnjspw^n^ml^ 
ifiuuu ^/mflifoz/^ ". Alisan, Siralc, p, 31. (306, 1) 

is Joh. Kat\, xix, p. 103-104, " ifowLfip JJw^mnnJjp fyiiLpwiiiuiiiwin^* fi^fuwu 
Zwjnij*** ^fihlp fwl[ hlfkqhgjili qbqb^mjmpdwp x np ft gwnwgwn.bnh Ifpbu": 
[Cf. Toumanoff, Studies, p. 214]. (306, 2) 

18a [Arm. Geogr., p, 34/45. Cf. Eremyan, Armenia, pp. 35, 42, 56, and Toumanoff, 
Studies, pp. 137 and n. 240, 201, 209, 218, 241, 309, 319, 321, 324.] 

19 hP\ xxxi, p. 186, " tfiuqlinunfi ", lxxiii, p. 266, " ftmnJilnmh ". Asolih, 
xxxviii, p. 266; xliii, p. 278, " tywqfinjnmh "• The more correct form is \?wnl[lnmh 
< btuql/iaj — nmh, " the foot of the flowery mountain ", cf. the present Ala-dagi, 
" the multicoloured ", or " pied, mountain ". [Cf, Vita 3b. OsJceanc, p, 60, also Hubsch- 
mann, Ortsnamen, pp. 435-436.] (306, 3) 

19a [Eremyan, Armenia, p. 59 ; Toumanoff, Studies, pp. 200, 202, 321-322 and 322 n. 77, 
342-344, 348; as well as Hubschmann, Ortsnamen, pp, 342, 441,] 

20 ZP\ lxxvi, pp, 452, 457, " * ♦ ♦ Puin.wunh fi hm^mhqfih Prnqpluiibqiiij, fi 
jw^iulinqfiwh ippfihu *np wiinnuhfi \,upuin, Sow fi uriLpp mkqfi Smmptuhh 
npuilu kptym. ^ajppiiljhgo^ ", PB, V, xliii, " ft pwqphuuhq quiminfi fi 
Pimfnunh WLiuufi, np I Sow jmtkpuil^uh QiuphCwiwhn. gwqwgfi *** Smfih 
wnuipft uiiLpp nnijbpmnh 8nlj£wuunL, np l{wjp jmjhd qlui ". (306, 4) 

20a [See above, n. 19a,] 

21 LP\ xxxi, p. 186, " fi n.WLwnu *np fyn^fi Xymnl^nLinu^ ifbp& fi pkpqb wdmp x 
nnp lXuq,nti whnLwhbh". Ibid., lxxiii, p. 428, " fi tyuinjj£nuiu * + * ft ^bpSIffiu* 
np fyn^fi ^nup^mljfi ", Angel is mentioned by Procopius, Pers., II, xxv, 15 [L. I, 
482/3], " MyyAcov ", by Sebeos, xxii, p. 74, " fi tyuinj[nmw&lti fi qf**-nb np tyn^fi 
JXbqnh "; so also Vita Sb. Osheanc. [Angl of Calkotn should not be confused with 
the great southern fortress of Angl-Karkathiokerta, see next note], (306, 5) 

22 MX, II, lxii, " nwLinhu Swmhnbu ". [Cf. Eremyan, Armenia, pp. 35, 85, 
Toumanoff, Studies, pp. 310 sqq.], (307, 1) 

23 [JFB, IV, lv, lviii], " The birthplace of the founder of the T'ondrakeci ", " JJfipww 
ft l ntiq.pwljtijij umw^fihu fi Qiupb^WLwh ^bq^i fi Xfwnjpnmh n.iuLwnf; ". [Asolih, 
III, iii, p. 160, Cf. Gregory Magistros, Letter to the Syrian Kafotikos, p. 153-154, 

" XfSpmm wju* + * mnwun gui£whwjni.pbwu fykpujfiL gwiiiuhwjfc, •*♦ A^A«/^ 
fi q.im.wn£n. XFwnfinwul; fi qk?l$t Quipk£ni.wu^ kl^km^ phwliknwL fi fonun.pwliu 
k nLum^mhkj_ ufyuiuL nwifkuiujh £wpfiu p n l n p-> * * * "]• (307, 2) 

24 ]pB, V, i, " « phpnh ^tupwdifitj np i jkpfypfth fynqwj, mp Ifwjfiu q.iuu&g 
HpfiulinLukwrju ". [Cf. Toximanoff, Studies, p. 322 n. 77]. (307, 3) 

25 Sebeos, xix, p. 68, " qpmpSfih hnpui inwpiuh*** fi p^fi!} £whquwwpwh L 
hqfih ft iniuiqiuufi fi n.kqh *}*ujpfiLUu, np j; fi %n^ni[fim n.uiLwnfi. 

Ibid., xxxii, p. 116. [Cf. Toximanoff, Studies, pp. 342-344 and 344 n. 16,] (307, 4) 

26 PB, V, xliv, [Cf. Toximanoff, Studies, pp. 321 sqq, and 321 n. 76,] (307, 5) 

27 Erom *Bagrat-vanda, where vanda is the ancient form of the later gund, incidentally, 
with the meaning " houses "• Cf, ij[wun.~uil{. Theoph, Sim., Ill, v, p. 117, 
mentions the city of the " Bwhoaafiopwv " which scholars have identified with the 
Gundesapur of Arab sources, Cf. Noldeke, Tabari, p. 41. [See also, Hubschmann, 
Grammatik, pp, 113, 130; Ortsnamen, pp, 380, 411 ; Markwart, Sudarmenien, p. *11, etc, ; 
and Toumano££ Studies, pp, 318-321 and notes,] (307, 6) 



NOTES : CHAPTEK XI 



457 



27 a [LP\ Ixxv, p. 457 and xviii, p. 110, cf, MX, III, lxvi-lxvii. See also In5i5ean, 
Description;, pp. 406 sqq., Eremyan, Armenia, p. 42, Toumanoff, Studies, p. 218.] 

28 [Arm, Geogr,, p. 31/42. Arm, Geogr,, II, pp. 360/1-362/3. See Appendix IV B- 
C], The last two names are missing in the long version of the Geography in spite of its 
statement that there were 16 districts in all. We have completed the list from the 
short version. (308, 1) 

29 Tacitus, Ann,, XIV, xxiv, [L, IV, 146/7], " in regionem Tauraunitium ". Procopius, 
Pers,, II, xxv, 35 [L, I, 488/9], " rd im Tapavvtuv xcopia ". Const. Porphyr., DAI, 
xliii, pp. 188/9-198/9, " Tapd>v ", Tauraun-itis is unquestionably connected with the 
Taurus mountains, Sem. tvr. The later toponym Smpntphpinh < Suuprij -pkpwh 
(cf, Smmsk^ < SmLm^hj^ = vulg. $nLpm.pbpiuu has a similar origin. There is no 
need to correct Tauraunitis into Taraunitis as this is done by Hiibschmann, Ortsnamen, 
p. 325, and de Lagarde, Ag,, p. 46. [On Taruberan, see Hiibschmann, Ortsnamen, 
pp. 251-254; Eremyan, Armenia, pp.85, 116; Toumanoff, Studies, pp. 129, 148, 199, 
209, etc. On Taron, see Hiibschmann, Ortsnamen, pp, 325-327: Markwart, Sudarmenien, 
pp. 204 and n, 1, 220; Eremyan, Armenia, p. 85; Runciman's notes to Const. Porphyr., 
DAI, II, pp. 157 sqq., especially p. 159 where he rejects but does not discuss Adontz's 
etymology of the name; Garitte, Narratio, p. 245; Toumanoff, Studies, pp. 138, 202, 
209-210, 215, 218, etc., see following notes. On the later history of Taron, see Adontz's 
own study, " Les Taronites ... "]. (308, 2) 

30 Arm, Geogr,, p. 29/40, u SwpnLpbpwh np t Smpoa ". (309, 1) 

31 FB, III, xviii, " ••* wpg fi lTiui}Jil{bwu wn£iii*" fci. £on.wu jw^fuiup^u 
jiLpbwurj jwSnLpu Swj n g* h- m<UI l uumwu inSu pruani.iiu* + * It pn^Ii ntnjj^ nmm.u 
fiLpbwug ". Ibid,, IV, ii, " qmiuubp rjhnuw [rjlfmiSjil^nhbuihu] jw^mpu w^Jvwp— 
{fiu Swjoij jfiLpkwhtj w^juwpZjih ", Ibid,, IV, xviii, " q.mmhijiu tjhw jftLpniS 
n,uiuimfih fi S&ju<> jjitp uiSnLp pbprifiu* npmj) wumu ^puifuwltji ^n^ ", 
LP', lxii. "... *L\wpt} \JTwiffiIinhkiuh'] np lp i&wijkiu^ wnmj fi qwjbuilfu jitp fj 
Swju". Cf, Ibid,, xli, pp, 231, 234; lxviii, p. 393; lxxv, p. 440. See Toumanoff, Studies, 
pp. 172 n. 97, 335 n. 143, 452, 498.] (309, 2) 

32 FB, V, iii, " IfnL^kq* . ♦ £p fi uSfiu rj.WLmn.fih Srupoh jfiLpniS pbprifiV 
npnL.d flqwlpnh lfn\bu y np Irmj Ji i[bpwj qbuinju h^prummj ", [Cf, Toumanoff, 
Studies, pp. 138, 209, and below n. 56,] (309, 3) 

33 See above, n, 31 [first quotation,] (309, 4) 

34 FB, IV, xiv, " £2 Zwjp Stupquikw fi n.uiLiunu Suipwrunj pun. jup rj.ki.nuh 
mburnhb^ ♦ • n.hf;p p^imSmhu $pn.wwwj w pP w jf J ' ' • QJiuipn., mu^ nwjumpup 
mhnjiu l^whmhgm^uihqbph. Smpriljiuh mnLbj_ £* k n£ wpwhrj ". The patriarch Sa- 
hak I was buried, " • ♦♦ fi riiuLiitnu SwpWLunj, fi pufjfy ^ OI fi ^M/ 7 uhw^wl^wh^ 
juiunLwuhwip > jJX^ujfj^uum ". LP\ xviii, p. 112, [The words in brackets are 
missing in the Venice, 1933 edition of Lazar. Cf, MX, III, lxvii, " fiwjrj nutwwnL— 
ml^wu dwpifyu ♦♦* t)pkdjmjjfi £wuqhp&*** wfilfuuiLu ^w^Jilinubuji^*^ mmphm^ 
£iuuq,nLgjiu jfjLpkwurj rj.kqh jlX^mfeww, np £ jj riwuitnfju §wpofi f \ (309, 5) 

3 & MX, II, xii, xiv, [Toumanoff, Studies, pp, 215, 218.] (310, 1) 

36 Ibid,, II, lxxxiv. [On the Slkuni, see Toumanoff, Studies, pp, 212, 215, and 
notes], (310, 2) 

37 MX, II, Ixxxi. The fortress of f)rjuil^mh — modern Akcan on the Euphrates not 
far from the village of Snluk north of Mus, where there is a ferry across the river. Xore- 
naci's account is undoubtedly based on a connection between Slkuni and Suluk. [Cf, 
Eremyan, Armenia, p. 74.] (310, 3) 



458 



NOTES : CHAPTER XI 



38 FB, III, viii, " km ^tvslft nmLpujijwpft ^ nwlpnuu Qiub^iubwlfftb k 
n&pwpih^ftiftu h n8\pi- q[pLJu wilkbwjh q.wLwnwliuiLgh £whj].kph. '\ The 

position of these localities is unknown, In the Vita Sb, Oslceanc, p, 65 Qprnpuirfu = 
the Sukawet mountains of Bagrewand. [On these mountains, see Eremyan, Armenia, 
p. 81 and Arm, Atlas, p. 104], 8^i- ^i/ u / u * s presumably a translation of Taurus 
(Tavpos — g*iL]_ " hull ") pro SwLpnj Q-inLfu cf, SwLpnj pkpwft, (311, 1) 

39 ZP\ xviii, p, 111 [See above, Chapter YII n. 55 for the text], (311, 2) 

40 See above, Chapter I, p. 18. [On the Paluni, Toumanoff, Studies, p. 212]. (311, 3) 

41 JPB, TV, xv, " -**j]Xp^wSnLukm^ mknkwrju ft hw^mhnih SwpwLU qwunniu ". 
ZP\ lxxvi, p. 485, " + *+n.uwjp ft uui^Smhu Zw^wkhftij*" 4 iwukwj^ pivhw^p 
ft ywuimfih JXp^mSnLhkmg ", On the same page [p, 483 of the Venice, 1933 edition] 
we should read IXp^wpn diking for IXp^uifimukmn in the passage, " «♦* mkukm^ 
ftjnLpuwj, np lp ft qwLiimib typ£Uiiim.nkuiij ft qkn^iu tjftpSmij qinjkwlinpqft 
mkmnuh tjftpmlimj ^ImSumpui^muftu ", Not only did the Kamsarakan's have no 
connexions with ArsamuniV, but we know from AsoUJc, III, xvii, p. 197, that Sirim was 
located in Arsarunik', " ♦♦♦ ft qipifuu JXp^wpndikwij quiLumftu^ npm.fi Cftpftifii 
hn$>-p>\ (311,4) 

, . 42 Babelon, Les rois de Syrie, pp, 193, 211. [Eor more recent publications on the 
RTimrud-dag monuments and their inscriptions, see Honigmann, Kommagene, Toumanoff, 
Studies, pp, 278 sqq. The inscriptions have been published by Jalaberb and Mouterde 
Inscriptions, The final report by Th, Goell and E.&. Dorner, Nemrud Va§, is announced 
but has not appeared to date], (311, 5) 

43 Asolih, II, ii, p, 81, " QniftCwh Xfmhqm!^ntuft y np £p ft ipnuitnli Up>~ 
wSntSikuitj ", [Of, Eremyan, Armenia, p. 40, Toumanoff, Studies, p, 212], (311, 6) 

43a [Arm, Oeogr,, p. 31/41, cf, Eremyan, Armenia, p. 38,] 

44 Tov, Arc, II, yii, p, 121, calls the inhabitants, " n.wqwuwpwpnjg^ wpftinm- 
ppni-g ", and believes that, " jwqwqu ftjppbft k wu£kinuinQinkj_ft ftiouftnh k 
pwpnLiju Ijn^fth }jnLp+ + + ", Of, Georg. Oypr,, p. 48, " dm Se /ecu oi oikovvtgs zls to 
opos tov Tavpov irAijcriov tov clvtqv jtAejjuaTO?. (fx^ydXTjs Appspias) Xaol j8' ovojioXp^voi 6 jikv 
sis XoBtuTai, 6 Se h-spos Uavaaovvirai ". In the days of Yovhannes Mamikonean, 
XoyV belonged to the princes of Taron, that is to say to the Mamikonean, J oh. Mam,, 
i, p. 13, " urnkw^ nJfnL^kn^ omf;pu lT>nj k J\Jni.pwj $wpounj ftyftjuiub k 
JJmuunj ", (312, 1) 

45 [Arm, Geogr,, p. 31/42], JPB, V, xxix, " IXnjiftiiihnu kinftulfninnuftu ]fwuwq-> 
J^kpmnj ", Ibid., YI, ii, " ft ^kq^k JTujhmLwnljhpmnj ", Ibid,, III, iv [See below 
n, 47 for the text of this passage], Z,P\ xxiii, p, 134, " Sip XFkiftmi Xfwuhlikpmnj 
kuifwlinuinu". Const. Porphyr,, DAI, xliv, I, p. 198/9, " to JWavEwfepr ", [Of, 
Bunciman, " notes " to Ibid,, II, pp. 167-169. Hubschmann, Ortsnamen, p. 328, 449- 
450; Honigmann, Ostgrenze, pp, 149, 169-170, etc, speaks of Manazkert as being in 
Apahunik', but he is speaking of a later period, see Markwart, Sudarmenien, p, 15, 
" „. an den Gau Hark', der im westlichen der Ebene Eulanyk entspricht und ursprunglich 
Manavazkert (Melazkert) einschloss, „. . Im 9. und 10, Jahrhundert aber gehorte 
Manazkert (Manckert) zu Apahunik'". Also pp, 78, 454, 505-506 n. 7. See below, 
n,51], (312,2) 

4 5a [^ee Toumanoff, Studies, p, 218.] 

46 MX, I, xii; II, viii, ".♦*• nhwftjwpwpni.Jo x ftLuu IXiuiu&Lukiuij k tjlfmum^ 
Lwqkwuft h pnhmukmhu ft hnjh nmLwl^mgh Z^J^J "♦ [Of* Hubschmann, 



NOTES : CHAPTER XI 



459 



Ortsnamen, pp. 411, 435, 449; Toumanoff, Studies, p. 110 n. 173, and below, mi. 48 sqq,]. 

(312, 3) 
46a [Toumanoff, Ibid., p, 231 n. 283, and 233 n. 290, is of the opinion that the Abra- 
hamean house never existed.] 

47 FB, III, iv, [" JtfnupniJ Pius^wLnph Z m J n $*** km nwuuhh L n^pmb tj.ujinjjin 
qknh hiu^minhmjih IfmhuiLiunkhfin hu^jiul^ntuTiujih JXq^pjiwhnu]?h jkfyhqkgji, 
ijlfuihmLiuqml^kpm wdhhwjh uw^Smho^ L qiummwIfwiSi iiuhqkpti np >ntp9 
qhngog £p, np J^mj ft l^nnSwhu n.hwnjh bifipwmwj ; fci bin nth t^pniSi qlxnh 
flpqnLiiLnn, npnj whniSj £p fjpQnpnu nLuinji ku^nl^ninnu PmuiuhnL wilhhwjh 
uw^Swhogh £whn.hp&, np Jihgh jailf £ J^pkpl 1 ^ Pwuwhni. ". Of. Toumanoff, 
Studies, pp. 218-219.] (312, 4) 

48 FB, III, viii, " pahrnhfig " is listed among the ecclesiastical lands in Ibid., IV, 
xiv. [Of. Toumanoff, Studies, pp. 199, 216, and 209-210, 213, 323-324 n, 81, for the 
later history of Bznunik'], (315, 1) 

4 » FB, III, xx, " IXjl J^plfpfiii Uinw&Lhkwn, jnmfihl] llbbfi [hpfih 
ITwubwn, inhqfih qinjli npuiminnLg, wknLnjh x npni.d uihniSi ^n^ji guinwg ]Xqjinpug: 
The word gwqwg is obviously used here in its archaic sense of " fenced area, hunting 
preserve ". Aiiorsk' was located in Aliovit, the Unjij—£nijjwi of Eaustus, Ibid., 
IV, lv, and the " IXnJun—Cmlfiin " of the Arm. Geogr. [p. 31/42]. Both forms are 
correct, " wn^fim " or " « mqji " means " salt deposit < wnji " = " salt ". [Of. Hubseh- 
mann, . Ortsnamen, pp. 329-330, 396. Markwart, Siidarmenien, pp. *14, 15, 74 n. 2, 
77-78. Eremyan, Armenia, pp. 33 (2), 36. Toumanoff, Studies, p. 199.] (313, 2) 

50 Sip'an = ancient XfJHJiwh, was first the name of one of the small islands in Lake 
Van (so in Arm. Geogr., p. 31/42), and only later was transferred to the mountain named 
\fkfu JJ*uwfiu. [Of. Markwart, Siidarmenien, pp. *11, 15-16, Eremyan, Armenia, 
p. 72,] (313, 3) 

50a [Arm. Geogr., p. 31/42, " ♦•• IXmw^nLhfig. phq npnj fij;£h whnwh£ IXpw- 
bwhp jkqjih pnhm.hkwtj ", Of. Hubschmann, Ortsnamen, pp. 328-329, Eremyan, 
Armenia, p. 45, Toumanoff, Studies, pp, 213, 216.] 

51 BL, p. 223, " t dnnnil ji Wwhwqlfhpm qkiULii fi um^SmhuimpLJu tjjs&ml^jih 
Ziupgwj ". Idem., " fi lift i[wjp jJXinui£m.iibuin n.imum fi q-knji ITiufuuqlfkpin " 
ZG, p. 40, " Ji qujLiiinh Hinm^nthhinn fi gwqwghqh ITwh&fyhpw ". Asolih, xliii, 
p. 277, " ji qui Liu ah Zwpg Ji Ifwhuinl^kprn gwrpug ", [See above, nn. 45, 49]. 

(313,4) 

52 Const, Porphyr., DAI, xhv, I, p. 200/1, " to Kavrpov rov MavliKkpr /*erd rijs x^P as 
tov 'ATTaxQwrjs /ecu rov Kopiq /ecu rov XapKa ". Ibid., p. 202/3, *' to Mavt,it<UpT ... to tc 
*Airaxovv7}$ /ecu to Kop-q /ecu to XapKa ". [See Runeiman, " notes " to Ibid., II, p. 170. 
Hubschmann, Ortsnamen, p. 330. Toumanoff, Studies, pp. 208-209. See above n. 45] 

(313, 5) 
52a [Arm. Geogr,, p. 31/42. See next note.] 

53 The reading %uwwmjnp is also found for ^uwhwujp. Snuupbw — mmip = 
in Arm. " shepherds' field " is equivalent to the Karayazi ovasi. %wjmp — " greenery" 
is the same as the Elmali plain = " apple plain ", <Xhus < mid. Arm., hjlinJifi^^ 
Jtfhnihfiu, Alishan, Hayapatum, p, 550. [Of. Hubschmann, Ortsnamen, pp. 327-329, 
470. Eremyan, Armenia, pp. 48-49, 51, 65, 82 (3), 86, 116, (314, 1) 

54 We do not know the relationship between Vara&mnik' and the princes of the 
same name found in Ayrarat. According to MX [II, xxii], Aliovit had been granted 



460 



NOTES : CHAPTER XI 



to a collateral branch of the Arsaeids ; might this statement he a reminiscence of the 
fact that several houses from Ayrarat, apparently drawn from the ostaniFs, had moved 
to Taruberan ? Hubsehmann, Ortsnamen, p. 328[ ?] presumes that the Bajunis ^^. [ 

of Arab sources should be identified with ^uidhmjjfig rather than with Uww^nLujig, 
an opinion which is shared by Ghazarian, Armenien, pp. 21, 74, It might seem more 
probable to identify Bajunis with the better known fiifiimJifiu, but this district seems 
to have been known to the Arabs by the name of its chief city, Xlat', which is mentioned 
together with Bajunis by Ibn Khordadhbek, p. 122. We should read Pahunis = < H>- 
wui^nLujiu, (314, 2) 

54a [Arm. Geogr., pp. 32-33/43-44, Arm. Geogr., II, pp. 362/3-364/5, which speaks 
of 37 districts, See also, Hubsehmann, Ortsnamen, pp. 261-263, 339 sqq,, Honigmann, 
Ostgrenze, pp. 169 sqq,, Eremyan, Armenia, pp. 82, 117, et ah Also, Appendix IV for 
the test of the two versions of the Geography,] 

5413 [This pattern is also suggested by Hubsehmann, Ortsnamen, p. 347, whom Adontz 
follows quite closely throughout this discussion,] 

54e [The Arm. Geogr. lists Darni for Garni and places it before Arberani.] 

55 This district is placed after fcpnuuuqnLujig in the long version of the Armenian 
Geography, and after finLdnLuJig in the short version [See, Appendix IVB-C]. The 
latter version is preferable since JXnunj—nwh " the foot of the Arnoy mountains " 
lay south of Anjewacik 5 . The mountain chain near Mokk' is still known as the Arnos 
[Arnas] dagi and is probably a derivation from Unbnj ump in which the Arm. ump 
has been replaced by the Turk, dagi, and the name of the mountain itself was mistakenly 
considered to have been JXnhnju ump. This was the ancient name of the range 
stretching from Mokk' to Julamerk. The part of the chain directly above Julamerk 
was the one known as Ununj nwu* [Cf. Hubsehmann, Ortsnamen, p. 402, Eremyan, 
Armenia, p. 37.] (215, 1) 

56 The Arm. Geogr. has ]Xp£fem£mlfiw instead of lXp£[j£Uil[m£lwi y which is found 
in the short version [See, Appendix IV B-C]. The modern form Ercek is a contraction 
of Hpifi^m^ the diminutive of Ilp&t> which is the name of a city in the northern 
corner of Lake Van. Part of this lake was called the bay of Axees in antiquity, and 
subsequently, the name was given to the whole of the lake : Strabo, XI, xiv, 8 [L, V, 
326/7], " TjSe'ApoyvJi, tjvkclI ©comnv xaAovmv ... ". This passage should be read, 
" 'Apcnjotf koX ScvoiTiriv ". " Bwvttitiv " = ArSes and Tosp, the name of Lake Van. 
Pliny, NB, VI, xxxi (127) [L. II, 434/5] gives the name in the form " Arethusa " where 
the ill corresponds to a palatal, [Adontz's thesis is followed by Eremyan, Armenia, 
p, 37, but Rakham in his edition of Pliny, loc. cit, gives the name as " ... lacum Are- 
tissam ... ", Cf. also Ibid., VI, xxxi (128), [L. II, p. 434/5], " alterum deinde transit 
lacum qui Thespites appellator ... ", The small lake above the city of Van [Ercek 
golu] is called JXp£fi^ml^ " little Arces ", to distinguish it from Arces, Perhaps mp£^ 
had the general sense of " lake " in the language of the pre- Armenian population, 
[On these districts and their respective positions, see, Hubsehmann, Ortsnamen, pp, 339- 
341, 405, 476. Markwart, SMarmenien, pp. *33-34, *59, 4, 29-32, 77, 232, 345, 358, 
370, Honigmann, Ostgrenze, p, 21 n. 3, 170, 207, 209, Eremyan, Armenia, pp. 39-40 
79, 86. Toumanoff, Studies, pp. 50 n, 44, 160, 205, 213], (315, 2) 

57 Zewond, xxxiv, p, 146, " ♦ • ♦ ji qwLwn l£npkpmufi ft qkqh Phpfypji ". 
Sebeos, xxx, p. 108. Const, Porphyr., DAI, xliv, I, pp, 198/9-204/5. Might Gure be 
related to Qmnltji, and Giresor, a village at the foot of the mountain, to fywnufi&np ; 



NOTES ; CHAPTEK XI 



461 



Cf. also the name of the river at Bayezid, the Gernevik [Geraaoksuyu], On the Garni j or 
mountains, see Eremyan, Armenia, p. 46. Mt. Gure (e, 39°25 } N x 43°58'E) is given 
by Lynch, Armenia, map, and the village of Gireaor (c. 39°14'N x 44°02'E) is found 
both in Lynch, Loc tit, and on the USAFM map 340 B I, but neither can be other- 
wise identified. On Arberani and Garni, see also, Hubschmann, Ortsnamen, pp. 341- 
342, Markwart, Siidarmenien, p, 459, Eremyan, Armenia, pp, 37, 46. Runciman, 
" notes " to DAI, II, p. 167. Toumanoff, Studies, p, 205], {316, 1) 

58 Tov, Arc, III, xiii, p. 197, " ••♦ [i qipLfuu knpnju Zwjntj" (and not Zwyntj, 
as in the printed version), " np £wu£ jJXpnLWUpnLufi qiuLumu ", MX, I, x. 
This is still its name at the present time, Cf, Ineicean, Geography, p. 144. [Eor these 
districts and their respective positions, see, Hubschmann, Ortsnamen, pp. 342-343, 
400, 443. Markwart, Siidarmenien, pp. *37, *53, 359-389, 422-423; Honigmann, 
Ostgrenze, pp.147, 170; Eremyan, Armenia, pp, 36, 45, 51, 86; Toumanoff, Studies, 
pp, 198-200, 204 and n. 232, 221, 235, 310, and below n. 75], (316, 2) 

59 MX, II, lii, " ♦♦♦ ft £iupwuij jiukb^fiij fynLut JTwukwi], np Ifn^ip Cuilw— 
pyiuljwh quiLiim^ ft *[kpwj upuQijni^ ^/^/^ whmSih IXpmtun^, gwuq^fj k w^fuwpffi 
nmuifj n.hphnwhu* Upwwq^ ^*v/f uJin^L ywjuop dwdwuiuljfi ", LP\ xei, p, 532 
" ♦ •• [j qwunnu np tyn^ji IXpmuin^ fj n.hoqh np mhrnhkh l}qjihn. ♦♦♦ ". Ibid., xxxviii, 
p. 217, " ••• Upwuiq^ £nLiu ft qkonh np l^n^ji JXttupwjp [j ^p2 wl J! w k w %/' 
SqpnLin qm^mfiu ", Vardan, Geography [p. 16], " IXpuiwq^ WwfynL £, mp fyiuj 
up umwgkwjh piuq.£nu ", Cf. Tov. Arc, III, xxix, £. 259, " * * • [i £wun.uwfj 
uppnju fowqinufj wnm^hjnj jJXpqnqmljmh n,wuunjj ". [See also, Hubschmann, 
Ortsnamen, pp, 343-344, 451. Eremyan, Armenia, pp. 40, 65. Toumanoff, Studies, 
p. 169, and below, nn. 74]. (316, 3) 

60 Tod. Arc, III, xxii, p, 232, " *♦* qwSnLph X^Ijinn fj ffcnnuwLuiu {pro fonupwLWU 
of the Arm. Geogr., p. 32/43) qumimji, k niuSnLph ^nmnp ft &npu Hhbwjfiij^ nmSmph 
Vkwiifi fi lpnpu idpwj". {Cf. Ibid., IV, i, p, 275, " ♦•• /r qwunnph XntSpwj 
l^n^hghwi <f>npwli"); Ibid., Ill, xx, p, 226, " ••♦ &np& Qubwjfjg y npnj hjj>u 
phq q.WLwnu Z&p £ mi l m £p "• Ibid., Ill, xxix, p. 264, " * * • pU£ &npu IXa^m^nf 
kljU 12 I 1 hjkpwlptiuh i^n^kijkmi qw^m", Tov. Arc Cont., IV, ii, p. 271, tc »"/f 
phpqu Mnmnpntj p &npu l£u&iu[unj", LP\ xxxix, p. 228, " ♦ ♦• jmn^lu Qhbwj- 
ULnj.", Blise, V, p. 100, " £udwj[u]wgli", p. 116, u Qubwj[iu"; p. 120, 
" Qtibwjunij ". Einally in the GahnamaJc, " Qhbwjkzjji " and in the Military List, 
" Qhbwgfch " instead of mhdwfupu = pubw£fiu. [See, Appendices III A, B]. 
The correct form apparently was phbwjfih, — uij (the phbmjununj of Lazar should 
be read pubwjhnj) and the other forms sprang from the erroneous phbta^jih. In 
Tov. Arc, I, vii, p. 51, we find an attempt to provide an etymology < wh^kiufu^ 
&iujuhfe but it is incorrect, and the whole passage may be a late gloss. 3STkan = Nagan 
(though Hubschmann, Ortsnamen, p, 395, repeating Incicean, Description, p, 209, 
asserts that " Die Lage der burg NJean ... ist " unbekannt "). The position of Nkan 
justifies the transfer of T'ofnawan from the basin of the Araxes where it was placed 
by Hubschmann to that of Lake Van, [Although Kkan is marked in A A pp. 105-107, 
c. 38°40 ; ISr x 44°10'E, and is discussed by Markwart, Siidarmenien, pp. 311,466 and 
n. 4, Hakobyan, Geography, p. 187, as well as by Laurent, Armenie, pp. 52 and 95 where 
he gives its approximate position, " .,. la forteresse de ISTkhan, situee a peu pres a egale 
distance des lacs de Van, d'Ourmiah et de Sevan ... " its precise position cannot be 
determined since it is not indicated on any other map to my knowledge, either contem- 



F 



462 



NOTES : CHAPTER XI 



porary to Adonta's writing, or more recent]. The other fortress, Sewan — Seyvan 
kale is placed closer to Kotur on the Prussian maps and eloser to Van by Lynch, Armenia, 
map, [As in the case of Nkan, there seems to he some confusion in this case: Seyvan 
is not located on the Mehmedik river, as Adontz claims in his text hut on one of its 
southern tributaries, See, Lynch, Ibid., TJSAF map, 340 B IV, et at On the districts 
discussed in this section, see also, Hubschmann, Ortsnamen, pp. 344-345, 400, 430, 
Markwart, Sildarmenien, pp. *59, 357, 390-392, 426. Honigmann, Ostgrenze, p. 170. 
Eremyan, Armenia, pp. 32, 33, 36, 117. Toumanoff, Studies, pp. 170 n. 85, 181 n, 143, 
197, 199-200, 219-220, et al'] (317, 1) 

61 Tov.Arc, HI, xxix, p. 264, " ..♦ phn Znph Ufc&mhOif klfi £P ft fakpiulfiiibh 
l^n^k^kw^ nw^tn, * * * ft biflspnijuh wpj>nLbft, ft nwiwnh 2?/uu/£ jmimhh Ifumwlfwh " 
J bid t , p. 254 ** * ♦ * . ITuinmlfmhh tumih^ urn ^wpilftph fyn^knhw^ nkmnift^ np 
ft2 Uf ^t ft tykmh Jjpwufti ". In antiquity, the river bore the same name ; 2$nmj£ — 
nnm i.e. ** the Oowas river ". [See also, Hubschmann, Ortsnamen, pp. 345-347, 448, 
Markwart, SUdarmenien, pp. 205-206, 208-209, 311, 313, 401, Honigmann, 
Ostgrenze, pp. 166-167, 170. Eremyan, Armenia, pp. 63-64, Toumanoff, Studies, 
pp. 219, 305 n. 119], (317, 2) 

62 According to Tov. Arc, III, xxix, p. 260, from Mardastan, " ••• ft jwhrjfth nwn- 
hwh ft £nLW} (on the left), and ft fonnftwLwh qivLuin (on the right), mhnktui pin 
yptiniSikwg nwLwnft ". Krcunik' is not derived from yftp£—nihft as might seem 
probable, but rather from *lpwft£~m.bft* ypm—ft£ has the same formation as 
lXwpiqwm~ft£ and means "Kurd", It is interesting to compare ^p£nthft with 
the ynpm£wj t p (< 1jnpmft£—wj 1 f>) of FB, which he uses [IV, xlviii] for the region 
of Salamas adjoining Krcunik'. [Of, Hubschmann, Ortsnamen, pp, 345-347, 442, and 
Eremyan, Armenia, p. 61, where he assimilates the name of KrBuniV not only to ICu- 
rucan, but also to modern Kurcivik. See below n. 75a]. (318, 1) 

63 Lewond, viii, p. 26, " ■*** ft ij.wi.wnh ffx^wmSikwij ft nftujh, npm.il ^ml^whuh 
Ifn^kh ". Of. Tov. Arc, in, xxix, pp, 251-252, " ♦•• <Snntq> n^mniSift^ 
PnnmSiftg) %mnwh * * ♦ JXpww^ukwh ",: {See also, Hubschmann, Ortsnamen, 
pp. 345-347, 409, 420. Eremyan, Armenia; pp. 41, 48.] \ ^ (318, 2) 

64 Among the doubtful districts, Vlwwntnwpmljft^ JXpmwiwhkwh^ Pwnwh y 
%wikpwh are [un]known . even to Tovma Arcruni, [Since these districts cannot be 
found in the listing of Tov. Arc, HE, xxix, pp, 251-252, see next note, it seems likely 
that the negative was accidentally omitted from Adonta's sentence. Cf f Hubschmann, 
Ortsnamen, pp, 345-347. Eremyan, Armenia, pp, 41, 44, 46, 77, both give alternate 
forms of these toponyms, See below n. 76.] (318, 3) 

65 Tov. Arc, III, xxix, pp. 251-252. [jSfee, Appendix IV for the text]. Tovma says 
of the last districts that they were won from Parskahayk', which is the situation depicted 
in the Arm, Geogr,, [pp. 32-33/43-44]. Tovma goes on to explain the absence of Naxca- 
wan and Gfolt'n by the fact that they had been lost to Vaspurakan, the former 210 [sic], 
and the later 186 years earlier. Since both districts are listed by the Arm. Geogr, in 
Vaspurakan, we should conclude that the redaction of the Geography dates from a 
period 210 years before Tovma. On the other hand, in view of the doubt we have 
expressed as to the authenticity of these thirteen districts, it is possible that Na^cawan 
and Golt'n are among the thirteen districts included in Vaspurakan on the basis of 
Tovma's commentary. In such a case, the redaction of the Geography must postdate 
Tovma's History. [On the problems of the Arm. Geogr,, see, above, chapter X, n, b, 



'*ww&^~ 



NOTES ; CHAPTER XI 



463 



and for the chronology of the commentary of Tovma Areruni, Brosset, OB. A, I, p. 203 
n, 4. On Goltn, Ka^cawan and the presumably transferred districts, see Hubsehmann, 
Ortsnamen, pp. 338, 346-347, 419; 427, 455, Eremyan, Armenia, pp. 48, 52, 53, 72, 84, 
117; Toumanoff, Studies, pp, 170 n, 85, 203-204 and n. 230, 293, 305 and n. 119, 310 and 
n, 32, 323. [See below n, 76]. Eor the entire discussion of the title of Mar&pet and its 
identification with the Areruni house, see, IUd., pp. 131, 169-170 and nn. 81, 86, 176- 
178 and nn. 115, 118, 199-200, 220, 231 n. 185, 233 n, 290, 237 n, 305, 248, 314; also 
Hiibsehmann, Ortsnamen, pp. 343-344, 541; Markwart, Mran, pp. 166-167, and Genea- 
logie, pp, 34-41; Garitte, Agathange, pp. 224-225; Manandian, Feudalism, pp.46, 61, 
63-67, 271-273, 276; Sukiasian, Armenia, p. 190, ei al Also below Chapter XV n. 36a, 

(318, 4) 

66 in BL, Gregory is called, " ♦♦♦ ukiy^wfywh qhqph ku^fiu^nt^nu", p. 73, but 
" ♦♦♦ Ifiupqiqhinm^wiU ku^jiul^nu^ou " on pp.70, 76, 81 [NB, both on p. 70 and 
on p, 76, though not on p, 81 we find a double listing; " ... (70) fypfynpfi Upb- 
pnihkwrj kuifiufyniqnu^ *** fi %pjijpipi Ifwp^iqkmm^mh kiiffiul{mifnu£;*** (76) 
fypfynp ITwpqiqkmmlimh kufliuliniifnijfi* thpjuj.npliUpbpmStktug hjqpu^ni^nu[i^," 
Similarly, Theodore is given as bishop, " *** uhui^m^uib qhqfi *♦• ". Ibid., 
p, 151, but as bishop " •♦♦ Ifwpqiqkumilpiih ", pp, 146, 196. [Here too a double 
listing appears in one case. On p, 146, Theodore is given as bishop of Mardpetakan, 
but the name of the Axcruni bishop is John], (319, 1) 

66a [On the toponym Vaspurakan, see, Hiibsehmann, Ortsnamen, pp. 210-211, 253-254, 
261-263, and Grammatih, pp. 60, 80-81. Justi, Namenbuch, p, 359a, OJiristensen, 
pp. 100-103, 108-109 n, 3, 123. Manandian, Feudalism, pp.50sqq,, 65 n, 1, 77, etc. 
Garitte, La Narratio, p. 244, Toumanoff, Studies, pp, 331-332, and below, Chapter XV 
n, 26-27a.] . 

6< 5*> {See below Chapter XIV n, 57,] 

67 FB, in, xviii, [Of. above, n. 65, on the Mardpet with the Areruni.] (319, 2) 

68 Ibid., IV,xiv, . (320,1) 

69 <&P', xxxix, p, 225, " *** tjwipb \Xpbpm5thwtj h qdwpfj.u^bmh IffiCp- 
£WUfnL£". Mlise, p. 193, " SmTj^^h Jlpbpm.hhwy XThp}wupii.C* * + ". [Of. above, 
n, 65], (320,2) 

70 ZP\ xxxir, p, 199, " * * • qmbq^p fj qopuigb JTwpqmhmwIjwh wjpni&Jinj ". 
lBU&e 9 m, p. 74, " QqnLhqb wJiw^fjb miujltb g\,kp£tuiiinL£ flrdpnuhwh^ k 
^.nL^mplfih qhw upm^mixpaiU m-^ump^ih^ ilkph Jj uiu£fiuihu limp u^m mwfyiuh 
w^fuwpZfih ", [Of. Ibid,, pp, 43, 99, 193]. Nersapuh is the popular form of Mihr- 
Sapuh, and botlT forms are found in ZP\ pp, 135, 144, 225, 236-237, etc, [Of. Hiibseh- 
mann, Grammatih, pp. 54, 56, Justi, Namenbuch, pp, 206, 228, Toumanoff, Studies, 
pp. 231 n. 285]. The meaning of the name f^Spnuhmh is unknown. (320, 3) 

71 ZP\ iv, p. 16, Tov. Arc, U, vi, p, 116, " «♦♦ kljh hStiLtn pbij. I^nqifti JXmpu^m- 
mmlpuh jJXtj^pwijL qui mm jkpfyfiph ^muu^nipmljmh^ k pmhml^k^mt fi 

jllqwjiuilfkpui mjwwhjjh Hpbpm.fjkuiij*++ " ei al. {Of. Hubschmann, Ortsnamen, 
p. 442, Markwart, Sudarmenien, pp. *59, 426. Toumanoff, Studies, pp. 199-200, 304.] 

(320, 4) 

7 2 £F, slii, pp. 236-237, " ♦ ♦* %fejuwbh ilhb UpbpnLbhujij ^X,kp^aa[ni4 
«•♦ tipifurnhh IXpbpniSikwg qjlmpuwd". Of. Mlise, pp, 99-100, and 193, " Sm^qih 
JlpbpmSihwq Ifhp^imifnLi It &WLwmq L Uhqfih k lfk£pm.dwh k fywptfk h 
Sw&wm *** QwpbpnLtikwij mn^dlh IXu^pumS." Nersapuh was prince of Albak, 



464 



NOTES : CHAPTER XI 



{LP\ xliv, p. 248, " ♦ •♦ Vm^i ujiipwl[wijli<> np lp qpwh kp£rf jj^jviuhjih 
IXpbpnrfikwg \,bp£Wiq£nj". Cf xlii, p. 237; xlvii, p. 272). The branch of Aprsam 
must have ruled Mardastan but contrary to expectation, it is Nersapuh and not Aprsam 
who is given the title of Mardpei. Evidently, with the passage of time, this title had 
become generalized among the Arcrunis. [Cf, Toumanoff, Studies, pp. 199-200, 231 
n. 285, and above, n. 65.] (320, 5) 

73 Unless Snutun^mnLufig in Tov, Arc, III, xxix, p, 251 is taken merely as an error. 
[See, Appendix IV, and Totunanoff, Studies, p, 213.] (321, 5) 

74 It is evident from Tovma Arcrtmi that the Amatunis first lived in Artaz, since 
he speaks of the bishop of Artaz as " * • * hmfiulfnuinuh Qpfiqnp IXd win nibbing 
wwhu, np humj-p [i £ujun.uwfi uppnju fowqlniifi mnw^b^nj jH,pq.nnu3l^mh 
qwutmfi ", Tov. Arc, III, xxix, p. 259, As we shall see, ecclesiastical subdivisions 
•usually reflected an earlier state of affairs. The Amatunis were one of the oldest princely 
houses, and its origin points to an ancestral domain on the frontier of Atrpatakan. 
See above, n. 59, and below Chapter XIV n. 63. (321, 2) 

75 Tov. Arc, II, vi, p. 109, " •♦. {i inn&t QuinnLubwrj*.. " should be read " ... 
*♦♦ ft mnCSi SpnjSikwtj*** " as it is given Ibid., Ill, ii, p. 134, Tir-uni is apparently 
a contracted form of Tirpat-uni. [Cf Hiibschmann, Ortsnamen, pp. 343. Toumanoff, 
Studies, pp. 221, 235 n. 301. See above n. 58,] (321, 3) 

75a [On these principalities, see above, nn, 57, 60, 62, and the next note. Also, Eremyan, 
p. 45, and Toumanoff, Studies, pp. 110 n. 173, 130 n. 229, 197 and n. 222 on Ake; 
pp. 132, 199, 205-206 n. 284, 298, 303, 318, 327, 453 n, 63 on the Gnuni.] 

76 Pwgwh should be compared with Pahgan, a village south of Maku [Lynch, 
Armenia, map], ^tuqpfilj—wh < ^mqppli — with the Pers. suffix -an — the 
Arm. ^hun^pfilj—^ of Tovma Arcruni, who gives " ^mnjifiljh JXwnL^wpwn^ ", Tov. 
Arc, II, vi, p. 109, but " ••• ^wnjifi^h lXwnLub[3 k ^uwntlf k <^w£wu". 
Ibid., Ill, v, p. 146, ^fidwunLufi is given as " ^wdrnhfi " in the Arm. Geogr. 
and " *luipwdhm.hfi " in Arm. Geogr., II, the last form apparently being the more 
correct one [See Appendices TV and above n. 64. The name Varaznunik' in its various 
versions appears in Ayrarat and Taruberan, as well as in Vaspurakan, cf Eremyan, 
Armenia, p. 82], The Hu^ubwu of the Gahnamah is apparently a distortion of 
jfimuip^bwu. S^J^phtuh might be compared with Deir [now Der-§ikefti, see, G 46, 
p, 178], a village north of Baskale, [Eor these minor districts, see, Toumanoff, Studies, 
pp. 230-233 and nn, 281, 283, 286, 289, p. 236 and n. 303.] (321, 4) 

77 [See preceding note], For example, Tov. Arc, H, vi, p. 109: JXwwCnLUJi^ 
Zuuhm-hfig, ^mpmjbtjbh. Ibid., Ill, xxii, p. 235 the JXpbnbufig mlp is given in 
Vaspurakan, as is prince *fiw^pbpnih]i. Ibid., II, vi, p, 109. [Cf Hiibschmann, 
Ortsnamen, pp, 329-330, 363. Toumanoff, Studies, pp. 206, 220-222. (321, 5 ) 



NOTES : CHAPTEK XII 



465 



CHAPTEK XII 



a [Tor more recent discussions of the place of the Church in the naxarar system, see 
Manandian, Feudalism, and Sukiasian, Armenia, Garitte, Narraiio, also individual 
notes below.] 

1 Ter Mkrttsehian, Die Paulihianer, p. 55, " Die armeniseheKirche aber ist yon Anfang 
an als eine selbstandige, nationale entstanden, in ganz anderen politischen und kultu- 
rellen Verhaltnissen als die grosse katholische Reichskirche ", 

la [Marr, Arhaun, pp. 1-2.] 
110 [See above, Chapter I n. lb,] 

2 Agaf, cxxiii, p. 631. [Of, Appendix III K for the text of this passage. Also, 
Garitte, Agathange, pp. 321-323 for the variant versions. On the problems connected 
with the text of " Agafangehs ", see above, Chapter X n, 89a,] (324, 1) 

3 A gat 9 , exxi, pp. 623-624. [See, Appendix HI £ for the text.] (324, 2) 

4 Idem, (324, 3) 

5 Uytanes, I, lxx, [Of, however Ixxxix. See, Appendix III K, vi, for the text. On 
U^tanes' date, see, Toumanoff, Studies, pp. 19 and n, 21.] (324, 4) 

. Orb,, vii, I, pp. 64-65, [See, Appendix III, v, for the text.] (325, 1) 



7 ,AsoUh, HI, vi, p, 174, " *♦* jjdwuinwljh Un/ifoz.^ pwqfiwfhnpCh jt 
tj.pmnLWpthu ^hpng uppnrj k jhpqu hpwd^mnLpbiuig, " See also, the " His- 
tory of the Vivifying Cross ", in Alisan, Ayrarat, pp. 547-549, (325, 2) 

8 Asolih, HI, vi, p. 174. The abbot preceding Samuel was Polycarp, who was 
present together with Xaeik at the election of Kat'oKkos Step'annos III Sewaneci in 
972, Ibid,, 3H, viii, p. 181, This same -Xa^ik was elected Step'annos' successor two 
years later, A,E, 421 = AJD, 974, Ibid,, HI, ix, pp. 184-185. [Adontz follows the dates 
given by Asohk, but the chronology of the period is contradictory and confused, De 
Morgan, Hisioire, gives the dates A,D, 969-971 for Step'annos HI, and 972-992 for 
Xagik I, whereas both Ormanian, Azgapatnm, I, pp. 1117 sqq, and " Index", p, xxxi, 
and Kogean, Armenian Church, pp. 323-331, date the two kat'oKkoi respectively AJD. 
969-972 and 973-992. Of, Grousset, Armenie, pp. 486-488. For the shift of Vxtanes 
from the Xth to the Xlth or even XHth century, see, Peeters, Sainte Soussanih, and 
n, 5 above], (325, 3) 

9 See above, Chapter X n, 6, (326, 1) 

10 This thesis is not affected by the latest conclusions on the composition of 
Agat'angelos based on a comparison of the Armenian and Arabic versions, cf, Marr, 
Ghristianization, p, 182, If our version of Agat'angelos cannot be dated earlier than 
the YIHth century, as is claimed by Marr, the author could have used Zenob's work 
in either version. In such a case, Zenob's naive claim that he had written his work 
before Agat'angelos, ZQ, p, 19 " hu utnm^wqnjh qpktjfj giuh ♦♦♦ qlXqwPwhqkqpu ", 
is less ridiculous than it first appears, [Eor the date of " Agat'angelos ", see above, 
Chapter X n, 89a, for that of the Pseudo-Zenob, Toumanoff, Studies, pp. 18-19.] 

(326, 2) 
ii We should note in particular the information that there were 620 provinces in 
Armenia. According to JSG, pp. 12-13, the Armenians wrote among other things to 
bishop Eleazar that, " wiJkhuijh qwuimwiju kuifwlfmifnu uffjinnj hh h gw^whwju 
qp pi^m nSmfi^ npo nmmkj> nmwkg klfkut^ kh dnqnijkujjj pwjg ^/^<? 
kb wjungfjfy um i]krj£wpki.p h guwh qwuimu £uijnij. np h qmimnjih Sffi lift 



466 



NOTES : CHAPTER XII 



gw^whuij l^wS kplfm jpL^ p£ £whnfiwfj ". This passage is repeated in the 
Letter of the Kat'oHkos Constantine I to King Het'um I in A.D. 1248, " f) k Ji 
nwLwng hh Z W J£ jbptfiliwjib [i i[kp h fywh np WL^p hh li l^wh np ft 
bwnwjntpkwh. pi Ij^nL guiLJih pun Zwjng, Jdnq^ kpfiwh wnwwbh ", BL, 
p. 509. It is interesting to compare Zenob's account with the passage in M U, lxxxix, 
pp. 184-185, dealing with ecclesiastical provinces under Kat'olikos Peter I at the end 
of the Xth century [Petros I Getadarj was kat'olikos A.D. 1019-1058]: " jnpdwfi lp m^p 
tybwpnu jiupnn ^wjpwwbwmpbwhh k fi Zwjuwwtub w^fuwptyu, ni.u£p 
^w^nLpjiLh £wjpbubwg y wmbw^ fj pwnwtnpwiju Z w J n ^ ^fi^n, iwpftLp 
qbq wunLwnfj k ifbbwufig k }w£wiJ;w pbwfjp puwfjp, unjuw^u k bwfiul^nwnu^ 
£fihn £wpfiLp ifiwnwLnpg, k nwLwnwwbumw fi tfbpwj £fibn £wpfiLp pbiffi 
^whwwwn. bplfnwwuwu bwfiu^nwnu^ L ^npu tfwpnwwbwg fj wwu £wjpwwbw[iu 
wbupnlfiuu tfrbtp k ifwpumu kp£r/ fi l^pohwmpwn k jw^wp^wl^wbwn £fiun ^wpfnp 
k dp hnuuuuj mpnn £wjpwwbwni.pbwuu guih PwnwLnpnLpbwhu Zwjny ". 
Any locality having a representative of the ecclesiastical authorities, nwLwnwwbunLS 
is considered to he an ecclesiastical province in this passage, but since these subdivisions 
characterize a late period, they are useless for our study. (326, 3) 

12 Uxtctnes, I, lxx, [See, Appendix III K, vi, for the text]. The last name in the 
printed text is u IffrnLUWLnpbwij " rather than " JfiLU IXunpbwtj ", which we believe 
to be the correct reading. (327, 1) 

13 Agat\, cxx, p. 621, " fez. wjuwfru pun wilbhwju bpl^ftpu ZwjntJ* fi bwnwa 
dfiusk ft bwnu &ntp wwpwbwbfcp qp}wlini.pjii-h ^wpnnnipbwhh k wihwwpw — 
untpbwuu* fi ]J www qui ijLnn gwnwDJ;u i/fihsk wn w^fuwpCwLU hjwnwbwrj, 
dftuik wn ^wnwp^o ^ g^ z^&s fiuwwn fi uwCdwhu JTwugpwa^ iffihtk fi nm.nhu 
\Xpwhwij, dfiU£ fi uw^Swhu Ijwuipfin, fi tpwjwwfywpwu gwnwg wpgwjnL.pbwuu 
Zwjnij* k jJXSnwtjLny gwqwglh rifib}k wn Jfbpfiu gwqwgWLy ,pbpip wn uw~ 
Cdwhogh Hunpnij) wn \,np Dftpwtywu bpfypwdi, k wn IjnpnLO^ Jfiu^k jw— 
Smp hplfjipu Ifwpwij, ilfih^k wn mwSph Ifw^pwwu fefvwufiu, ifftbsk jlXwp— 
wwwwlfwb tytp wwpwbnrfjj-p nwi.bwwpwunL,pfiLUU fnp ", The expression, 
" **. rf>wjwwl[wpwu gwqwg wpgwjnLpkwhu Zwjntj •♦• " points to an Armenian 
city of the royal period. 

14 [V®]> Marr, ChrisiianimUon, p. 136. [See, Appendix III K, iv, for the text], Marr's 
edition has ^cJurL) jJu * s the land of Gilan " and \\^.\ ill" to the Abkhaz ", both of 
which we consider to be incorrect. The Abkhaz had already been mentioned in the prece- 
ding paragraph, both Abkhazia and Gilan lay outside the boundaries of Armenia, and 
consequently had nothing in common with the Armenian districts of Greater and Lesser 
Cop'k', Hasteank, Mokk', and Mardpetakan, which are listed after them. The first 
name is more correctly read, ^^vJuJ- 1 = 'Ayy^rjvjjv, and the second should be taken 
as a distortion of Arzanene, [This correction is born out by the new Greek version 
published by Garitte, who accepts Adontz's reading, Agathange, clxxi, p. 171, also 
pp. 200-201, 215-216.] (328, 2) 

14a [Garitte, Agathange, Fez, clviii-clix = Ug, clxx-clxxi, pp, 101-102. See, Appendix 
III K for the texts.] 
Mb [Ibid., Ya, clx = Vg, clxxii, pp. 102-104. £ee, Appendix III K for the text.] 

15 Marr, GhrisUanizaiion, p. 136 [= Garitte, Agathange, p. 103], The first syllable 
of t3>£^2^ m Parr's text should be separated from the rest of the word and read 
jjj "land", whereas (Jaj^ai) seems to be ^nn, a district in Tayk\ although we 



NOTES : CHAPTER XH 



467 



should expect TcoV since the Arabic version is a translation from the Greek. . On historical 
grounds we should expect thi3 to he distortion of the bishop of Vanand, [Of, Garitte, 
AgaiJmnge, pp, 221, who gives Kcorcov — Kotayk', 221.] (328, 3) 

16 Marr's edition has £yCj \jS " of the Kuanites " but the correct reading is £jC«j I 3 
= Kapyvtrw = Arm. ^wpfih [Adonis's correction is supported by the new Greek 
version, and accepted by Garitte, Agafhange, pp, 103, 21 8 J (328, 4) 

16a [Garitte, AgatJiange, pp. 103, 196-198, 235-236, gives Albios as bishop of Taron 
and Tayk' rather than Taron and Bsnunik', as suggested by Adontz. Garitte's con- 
clusion seems more likely both in the light of the new Greek version, and because Albios 
was associated with the Mamikonean house which had bishops both in Taron and Tayk 5 . 
See below, pp. 263,266-267,271-272,288, also Toumanoff, Studies, pp, 138, 172 n, 97.] 

17 ZP\ xxiii, pp, 133-134 = fflise, U, pp, 27-28, [See, Appendix III for the texts.] 

(329, 1) 

18 BL, p. 41. [See, Appendix III for the text. On the date of the Council, .see 
Garitte, Narmtio, p. 152], (329, 2) 

19 BL, p. 73, [See, Appendix HI for the texfe, From the ** Letter of Accusation " 
of the KafoSkos Kerses II [BL, pp, 70-71] we see that the following bishops were living 
during his pontificate, but were absent from the Council; 

JQnupml IXdwrnnthkuji] Znnniiwhnu lAfyfmj 

JJiu^ml^ rt*}mni.hhwn Jffipuiji^ Qwph£wLuihJin 

JJnqni[^nh ITnlputj JXhrnmnip PdhriLhkmg 

JJmkjjimhhnu JXhhkmghwq 
The eleventh signature at the Council of 555 is given in the text as Markos of Bagrewand, 
but Bagrewand was already represented [in the fifth signature], so that this must be 
considered a mistake for the missing bishop of Bznunik' or Tayk', [Of, Garitte, Narmtio, 
pp, 130-175.] (330,1) 

w* [See above, chapter SI n. 66.] 

1913 [Of. Garitte, Agathmge, pp. 103, 238-239, and Toumanoff, Studies, pp, 208-209.] 

i9c [gee above, chapter XI, pp, 236,242 sgg.] 

a9d [See above, chapter SI n>41.] 

30 BL, p. 78, " %kpu pwukhnj h ITapqwqnj huftuiptuinu ", [Of. Ibid,, p. 73.] 

(333, 1) 

20a [On the date of the Council of 644-648 ? See, Ormanian, Azgapaium, I, p. 711. 
See also next note.] 

21 BL, p, 146, [On the election of the successor of Kat'ohkos Movses I, see, Ormanian, 
Azgwpalvm, I, pp, 607-610, 615, Garitte, Narraiio, pp, 223, 258-259. For the text 
Appendix IV L, iv.] * (333,2) 

33 BL, p, 149, [See also preceding note, and Garitte, Narmtio, pp, 259-261, Ap- 
pendix HI L, iv.] (334, 1) 

33 Ibid., p. 151, [Of, Ormanian, Azgapatiim, I, pp. 817-618, The text of Adontz 
gives 571 as the date of the treaty, but this must be a misprint. The treaty between 
Maurice and Xusro II was signed in 591, See above, Chapter I n, la, and Garitte, Nar~ 
ratio, p. 260.] (334,2) 

24 Idem. [See Appendix HI L, iv for the text]. The eight signature, that of 
Snijimhliwj Ijfinj has been distorted to &n£wh Mwjqpj and tjwjhjnj in Uxtanes 
" Schism ", IT, pp. 64 and 57 respectively. (334, 3) 



468 



NOTES : CHAPTER XII 



25 £^ or }}}, more correctly hl^> is unquestionably a contracted form of the JXjjJi 
listed in the Arm. Geogr., [p. 32/43] among the districts of Parskahayk\ In the Geo- 
graphy Ayli is identified with Kufiean '* ^Xkjifi np ^ n >p "^nutpSmh " = modern 
Culiean in Gever, Cuinet, II, p. 717. Ami is likewise listed in Parskahayk'. The 
BL gives IXnmhwj p, 151, and Unhiujnj p. 147, instead of the Unlifi found in the 
Arm. Geogr. (334, 4) 

26 Dwin Canons. [See, Appendix III L, for the test]. Qamgean, History, II, p. 345, 
gives the same names, but in a different order, moreover, the bishop of ArsarunuV 
is given as Theophilos and the bishop of Arsanxunik' as Gregory, in reversal of our 
document. In view of the fact that we find a Theophilos, bishop for 36 years, in the 
List of ArsaruniV bishops given in Dashian's Catalogue, p. 656 (according to a copy 
of the Paris MS), we must give the preference to Qamgean's reading, There is a Gregory 
in the List of Arsarunik' bishops, but this is the famous scholar who participated in the 
Council of 726. (334, 5) 

37 flotak = Jlostak, was a district of Salamas according to J oil. Kat' ., lv, p. 363, 
** kp[i}wj[ih"* Ejl^nqSwSp.^h f}*nww!fwj ^wLmnjiIi jj Zkp A \Jwnwfiwu, li mhmji 
ji gwrjwg JXwpupnmmliwhji *** ". [Cf, Hubschmann, Orisnamen, pp. 260-261, 
Eremyan, Armenia, p. 63.] (335, 1) 

2 ?a [See, Appendix III L, v.] 

28 Joh. Kai\, xx, pp. 118-119," lJiu£w![*»* hiuju Ifiupiikw^ lp jhiqjiul^nu^nunL- 
Pjitbh f}*nmml^w^ k t[fyhfi wupu IfnsbijujL jutprrn uppnjh ^pfjqnpfj ". (335, 2) 

29 Mich. Syr., II, 497; BL, p. 223; Dashian, Catalogue, p. 768; Qampean, History, 
II, p. 398. [See, Appendix III for the texts]. The Syriac version dates the Council 
in 1037 of the Greek (i.e. the Seleucid) era, and incorrectly in the 135th rather than the 
175th year of the Armenian era; the latter date being the one given in the Armenian 
version, Kir. Ganj., [p, 69 — p. 39 of the 1865 Venice edition] gives the year as 156 
instead of 175, wbk pro w£k [sic]. One of the MSS in Dashian's Catalogue, p. 108 
gives the date as: "jwilfi whmtih n£k" instead of " jwdfi Zwjn$ w£h". [There 
appear to be a number of misprints or confusions in this note; the 1865 Venice edition 
ofKirakos gives the date as 155, t( ••• i//yi/ihz/£ ji ^uipfnp jjiumh k ifihq. pHLwIjush^h 
iffih^k fj i£kg £tupljLp fjhhiwih, np wjdS ifkp uuinLpu I ". The last number being 
in fact the one given by Adontz in Armenian, since h — 5 in the Armenian numerical 
system. The critical edition of Kirakos, p, 69, corrects the date to 175 = " . . . ul^uhmj^ 
fi 2$Zb Pnuvfyiuhfrh ... ". Finally the correct symbol for the number 100 in the 
Armenian system is £ and not m, as given by Adontz.] (337, 1) 

3° Theodoros of \ARMN [Mich. Syr., II, p. 497] should be read Theodoros [ostan] 
ARMN = " Duwuih Z^jng " [Cf. BL, p. 223], The preceding name is given in the 
Armenian version as Zwpgwj — AKKI-WS, in the Syriac version ; might the final ws 
of this name = ostan and have been suffixed on Arka by mistake [Cf. Markwart's 
notes to Mich. Syr., II, p. 497.] (337, 2) 

31 Step'annos, Incorruptibility, p. 30, Bzunik', one of the southern districts of Vas- 
purakan, was derived from Baz ; the mountain district south of Julamerk is still called 
Baz, Cuinet, H, p. 651, and is still inhabited by one of the five autonomous Syrian 
tribes named Baz. Both the Armenian form pwdjuhfiy found in JJytanes, and the 
form Pdnihji, which reflects the influence of P^hndip < Baz. [See above, n. 32 
and Chapter XI, nn. 46, 48, 54-55.] (337, 3) 

3ia [Mich. Syr., II, p. 497 and notes.] 



NOTES : CHAPTER XII 



31t> [On the evolution of the Armenian Church and its relations with Syria tip to the 
Council of Manazkert, see particularly Ter Minassiantz, Die Armenische Kirche, also 
Ormanian, Azgapaium, I; Honigmann, Bviches; Garitte, Narratio, pp. 108 sqq.] 

31c [Eor a discussion of Adontz's dating and interpretation of the Arabic version, 
see, Garitte, Agathange, pp. 351-353, who expresses some reservations as to Adontz's 
conclusions. Eor the religious situation in Armenia during this period, see the preceding 
note ; and Gouhert, VOrient, pp, 211-246; for a more recent review of the material, see van 
Esbroeck, Chronique.] 

32 " Letter of Accusation of the Kat'o&kos Nerses ", BL, pp. 70-71. [C/, Ormanian, 
Azgapattim, I, pp. 547-548. " Nestorian " is a term commonly used by Armenian 
sources in this period to designate the adherents of the Council of Chalcedon and should 
not be taken literally in all cases. See below, n, 39a.] (339, 1) 

32 * [BL, p. 73, see Appendix III L, iii.] 

33 Ibid., pp. 76-77. (340, 1) 

34 Ibid., pp. 78-80," fei. ifiuuh qjt * + * jpLiug kpt iqkqb Xrhumnpuiljuihg pwipnL.iig 
h fj <kkp lu^fuiupijjq p.biufyhiiijL hh y L qnLj> qgnuuj fi umpp hljkqkgjiu ptiqni.hfj£, 
L phtj. qnum ^luqnpqfj^. * . . (79)", (340, 2) 

3 5 Ibid., pp. 81-84. 

3 5 a See above, Chapter X, n. 100. (340, 3) 

35lD [See, Appendix IIIK for the various versions of this passage of "Agafangefos", 
also behw n. 39a on the " JNestorian " bishops.[ 

35c [Tji e text does not seem very clear at this point, Adontz himself admitted that 
the Arabic version was translated from the Greek, see above n, 15, which is the opinion 
accepted by most scholars, it is therefore difficult to see what he intends by " the Ar- 
menian original of the Arabic version ". On the relations of the various versions to 
one another, see, Garitte, Agathange, passim, and particularly pp. 260-261, 311-323. 
Also, below n. 36.] 

35d [Garitte, Agathange, pp, 196-198.] 

36 Some of the stylistic characteristics of " Agat'angelos " link it with Koriwn's 
Life of Si. Mesrop, so that the hypothetical earlier version of the Life of St. Gregory 
might be attributed to him, but since Koriwn's work has also suffered alterations through 
a curious quirk of fate, its surviving version cannot be considered the original one. 
Lazar Parpeci made extensive use of Koriwn's work, but evidently in a different redac- 
tion, since it is unlikely that he would have altered Koriwn's style to such an extent 
while borrowing entire pages from him, We believe that both these documents: the 
Life of St. Gregory and the Life of St. Mesrop were linked from the start and suffered 
the same vicissitudes. Recently Marr, Christianization, pp, 157, 180 sqq, has presented 
an illuminating hypothesis on the existence of a link between Mesrop and the Life of 
St. Gregory, though his presentation of the problem needs some rectification, [On 
Marr's thesis, Gf. Garitte, Agathange, pp. 338-350], Mesrop's improvement of the 
art of writing in Armenia proved a new and powerful instrument for Christian prosely- 
tism. As the continuator of the work of the Illuminator of Armenia, Mesrop was 
probably the first to collect hagiographie materials concerning St, Gregory and the 
martyred virgins Hrip'sime and Gayane, and to compose a " Life of St. Gregory " 
which is the prototype of the present " Agat'angelos ", Mesrop's disciple, Koriwn, 
made use of his teacher's works, the Life of St. Gregory among them, in writing the 
biography of his teacher, and this is the fashion in which the points of contact between 



470 



NOTES : CHAPTER XII 



the two Lives should be explained. At the turn of the Vlth and Vllth centuries, 
Koriwn's Life of Si. Mesrop was rewritten in accordance with the new mood of dogmatic 
dissentions which resulted from the altered political situation brought about by the 
events of 591, The author of the new version was probably Eznik the Priest, known 
to us through a brief work surviving as a supplement to the History of Agat'angeios 
in one of the MSS of the Bibliotheque Rationale [Macler, Catalogue, cxii, p. 55]. This 
work is entitled " \,£whujqpp Ifwpqwg ^^pg bqpfynihh kpftgrn. " in the copy 
of MS 51 made by me (edited by Ter Mikaelian as an addendum to Sam. Aim,, p, 266). 
[MS, li ** cii, cf. Macler, Catalogue, p. 182], It gives a brief account of the kat'oKkoi 
and kings of Armenia from Trdat III to the pontificate of Komitas I [615-628], Eznik 
was Komitas' contemporary and an eyewitness of the relics of St. Hrip'sime. The 
work closes with the words, *' kqjigp jfewwwli bijhwljh ft pmpt^Smhk^ q/ffip** 
maw " (the next section of the MS is no longer part of Eznik's work). Abbe Martin 
who composed a catalogue of the MS in the biblioth&que Rationale [See, Macler, Cata- 
logue, p. xxiv] suggested in his description of MS 51 [= 112] that our version of Agat'- 
angelos was the work of Eznik, an opinion shared by Langlois, CHAM A, I, p. 103 
" note additionelle ". Marr, CJirisiianizaticm, p. 152, rejects this thesis and believes 
that Eznik's colophon refers only to his translation of the List of Armenian kat'ohkoi 
and kings and not to the Life of St, Gregory. The difficulty is that the List of kafohkoi 
can hardly be a translation, so that the words, " let Eznik the translator of this book be 
remembered ", must apply to the work of Agat'angelos. It is, of course, possible that 
BrnpaSuthhi is used here in the wider sense of " comment, compose (Gf* Qmjjjwhkwh, 
\fwjuhhmg nmSliopihy pp, 179 sqq.). While altering the work of Agat'angelos, 
Eznik likewise retouched the Life of St Mesrop, which was apparently added as a supple- 
ment to that of St, Gregory. The common passages linking ELoriwn's work with the 
prologue and epilogue of Agat'angelos are due to Eznik's pen. Hence we must acknow- 
ledge that the Life of Si, Gregory went through three stages and three re-workings: 
the first stage belonged to Mesrop-Koriwn, the second to Eznik at the beginning of the 
Vllth century, and the third, the nationalistic version dates from the early part of 
the YXHth century. [On Koriwn and iazar Parpeei, see, Abelean, I, pp, 157-176, 
325-359. On the problems of the versions of " Agat'angelos ", see above, nn. 31c, 35a-d, 
and Chapter X, n. 89a, Garitte, Narratio, pp, 50, 254-277. On Eznik, see also Adontz's 
later work, Nsanagir.'] (341, 1) 

36a [Koriwn, X, 2, p, 30, and 90 n, 39,] 

3? J oh. Mph., II, xvii-xxix, pp, 57-64. [Cf. Stein, Studien, pp, 23-24; Grousset, 
Arminie, pp, 292-294, Garitte, Narratio, pp. 175-225, 225-254.] (343, 1) 

38 Sebeos, iii, pp, 36 sqq. [Cf, Goubert, VOrient, pp, 191 sqq,] (343, 2) 

39 Diegesis, according to my own copy made from the Codex Parisinus 900, fol. 144. 
[Cf. Garitte, Narratio, ci-cvi, pp. 40-41, 242-246.] (344, 1) 

39a [There seems to be some confusion or lack of clarity at this point. Already earlier, 
in his discussion of the Council of Dwin of 555 {see above, n, 32) Adontz seemed to take 
the term " Kestorian " found in the Armenian sources too literally. Erom the Vlth 
century on, this term is commonly used by the adherents of the Armenian Apostolic 
Church as a pejorative synonym for the partisans of the Council of Chalcedon, and not 
in the strict sense. Insofar as the Chalcedonians had accepted the Council of Ephesus I, 
they had undoubtedly condemned Nestorianism, but in the eyes of their Armenian 
contemporaries they had fallen back into the same heresy. While Nestorians were 



NOTES : CHAPTER XH 



471 



indeed to "be found both in Syria and in the southern districts of Armenia, the particular 
Syrians who appealed to KaVohkos ISFerses II in 555 represented the opposite extreme 
on their dogmatic position and seem to hare been advocates of the beliefs of Julian of 
Haliearnassus, Consequently, their influence could hardly have pushed their neigh- 
hours, the south Armenian bishops into the rival Nestorian camp, as Adontz seems to 
be arguing, Finally, the creation of a Chalcedonian patriarchate in Armenia during 
the reign of Maurice can hardly be taken as the work of the " national party ", the 
precise reverse being in fact the case, with the t( uniate " Chalcedonian patriarch John 
of Bagaran considered a heretic and rejected by the national Church, Cf, Ormanian, 
Azgagaium, I, pp, 577-579. Garitte, Narratio, pp, 130, 144 sqq,, 246-253 et gassim*] 
39Td [On the early Armenian Church see, Ormanian, Azgapatum, I. Ter Minassiantz, 
Die armenische Kirche, Ter Mikaelian, Die armenische Kirche. Markwart, Die Mnt* 
stehung, Peeter, Becherches, I. Garitte, Agathange and Narratio. Honigmann's 
articles on the Conciliar Lists should likewise be consulted for early Armenian eccle- 
siastical prosopography, and on the Council of Chaleedon, Sarkissian, Chaleedon.'] 

40 IB, IV, iii. (345, 1) 

41 [IB, VI, ii-xv], Cf. M x , HI, lxv, (345, 2) 
41a pTor a recent review of the question of the apostolic mission of Thaddeus and 

Bartholomew in Armenia, see, van Esbroeck, Ohronique, pp. 425-432, See also Duval, 
Idesse; Hayes, Idesse; Sarkissian, Chaleedon, p. 76 and n. 2; Voobus, /Syrian Asceticism, 
pp, 31-61,] 

43 Euseb. Caes., EM, VI, xlvi, 2 [L. II, 128/9], " .,. kcli tois /card 'App&iav 
waavrcos sre/H ^ravoias iiriGTiXXst wv ewetx/coTreue Mepovldvijs ". [See below, Chapter 
XIV, nn, 49, 51,] (347, 1) 

42a [Gelzer, Anfange, p, 172,] 

43 Gutschmid, Konig. Qsrohene, p. 16, [Duval, Mdesse; Hayes, Mdesse*, Tournebize, 
Arminie, pp. 36, 48 ; Ter Minassianz, Die armenische Kirchen, pp, 2-3,] (347, 2) 

4 ^IB,UI, iv, " *••. np Ifwj j] ijiiq^fiwhu qkuwjh kippwmwj", [See above 
Chapter XI n, 47 and Appendix III K,] (348,1) 

Ma [IB, III, xiv, Ter Minassiantz, Die armenische Kirche, pp. 5 sqq,] 

45 Cass, Dio, lxxviii, 12 [L. IX, 304/5], [Cf Toumanoff, Studies, p. 284 Marieq, 
CaracaUa]. (349, 1) 

46 Harnack, Mission, II, p. 163. (350, 1) 
4? Harnack, Mission, p. 166. (350, 2) 
47a ^cf. however, Ananian, La Data,] 

48 IB, III, ii, xi, xii. [Of. Peeters, Intervention, pp. 237 sqq,] (351, 1) 

49 See above n. 44a, (351, 2) 
49a [IB, III, xvi, " • * • Tj^>tan^h nifii kp}g, p quiLumkh Swpohnj* * * ". Ibid. 

Ill, xvii, ss ***{}w£wl]h+" hfiwhkwij i[wpni^It <j>wnkhwj", On the two foci of 
Armenian Christianity in the IVth century, see Ter Minassianz, Die armenische Kirche, 
pp, 4-29; Conybeare, Key of Truth, pp. cx-exvi; Peeters, Alphabet and Intervention; 
Garitte, Narratio, pp. 59-61, 420-421 ; Sarkissian, Chaleedon, pp. 80 sqq,] 

50 Might the so-called " Danielian " characters [Koriwn, vi, pp. 16-17] be associated 
with the name of this missionary ? Just as Mesrop, the representative of Greek Chris- 
tianity in Armenia, concerned himself with the improvement of the written language, 
for the purpose of proselytism, so the representative of Syriac Christianity must have 
been concerned with the self-same problems. It is very likely that it was Daniel who 



472 



NOTES ; CHAPTER XII 



adapted the Syriac characters to Armenian, though they were soon abandoned as 
unsuitable. Then, with the re-awakening of interest in this script, the name of its 
creator was again remembered. Since there is much that is legendary in the account 
of the invention of the Armenian alphabet, it is not surprising that Daniel should be 
described as a contemporary of Mesrop living somewhere in Mesopotamia. There 
would have been no reason for a Syrian bishop to have preserved the Armenian alphabet. 
[See Adontz's later work on this subject, Mastoc, also Markwart, Armenischen Alphabet; 
Peeters, L' Alphabet; Sarkissian, Chalcedon, pp. 85-97; van Esbroeck, Chronique, pp. 435 
sqq.] (351, 3) 

50a [There is no evidence that Daniel belonged to the house of Albianos of Manazkert. 
The claim is made for Yusik, 2awen, Sahak and others. See, Garitte, Narratio, pp. 418- 
421, 425), but Eaustus specifies that Daniel was a Syrian by birth, FB, III, xiv, " J}l kp 
um ^mhji^ wqjfUiL wunpjj ". Cf. Ter Minassiantz, Die armenische Kirche, pp. 5-8, 
13-17. Sarkissian, Chalcedon, p. 83 and n. 2.] 

51 MX, II, lxxiv, " h tuhrj. muhu tjjqpLpjii.fi Sop uppnj k i/kbfi l^nuiwuip^fih : 
H^wuh npnj k qhnpfiu JXnm^k^nj ^Lnp^h phl^mjhmy np wn ^whquinuipmhwL 
unpfiu q^hkjnLpjiLbh fam, qlwpjiu k^fiij qCnqhnp S^wtytiLpkujuu mml^munh 
PfjLh". £G, p,21. (352,1) 

52 JB, IV, ad. (352, 2) 

53 Ibid., TV, xv. [As in the case of Daniel (see above n. 50a), there is no reason for 
associating Cunak with the house of the bishop of Manazkert; Eaustus makes it amply 
clear that Cunak was not a man who could boast of his noble lineage, " • . ♦ Qmuuilf nh% 
whrnSj*" umpni.il fi umpjiL^m^ wpgrnSifj ". Cf. Ananian, La Data, pp.356, 
359-360.] (353, 1) 

54 FB, VI, i, (353, 2) 

55 Jbid., V, xxxii. (353, 3) 

56 Ibid., V, xxix, [See above n. 50a.] (353, 4) 

57 [Ibid., VI, ii-iv] Sahak is called ^np£kwj. If this reading is correct, Sahak must 
previously have been bishop of Korcek'. It is more than likely that he. too was from 
the house of Albianos, [See above n. 50a, also Ananian, La Data, p, 360.] (353, 5) 

57a [Though occasionally oversimplified or overstated, Adontz's thesis of the alter- 
nation of two parties in the Armenian Church during the IVth century: Syrian- Greek, 
Sasanian- Byzantine, Albianid-Gregorid, is both illuminating and borne out by the 
sources, as I hope to show in my forthcoming study on Armenia in the Eourth Century. 
Of. Sarkissian, Chalcedon, p. 80 and n. 1 et sqq.] 

57b [On the organization of the Church at the time of the Council of ISFieaea I, see, 
Fliche- Martin, pp. 437 sqq, ; Dvornik, Apostolicity, pp. 3-38 ; Jones LRE, II, pp. 873 sqq.] 

57c [Jones, LRE, I, p. 47.] 

57d [Cf. Dvornik, Apostolicity, pp. 17-18.] 

58 [Mansi, III, col, 560], Benesevic, Syntagma, p. 96, 

" Tovs VTrep SwiKTjGiv imoKOTTQVs reus VTTGpopiois GKKAyoiais pi} hriivai, prqhk avyx^ lv 
Tas e/cA^crias' dAAct Kara tovs Kavovas rov pkv ^AA^avdpeias emWo-nw ra iv 'Aiyvirrtp p6vov 
oiKovopdv, tovs Se tt}s dvaroXTJs emoTcoTrovs ttjv avaroXyv povqv Biqlkglv, <j>vAaTTQp4wov 
rwv cv tois Kavovi rots Kara NiKaiav TrpeajSeiW tt} *Avtioxgwv e/acATjcria, /cai tovs rrjs *Amav7}s 
hwiKiJGZCOS £tti(?k6ttqvs to. Kara ttjv *Aoiav povijv oucovojuetv, kclI tovs ttjs IIovtw7)S to, tt\s 

HoVTtKTJS p.QVOV, /Cat TOVS T^S 0pq,KT\S TO, TT)S SpdxTJS pQVQV olKOVOpSlV ". (355, 1) 

59 [Mansi, II, col. 669 sqq,] = Benesevic, Syntagma, p. 86, 



NOTES : CHAPTER 331 



473 



" Td dpxaia Wt] Kpardrco, rd iv AiyvTrrtp /cat AiftvTj /cai i7evra7roAet, ware rov *^Aef av- 

Sjoetas iiricfKOTtov tt&vtwv rovrcov f^eiv t^v i^ovuiav, i'JTGihr) koX rep £v t Pu>jji.7j €mewco7ra) tovtq 

ovvvr}94s iariv voy.oicos 3e /cat /card tt)v '^vTiox eiav * a * ^ Ta ^ d^Xais £-napxlais t tci irpecr/feia 

aco^ecrflat tciis iKKXrjcrlms ". (356, 1) 

60 Markwart, Staatsverwaltwig, I, p, 322. [Seston, Diocletien, pp. 294-351' e£ «?.]. 

(356, 2) 
eoa [(?/, Dvornik, Aposioliciiy, pp. 6 sqq. ; Jones, -LR5? II, 883 sqq.] 
6013 [16R, pp. 11-15,] 
si .flfcwwi, YII, eol, 428 = A.C.O., II, i, 3, p. 88 sqq, 

** ... KoX TOVS T7JS IJ0VTIK7)S KCM TT)S *AoiaV7}S KO.I TT]S &pO,KlKlf\s 8wiK7}0<ztQS ^TpOTToXtraS 
[IQVOVS, €Tt 0€ ATat TOVS *€V TOIS fiapfiapiKQlS imVKOTTOVS TtDv TrpOGlpT}p.4}>Q)V BlOlKTJfJSCOV X^tpOTO- 

vsiudai vttq rov irpoGt,p7)fji,4vov dyicordrov Bpovov rrjs Kara KmvoravrivovTToXiv dytwrdrrjs 
iKK/tymas* S^AaS^ £k&vtov p^rpoTToXirov rwv / 7Tpo^ip , qp t ivo>v Sioi/c^crecov fisrd rtov rfjs iirapxicts 
hriGKQittAV ^eipoTovovvTos 1 rovs rr\s iirapxias iiriuKoiTovs, KaBcbs rots dsiois kovqui Bnjyopzvrar 
XtipoToveiodai Se Ka9cos ^tpTjrai, rovs p^rpoiroXiras rtov Trro$ipT}}i4vQ>v bioiKTJoewv irapd tov 
KovaravrivovTroXecos dpx^^^K07Tov ^^tajuarcuv ov^wveov Kara to %$os ytvop,4vcov Ka\ 
€w' avrov ava<j>epQtx4va)v ". [(7/, Dvornik, Apostolicity, p. 82 sqq.] (358, 1) 

61a [There must be some misunderstanding here, the patriarch John the Faster was 
not a contemporary of Justinian but of Maurice, he occupied the patriarchal throne 
from 582 to 595 and took the title of (Ecumenical patriarch ca, 587, though the title had 
been in sporadic use before. The patriarch of Constantinople in 558 was Eutychius. 
See, Danielou-Marrou, BEE, I, p. 441 ; CMH, IV, 1, p. 800; Grumel, Beqestes, I, eelxiv, 
p. 105, et ah] 

63 Bee above p, p. 72-74 for the residences of these metropolitans. The ecclesiastical 
divisions of Armenia were not altered by Justinian, see above p, 73. (360, 1) 

63 Gelas, Cyz.y p, 1310 gives the following provinces as being under the jurisdiction 
of the bishop of Caesarea, " Ilovrov IJoX^iiaiKov, > Apfuzviav p-wpdv koi p&yak'qv ", (360, 2) 

64 FB, V, xxix, " &t jipgu uiju fik& qmpSwhwipil rjwuilwh hlfkwj^ £wj— 
pwtqkmfjh ^humpnL tjwuh jiwhfm mjunpft^ li kqf*L dnqpi^ kuijiulini^numgh 
ufuh^nqnufjh hm^whqjih ^humpnu wnwhtj ^mjpmuihmft^*** k qphgfih Pnuip 
mn pwij.WLnph tywiq, k iptbfji qJj^jumhnLpfiihh ljiupnQ[jl}nunLphwhh, zjji 
np ifyfib imjpmu^hm Z^jntj. hw qpmhh wpgnLtiji fefukugt q^mgp opthb^ 
L dp p^fuhuift ^hnhmqph^ qhmjiuljnu^nuh Zwjnjj) "P^t 11 unjfnpnLpjjii £p fi ph£ x 
fci. jmjhS £hw£ pwp&uiL fefuwbiiLpfiLiih Z&]nj] qkmfiulpiuimiij khnhmfj.ph^ 
W JL n P L^Hth huffiu^u^ni^nu^ wdhhwjh ipnunnwij quiLuinwij Z^jng h l^nqSmh^^ 
hnqpmhq^ wjIwL^htnL jwjbii £kw£ np ifjimhii.mS ^Jih£fjb kuffiulfnufnug uw£dw— 
hwijh Zwjng** kppiuj[iii ji gwquigh i^humpmging^ k whq ^h^fili hu^fiul^nu^nu^ : 
^wuh qfj jwfifi Qiml pwp&WL fi^fuwforiLppLhli jkpfyp£h Zwjnijy wjhfiL^kwhi. n£ 
hyjuifih hu^ul^niqnu^ khnhmqphij p m J9 n P Ll 1 ^P mLUJI l kiq^u^niqnmn^h^ fi 
t[kpiaj tffiwjh huw£p, L £iutj op£h£p pwqmuipwijh », (362, 1) 

65 St. Basil, tip., scis [L. II, 170/1-182/3], exx-csxii [L, IV, 244/5-252/3], Note 
p. 248 n, 2, that the editor seems to have mistaken the proper name, Pap, for the title, 
pope]. Gelser, Anf ange, p. 159. (362, 2) 

65a [2B, IV, iii; VI, v.] 

65b {Cf. Garitte, Narratio, pp. 56-57,] 

66 See above, Chapter II, n. 22. (363, 1) 
66a [On Nisibis and Amida, see, Duval, Mdesse, pp, 138-146; Peeters, Legende; Voobus, 

Syrian Asceticism, I, pp, 142-143,] 



474 



NOTES : CHAPTER, XII 



67 Mansi, IX, 391, " Gregorius ... ep. Iustinianopolitanorum civitatis magnae Ar- 
meniae ". Ibid., p. 191, the name of the same personage is given as *' Georgius ", 

{363, 2) 

68 Mov. KalanJc., p. 212, (364, 1) 
68a [Koriwn, xiv, p, 38, " ... BiipninnLmlimh biqliuJjnuimili ^bp^whwj^ npmS 

rnhndi l[n^ip Qfihp, ... ", cf. p. 97 n. 63. Garitte, Agathange, p. 209.] 

69 On Theodore, see above. Chapter VI n. 53. On George, see above, Chapter III 
n. 20, also, Mansi, XI, pp. 613, 993. On Marianos, Ibid,, p. 1005, " Mapiavos avag en, 
Kidapi^cov Tijs TrpwTTjs {sc. TGTTtipTTjs A pro A) tcov 'Apfj,€vicov iirapxias ". [Cf, Garitte, 
Agathange, pp. 209, 212-213.] {364, 2) 

69a [See above n, 64.] 

70 JfX, III, lxv. Tov. Arc, I, xi, p. 74, " JXhwrnnqji qopwi[wph QmSiwg 
q£wu}iJi ntfu '^ni^nQuj^ji Ipunnjt] Smufihu jmSiwy busjiulinuinuuiiubw. k 
wjuujiu wqSIfkwi* k £ijinpkw[_ bpllfipu ZwjntJ ^ m JP uiuifywpwunLpbuiiip, j^ji 
wiiljuipifiiLphwiip ". (3y5, 1) 

7 °a [See above n, 64.] 

71 Mansi, II, 669, " * Ettiq-kottqv TTpooTjKZi ixaXiara p,h> vtto irdvrtiov rwv iv rrj iirapxio- 
KaBioraadav ei Se Sw^epes s?7) to toioutov, t) 8td KOTGirziyovvav avdyKTjv rj Sm p,rJKOs 
obov, i^amavros rpsls em to avro uvvayofAGvovs . avfjupTj^cov yivop,evcov rcov amovrmv /cat 
uvvTiQsjisvwv bid ypafj,p,drwv, tots t^v x<=iporoviav noislaOai, to 8e Kvpos rtov yivop4vwv 
St8dcr0ai Ka0' iicdoTqv iirapxiav tw p,<zTpo7To\iTj) ". [Qf, Dvornik, Apostolicity, p. 6, 
Jones, 2JUB, II, p. 880.] {365, 2) 

?2 #5, IV, xv, (366, 1) 

72a [This conclusion cannot be reconciled with the evidence of the sources on the 
presence of other bishops in Armenia, and particularly with the list of Gregorid consecra- 
tions given in the various versions of " Agafangehs " some of whom, e.g. Albianos 
of Manazkert are authentic IVth century figures, Cf. above nn. 14a-b. If the Ar- 
menian hierarchy was still almost inexistent in the Kat'olikosate of JSFerses I, who would 
have been the dignitaries summoned by him to the first Council of Astisat c. 365 ? Cf. 
FB, IV, iv, " fci fihgh\\>bpulu\ ... dnqnt[ip um fchgh nuiilbuwju buj^ulinuinu- 
rn.hu Zwjng wyJvwpCJiu : (frnqnilbijwh ft olinji H^mfi^iuin^ nLp n^umui^fiuu 
n^klfbiibgfiu £p ^fjubwi^ q^fi hm £p Swjp blibnbrjbwtju, k mbqji jbwi^ huiju- 
ubuitjh dnqnijntj ufiLu£nq.nujiu : fr dnnni[ iuiLiuum-PfiLuu blfbiu^ wifbubrjnLu, 
L junpimpn. pwhbwEj fi z?^ wplpuhj;fiu, ipumuipbi^ ivuq. t^iu^fump^mljusu 
liwpn.u blfbnbzjLnjh It dnqnijh CuiLiuwnzjh /iwLwmupnLpbwh ". Cf. Garitte, 
Narratio, pp. 99 sqq.] 

73 Ibid., V, xxi, " pwndwtjnjt] qljiupqu upu-£uioul;fizj"* b jwifbhuijh qui— 
Liunu Ifuitjnjrj wbrniLUi biufiuljnuinuu ". Ibid., V, xxxi, " z^u/jw^i. £wpnu 
buifiul[niqnuujn jwilbuwju rj.WLiun.ij £uijnEf jw&w]vip puui mpdmuji fiLpkwbrj ". 

(366, 2) 

74 Ibid., V, xlii, " jwdbhiujh qwLwnu IpugnLijiuulp IftuhnL^ uiu£wuibuw k 
mbwpu qwLwnwtj q.WLwnwt} ", {366, 3) 

75 The Greeks used this argument against the Armenians in the period of intensified 
ecclesiastical controversies. According to the famed learned monk Solomon cited by Mov, 
Kalanh., II, xlviii, p, 309, 312 the Greeks pointed out that the ecclesiastical hierarchy 
should comprise nine degrees; namely, " ^mmpjitupq. np £ £uijpiuufhw k mp^buj^jiul^ 
niqnu np bin^i buifjul^nuinumuibm h Ifwpnqfilfnu, dbwpniiin^fiin k biqfiulinj^nu^ 



NOTES : CHAPXEE XII 



475 



gw£iuhwj, uuiplfmLmq, l^kuuwpl^wLm^ phpkpgnfy k lfiuwqg »♦♦ *\ and said, 
" bpt nq2SmriLpkmSp ^wLWtnwjg, jttnuwnifwh jhpnLg, kpi m[ £ iqwinp- 
fiwpqh &hp ". They asked which of the four original patriarchs had had jurisdiction 
over Armenia, since Gregory had heen consecrated at Caesarea merely as archbishop. 
Influenced by theses arguments, and " ♦•♦ ijmuh tlhbmpmhnLpkwti QjilIjujij^ . 
np ifmgpliwtjtJLtjwhhjL ^uihiujlih lty w J u £uihquwhwh f&wqinufj umwghjpf 
^mSikjL iwjfiulfniifnuwiqkm k ilhmpnu^n^jun, ", Idem, the Armenians attempted 
to fill in the missing degrees and pointed to the bishops of Mardpetakan and 
subsequently of Siwnik' as metropolitans. According to another source, the Arme- 
nian kat'oEkos "was considered to be a patriarch, the kat'olikos of Albania was an 
archbishop, and the bishops of Siwnik', Iberia and Mardpetakan, simple bishops. 
Joh. Kaf, xii, pp. 62-63. Stqph. Orb., vi, I, pp. 60-63, [See also next note]. (367, 1) 

76 In the Armenian translation of the Canons of the Council of Nicaea, the word 
(XTjrpo'TToXiTTjs is translated ^uf^?j?^r^j7ir cf, MehV-Tangean, Canon Law, p. 256. In 
JFB, IV, iv, bishop Eusebius of Caesarea who consecrated Nerses I is called " Ipupniiji- 
fynuiug l^mpnqji^nu " which corresponds to the title metropolitan of metropolitans, 
which was in fact the position of the bishop of Caesarea. Scholars believe that the term 
KadohKos was originally used to designate a financial official, and was only subsequently 
■used to designate the heads of the Armenian, Iberian, and Syrian Churches. Gelzer, 
Anfdnge, p. 139, Incicean, Antiquities, III, p. 251. Frocopius, Pers., II, xxv, 4 [L. I, 
480/1] understands this title literally, " rov re t&v xpiaTiavwv lepia KaBoXtKov xaXovcn rfj 
5 JSXK^vcov tjxjwf}, ori 8^ ifpevr^Kev els cuv airaai rots ravrji x w pi°*$ "• ^ * 3 altogether 
possible that the title Ka96XiKos for KadoXiKos etrlaKcmos was the original title of the 
chief bishop of an eparchy, who was later called emcwoiros p>7}rpo7roXirT}s f because he 
resided in the metropolis of the eparchy. [C/. above n. 65b. Also, Kogean, Armenian 
Church, pp. 109-112, and van Esbroeck, Chrmique, p. 435,] (367, 2) 

77 See above n. 64. (368, 1) 

78 Mansi, XI, 957/8-959/60 " ^JS-jreiBTJirep eyvu>p,ev h> rf} *App.ev{tov x&PQ P>ovovs h> ^A^/jcw 
rovs £k yevovs iepartKov Karardrreo9at lovBaiKots Weeny hrop,e r v(jov rwv rovro irpdrreiv hnx &r 
povvrcov, rivas Be avr&v xal p,r) aTTOKeipopevovs teporpdXras /ecu avayvwaras rov Belov vqjmov 
KaBtoravdai, vvveiBop,ev, ware amo rov vvv firj efeiycu rots els kXtjpov fiovXop.evoi$ irpodyew 
rwas els to yevos cwro/?Aejmv rov Trpo^tpi^o^eVou, dAAct BoKip.dlovres, el aftot elev Kara rovs 
reBevras e> rots lepots xavovw opovs ev /cA^paj «raraAey^vai, rovrovs l/c/cA^cnacm/ccos wpoxeipi- 
£€<70atj etre /ecu eK irpoyovoov yeyovaow lepio>v elre /ecu p/rj *\ (368, 2) 

79 Gelzer, Anfange, p. 140, also mentions Judaizing independently of the Council 
in Trullo, " ... der Katholikat hat thatsaehlich wenig Aehnliehkeit mit dem christlichen 
Episcopat, sondern erinnert vielfach an jiidisehe und heidnische Vorbilder, Dahin 
gehort vor allem die Erblichkeit des Hohenpriesteramtes ". This characteristic was 
due to the influence of nayarar practices and not to Judaising tendencies as was believed 
by this scholar followed by Weber, Kathdlische Kirche, p. 218. (369, 1) 

so LP\ Ixii, p. 354. [Cf. Garitte, Narraiio, pp. 99-100, 422-427.] (369, 2) 

80a [While there is no doubt that the Mamikonean bishop played an important part 
in the Council of Dwin of 555, and he signs the Acts of the Council of 505 immediately 
after the kafohkos, the precise status of the bishops does not seem to have been rigo- 
rously fixed in the ecclesiastical hierarchy. In the listing of the bishops at the Council 
of 450, the second place is given to the bishop of Siwnik', while the Mamikonean bishop 



476 



NOTES : CHAPTER XII 



of Taron is found in fourth place. On the other hand, the bishop of Siwnik' has dropped 
to ninth place at the Council of 555. [See, Appendix III L, iii], Nevertheless, there 
seems to be no question but that the great houses had their own bishops who shared 
in their prestige.] 



NOTES : CHAPTER XHI 



477 



CHAPTER XUl 



a [For a critique of this entire section of Adontz's work, see Toumanoff, Studies, i, 
particularly p. 70 n. 76. On the Iranian social structure, see, Benveniste, Les classes 
societies; Ehtecham, Vlran; Ghristensen, *' Introduction " et passim,* Frye, Persia, 
pp. 49-52, etc. For a summary of Dumezil's controversial thesis, see, his Idiologie 
tripartite, and for his application thereof to Iran, his Naissance d'archanges. See also 
the bibliographical and historical references in Duchesne-Guillemin, Religion. Many of 
Adontz's Iranian transcriptions and a number of his interpretations are now out of date 
and should be checked against recent scholarship,] 

1 Montesquieu, De V esprit des his, XXX, i, II, p. 249, " C'est un bean spectacle que 
celui des lois feodales. TJn cheque antique s'eleve; Tceil en voit de loin les feuillages; 
il s'approehe, il en voit la tige ; mais il n'en apercoit point les racines : il faut percer la 
terre pour les trouver ", (373, 1) 

2 Kuhn, Verfassung, " Eorword ", p. ix, '* Das romisehe Reich ist zu denken als axis 
Stadte bestehend, welehe der Kaiser beherrscht ", On the other hand, Spiegel, on of 
the specialists on Iran, insists, Iranische Stammverfassung, p. 685, that " die Verfassung 
der alten Perser und des alten Irans uberhaupt eine Stammesverfassung war ". (373, 2) 

3a [Ehteeham, Vlran, p. 18, " Les Iraniens, bien avant leur separation des branches 
indo-europeennes, e'est-a-dire anterieurement a la fondation de TEmpire des Aeh6me- 
nides, ont constitute tine sociSte patriarcale du type gentilice ", Ghristensen, p. 15, 
" Les Iraniens ont formes des les temps les pins anciens, une societe patriarcale ... . 
Dans Tlran occidental, la base patriarcale de la society se cache en partie sons nne surface 
empreinte de civilisation babylonienne .... Mais l'organisation patriarcale n'avait 
jamais cesse d'exister ". Ibid., p. 17, " ... le royanme des Arsacides, malgre son vernis 
hellenique, est reellement d'un iranisme pins pur que celni des Achem&oides ... , Avec 
cette predominance des Iraniens septentrionaux Tancien regime patriarcal reprend 
vigueur. La notion de la filiation gen^alogique de la soci6te a' est eonservee pendant 
bien des siecles, meme apres la chute de Tempire des Sassanides, dans la eommunaute 
zoroastrienne ",] 

3 Weissbach, Keilinschrifien, I, xii, p, 14/5 [Kent, Old Persian, pp. 117, 120, " 0atiy: 
Darayavaus: x^y a 9iya: aita: xs a 9am: tya: Gaumata: hya: magus: adina; Kabujiyam: 
aita: xs a 9 am « haea: paruviyata: ama^am: taumaya ... Saith Darius the King: This 
kingdom which Gaumata the Magian took away from Cambyses, this kingdom from 
long ago had belonged to our family "]. Weissbach, Keilinsckrifien, I, xiv, p. 14/5, 
[Kent, Old Persian, pp, 118, 120, " #atiy: Darayavaus: ^saya^iya: xsacam: ^y a: h ac g, ; . 
ama^am: ta umaya: parabartam: ... Saith Darius the King: The kingdom which had 
been taken away from our family, that I put in its place: I reestablished it on its foun- 
dation "], Darius characterizes his reign as follows, Weissbach, Keilinschrifien, IV, 
IV, LXII, p. 28/9 [Kent, Old Persian, pp, 129, 132, u 0atiy: Darayavaus: xsaya fliya: 
avalM/arac%: Auramazda: upastam: abara: uta: aniyaha; bagaha: tyai#: Ttatiyi yaB&i 
naiy: ari&a: aham: naiy: draujana: aham: naiy: zurakara: aham: naiyi adam: mtimaiy: 
tauma: upariy: arstam: upariya#am; naiy: skaurim: naiy: twnuvatam: zura: akunavam: 
martiya; hya: hamataxsata: mana: v£0iya: avam: ubartam: a&aram: hya: viyanafl&ya: 
avam: ufrastam: aparsam: ,„ Saith Darius the King: IPor this reason Ahuramazda bore 
aid, and the other gods who are, because I was not hostile, I was not a Lie-follower, 
I was not a doer of wrong — Neither I nor my family. According to righteousness 



478 



NOTES : CHAPTER XIII 



I conducted myself. Neither to the weak nor to the powerful did I do wrong. The 
man who co-operated with my house, him I rewarded well; whoso did injury, him I 
punished well ", Of, Erye, Persia, iii, pp, 83 sqq. etc.] (374, 1) 

4 Strabo, XV, iii, 2 [L. VII, 156/7], " ol yap JJipoai Kparrjuavres MijBwv /cai o Kvpos, 
... ivravBa \Sovuis] Wgvto to ttjs rfyepovtas fiaoiXziov' dfia Kai to 6/j.opov ttjs X^P * dTroSe- 
£dfj,izVOi kclI to d^imjxa ttjs TroAecus* /ecu KpeTrrov to /x^SeiroTe /ca0* iavTTjv ttjv JEtovviBa 
TTpayiAaTWV fxeydXcov iTTTjfioXov ysyovivai, d/U' ad u<£' iripois virdp^ai Kai iv pipst Teraj^ai 
woTrjfiaTos {jl^I^ovos, ttXtjv €i dpa to TraAcuov to Kara tovs yptoas ". The Persians 
shifted their capital to Susa so that the Susians should not plot to defect, but rather 
remained within the great confederation and not as they had lived in heroic times, 
[Adontz translates " ovcmjpa jxei^ov " as " great confederation ", whereas Jones, 
L, VII, p. 157 gives "larger organization". Of, Toumanoff, Studies, pp. 36-37,] (374, 2) 

4a [This interpretation is questionable. See Kent, Old Persian, p. 190, dahyu, dasyu.] 

5 Weissbach, Keilinschriften, pp, 44/5, 40/1, 42/3 [Kent, Old Persian, pp. 137, 
148-149, 152-153, 190, 211. See also above n. a, and below n. 8.] (375, 1) 

5a [Ehtecham, Vlran, pp. 24-25, 40-41, 91 sqq., 110 sqq. Erye, Persia, pp, 66, 70-71, 
90 sqq,] 

6 Herod,, I, 125 [L. I, 164/5], ,s eun Se JTepo-eeov xro^yd yivza, ... sort Se rdBe, ef Sv 
cSAAoi irdvTss dpriarai Uipaai, UaaapydBai Mapd<j>ioi Mdumoi. tovtwv IJauapydBai 
apiGTOt, h> Toio-t Kai * Axaiy.zvihai elm ^pTJrprj, ev$ev ol jSaaiAees ol TJepaeiSai ysy 6vaoi,aXAoi Se 
Tlipaai €ial otSe, HavdiaAaioi AypovmaToi r^pfidvioi, odroi ftev irdvrzs dporrjpes ewri, ol 
Se a'AAot vo/^aSes, Adoi MdpZoi ApomKoi Uaydprioi ", Ibid,, I, 101 [L. I, 132/3], " eem 
Ss Mijbcov Toaa8e yivsa, Bovvai IJapyraiajvoi ZVpou^ares 'Api^avroi Bovhioi Mdyoi, yivsa 
ph> br) M-rjScov iorl roadbs ", (376, 1) 

6a [EhtScham, IS Iran, pp, 20 sqq,, 26, 39. Erye, Persia, p. 50 table,] 

7 Weissbach, Keilinschriften, III, si; II, xxiv. [Kent, Old Persian, pp. 123-124, 
126-127, Weissbach, p, 23 translates " das persische Volk das im Palaste war ", 
whereas Kent, p, 127 gives, " the Persian army which (was) in the palace " the same 
translations are given for the Medes. Cf. Kent, pp, 179-180, 208; Ehtecham, Vlran, 
p. 63 n, 5.] Note likewise that the entire Median people is called hara while the dynasty 
of Cyaxares is called ianma [" Uvaxstrahya: taumaya ", Kent, p, 121. Of, Erye, 
Persia, pp, 91, 103]. (376, 2) 

8 These terms are often found in Avestic literature, the most valuable passage is 
in Yasna, XIX, 18, where the answer to the question " Who are the lords ? " lists four, 
and a fifth who is Zara lustra, " kaya ratavo, nmanyo, visyo, zantumo, danhyumo 
Zara #ustro pu^to ". ' Yasna, IX, 27 gives the synonyms: Nmanopaiti, vispaiti, zantu- 
paiti, danhupaiti, spananha vaedyapaiti. The last term " head of the religion " = 
Zara lustra. 

tauma zend. tao^man, " family, offspring, kinship " skt. iakman, gk. tIkvov 

" child ", Tojcevs " parent ". 
zantu, skt. jantu, lat. gen<ti>s, gk, yivos, arm. bjth, jan, zan, whence also 

zana, arm. £biu — qwhq, 
nnmnu, O.P, manya < man, 

vi$, skt. ves-as, vis, lat. vic-vs, gk. oIkos = oIkos, slav. 6bcb. 
danlm, skt. dasi, pers. deJi, 
[Of, Kent, Old Persian, pp. 185 and 218, 211, 202 " maniya ", 208, 190 "dahyu". 
Markwart, Eran,, pp. 122-124; Ehtecham, L'lran, pp. 40-41, etc.] (377, 1) 

9 Eor example in Darius' inscription, vi 9 is used as a termiwis gentis, " vi to: 



NOTES : CHAPTER XIII 



479 



tyam: ama^am: ga#ava; avastayam ... ", "I reestablished our royal house on its foun- 
dation as (it was) before ", [Kent, Old Persian, pp. 118, 120,] (378, 1) 

10 Weissbach, Keilinschriften, p. 34 [Kent, Old Persian, pp. 135-136, Ehte"eha:m, 
VIran, p, 21 and n. 3. Prye, Persia, pp. 18, 51-52. This interpretation is doubtful,] 

(378, 2) 

ii Xen. Cyrop., I, ii, 3-15 [L. I, 10/1-24/5]. See, Appendix II for the test. (380, 1) 

i 2 The word sabha has survived in German in the form sippe meaning t! kin " a 
semantic change showing that the sabha was an assembly of groups of kinsmen. [Cf. 
Vernadsky, G., Kievan Russia (New Haven, 1948), p. 134. The form abicaris is proble- 
matic. See, Kent, Old Persian, p. 68.] (381, 1) 

i3 Xen. Cyrop., II, i, 14 [L, I, 142/3], " ev p.ev rfj 77077x81 ov juere^sre twv I'crcuv rjiMv, 
ovx vfi -rfixwv amskaSivrss aAA' vito rov raTTtTrjBeia avayKTjv vpiv etvai iropi^oBai ". (382, 1) 

1 4 Ibid,, II, i, 2 [L, I, 132/3-134/5] " ... Cyaxares asked Cyrus how large the army 
was that he was bringing, 

' Thirty thousand ', he answered, ' of such as have come to you before as mercenaries ; 
but others also, of the peers [o/aoti/aoi], who have never before left their country, are 
coming \ 

* About how many ? ' asked Cyaxares. 

* The number ', said Cyrus, ' would give you no pleasure, if yon were to hear it; but 
bear this in mind, that though the so-called peers are few, they easily rale the rest of 
the Persians, many though they be '. [ odroi ol ofiorij^oi KaXovfievoi ttoXAcqv ovtwv rwv 
aAAcov IIzpacQv pahlws apxovmv "]. 

Ibid,, I, v, 5 [L. I, 78/9] t! ... Cyrus had by this time completed his ten years among the 
youths also and was now in the class of mature men. 

[5.] ... and the elders in council chose him commander of the expedition to Media. 
And they further permitted him to chose two hundred peers [BiaKomovs ra>v o/aotijuW] 
and to each of the two hundred peers in turn they gave the authority to choose four 
more, these also from the peers [4k rwv 6jioTip,o>v\. That made a thousand. And 
each of the thousand in their turn they bade choose in addition from the common 
people [e/c rod h-qixov] of the Persians ten targeteers, ten slingers, and ten bowmen. 
That made ten thousand bowmen, ten thousand targeteers, and ten thousand slingers — 
not counting the original thousand. So large was the army given to Cyrus ". [Cf. 
p. 79 n. 1]. Here the o^drt/xoi are opposed to the St^ios., The term ojaqtiixoi designates 
the men picked from the same class as the chooser. Prom Cyrus' point of view all 
thousand men were d^drt/zot i,e. [peers] from the class of mature men. This is also 
evident from the fact that they carried the same wapons as mature men, Ibid., II, i, 9 
[L. I, 138/9], " 9a>pa£ jjl£i> irspi ra oripva 3 yippov Se eis tt}v apiarGpav, kottis Be 77 aayapis ets 
rqv Beg lav ", cf. I, ii, 13 [L. I, 22/3, see, Appendix II]. Might ofionpoi be the Greek 
translation of the Iranian word preserved in Armenian as CwdwCwpii < harz, arz — 
<j* * I and iupd — uthfi—g^ with the sense (t equal in worth, or equal in honour ". [Cf, 
Hubschmann, Orammatik, p. 177.] (382, 2) 

14a [Xen. Cyrop., I, ii, 2 [L.I, 10/11-12/3], "The hucksters with their wares, their 
cries, their vulgarities are excluded from this and relegated to another part of the city 
so that their tumult may not intrude upon the orderly life of the cultured ". See, 
Appendix II for the text.] 

1413 [Cf. Hubschmann, Orammatik, p, 242. This relationship is very doubtful.] 



480 



NOTES : CHAPTEE XIII 



14c [Coulborn, Feudalism, 108-119, Chrisiensen, pp, 15-16; Erye, Persia, pp, 90 sqq,, 
106-107, 182-188, etc.; Ehtecham, VIran, pp. 18-21, 40, 47, 110-115; Toumanoff, 
Studies, pp. 34-40 and nn. 13-14.] 

14d [This thesis which was accepted by Bostovtzeff, was rejected by Brundage, in 
Coulborn, Feudalism, p. 110, " In theory the king was absolute; in practice the Aehae- 
menian state was a fairly stable symbiosis of crown and aristocracy. [The] rebellions 
nobles ... never ... attacked the integrity of the crown itself. ... feudalism never emerged 
out of the disintegration of Achaemenian times. ... Bostovtzeff says, * The feudal 
structure of the Parthian Empire was inherited by the Arsacid from the Achaemenids 
and was transmitted by them to the Sassanian kings \ This comment must be rejected 
for the reasons which have just appeared ", This point of view is shared by Chrisiensen, 
p. 16, " ... le flodalisme ne s'6tait pas encore developpe* sons les Aehemenides "♦ Cf. 
EhtScham, VIran, pp. 47, 52 sqq., etc.] 

15 Xen. Oyrop., VIII, vi, 1 [L. II, 408/9]. [Herod., Ill, 89, L, II, 116/7]. (383, 1) 

15a \Pf* ^rye, Persia, pp. 182 sqq., and above nn. 20-21. Debevoise, Parihia, 
pp, xxxviii-xxxix. Debevoise, PariMa, pp, xxxviii-xxxix.] 

is Herod*, I, 125 [L. I, 164/5, Of. Erye, Persia, pp. 48, 172-175,] {384, 1) 

17 Strabo, XI, viii, 3 [L. V, 260/1], " ...rcbv Aawv ol p,h> irpoaayopsvovrai "Airapvoi, ol 
SI EdvBwi, ol Bk IJlaaovpoi ". Cf. Ibid., XI, vii, 1 [L. V, 248/9]. Tomaschek, Aparnoi, 
takes this name to be avest. aperenayuka, 0,P, *aparenayu < a -\- parena — " not -f- 
full " (1SF.P, burna, slav, FFK/l- H) with the sense, " unripe, immature ", [See also 
preceding note.] (384, 2) 

I'a [Cf, Chrisiensen, pp. 16 and 220 n. 2, and above n. 16.] 

18 Sirabo, XI, ix, 3 [L. V, 274^-276/7] tells us that the entire sixth book of his His- 
torical Sketches had been devoted to the political institutions of the Parthians, " ... d/^- 
/coVes Se iroXXa irspl rcbv IlapdiKCov pojiu/acuv iv rfj I/ctt? tojv laropiKwv vTTop.vr]ixdrwv pt($Xa>, 
Bzvrzpa 8e Tcov ji^ra JloXvfiiov ... ". Not even quotations from this work have survived. 
[Cf, Erye, Persia, pp. 87, 127, 135, and Ehtecham, IS Iran, pp. 110 sqq., etc, on the 
Achaemenid satrapies,] (384, 3) 

19 Justin., XLI, ii, 1-2, " Administratio gentis, post defectionem Macedonici imperii, 
sub regibus fuit. 2, Proximus majestati regum probulorum ordo est; ex hoc duces 
in bello, ex hoc in pace rectores habent ". (385, 1) 

20 Sirabo, XI, ix, 3, [L. V, 276/7], ** ... rcov IIap8va(cov uvvihpwv <f>7)viv stvai IJooeiBcovios 
hvrrov, to juev avyysvmv, to 8e ao$wv koX ixdycov, e£ Sv apfow ftamAeis KuBhrauBat *\ (385, 2) 

2i Justin., XLII, iv, 1, " Igitur Mithridates, rex Parthorum, post bellum Armeniae, 
propter crudelitatem a senatu Parthico regno pellitur ", Ibid., XLII, v, 4, " Qua 
victoria insolentior Phrahates redditus, cum multa crudeliter eonsuleret, in exihum a 
populo suo pellitur " [ital, Adontz's], Here both senatu and populo refer to the ordo 
probulorum [cf. above n, 19], The actual MS reading is " ordo populorum ",-the recti- 
fication " probulorum " was made by Gutschmid, GeschicJite Irans, p. 57, n, 3, and was 
accepted and included in the Teubner edition, Saint- Croix, Memoir e, p. 60, note y, 
had already suspected that the MS text was incorrect, and suggested the emendation 
" optimatum ". In view of the passages of Justin just cited in which populo is un- 
questionably the equivalent of the disputed ordo, the proposed rectification seems 
unnecessary, or else populo in this passage should also be corrected into probulis. (385, 3) 

2 i a [Justin., XLI, ii, 5-6, " Exercitum non, ut aliae gentes, liberorum, sed majorem 
partem servitiorum habent, quorum vulgus, nulli manumittendi potestate permissa, 



^ 



NOTES : CHAPTER XIII 



481 



ac per hoc omnibus servis nascentibus, in dies crescit. Hos pari ac liberos suos cura 
et equitare et sagittare magna indxtstria docent. 6. Loeupletissimus ut quisque est, 
ita plxires in bella equites regi suo praebet. Denique Antonio bellum Parthis inferenti 
cam L milia equitum oceurrerent, soli CCCC liberi fuere ". [See also next note.] Pint. 
Grassus, xxi, 6 L. Ill, 378/9 " ... Ittttgis Se Kard^paKroi xAioi, irXdoves Se rwv kov$wv irap~ 
iir^pmov, el^e 8e rovs vvpmavras Iitttgis dfiov ireKdras re kclI BqvAqvs puvpiwv qvk dirohiovTas '*. 
This problem has been extensively debated, see, Prye, Persia, pp. 184-187 and Perik- 
hanian, Slavery, Manandian, Feudalism, pp. 117-118, categorically rejects the thesis 
of a " slave " cavalry, " JXju ujhrj.nLiHihp^** i/ftrnhqw^mjli ufuiv^ kh : ". Adontz 
seems to have followed the contradiction inherent in Justin who characterizes the 
horsmen as both slave and free. [See, next note,] 

32 [See preceding note for alternate Justin, text and discussion]. Justin,, XLI, iii, 4, 
" Equis omni tempore veetantur; illis bella, illis convivia, illis publica ac privata officia 
obeut ; super illos ire, consistere, mercari, colloqni. Hoc deniqne discrimen inter servos 
liberosque est, quod servi pedibus, liberi non nisi eqnis incednnt ", [Eht6eham, VIran, 
pp. 64-65.] (386, 1) 

22a [Prye, Persia, pp. 182 sqq.] 

23 Pliny, NS, VI, xxix (112) [L, II, 422/3], " Pegna Parthorum dnodeviginti stmt 
omnia ... ". AM, XXIII, vi, 14 [L. II, 356/7, " Sunt autem in omni Perside, hae 
regiones maximae, ... Assyria, Susiana, Media, Persis, Parthia, Caramania maior, Hyr- 
cania, Margiana, Baetriani, Sogdiani, Sacae, Scythia infra Imaum et ultra eundem 
montem, Serica, Aria, Paropanisadae, Drangiana, Arachosia, et Gedrosia ". Cf p. 356 
n, 1], Hence the Sasanian empire likewise comprised eighteen provinces. Is this 
an accidental coincidence or an imitation of Pliny ? [Cf. Prye, Persia, pp. 202-206.] 

{386, 2) 

24 Isidore of Char ax, p. 2, " MeooTroraixias ko\ BafivAwvias ... > ATroXXtovidridos ... 
XaXwvinhos ... MyBtas ... KafifiaZrivrjs ... MtjBias tt}s dvco ... x Payiavf}s Madias ... Xoa- 
pTjvrjs ... KQjMorjVTJs ... 'YpKavias ... ^AaravTjvTJs ... HapSvTjvfjs ... 'AiFavapKTuaprijs ... Map- 
ymvrjs ... *ApGias ... ^Avavijs ... Zapayyiavfjs ... UaKaaravrjs ... 'Apaxcoulas ... " et passim, 
[Cf, Prye, Persia, pp. 176-182 and next note.] (386, 3) 

24a [See preceding note. Pliny, NE, VI, xxix (112-114, L, II, 422/3-424/5, " Kegna 
Parthorum ... ita enim dividunt provincias circa duo, ut diximus, maria Eubrum a 
meridie, Hyrcaninm a septentrione, ex his XI quae superiora dicuntur inciprunt a 
confinio Armeniae Caspiisque litoribus pertinent ad Scythas, cum quibus ex aequo 
degunt. reliqua VII regna inferiora appellantur. quod ad Parthos attinet, semper fait 
Parthyaea ... habet ab ortu Arios, a meridie Carmaniam et Arianos, ab occasu Pratitas 
Medos, a septentrione Hyrcanos, undique desertis cincta ... . Media ab occasu transversa 
oblique Parthiae occurrens utrasque regna praecludit. habet ergo ipsa ab ortu Caspios 
et Parthos, a meridie Sittacenen et Susianan et Persida, ab occasu Adiabenen, a septen- 
trione Armeniam ". Ibid., VI, xxv, (92), L. II, 408/9, " regio ... Bactriae; Arianorum 
deinde ... ", Ibid,, VI, xviii (48-49), L, II, 372/3-374/5, " ... Bactri „. ultra Sogdiani 
... ».] 
24D icf, Prye, pp. 134-135, 139-140, 162-164, 172 sqq. Debevoise, Parthia, pp. 8-9,] 
24e [Prye, Persia, p. 177, attributes the assumption of the new title to Mithradates II 
(ca, 123-87 B.C.), Mithradates I had merely earned the title of " great king ", Ibid,, 
p, 176. Por possible Urartian influence on Achaemenid titulatur, see, Melikishvili, 
Inscriptions, pp. 112-114; Toumanoff, Studies, pp. 43-48 and 47 n. 36, 50-52 and nn. 43- 
44, 107 nn, 162-165, etcJ 



482 



NOTES : CHAPTER XIII 



25 [Bamzah al-Jsfahdm, p. 28, " When Alexander perished and the country fell into 
the hands of the tribal kings ... , There were in all ninety of these tribal kings, and they 
respected the king who ruled over Iraq and resided at Ctesiphon, which is the same as 
Mada'in. Whenever he corresponded with them, he began with his name first ",] 
The figure given is an exageration, or else the result of a misunderstanding. Under 
Seleukos I Nikator (312-280 B.B.), the Achaemenid satrapies were split into smaller 
units numbering up to 12 altogether, whereas Parthia consisted of 18 provinces. Perhaps 
the figure 90 was derived from 72 + 18. [The figure 72 is found in App. Syr,, lxxii, 
" aarp arret at §€ fjaav vtt' avra> [ZgXgvkos] hvo kglI ipbofiTJKOvra ", but Bickerman, In- 
stitutions, pp. 197-203, concludes that, " II serait vain de vouloir dessiner la carte ad- 
ministrative du royaume remaniee mille fois par les evenements, ou de donner Tenume- 
ration des divisions territoriales, qui auraient correspondu a une epoque d£terminee. 
Les indications dont nous disposons ne sont pas coh^rentes .... II est evident d'autre 
part que les circonscriptions pouvaient bien changer pendant la dur^e de plus de 250 ans 
de la domination sSleueide ". Ibid,, p. 199. Tarn, CAH, IX, p. 590, reverses Adontz's 
pattern and argues that the Parthian provinces varied in number, but " The big Seleucid 
satrapies were broken up into smaller units ". )387, 1) 

28 According to Tabari, [Zotenberg edition], I, cxiv, " Les rois de provinces lui [Aschk] 
obeirent et reconnurent son autorite" ; car il etait f ils de Dara et il avait des droits au 
gouvernement. Ils lui envoyerent des lettres et inscrivirent son nom en tete avant 
leurs propres noms. II en fat satisfait, mais il ne put leur enlever le gouvernement ..." 
(p. 592), cf. p. 526, also I, cx-cxiv; II, i.] 

27 Xen, Cyrop., I, i, 4 [L. I, 6/7], where Cyrus is distinguished from other kings: 
* f ... Snjvey/ce rcov aAA<w jSacrtAeW, /cat tojv Trarptovs apx&s TrapuATjcftOTCov /cat rwv St 
iavrcov KT7)aap.4va)v, ... ". [Cf. Toumanoff, Studies, pp. 41-48, et passim.'] (388, 2) 



NOTES : CHAPTER XIV 



483 



CHAPTER XIV 



a [The bibliographical indications given at the beginning of the preceding chapter 
are also relevant here. Eor a critique of Adontz's periodization see Introduction n, 2. 
The entire chapter must be considered in conjunction with ToumanofTs more recent 
Studies, on this subject. See also Trever, Armenia, and the next note for the inscriptions 
of Artaxias I.] 

1 The pre-Arsacid period in Armenia is usually considered to have begun with Artaxias 
and Zariadris. According to Strabo, XI, xiv, 5 [L, V, 322/3-324/5], they were generals 
of Antiochus the Great (223-187 B.C.) who divided Armenia after Antiochus' defeat 
in 189 at Magnesia and proclained themselves independent. In fact, however, the 
division of Armenia had already existed for a long time and is shown by the components 
of the XIII and XVIII satrapies of the Achaemenids, Xenophon, Cyrop,, III, i, 2 
[L. I, 216/7, 220/1 sqq.] speaks of an Armenian king contemporary with Cyrus and 
Astyages who had two sons; Tiyp&vqs and Zafiapis. Xenophon does not give the 
name of this Armenian king, but the Armenian tradition calls him Tigran and makes 
him the ally of Cyrus against Astyages [MX, I, xxiv-xxxi]; the great descendent 
of Artaxias, Tigran II, likewise bore his name. The house of Artaxias undoubtedly 
sprang from Tigran the contemporary of Astyages, and Zariadris, the king of Sophene, 
was also descended from an ancient royal line, since Antiochus III had fought with an 
Armenian ruler named Xerxes at Arsamosata, Joh. Antioch,, liii, p. 557. As early 
as the reign of Seleukos I (312-280) we hear of an Armenian king whose name is given 
as Ardoates or Artaones ; the latter form corresponds to 'Apr&vTjs, the name of one of 
Zariadris' successors in Sophene. This name was evidently common in the Zariadrid 
family, and the names Xerxes and Zariadris, Arm. fcwLiup^ and QwpkC were also 
favoured by the Arcruni house which came from Sophene. The Anonymous History 
speaks of kings named Qwph£ fiuiLinp^ and JXpSnq, ( — \Xpmnli cf. ^Aprdnjs) [Sebeos, 
p. 6] cf, Markwart, Eran, p. 177. The original division of Armenia into two portions 
goes back to remote antiquity, and in my opinion, should be connected with the two 
great kingdoms of Urartu and Manna mentioned in cuneiform inscriptions, and corres- 
ponding on the whole with the Eastern and Western Armenias of the satrapal period. 
Under Tigran II, the kingdom of Sophene was fused with that of Artaxias. We have 
called the pre-Arsacid period in Armenia " Tigranid ", in honour of Tigran, the first 
known representative of the dynasty, as well as of the famous conqueror, Tigran the 
Great. [On the Artaxiad and Zariadrid dynasties, see Toumanoff, Studies, pp, 69 n. 71, 
72-75, and particularly 277 sqq. On the Orontid dynasty, see also Manandian, Armavir 
Inscriptions and Trade, pp. 33-43, Trever, Armenia, pp. 104-174, Eor the ISTimrud 
dag inscriptions, see, above Chapter XI, n, 42. Eor the Aramaic inscriptions of Artaxias 
I, see, Perikhanian, Une Inscription arameenne, which gives the earlier bibliography. 

!» [Cf, Toumanoff, Studies, pp. Ill nn. 174-176.] 

2 Xen, Anal,, IV, v, 9-35 [L. II, 46/7-56/7], " o Se aAAo /xev ovUv iUx^o, ottov Se 
rtva T&v myyyGvajv ?Boi f vpos iavrov del eAaftjSavev ". Ibid,, IV, v, 32 [L. II, 56/7]. 
In one of the villages, the Greeks seized the Jcomarch and seventeen foals destined as 
tribute for the Persian king, "... KaraXaiifSavsi iravras IvSov rovs Kcop.TJras xai rov 
KWfAapxov, Kal ttwXovs ds SaajuoV /JacriAei rps(f>ofxivovs €77Ta/cat6e«:a, ... ". Ibid., IV, v, 
24 [L. II, 52/3], u ... Kal wdXiv rjpwrwv tivi oi ittttoi rpi^ovrai. 6 S' efAeye' on jQacriAet 
Sao>oV ". Ibid,, IV, v, 34 [L. II, 56/7], Strabo, XI, xiv, 9 [L. V, 330/1] says that 



484 



NOTES : CHAPTER XIV 



X 



the Armenian satraps sent each year to the Persian king 20,000 foals for the Mithrakina, 
",., o aarpd'TTTjs ttjs y App,Gvlas tco IHpaj) kwt Itos hiupivpiovs ttwXovs tois Midpaxivois 
eTTejaTrcv ". Xen, Anab,, IV, v, 35 [L. II, 56/7] says that the horse taken by him was 
dedicated to the sun, " .,, ort tjkovgv avrov Upov stvai rov "HAiov ... ". The infor- 
mation of Straho refers to an earlier period since in his own time Armenia was an in- 
dependent country ruled by its own king and not by a satrap. If one village sent 17 
foals, a tribute of 20,000 should indicate some 1,200 JcomarcJi3. [Of Toumanoff, Studies, 
pp. 69-70 and 70 n. 76, 277.] (391, 1) 

3 Pliny, NH, VI, x (26) [L, II, 356/7]. [Of. Bengston, Strategie, II. Ehtecham, 
VIran, pp. 113-115, 184, Toumanoff, Studies, pp.79, 111-112 n, 176, 136 n. 238, 
156-157, 290, 291 n, 58, 292.] (391, 1) 

3a [Of, Manandian, Trade, pp. 40 sqq., Feudalism, pp. 241 sqq,, Toumanoff, Studies, 
p, 108, etc.] 

313 [Hubschmann, Grammatih, pp, 101, 208, 242,253. Toumanoff, Studies, pp. 114 sqq, 
and notes, See also below nn. 23, 25-26 and Chapter XV,] 

3c [Gf. Toumanoff, Studies, pp. 62-70. See also next note.] 

4 Herod., HI, 93-94 [L. II, 120/1-122/3], " otto IlaKTviiojs §e /ecu *App&iwv Kal <tw 

7TpO(7GX^V ^XP l T0 *> TTOVTOV TOV Ev^lVOV ... VQfZQS TplTQS KCLl BsKOTOS O^TOJ. ... MctTVqVOiai 

Se /ecu Sacmeipai koX * AAapohlotvi ... vopuos oyBoos Kal heKaros o£tos. ... "• [Of, Touma- 
noff, Studies, p. 68 n. 65. Ehtecham, VIran, pp. 121-184.] (393, 1) 

5 Herod*, V, 49 [L, HI, 52/3], " KiAikcov Se TtDySe %x 0VTai *App,£vioi oiSe, /cat ovroi 
iovrzs TroXvjTpo^aroi, *App.Gvitt>v Se Mariijvol x&PW ' r W^ I^ovTes" ". 

Ibid,, V, 52 [L, III, 58/9], " ofipos Bk KiAuciys Kal rys 'Apiizviys ion •7Torap,6s vrprnTri- 
prjTos, toj ovvofxa Ev(j>p7jT'r]S. ... Trorafiol Se vyjvvnrzpjjTOi retxerepes Sid tclvttjs piovui, rovs 
Tracra dvdytcq Biairopd^vuai eVi, irpcoros juev Tiyprqs, fterd Se hsvrepos tc Kal rpiros wvtqs 
ovo/*a?ojU€VOSj ovk cavros icov Trorapos ovSe e/e rov avrov pio>v* 6 fj,£v yap TrpoTGpov avrwv 
KaraXzxQzt'S *£ *Apfi£vltav p£$h o S* varspov e/c MariTjvcov ". (393, 2) 

« Ibid., VII, 62-79 [L, III, 376/7-388/9]. [Of. above n. 4,] (394, 1) 

7 Xen, Anab., HI, v, 17; IV, iii, 4; IV, iv, 4; VII, viii, 25 [L. I, 490/1; II, 24/5; 
38/9; 370/1, cf, p. 370 n. 1 for later additions.] [See also Manandian, Trade, p. 20 and 
36 sqq., Toumanoff, Studies, pp. 277-279.] (394, 2) 

8 Xen. Anab., VIA, viii, 25 [L, H, 370/1], this passage is considered to be a later 
addition by the editor, cf p, 370 n. 1. See also, Toumanoff, Studies, pp. 59 n. 58 and 
68 n. 65. (394, 3) 

9 Arrian, Anab., HI, viii, 5 [L. I, 246/7], " ... '^pjueviW Se 'Opovrys Kal Midpavarys 
?jpxs> ... M [Manandian, Trade, pp. 36 sqq., Toumanoff, Studies, p. 279.] (394, 4) 

9a [#ero#., yn, 73. L. HI, 384/5, " 'Appivioi Se Kara irsp 0pvyss iusv&xaro, Uvres 
$>pvyG>v amoiKoi, tqvtwv avvap t ff>or4pcov vjpxz 'Aproxpys Aapeiov %xo>v Bvyaripa ". [Of. 
Toumanoff, Studies, p. 53 n. 49.] 

w> [Ibid., VH, 68, 78, 79 [L. HI, 380/1, 386/7, 388/9]; Ibid., VII, 62 [L. HI, 376/7], 
" MtjBoi bk tt)v avr^v ravrrp* ecrraA/iO'Ot ecrTparevovTO* Mr\hwrj yap avrrj i) gkgvt} am Kal 
ov U^pmKrj ". [See above n, 8, afoo, Toumanoff, Studies, pp, 55-62.] 

90 [See above nn, 4, 6-9b. Also, Toumanoff, Studies, pp, 129, 233 n. 291, etc.] 

10 Strabo, XI, xiv, 5 [L. V, 324/5], " ... <2>ars irdvras opoyAmTrovs dvai ". (395, 1) 

11 Idem. The printed text has Tapcovlns but Xenophon's description shows that 
the district was already in the hands of the Armenians at the time of the retreat of the 
Greeks in 400 B.C. The more accurate reading is, therefore, Taixwpins — S^tipfi^ 



NOTES : CHAPTER XIV 



485 



[See the notes L, V, p. 324, for alternate readings. Eor the acceptance of Adontz's 

correction, see, Toumanoff, Studies, p. 323 n. 78], (395, 2) 

is Sirabo, XII, i, 2 [L. V, 344/5-346/7], [Cf. Toumanoff, Studies, pp, 56 and 57 

n. 54,] (395, 3) 

13 Herod,, I, 72 [L, I, 86/7], " 01 cU ifa-jra-aSo/cat xmo t EXkfjvoiv Hvpwi dvofidtovrav „, ". 
Ibid,, VII, 72 [L, HI, 384/5], " ot Se Uvpioi ovroi wo Jlepaicov KainraBoKai /caAeWrat ". 
Sirabo, XVI, i, 2 [L, VII, 192/3-194/5], " ot yovv KairirdBoKGS dju^oTepoi, oits irpos 
tw Tavpto /cat ol irpos r<x> JJqvtw, fte^pt vvv AsvKouvpoi /caAowrat, ,..**. (395, 4) 

14 Marr, Tables. [On the origins of the Iberians, and Man's controversial thesis, 
see, Tonmanoff, Studies, pp, 55-66 and notes, also, Thomas, Marr.'] (395, 5) 

15 Sirabo, XVI, i, 2 [L, VII, 192/3], " Aokgi Se to tcov 2vpa>v ovop,a StaTetvat dVo fteV 
ttjs Baf$vho>vias jU€#pt T °v 9 Ivuikov /coAttov, otto Se tovtov ^te^pt rov Ev^ivov to 
iraAatoV ". (396, 1) 

15a [jSfee a&cwe n, 9a,] 

16 Sirabo, I, ii, 34 [L, I, 152/3], 

ft to yap tcDv *App.€v(cov £9vqs Kal to t<£v £vpo>v /cat *Apdf$wv irohXijv 6jxo(f>vXiav ep^atVet 
Kara T€ ttjv StdAe/CTOv /cat tovs jStovs /cat tovs rwv acop,drwv ^apa/crqpas, Kal pdXiora Kado 
irXijmox^poi etcrt. S77A0? S* tj MecroiroTajiua e/c rwv rpimv fjvvGQTtoua tovtwv iBvwv* pdAiara 
yap eV tovtois y ofMoloTqs Biatjtaivzrai, et Se rts irapd to KXipara yivzrai, hiatftopd rots irpotj- 
/Joppots iirl ttAsov npos tovs juecn^/Jpwovs Kal tovtois irpos jiiaovs tovs Svpovs, aAA* em/cpaTet 
ye to koivov. Kal oi y Aoavptoi Se Kal ot y Apiavol /cat ot \<4pajuftatot irapairbqoiws mus Invert 
Kal irpos tovtovs Kal irpos dAA^Aovs. et/cd£et ye Btj Kal ras tcov iBv&v tovtwv Karovofiaaias 
e^epets dAA^Aats etvat, tovs yap v<$? ojficDv Zvpovs /caAov/xeVovs vtt* avrtov rwv Svpmv 
^Apifxaiovs, koX ' Apap>p,aiovs /caAeicrflar rovrtp 8* eoi/cevai tovs y App,sviovs /cat tovs "Apaftas 
/cat 'EpGfifiovs? rdxa raw irdXai *EXArjv<jQiv ovtcu koXovvtcov tovs *Apafia$ **, 
[Ctf. notes p. 152 for variant readings. Adontz gives *App,ivioi where Jones has 
Mpa/^tatot and > Apip,aiovs.~\ (396, 2) 

" Ibid., XVI, iv, 27 [L. VII, 370/1], " fam U [TJocretSwwos] TavTa rpia Wvtj, ovvc^ 
dAAojAots lBpvf&4va, ojuoyevetav Ttva ip,tf>aivGiv irpos aXXrjXa, Kal Bid tovto irapaK€ip*ivois ovop,aQi 
jce/cA-^a^at, tovs j^ev Mpjuevtovs, tovs Se *^4joa^atovs, tovs Se '^paftjSovs* c5a77e/> Se a7ro !&>ovs 
eVos v7roAa/zj5avetv Icttiv eis Tpta BiTjpijoOai Kara ras twv /c/UjudVeov hia^opas del Kal ^taAAov 
e^aAAaTTOjLtevaty, ovtco /cat toZs ovo/zaat xW eracr ^ ai TrAetootv av0* evos **, (397, 1) 

18 Ibid., I, ii, 34 [L, I, 150/1], " diro rijs tcov iBvcov ovyyevsias, Kal koivott)tos 
iTVjioXoycbv J> , (397, 2) 

18a [On the place of Armenian among Indo-European languages, see, Meillet, Qrammaire 
Gomparie, Solta, Die Siellung, Tonmanoff, Studies, pp, 63-55 n, 49, 62 n, 59, et at While 
the Indo-Enropean character of Armenian is generally accepted, its independence from 
other Indo-European sub-groups is generally stressed at present, Cf. Meillet, Gram- 
maim comparee, p. 9, " L'armenien est un rameau de la famille indo-europeenne aussi 
ind^pendant de tous les autres que le sont par exemple le grec on le germanique, II est 
de plus isole", n'etant pas aecompagne" d'une langue d'aspect analogue, comme le slave 
Test du baltique, ni meme d'une langue offrant des innovations importantes en commun 
avec lui, comme 1'italique Test du celtique ", See below n. 62 and Bibliographical note], 

19 On this basis we may raise some questions concerning some of the peoples living 
to the south and north of Armenia: the hardu-Kapdov~x-Qi> /copSv-crW, yophv-ijvoi, 
and the Far du, ^^co3"OC? ' j° ,ftc ° r 3 1 '\3C? * (The hypothesis that the Karduehians 
were the ancestors of the Kurds has now been abandoned. Ancient sources refer to the 



7 



486 



NOTES : CHAPTER XIV 



Kurds under the name Kvprioi, Noldeke, Kiepert Festschrift, p. 73, and Hartmann, 
BoMan, p. 90). The p,6u~x 01 are related to Arm. U*nL£> ITnLy—k^ where 
— hi£ — Georg. Q(™o. The southern x<^S-aiot are related to the northern JOwrjinfc—g 
or x^-vfi-zS' Sirabo, XI, xiv, 9 [L. V, 328/9] distinguishes two districts named 
Hvompiris, one on the Qoruh = the Zacnretpes of Herodotus [I, 104, etc.], the ' EaTTSptrai 
of Xenophon, [Anab., VII, viii, 25, a later addition according to the editor L. V, p. 370 
n. 1], the other to the south near the Tigris. The former is the presentday Sper, the 
latter = Supri, see, Markwart, Eran, p, 159. Arm. Ifnlf—g is related to Mvkoi 
> Mughan, and Ifjuwhp in Arcax- [See above n. 9b, and Toumanoff, Studies, pp, 
321-323 n. 76; below nn. 58, 77.] (398, 1) 

20 Marx has found it possible to derive armen-ius from Aram without the help of 
ancient sources, see his Review. As for £wj, some derive it from pati, Bt head " with 
the sense of conqueror, others from Hati {cf. Jensen, Hettiter und Armenien, 1898), 
or finally connect it with the pai in Flaiovia, a district of Macedon, and the original 
home of the Phrygians (Hommel, Grundriss, I, p. 31), Sirabo, XI, xiv, 12 [L. V, 332/3] 
argues that the dress of the Armenians points to a Thessalian origin, " ko\ -rqv iodyra Se 

T7JV *ApfJ.€ViaK7)V ®GTTOj\lK7)V (jxiOlV, OlOV TOVS fidd^lS X lT ^ va S> OVS KaXoVGl ®€TT(xXlKOVS £v TO,ls 

rpaytp&iais, /cat ^vwvovai 7T€pl ra GTT\Bf}, koX i^airrihas ws Kai rcov rpa/ycahwv nip/quafiivoiv 
rovs S^rraXovs ... /ecu tov ttjs Itt7tik7}s EfiAov <j>aaiv eivcu <5>€ttcuU/cov /cat tovtois qjj,o(cos 
Kai Mtj&ois- ... ". [On the ethnika " Armenia " and " Hay " their sources, 
and the recent views on the subject, see Toumanoff, Studies, pp. 55 n, 49, 59 n, 58, 
108-109 and n. 163, 294-297, etc. Adontz developed the question further in his Histoire 
d'Armenie. Also, see below n. 62,] (398, 2) 

20a [Qf, Manandian, Trade, pp, 42-43.] 

2i Herod., V, 49 [L. Ill, 52/3]. [See above n. 5 for the text.] (399, 1) 

22 Sirabo, XI, iii [L. V, 216/7-220/1] Iberians ; XI, iv [L. V, 222/3-230/1] Albanians ; 
XVI, i, 1 [L. VII, 192/3] Assyrians; XVI, i, 26-28 [L. VII, 232/3-236/7] Arabians. 
[Cf. Toumanoff, Studies, pp. 84 sqq., Trever, Albania,] (399, 2) 

22a [Cf, Garitte, Agathange, p. 223, Toumanoff, Studies, pp. 208-209.] 

23 This title belonged exclusively to the Bagratuni: *' JXuujkinfi tlhb u^mppiih h 
ujuj^iwLh ", Sebeos, p. 34 the correct form should be \Xu \uimpuf\ unbuilt as it is 
given in Ibid., p. 36, " qlXuiqujpwiqkmh lfkb q^ufiuppidi k tp^w^j^iuLU ". [Cf, 
Toumanoff, Studies, pp, 324-326, for a critique of Adontz's etymology, also below nn, 25- 
26, and Hubschmann, Grammatik, pp. 13, 109; Benveniste, Titres, pp. 9-10.] (400, 1) 

24 Marr, Etymologies. (401, 1) 
24a [Cf. Toumanoff, Studies, pp. 92 n. 132, 115 n, 186. [Benveniste, Termes, pp. 9-10.] 
24b [ F b, VI, iv, xv.] 

25 Kir, Ganj., pp. 23-24, [Where the initial *£ has been removed, altough the form 
^muu^nLpwIj^u is preserved in the Venice, 1865 edition, p. 14], Narratio, ciii, 
" ... ttjs *AuT7ovpaKav x^P - 5 °* iTrioKOTTOi ... ". [Cf. Garitte's " commentary ", Ibid., 
pp, 244, also 403, 406, 418-419]. T'aiqasvili, Three Chronicles, p, 122 " CHBHeTifl H 
AcnaparaHt, ", Mov. Kalanh., p, 69, one of the participants at the Council of King 
Va5agan of Albania was called IXuu^uilj—iu. [Cf. Mos. Da$x*> I, xxvi, p. 54, which 
gives the name as "Sprakos" with the variants "Asprakos" and " Sparakos ".] (401, 2) 

26 Although the form vaspur is closer to the original, aspur is the older form from the 
point of view of Armenian phonetics. Armenian borrowings from Iranian in the pre- 
classical period necessarily lost the initial v since Armenian, like Greek, did not pronounce 



NOTES : CHAPTER XIV 



487 



this phoneme in initial position, In roots, the v > g, while in borrowed words it dis- 
appeared altogether. [Of. Meillet, Grammaire comparie, pp. 48-50]. Iranian words 
with an initial v found in Armenian entered this language in the Classical period. This 
clarification should help to remove Hubschmann's misgivings, ZDMQ, XLVI, p. 327, 
as to the disappearance of the initial m~ in sepuh. As shown by the Greek equivalent 
oIkos = viB, the Iranian vi Ba-puhr < *vai Ba-puhr > *vase-puhr and *Vasia-puhr, 
the prototypes of the Armenian ase-puhr and ase-puhr (depending on the accent) without 
the initial v-. The alternation as - se can incidentally he explained by the dropping 
of the vowel in compounds, especially in the case of an initial sp-, e,g, aspanj, Arm, 
wuwu^mlimu, Pers. sepanj; Aspahapei, Spahpei, Ispahbed; aspar and spar "shield", 
etc. {See above n. 24a.] ■ * (401,3) 

27 It is probable that l[Iu " proud " also = vi Bya, *vai da < vid with the suffix -ya, 
*l\}u, like ipukfi, originally indicated membership in a clan. Cf. mn££, mn^i/p^ 
In its archaic sense, vsam is found in Thorn. Arc, HE, xxii, p. 231, " ***kpmukjp 
hufpufynufnuu fyppiinp f}^uiniSikm^ ijuwS"* In the Canons of the Council of 
King Yacagan, Mov. Kalanh, p. 68 we find the injunctions that " ***qjCnw kfyhqkijLnj 
p ifumS hlfkqktjpit mmtjkh". [Can. XVII] and " ♦ ♦♦ p fypLpmlilp mj;p k 
bmnmj p ijumfi hfykqkijpu hpPpijbu jmnpPu \ [Can. XIX]. In Canon 
XVIII of the same Council, the decree, " np mmumhnpriu mmh mqmm Smpiipg 
nJl^uu pmh kljhrjh^jih mmyhh, k nl^tuh jpipkmuij htykqkijpu ". thus opposing 
the pni.h klfbqkijli to the azat or noble church, [Cf, Dowsett, Mov, Das%*> H> xxvi, 
p. 53 n. 1 for the possible reading wS lJu^ with the contractions, — ijmuu imlkhltjmSi 
klfkqhtji.nijh'], One of these churches is taken as being the ijumfi hlfhqkijp. If we 
admit the existence oVvi Bama-pati, we can derive from it the favourite given name of 
the Aspetsi Uil-^u^mm < H-sempatL [See above nn, 23, 24a, 26,] (402, 1) 

28 The name Mamak was common in the Mamikonean family as early as the Vllth 
century, Of. Sebeos, vi, xi, xiii, pp. 48-49, 56, 58. The suffix — niup or — mh = 
Georg. 06, JfmiJp—nu and IfmSj^—nLu are alternate forms of -an, the sign of 
the plural, -kmu is a tautology as an equivalent suffix, [Of. Toumanoff, Studies, 
p. 211 n, 238.] (402, 2) 

29 JPB, V, xxxvii, " +"i/hp uwpihpg lp w l_ ip u piurj.iuL.npg mppjmp^pu 
ghhmy ". (403, 1) 

2 9a [" Primary History ", Sebeos, p. 12-13, " \lTmifpi)nuhmug\ n£ kh ♦♦. npqpg 
mnj^mbhphh lXpiuJ}ujuklj[Wji mjj^ hu kfykw^ p rfkuwumwui ♦*♦ <pmpwmwl[wu 
[pmj_ Jfrnilplju h ynuwlju nStwh mn mpgwju UppmfynjLUp* np uum^p p fiw4i 
Dui^wumwuff jhpfypph 'finL.yiuaujtj *** \wpgwjh UpprnfynLup] ns hm njtnnm p 
hhnu unpm \jShhpwlf^ni.pp"\ mjj^ n.pl mn urn uppm£ *** kmni. mmukj_ qhnurn p 
finL.mii mphuL k jhnjt kpfypj] p mktipu jmjh, np mpkzLmlfu p ilmjpu fimmul; ". 
See, Toumanoff, Studies, pp. 209-211 and particularly n. 238 for the origin of the Mami- 
konean, and the extensive bibliography on this subject, For the Tzans, see above 
Chapter I, nn, 46 and 46a, Chapter III, pp. 49 ssg, and nn, 26a-30, 32a,] 

2dh [On the Orbeliani and their Mamikonid antecedents, see, Toumanoff, Studies, 
pp, 211 n. 238, and 270,] - 

30 The Lat, deus and dies, Gk. Zzvs* §ios y Skr. djaits, " heavens *' and " god ", devas 
" god ", Zend, daeva, likewise belong here. According to Herod*, I, 131 [L, I, 170/1], 
the Persians called, " ,„ tqv kvkXqv ndvra rov ovpavov ", " Ala ". The same meaning 
is attached to the Armenian mp — in the word mp—hnhpg having the original sense of 



488 



NOTES : CHAPTER XIV 



"firmament " similar to the Persian, " Ala " and subsequently coming to mean the 
inhabited universe " oikumene " [k^kp^ " region " is attached to it as a reinforcement). 
[Cf, Toumanoff, Studies, pp. 90-91 nn. 128, 130, 114-115 n. 185 and the objections of 
Dowsett, <Ter, pp. 136 sqq,] (405, 1) 

31 According to Pomp, Trog., " Prologus ", xlii, " suceessores deinde eius Artabanus 
et Tigranes eognomine dens a quo subacta est Media et Mesopotamia ", The reference 
here is to Tigran the Great, how are we to understand the cognomen dens attributed 
to ham, might it not be a rendering of the Armenian in its sense of "Mug " or "god " ? 
The name Tigran may also be connected with the root mp~ and have the sense of " divin- 
ity " like the Lat, Diana, Gk, Ai&vt}* Hommel, Grundriss, pp. 39 n. 2, and 43 n. 1, 
already suggested that the initial syllable te~ in Techib, Tisup, Teisbas, Tispak, meant 
" lord '\ [See the preceding note, and Hiibschmann, Grarmnatih, pp. 87-88.] (406, 2) 

31* [Strabo, XI, iii, 6 [L. V, 220/1], 

" T4rrapa S£ /cat yivT] twv avOpccmcoy oi/cei ttjv ywpav ev p,h> Kai Trpwrov, ££ o$ rovs 
ftamX4as KaBioraai, kot ayxiorsiav Te koli yXwiav rov TTpsufivrarov, o Se hsvrepos 
BiKawhoTGi Kai arpar7}Xaret' ScvTepov §e to twv Upicov, o? emfieAowTat Kai twv irpos tovs 
Qfxopovs hiKaioov rpirov 84 to tcuv arpar^vofxivcov Kai yswpyovvTaov' TerapTov 8£ to tcuv Xawy, 
o? fiamAiKOi BovXoi curt /cat irdvra BiaKovovvrai ra 7rp6s rov fiiov. Koival 8* etcriy avrois at 
kttJvgis Kara avyydv&av, apx& 8e Kai TajUievet e/cdcmjv d TtpGufivraTOS. toiovtoi fikv ot "Iftypss 
Kai 7} ^tupa avrcov ", 

[Of. the following notes.] 

32 Ibid,, XI, iii, 3 [L. V, 218/9], Strabo distinguishes two types of population in 
Iberia, " ... the plain of the Iberians is inhabited by people who are rather inclined to 
farming and to peace, and they dress after both the Armenian and the Median fashion; 
but the major, or warlike portion occupy the mountainous territory, living like the 
Scythians and the Sarmatians, of whom they are both neighbours and kinsmen; however, 
they engage also in farming. And they assemble many tens of thousands, both from 
their own people and from the Scythians and Sarmatians, whenever anything alarming 
occurs ". This is the reason for which Strabo, Ibid., XI, iii, 6 [L. V, 220/1] defines 
the third caste as made up " ... rwv crrpaTeuojuevw Kai yzwpyovvTwv ... " and not for 
the reasons given by Java^isvili, Polity, p. 56. [Of. Toumanoff, Studies, pp. 90-96 and 
notes.] (407, 1) 

32f * [Strabo, XI, iii, 6 [L. V, 220/1], See above n. 31a for the text.] Of. also Toumanoff, 
Studies, pp, 90-91, 96-103, 106 sqq., and notes.] 

33 At first, the heir may have replaced the king only in cases of illness or absence as 
was the custom among the Sabaeans where, " the king has as Administrator one of his 
companions who is called 'brother' {KaXovpsvov dSeAt£oV) ". Strabo, XVI, iv, 21 [L, VII, 
352/3], [The reference in Strabo is to the Nabateans, who ruled Petra, rather than to 
the Sabaeans.] Of. Xen, Oyrop., VIII, vi, 16 [L, II, 418/9], " fiamMms adetyos, fiamXiws 
6(j>9aX}i6s " among the Persians. (407, 2) 

34 The union of the functions of judge and commander of the army in the hands of 
one man shows that the judge was in need of force. Among the Assyrians also, cases 
of thefb were tried before a military court [Strabo, XVI, i, 20, L, VII, p, 226/7, though 
this is not the precise sense of the passage.] This pattern is characteristic for a period 
of disintegration in the tribal structure, since as long as the bases of natural-legal rela- 
tionships remain strong, the society maintains its own equilibrium and has no need of 

Strabo's friend Artemidorus, the philosopher, was greatly amazed when 



NOTES : CHAPTER XIV 



489 



visiting the Nabatean [the Russian text reads Sabaean, but see preceding note] capital 
that " the foreigners often engaged in lawsuits, both with one another and with the 
natives, hut that none of the natives prosecuted one another, and that they in every 
way kept the peace with one another ". Ibid., XVI, xiv, 2, [L, VH, 352/3]. According 
to Strabo himself, the Nabataeans were still in the initial stages of development, [Cf. 
above nn. 31a-32a.] (408, 1) 

35 Java^isvili, Polity, does not raise the question of the interrelation between the 
four castes of the Iberians, and consequently the origin of the royal BovXoi is incorrectly 
interpreted by him, since he . argues on the basis of the Georgian term 9ca6c> = 
maniya, " slave ", that the institution itself had been borrowed, J bid., p. 74. If this 
point of view is accepted, a great deal will have to be acknowledged to be of foreign 
origin. [Cf. Toumanoff, Studies, pp. 94-96 and nn. 140, 142,] (408, 2) 

36 Strabo, XI, iv, 6 [L. V, 228/9], " vvvl (xkv odv els cwrdWcov a/^ei, TTporepov Bk /cat 
kclQ 7 iKauTTjv yXtorrav iSta ejSaatAetWro f jcaaroi. yXwrrai 8' eiaiv ef koX eucoat avrois hik to 
/x^ even ijiiKTOv irpos aAArjAovs ". [Cf. below n, 38.] {409, 1) 

37 Ibid., XI, iv, 7 [L. V, 228/9], " Uparai 8* avrjp ivTifMOTaros fiera yn rov jSacrtAea, 

TTpOCOTCQS T7]S Up&S ^COpa?, TToAXtJS KUl €V&vSpOV, KO.I aVTT]5 JCCU TWV IGpoBovAzOV, ,,.**, [Cf. 

Toumanoff, Studies, p. 96 n, 142,] (409, 2) 

38 Ibid., XI, iv, 5 [L. V, 226/7], the Albanians, " send forth a greater army than that 
of the Iberians; for they equip sixty thousand infantry and twenty-two thousand 
horsemen, the number with which they risked their all against Pompey '*. Plutarch, 
Pompey, xxxv, 2 [L. V, 208/9] gives twelve instead of twenty-two thousand horsemen, 
as in Strabo, (we should read " biaxiXiovs wrams eiri [Sia]fxvpiois " [the Loeb edition 
gives merely " \xvpiois "]), " They were led by a brother of the king, named Kosis (^yetTo 
§e avTvov fiaoiAscos aSeA^os ovop,a Kwuis) and not by king Oroizes [" Orgires " according 
to Appian, Mithr., XV, ciii]. Kosis obviously led the army as part of his office of 
commander in chief. It is possible that he was likewise responsible for the adminis- 
tration of justice, as was the case in Iberia, " Three chiefs of the Iberians [and] two 
of the Albanians (,„ riyspovt-s rpels 'iprjpcov /cat *AAf$dvwv Suo) " were found among the 
324 vanquished chieftains preceding the victor's chariot during Pompey's triumph, 
Appian, Mithr,. XII, xvii, 117 [L, II, 466/7], The three Iberian leaders were the king, 
the high priest, and the military commander; the two Albanians were the king and 
the high priest, since Kosis, the commander of the army, had been killed by Pompey, 
[Plutarch, Pompey, xxxv, 2, L, V, 208/9]. [On Albania, see Trever, Albania, and 
Toumanoff, Studies, pp. 84-86, 96 n. 142.] (409, 3) 

38a [Cf. Manandian, Trade, pp. 42-43, and above Introduction n. la.] 

39 Plut, Lucullus, xxi [L. II, 536/7], " /JacnAeis §€ 7roAAoi pkv fjoav ol QspcrnGvovTes 
avrov, riaaapGS 8e, ovs aet Ttepl avrov ei^ev tScrjrep diraSovs rj §Qpv<j>6povs, mTtOTT) }ikv eAav- 
vovrt tts^qvs 7Tapa$4ovras iv xtTcovtcr/cois:, KaBrip,h><u> 8e /cat xP 7 }} J " aT % OVTl wepteOTtoras' ztttjA- 
Aay/xeVats" St* dAA^Acoy rats x e P Q w> OTrep eSd/cet /taAurra rcov axr)p>druiv igofiohoyrjms 
stvai SouActas', olov d-Troho^ivcov t1)v iAevOspiav /cat to ucojxa rep KVpito TTapexovrtov ira^etv 
eTot/zoTepov rj Trot^crat *\ [In this section Adontz generally follows the argument of 
Markwart, Bran, pp. 165, 172 sqq. On Tigran the Great, see also Manandian, Tigrane II, 
and Toumanoff, Studies, pp. 76 n. 84, 77 n. 86, 82-83 and 83 n. 104. On the institution 
of the bdesxs, see also, Christensen, pp. 22-23, 101-102, 518 sqq.; Frye, Persia, pp. 97-98, 
186, 201, 273 n, 9; Toumanoff, Studies, pp. 76, 154 sqq.; I^ap'ancyan, Hist. -Ling.- 
Studies, pp. 467-sqq,], (410, 1) 



490 



NOTES ; CHAPTER XIV 



40 Since Markwart's clarification of this point, Bran, pp, 172 sqq. Gntsehmidt's 
hypothesis, Geschichte, p. 85, that the kings of Atropatene, Gordyene, 
Adiabene, and Osrhoene are intended here, mtist "be completely abandoned. This 
hypothesis was already shown to be incorrect by that fact that these kings were not 
with Tigran at the time of Imcullus' expedition: Tigran was awaiting the kings of 
Adiabene and Atropatene, Zarbienos, king of Gordyene had been executed earlier for 
treason [Pint, Lucullus, xxix, 6, L, II, 568/9], and Cleopatra, qneen of Osrhoene, was 
being kept prisoner at Seleneia, Strabo, XVI, ii, 3 [L, VH, 240/1]. [Of. Tonmanoff, 
Studies, p, 82.] (411, 1) 

41 App., Syr., XI, viii, 48-49 [L. II, 196/7], " 6 Tiypdvrjs fee Zvpias rijs per Ev<f>pd- 
rr\v f oaa yh>v\ Svpcov p,4xpi Alyvirrov. 'rjpxs 8e 6p,ov kclI KiXmias (Kal yap rjBe rots UgAgvkI' 
Bats VTrqKovs); MayaBaTfjv arparTjyov imrd^as wrramv, em 'dry Teo-tfapecrxcuSe/ca. ,„. d 
MayaBdrrjS j]Gi fisrd rov curparov Tiypdvrj PotjBtJvwv, .,, ", The printed text has 
MayaBdrrjs for Bayaddrrjs but the MSS haYe both versions and the second is more correct, 
cf. Markwart, Bran, p. 174, [See also, Tonmanoff, Studies, pp. 202, 313-314, 320-321, 
and 420 n. 71, 324> and below note 44.] (411, 2) 

42 App, Mitfar., XII, xii, 84 [L, II, 398/9], " MiBpofiapidvyv rrpovirspmi- fterd Sicr^AiW 
i7T7t4wv 3 AzvkoXAov imax € w T °v Bp6}j.ov. MayKaiw Se TiypavoKGpra <f>vXdrrew eWTpe^rev, ijv two, 
ttoXw* <x>s p.oi TrpoGipTjTai, iirl TifiTJ rrj iavrov fiamXevs iv iKtivtp yeviaBai rtp x^piw owa>Ki£e, Kal 
rovs dptQTOvs is avrijv owe/cdA^i, ^rjp^lav imriBsis, dou pur) pLzracjtipoiGV, Be&rjfiGvoQat,. Tet^ 
r€ avroTs Trepie/foAe n^vrriKOvr afr^x 7 ! T ° ^fyfros, mtrouraQicov h> rep fidBst, y4p,ovra, Kal jSacreAeta 
Kal TrapaBziaovs Kara ro Trpodar^iov iirofci paKpovs, Kal Kwijydaia TroXXd Kai Xipvas' dyxov 
Se Kal (j>povpiov dviart] Kaprepov, Kal Trdvra rors MayKaicp ravr iirvrpiifsas, Trepiijet ctrpartdv 
dyeepcov. MiBpofiap£dv7]v p,ky o$v 6 AzvkoXXos svBvs £k rrjs irpu>T7)s ovp,f$QXrjs rp&ffdp,GVQ$ 
eSuo/ce, MayKaiov §e S^riXios is TvypavoKSpra /caTa/cAacas rd \xkv jSacrtAeta avriKa, drzixwra 
ovra, Sirjp'TTaoG, ri}v Be noXiv Kal to (j>povpwv dnzTa^pevz, koi pyxavds etj&icmj, koi virovofXQis 
avsKprjpw'r} to t^x°s ". 

[Of. next three notes, and nn, 50-51] (411, 3) 

43 Pint. Lucullus, XXV, 2-4 [L, II, 550/1] il irp&ros 8* avrtp rcoy <j>iXa>v eroA/^cre 
MiBpofiapldvrjs <j>pduai to dXrjBds. ov8* o$ro$ Se ^p^aTov ^viyKaro yipas tt}s 'napprqaias* 
iTripL^Br] yap <sv8vs ctti toV ^tov/covAAov avv lirnsvm Tpicr^tAtois, itg^ois Se TrajLtn-oAAot?, 
kgXsvuBgis rov juev o-TpaTijyov ayziv ^mvra, rovs S* aXXovs KaraTrarTJaai. ... d 27efTtAtos ... 
iftiduBr) 8* vitq rov MiBpofiap^dvov Bpavicos iir^Xavvovros €ts x € *P as i^Beiv. Kal yspopivys 
p-dxrjS o jtiev MiBpofiap^dvrjs hrzetep dya}vi^6p > evos, ol 8' dAAot <j>evyovres dtrcoXovro ttXjjv 
oXiyayv amavres ". 

[Of, preceding note, and below nn. 50-61] (412, 2) 

44 Ibid,, xxxii, 4-5 [L, II, 578/9], " ravrjjv [Ntmfiiv] el^ev d£iwp,ari. juev dSeA^ds 
Tiypdvov rovpas, ip.TTGtpiq 8e /cat SetvoT^Tt fxijxaviKf} KaXXipaxos 6 Kal ir^pl *Ap.wov irX&ara 
'jrpdyp.ara AovkovX\o> irapaux^v. f$aX6p,GVOS Se CTTpaTo-jreSoy Kal iraaav IBiav iroXtopKias 
iTrayaywv oXlytp XP° VC P ku™ Kpdros Xap,fidvsi rrjy ttqXw. koi Fovpa p,$v iavrov iyxsipicravn 
tjjiiXavBpco'TTws ixpfoaro, ... ". (412, 2) 

45 See above chapter XI, p. 242 and n, 27. The phonetic shift 8 > r in the name 
Bagarate < BagaSates, which is considered to be a characteristic of dialects from 
Atropatene according to Iranists, points in the same direction, Griindriss, I, 2, p. 355, 
cxYi, [Of. Tonmanoff, Studies, pp. 306 sqq,] (412, 3) 

46 Cf, Markwart, Bran, p. 174 [and the preceding note,] (413, 1) 

47 Sebeos, p. 6, " *♦♦& npyfrgli pwqwpiumwj dmniuliq.h^[ih qdutnuibqnL— 



T^*^*v^*S«3§tW3 



35FOTES : CHAPTER XIV 



491 



Pp&u fjLpkwfctj ft Ifnqpwbu wpkiiwfjij. wjufibgh £ JXbqkq mniSi. ijwuh qji 
l^n^k^utL Pwqmpwm h \}Jiqhq qnp fi dindwhmlijjh jmjhd^ mqq pwppiupnuwgh 
muinriLwb lin^fjh", [Cf, above n. 45,] (413,2) 

48 In this case we can postulate a connexion between IXhqkq in mh and JXIiqn i 
the village in Bagrewand which we proposed above as the ancestral home of the Bagra- 
tuni [See above chapter XI, p, 242 and n, 27], The Biblical JXhqq — Gk. NzpyiX 
(probably < hrohjaav ri)v < v> epy& pro ipysX) has no connexion with JXhqkq 
wm.h in spite of the Anonymous History's familiarity with it, which was accepted by 
Khalatiants, Armenian Mpic, p, 84. [Cf, Toumanoff, Studies, pp, 296-305, and above 
n, 45.] (413, 3) 

49 The MSS of Appian have the correct reading Midpofiovldvys, as shown by Justi, 
Namenbuch, pp, 208-209, MidpoftQvt&v7)s consists of Midpa + buzan, Arm, pmd — 
ly and has the sense' of "saved by Mithra ". [Cf. Hubschmann, Grammatih, pp, 
52-53, Toumanoff, Studies, pp, 299, 321,] (413, 4) 

50 According to Plutarch [Lucullus, xxv, L. IE, 548/9-550/1], the first man to bring 
Tigran news of Lucullus' expedition was decapitated, hanged, according to Appian 
[Miihr,, XH, xii, 84, L. II, 398/9], after which, according to Plutarch, no one dared 
speak of Lucullus to the king, " ,., those who flattered him ,„ said that Lucullus would 
be a great general if he ventured to withstand Tigranes at Ephesus, and did not fly 
incontinently from Asia at the mere sight of so many myriads of men ,., , The first 
of his friends who ventured to tell him the truth was Mithrobarzanes, and he, too, 
got no very excellent reward for his boldness of speech. He was sent against Lucullus 
with three thousand horsemen and a large force of infantry, under the orders to bring 
the general alive, but to trample his men under foot ", [See above n. 42 for the text]. 
According to Appian, Tigran's order was very modest, " „. to hinder Lucullus' march. 
... ", Mithrobarzanes had but 2,000 horsemen, and the battle ended in Mithrobarzanes* 
flight [see above n. 42 for the text]. Plutarch, however, raised his forces to " three 
thousand horsemen, and a large force of infantry " and had him fight not against Lu- 
cullus himself, but against his legate Sextillius, with the resultant death of the Armenian 
commander, These disagreements show the bias of ancient historians, and Tigran's 
history still awaits an. impartial study. Scholars have accepted all the information 
given by the sources forgetting that these were based on the false reports of the boastful 
Lucullus who described partisan skirmishes and bandit raids as pitched battles in which 
20,000 men defeated the enormous and brilliant forces of Tigran, which numbered up 
to 260,000 or 300,000 men. At the same time, it is claimed that whereas the enemy 
had lost more than 100,000 dead, and almost no one had survived out of a cavalry of 
55,000, Lucullus suffered losses of some 100 wounded, and 5, or even, 1 dead. The 
nature and value of Lucullus' accounts can be judged from the following classic example; 
in his report of the battle for Artaxata, Lucullus wrote that " Of three kings who together 
confronted s the Romans, Mithridates of Pontus seems to have fled most disgracefully, 
for the could not endure even their shouting ". [Plut, Lucullus, xxxi, 7, L. II, 574/5- 
576/7], This is told of a man unequalled for -his fearlessness, one whose mere name 
terrorized the Romans, and of whom it was said affeer his death that " ,„ in the person 
of Mithridates ten thousand enemies had died ", Plut, Tompey [XLII, 1, L, Y, 222/3, 
[Eor a recent attempt to clear Tigran's reputation, see, Manandian, Tigran II, who 
shares Adontz's opinion of the reliability of classical sources on this subject, cf, in parti- 
cular, Ibid,, p, 2 sqq,, also Toumanoff, Studies, p. 299 n, 89.] (414, 1) 



492 



NOTES : CHAPTER XIV 



si DM. Sic, XXXI, xxii, [L, XI, 370/1-372/3], c/, Polyb., XXXI, xvi (xv), 1 [L. VI, 
194/5], We have already seen that the "bishop of Sophene was also named Mspovldvys. 
See above, Chapter XII, pp. 270-271 and n, 42, [Also, Tonmanoff, Studies, pp, 292 sqq,, 
and stemma p, 282, The protector of Mithrobonzanes is nsnally considered to be 
king Ariarathes V Eusebes Philopator of Cappadocia (163-130 B.C.) and not his father 
Ariarathes IV, cf. Diod. Sic, XXXI, xix, xxi-xxii, L, XI, 368/9-370/1, and p. 369 n. 3] 

(414,2) 

51a [Stmbo, XI, xiv, 15 [L, V, 336/7], " tov Se ZapidBpios 6 ZwxjyqvQs 'Aprdvys %x wv 
to. vona iiipf] Ktd rovrcov rd TTpos Svaiv juaAAov, KaT^XvBf) 8* ovVcs vttq tov Tiypdvov, Kal 
irdvrwv KariuTrj Kvpios e/ceivos 1 ". Tonmanoff, Studies, pp, 292-294.] 

b» [Ibid., pp, 209, 321.] 

52 Markwart's hypothesis, Eran, p. 176, is very likely. The transfer of the Sophenian 
branch to Albak does not belong in the period of Tigran II, as he supposes but in that 
of the Arsacids, The etymology of the name is still nntraced, since all the attempts 
to interpret this name remain as yet unacceptable. [On the Arcrnni, see Tonmanoff, 
Studies, pp. 110 n, 173, 164-165, 170, 199-200, 298-299, 303, 305, 320-321 j on the ety- 
mology of their name, Ibid*, pp. 201 n. 228, and 298 n, 83; on the Bagratnni, Ibid., 
pp, 110 n. 173, 201-203, 306-354, etc.; on the Mamikonean, Ibid., pp. 110 n, 173, 209- 
210, 321, 325 n, 88.] (415, 1) 

52a [OntheZorsofnni,5ee, Ibid., pp.110 n.173, 208-209, 325 n, 88; Garitte, AgaU 
liange, p, 223,] 

52i> [See above n. 39,] 

53 Pint. ImviiVms, xxxi, 5 [L, II, 574/5], " ... tto)0\.oI yap rjaav fanreis Kal Aoydfes dvri- 
irapaTerayjue^ot, itpo 8* avrwv iVjroTofoVat MdpBoi Kal Aoyxo<j>6poi "IfiypEs ofs ndXiara rtbv 
£4vcov 6 Tiypdv7)S iTriarsvsv cos fia^tficDTaVoiS' **, (417, 1) 

54 Sirabo, XI, xiii, 3 [L, V, 304/5], "... oUvttj JHspaiBi Kvprwi Kal Mdphoi f/cae yap 
ovrco Myovrat ot "A/xapBoi) Kal ot iv t$ 'Appevia jue^pi vvv o^icuvv^cos Trpocrayopevofteyoi ttjs 
avTTjs €icriv ibias ". The Mardians are undoubtedly related to the people haying 
the same name (Mardi or Amardi), Hying near the Caspian Sea, whence were deriyed 
the Ami of the Arm. Geogr., p, 40/53 [See above, Chapter IX n, 8]. Ami < *Amrda, 
Andreas, Amardi, Markwart, Eran, p, 136. [0/. Tonmanoff, Studies, pp, 169-170.] 

(417,2) 

55 Xen. Anab., IV, iii, 4 [L, II, 24/5], " „, op&mv hirias ttov iripav tov iroraixov 
££wTTfaofjt,4vov$ ws KwXvaovras hiafiaivGw, ttz^qvs 8* itrl rais oxBais iraparsrayfjbivovs dvw rwv 
famiaw (hs KcaAvaovras sis Tip? > App,Gviav e/c/JatVeiv. ^uav 8* ovVoi 'Opovra Kal *Aprovxa 
'App,4vwi Kal MdpBoi Kal XaXBawi jxtQdo<f>6poi *\ (417, 3) 

56 Tacitus, Ami., XIV, xxiii [L. IV, 146/7], " Atqne illnm [Corbulonem] finis suos 
praegredientem incnrsayere Mardi, latroeiniis exerciti eontraque inrumpentem montibns 
defensi; quos Corbnlo inmissis Hiberis vastavit ... [xxiy] ... Unde in regionem Taurauni- 
tinm transgressns inprovisum perieulum yitayit .., '\ (417, 4) 

57 See above chapter XI, pp, 249-250; Sebeos, xxxy, p. 138, " ••• np krjkh fi 
XTwpqntjluhi "; Zewond, ii, p. 7 " •♦• fi utuCSwhu JfujpqnLgwjIi^ *♦♦ ", Both 
consist of tTwpqnij (nom. he) < murd, while the ending — J-g or — wjg giyes it the 
sense of an ethnikon; " the inhabitants of Mardoc ", which is similar to ^np£—l^ 
Zujjiuuminh — kiujg) etc. We belieye that, in addition to Mardastan, the name 
of the Mardians has also been preserved in the province of Mardahk' (see above, Chapter 
III n. 18), Perhaps the Mardians of Mardahk' are the deseendents of those Mardians 



NOTES : CHAPTER XIV 



493 



who attacked Xenophon at the Kentrites crossing together with the Chaldaeans [see 
above n. 55]. Still together they moved northward and occupied the districts named 
respectively Mardahk' and Chaldia. The Mardians who appeared in Mardastan pro- 
bably came at a later date and from the direction of Iran, [See above nn, 54-55], (417, 5) 
5?a [Cf. Toumanoff, Studies, pp. 169-170, 200, 231 n, 285, 237 n. 305,248; Garitte, 
Agathange, p. 224.] 

58 Pliny, NH, VI, x (28) [L. II, 356/7], " ... proximi Axmeniae sunt Menobardi et 
Moseheni ". [Markwart, Sudarmenien, p. *53, Toumanoff, Studies, pp. 182-183, also 
458-460 nn. 93a and 98, See above nn, 6, 19; Garitte, Agathange, p. 225, and Hubsch- 
mann, Ortsnamen, pp. 254-255; Eremyan, Armenia, p. 71]. (418, 1) 

59 Ijuptikg < ^npmfi£—mj^, where ^npm[i£ = " Kurd" ; the form is similar 
to mmpuimuj—^ pwnwu—ji£, etc. Andreas' derivation from Jcurti-hayh is un- 
founded. [Cf, Hubschmann, Ortsnamen, pp. 255-259, 333-335 ; Markwart, Sudarmenien, 
pp. *53, 353; Garitte, Agathange, pp. 219-220; Toumanoff, Studies, pp.57 n. 54, 60 
n. 58, 129-130, 169-170, 181-182, 197 n. 222, 468 n. 138; Eremyan, Armenia, p. 60; 
Hewsen, Armenia, p. 329 n, 36. See above n. 19]. (418, 2) 

60 According to the Anonymous History [Sebeos, p, 2] and MX [I, xii], ^wqfinu was 
the grandson of Hayk, the son of Armenak, and the ^Jutqilhwu house was descended 
from him. It is possible that Kadm is a misunderstanding of the Syr. Jcadun, " first " 
so that "£w«*' duimuhq.riLpfjLft ywn.iJhwj (Pnn.pL fjLpni.fi) npqnjh JXpilhbwifwj 
... " {Sebeos, p. 2] means, " he gave ... to the first or eldest son of Armenak .., ", On the 
other hand, the hypothesis that faf} — S-~nu is the toponym equivalent to the Syr, 
hasdim (plur. hasd) > hard the root of Kardu-%-oi, is also possible, Cf, Marr, Tables, 
p. 5, Hommel, Qrundriss, I, pp. 187 n. 4 and 244 n. 4. In such an interpretation, 
Ijwqifhwfo would become a name of the type f&npnjidkuih, JXpmifkwu^ etc. [Tou- 
manoff, Studies, pp. 224-225 and 224 n. 270, 236.] (418, 3) 

61 Herod., V, 52 [L. HI, 58/9] MavnyvT) was a country adjacent to Armenia on the 
river Zab, or, as Herodotus calls it, the third Tigris. According to his indication, 
the Boyal Highway from Susa to Sardis passed through Armenia and Mantiene, " In 
Armenia there are fifteen resting-stages, and fifby-six parasangs and a half " whereas 
in Mantiene there are " thirty-four [sic], and a hundred and thirty-seven parasangs ". 
[NB, Adontz gives only four resting-stages in Mantiene and deduces the relative size 
of this country and Armenia from this ratio], Sirabo, XI, vii, 2 [L, V, 250/1], puts 
Matiane together with Media Piol,, VI, ii, 5, mentions Marovardva < matu and stana, 
and Mapriavrj- According to Sirabo, XI, xiv, 8 [L. V, 326/7], Lake Urmiah was called 
st J) Mavriavrj, Kvavrj ipfi'qvsvddaa, ... ". In his description of Media, [XI, xiii, 2, 
L. V, 302/3] the same lake is called Siravra [cf, p. 302 n. 3 " Katravrav "]. In one case 
Strabo has followed Apollonides, and in the other Eratosthenes, moreover he mistook 
Kvavrj " blue " as the translation of Mavnavij,, whereas it belonged with i/n-avrav < 
[Ka}7Tovrav = yirnqnunwu of the Arm. Geogr. II, (Pers, /jl^^S') which does mean 

"blue", Cf. Hubschmann, Ortsnamen, p. 439; Markwart, Bran, p. 143; Eremyan, 
Armenia, p, 58; Hewsen, Armenia, p, 329 n. 39, also next note,] (419, 1) 

62 The etymology of manda and of the other forms has not yet been worked out, 
cf. Hommel, Qrundriss, p, 195 and Host, Untersuchungen, where the latter compares 
mada and madh, with the sense, " verstandigen, in verstandlicher Sprache redenden ", 
Ibid., p. 73, According to Strabo, XI, vii, 2 [L. V, 250/1], Media and particularly 
Mantiene were famous because, " ^ /zev yap ajtwreAos fter/wjT^v oivov $epei ... eV 8c rots 



494 



NOTES : CHAPTEK XIV 



S&cSpecrt ofiyvovpysiTai teal rwv tfyvXhwv diroppet p,4Xv ... ", See also. Ibid,, II, i, 4 
fX, I, 272/3], This fact may perhaps provide a basis for tracing mada and maii from the 
root madu, Avest, mada, Gk, pidv, Slav. MiAJh, meaning " -wine " or " mead ". [The 
entire question of the " Mada ", the part they played in the " barbarian confederation " 
or " Umman - Manda ■" of late Assyrian sources, their relation to the country of Manna 
and its inhabitants, and finally their connexions with the Armenians, has grown enor- 
mously since the time of writing of this work, and, indeed, is considerably developed 
hj Adontz himself in his late work, Eistoire d'Armenie, On these highly controversial 
problems, see also, Piotrovskii, Origin, and Kingdom of Van, D'iakonov, Assyro-Babylo- 
nian Sources, Media, and Collection, van Loon, Vrartian Art, pp, 1-28, et ah] (419, 2) 

63 MX, 11, lvii, " BuiiMLpu unpm muhh btybw^ f w^f mSmmnihbujg fi 
l^utipmhgh wpkbj_[ig Upbuig wyfrnup^fih : fiuijg kh ungw phutpbuiSp Zpbmj 
fi ffrnhnibutj Tjufhifut* *** Swpbw^g bh ungw ft jJlp^mlfiuj wnw^fih fi tJ}wppkwtj 
PwqwLnpkgkipjIj^ k jwnw^qfiiJnLpbujJip mhtiih jlXphwg w^fuuip^fih ft lywwfiL 
4mukm^p ft tynijpmhu JXCSmmwip : pwjy p} jfal u^wm^wnw^ qw[pLituj hn^ui 
uijup n£ qfimbd* wj^ vpumnhfih jJXpww^fiuf; ijhqJiLg k qwuwwlibpmog s h wh— 
niwhfib wSmmndifi^ npiq£u pi; biff : fet tying fi fywpufig Smtintbuihu qhnuui 
fyn^bb jwhmSt hmfuhmji ", Thus, the Amatuni came from Ahmatan and were 
the descendents of a certain Manwe, Xorenaei presents the Amatuni as bty d " immi- 
grants " (evidently from Pers. Jul)? but why then mention their ancestor Manwe, " on 
account of whom some of the Persians still called them Manwean " ? The legendary 
account presumably had spoken of Ifwwm. and Uwmm.bwh, (subsequently distorted 
into JfwhnL JfminLhwh), and Malm is an eponym pointing to the people having the 
same name. The initial a- in A-mainni is also found in the form A-madai for Madai, 
and seems to have been a characteristic of Alarodian pronunciation, Homrnel, Grundriss, 
p, 195 n, 2, [Of Toumanoff, Studies, pp, 61 n. 58, 169 n, 81, 197-198, on the Caspio- 
Medians and the Amatuni; 110 n, 173, 198 n, 223, 212, for the Mandakuni.] (419, 3) 

64 The Gahnamak has Jfmpw^hmh pro ITwpnijbiiih, as does Thorn. Arc. HI, vi, 
p, 109, MX, II, viii, insists that " bpfypnpij. pwqwuipnLphwfch l[*&ip fi ^ik/^ 
JX^mmimtywj Xfwpwrj bqbjnj ptu^minpfj^ np mjdS IfnLpmgmhq tyn^bh : 
'fiuihiifj n£ uwbh hmimujbmfj tvq^fih IfnLpivijwb mip^ wjj^ ifiwpwtjLntjh mf;p ". 
but he himself speaks of " JXpt^mS muihni.mf;p wqgfih Ifntpaigmh * * ♦ ", Ibid., 
II, xliv, and " IXpqmS hwCwmbw ITnLpiuzfwhujij ", I bid., Ill, xlvi. [Cf. Tou- 
manoff, Studies, pp.224 n, 270, 230-231 nn, 280, 284. Eremyan, Armenia, p, 65. 
Hewsen, Armenia, p, 331 and nn. 47-48,] (420, 1) 

65 Plut, Imc%Vms, xxvi, 4 [L. II, 554/5], " ,„ awfjkBov *App,4vioi koX rophvyvot, 
iravoTpaTiq. Se MijBovs K<zi > Ahia^vovs ayovres ol j8a<rtAets Traprjaav, $}kov 8e ttoAAch ph> wttq 
T7js h> BafivXwvi 9aM(ju7}s MjoajSes 1 , ttoXXoi hk wno tt}$ Kamrias *AXf$avol Kal "Ifiypzs *A*\fia- 

VOIS TTpQVOlKQVVTGS, OVK dXiyoi 8e TWV TT$p\ TQV *Apd^7)V VG}XQp,ivWiV dpacfiX^VTOi X&plTl Kal 

Bd>pots 7T€ia$svT€s aTTrjvTTjijav, ..,*';' (420, 2) 

66 Herod., ni, 93; VII, 68 [L. n, 120/1; HI, 380/1], Strabo, XI, vii, 1; XI, viii, 8 
[L, V, 248/9 ; 268/9], Pliny, NS, VI, xvi (42), " Atropatene ab Armeniae Otene regione 
discreta Araxe ". JPtol, V, ^ xii, 4, " ... wapa §€ t6v Kvpov Trora^ov ... i? 9 £nprj ", 
[Of. Toumanoff, Studies, pp. 68 n, 65, 110 n, 173, 129, 132, 219, 475, 482. Eremyan, 
Armenia, p, 75, Hewsen, Armenia, p. 333, Cf. Erye, Persia, p. 47; Trever, Albania, 
p. 46 etc] (420, 3) 



NOTES : CHAPTER XIV 



495 



6? FjjXm > modem Gilan = " 9*%- wg ", MlUe, v, p, 116; . " fykimiiQ " jtfew, 
Kalarik\, II, six, p. 140; " kpfyfrp ^hhijp—wj " Sebeos, xii-xiii, pp. 57-58; " thkrjdwhrj 
jhuinh " J£X, II, liii. (7/, %kqmS^ ^hquipgnLtifi* [Eremyan, -4r^ema, p, 47 ; 
Toumanoff, Studies, p. 214; D'iakonov, Media, p, 93 n, 1,] (420, 4) 

68 Strabo, XI, vii, 1 [L, V, 248/9], et ^tvtaVas 8* e> T77 Oviria r&xtacu woAw, ^v Aividva 
KaXeiaOat .„ ", Andreas, Ainiana, derives the Arm, j£iu&ji in Paytakaran from this 
city. But Strabo, XI, xiv, 14 [L. V, 334/5] also asserts that ** Aiyovrai Se /cat tcDv 
Alvidvcov rivis, ol fxkv ttjv Oviriav olidjaai, ol S* VTrspBs rcov ^Apjx^viwv virkp rov "Afiov /cat 
tqv Ntfiapov ". On the basis of this passage, the Armenian JXbfi should also be 
linked with the \4twa-wi. [Eremyan, Armenia, p, 62], (420, 5) 

69 Strabo, XI, Yiii, 4 [I*. V, 260/1-262/3], " [27<zjccu] amavrts 8' cfe «rt to ttoXv vo/wxSss ... 
TrapairX^mas g<J>qBqvs iiroiijuavTO roi$ Kip,p,$plQi$ /cat Tpypeai, ras ftev fta/cpOTepa?, ras Se 
/cat iyyvOzv* /cat yap r^v BaKrpiav7)v Kariaxov Kal ttjs > App f €p(as KarsKTrjaavro tt}v aplcmpf 
yfjvj, rjv koi e-nwvftov iavrcSv KariXmov ttjv SaKaafprqv, Kal p,i%pi KaTTirahoKcov, Kal 
fxdXicrra rwv irpos JSvijeiva), ovs Uovtikovs vvv koXovoi, TTporjKBov " '» [As in the case of 
the Mada, the migration of the Saka has received a great deal of attention from recent 
scholarship. See above, the bibliography given in n. 62, also Eremyan, Armenia, p, 73; 
Tonmanoff, Studies, pp.52, 54 n, 49, 60 n. 58, 80; Frye, Persia, pp. 41-42, 47, 67-70, 
152-160, 164-167, 177-178, 216, 252 nn, 50-52; Mellink ed., Dark Ages, Trever, Abania\ 

(421, 1) 

70 Hiibschmann, Ortsnamen, p, 457, is of the opinion that the phoneme s for 5 in 
fcwl{w>lh stands in the way of this etymology, but even if we were certain that saJca 
must > uwfy and not jw/^r in Armenian, it would still be possible to attribute this 
transformation to some peculiarity of Titian or Albanian pronunciation, Markwart, 
JSran, p, 120 n, 3, holds that the etymology of Sakasene, and of the festival tA cra/cata 
" ist eine blosse Konjektur " of ancient writers, that they probably had nothing to do 
with the Saka, and the etymology was devoid of historical value. This conclusion 
is correct in the case of the ad/cata which has no connexion with the Saka, and is 
probably < sdh " tribute " referring to the great giffcs made at the time of the festival. 
[See preceding and following notes.] (421/2) 

71 Arrian, Anab*, HI, viii, 4 [L, I, 246/7], " MrjScov 8e rjystTo 'Arpoirdrijs' fwsTdV- 
tqvtq Se Mrfiois Kahovoioi t€ Kal > Ahf$avoi /cat SaKtaivai ". Strabo, XI, xiv, 4 [L, V, 
320/1], " t) ZaKaarpr/j, Kal avrrj rfj 'AXftavta irpoa^opos ... ". Pliny, NH, VI, x (29) 
[L, II, 358/9], " .,♦ Moschorum tractus ad Hiberum amnem in Oyrum defluentem et 
infra eos Sacasani et deinde Macerones ad flumen Absarrum ". Ptoh, V, xii, st HaKa- 
vrprfj ". [See above nn. 69-70 and next note,] (421, 3) 

73 Be Lagarde, Gesam. AbK, p. 155 and Arm, Studien, pp. 135-136, 199/5, believes 
that the original name of the country was Si, which became Si-sakan (i.e. " Si of the 
Saka ■") after its occupation by the Saka — An unacceptable hypothesis ; composite 
names usually have a separate origin. The Armenian form JJjithji consists of the root 
ufj~ - and the ending ^rnhji, but the Persian form Jjpuwl^w'ij presuposes the root 
ujiu— , The family historian of the house of Siwnik' says that king VaWsak gave the 
supreme command of the army to the house of Sisakan, " &t ^pmStuji Ufiuw^uihmgh 
ipmSwhwmmp ifrhk^ fj i[hpiuj unlkhinjh mp^mhp ippmyh, k kplfpnpq piiiqim.- 
npnLpkmh [ii-pnj. k phi^iS J^tuj^ £whwuiuit[ u^mmkpmqSiUL Tj.pwhh £rihmy* 
npnj rnhmh wnw9bnjh Uhuiufy fynslup :. ", StepK Orb,, I, iv, I, pp, 54-55. 
A few centuries before Orbelean, the Arab geographer Ibn al-Eakoh (ca. 903) wrote that 



496 



NOTES : CHAPTER XIV 



the Arabs took the city of Baylakan under the caliph Ut'man (644-656) and sent their 
cavalry to conquer, a>j jU^gilj jl^T^allj 3j!j l)IjA*JIIj l y*^^ 
o ~»l£ f/L3L#\ BGA, V, p. 293 = Karaulov, Sbornih, xxvii. It is easy to iden- 
tify Sisan (instead of the Sisar of the text) Mezkowantf (\Jhbl^ni.mu-^ Utile' 
(/)lw7^— ^) ? Meciranh' (JJ hbfjpwh—g) among these rustaq. The last name is a dis- 
tortion of £wp&jtuh-g or Ifm^jumii] £wp£iwh—g. With the exception of the 
first, the remaining names are known from the Arm. Geogr., [p. 33/44, see Appen- 
dix IV] as districts in Arca^. The first name is to he identified with the Ufiuwu 
of Orbelean, according to whom this was the original name of the Gate of the Huns, 
i.e. the present Derbent. Ibn al-Pakih gives a detailed description of the pass at Der- 
bent which had been fortified with a long wall by Xusro Anosarvan, and says that 
there were seven passages opened in this wall at each of which a city was built, these 
were inhabited hj Persian troops and called siasihin, " JUj (**SaJI ,V» <UjIaII 
t^SvlwL^JI rtJ> "• Moreover, the author continues, "it is said that men 
from Armenia are required to guard this wall and these gates ". BGA, V 
p. 291 = SbomiJc, p. 23. These last words leave no doubt that siasihin is 
the Arab. plur. of sisalcan. The modern Arab pronunciation is sisajan (plur, 
siasijin, as given in BGA, V, p. 728 = Karaulov, Sbomih, p. 17), -Xusro built 
Vayc (^ojj) = ^wjng \&np~\ and other fortresses in the land of Sisajan and filled 
them with Siasijans, a warlike and brave people). The departure from the correct 
Arabic pronunciation shows that the author relied here on an ancient Persian source. 
Since the defense of the gates was entrusted to the population of Sisakan, in which lay 
the district of Sisan, we can understand the reason for which the gates came to be called 
Sisan. It is the echo of an historical event, namely, that these gates had been guarded 
by Armenians from Sisan (c/. the words of Vasak of Siwnik', LP\ XLV, p, 264, *' „. 
ku i/jiu £ *lpwij Smpquiwhh Ifi k T»jiiji& JXqpLwuJii} jfidnLiS &kn[ih lp, pwijnLS 
qopwqipLfug £>nhwij phij. jiu pmpkljwSmgmh * • » ". )Should we accept Sisan as a 
contraction of Sis<ak>an, or of Sis<aj>an, or should we take it as an independent 
formation (< the root sis + an, the suffix indicating the plural) and connect it with 
modern Sisian, a seetion of the Zangezxor district of Siwnik', although Sisian can also 
be taken as a contraction of Sisakan, The question whether Sis is the name of a parti- 
cular people, or whether it has the same origin as Siwnik' is not clear. The form V^i-hfj^ 
is older than Ufwwlfiuh; it is found in Eusebius, Praep. ]£vang. 9 VI, xxxi, " /ecu oM) t# 
> AXavia Kol *AXf$avia koX > Qt7]vtj kch Savvia ", It has also been suggested that &awi-ns 
should be read 2avvi~ns in Strabo, XI, xiv, 5 [L. V, 324/5], Procopius, Pers*, I, xv, 1 
[L, I, 128/9], knows of the Sunitai as neighbours of the Albanians [sic. The text of 
Procopius has Alans, " to 6e arpdrev^a tovto 77epaa/>/i.evtW re /cai Hovvirwv ^crav, ot 
Btj > AXavots slmv ojuopoi "], The Persian form of the name: Sisakan, appears for the 
first time in Zachariah of Mitylene, XII, vii, p, 328, for the year 554, where he gives 
Sisagan as a country separate from Armenia as are Arran and Gurzan [See above, chapter 
IX, p. 171]. In Armenian literature, Sisakan also appears later than Siwnik 5 . It is 
interesting that the Anonymous History does not include Sisak in its genealogy of the 
Haykids, where this is done by MX [I, xii, ***** bhirn. qnpqfi pLp qUfiuwlf *** 
uwCSwhu £wmwh]; hfiui dwnuiui^nLphujh jj bmj£h phq. wpwLWju rffju ^k grpii^w 
ilji^ np fj.kmh bpuwfu *** Ji£wb£ ji qm^mu : Hum phml^km^ Ufium^ jhuL 



NOTES : CHAPTER XIV 



497 



^jjhnLphmSp. quw^Smbu phinl[m.pkuih fri-pnj* h qm^fuwpih Ifn^i fiLpni£ 
whnLiudpu Xfjuhp^* mj^ tywpug jumwljuiipijh jiiii^ Djiuiul^uiu ty n >ku "• Does 

this omission indicate that the form Sisakan was not yet in use in the period of 
the Anonymous History, or does it reflect the period 571-660, when Siwnik' was separated 
from Armenia ? [Cf. Toumanoff, Studies, pp. 330 sqq., particularly 332, for the use 
of this term to date the History of Movses Xorenaei], The root sis- in Sisakan can be 
reduced to si-, in Siwnik', if we accept the hypothesis that sis- has the same relation 
to si-si~Jc\ as Pwqwu < wfywh > to Pwij—g, Moks-ena to Ifnfy — £, Akilis-ene to 
h1{kqji—g^ blfkqkwijy, etc. [Cf, Toumanoff, Studies, pp. 130 n. 229, 214 and n. 244, 
352 n. 50, and above nn. 69 sqq.] (421, 4) 

73 FBy III, vii, the " qopu pwqwu—£wtj " live north of the Kura. In Koriwn, 
XI, 5, p. 34, the " Puiqwuuilfwu Ifnqiiwug " lie on the road from Albania to Gardman 
and Tasir, i.e., along the Kura. Muslim sources are familiar with Balasakan, on the 
right bank of the Araxes between Ardebil and Vardan[akert], Markwart, Fran, p, 120, 
(cf. ptuqmh — nnw^ a district in Paytakaran [Arm. Geogr., p. 33/44]. We can see 
from this that the people under discussion came down into the valley of the Kura from 
the mountains and moved to the junction of the Kura and the Araxes. During this 
migration, some of them must have settled in the locality named Pinq—g^ Pwi^ — u y 
after them. Mlise, vi, p, 134, " ••• pwpkljiu Llwijkw^ ip uw phij. Zhnuiuinj 
Znbfi ilfmiprnhnLphmSp piujjmuml^mh wpgwjfiu ••♦ ". Since, in the same con- 
text, ZP\ xlvi, p. 268, mentions only the king of the Huns, we must presume that 
EKse's King of Baiasakan is one and the same person as Hefan, the Hun. In this 
case, it is possible that Zknwh is merely a form of IXnuiu = Albania, i.e., that Vasak 
had concluded an alliance with the Albanians, or the Albanian Huns who were ruled 
by the "King of Balasakan. The Balas probably were one of the Hunnic tribes. [Cf. 
Eremyan, Armenia, pp.43, 88; Trever, Albania, pp.75, 150, 191-192, 196-197, 208]. 

(423, 1) 

74 Pliny, NH, VI, xi (29), [L, II, 358/9], " ,„ ab Albaniae confinio tota montium 
fronte gentes Silvorum ferae et infra Lupeniorum, mox Diduri et Sodi '\ Ptoh, V, xii, 
p. 938, J£ohovK~7)vi] which reproduces the Arm. JJnwpg. The comment of Steph. 
Orb., I, iii, I, p, 51, " Qnppnpq UnPfjij rj.im.um, np uwlm 4whmu^wnppq.kuib 
pgng h qumuui^mu^ OI } n S u Ifn^kijuiL UnPj> : Zfrhfykpiipq lXqin£k£ tj.wi.um *** ". 
The IXqw^hr&g were apparently drawn from the people whose descendents in southern 
Daghestan are now called \w£Ji6* [Cf. Toumanoff, Studies, p. 182 n. 146]. (424, 1) 

75 FB, III, vii, " qopu ^nLrpupiutj " pro " ^uipquipinij " under the influence of 
^tiL^wpg the well known province of Gogarene. MX, II, viii, " Quipqwpwguig 
fefuuihnLpfiLu " . Thorn. Arc, III, x, p. 177, " •«* ji£ni&l fi qw^mu Qinpqw— 
puizjLntj, k kppkmj^ dwwul; ji guujiugu ifkb fywpmujL ••• ". Ibid., Ill, xviii, 
p. 216, " fflwpwwL jw^fuwpQi ^wpij.wpwijLni} ". [See next note], (424, 2) 

76 Strabo, XI, v, 1 [L. V, 232/3] rapyapeis. The similar rwyapyvrj, which is also 
known to Strabo has nothing to do with the Gargarians and belongs to Armenian nomen- 
clature. Gargar may = Assyr, JcahJcari, " country, province ". There is, incidentally 
no etymological or historical objection to the association of Qm.if.wpg with the 
^wpn-iupg, who would then be considered earlier migrants. This is all the more 
tempting that the neighbouring provinces of Sirak and Asoe have also preserved the 



498 



NOTES : CHAPTEK XIV 



memory of mountain tribes, the " UipaKcov teal , A6pawv <j>vAa ... ". Strabo, XI, v, 8 
[L, V, 242/3], who "were the most numerous and powerful of the peoples nomadizing 
between the Maeotic and Caspian seas. They moved from north to south (" foyahes 
t&v dvwrdpco ") and, gradually spreading carried on trade with Indian and Babylonian 
wares which they received from the Armenians and the Medes, " ... henop^vovro 
jcajuiJAots tov *IvBikov $6pTQv Kal tov BafivAwviov Trapa ts *App,svicQV Kal MrjBicov 
SiaSexoftsrw ". Idem, [Cf. the objection of Manandian, Trade, pp. 49-50]. In view 
of their close connexion with the Armenians, it would not be surprizing if part of them 
had moved to Armenia into the district later called fclipuify = 2ipax„ while IX^J = 
Mopero*., According to Pliny, NH, VI, iv (16) [L. IE, 348/9], the Cephalotomi lived 
next to the Seraci [NB. the text of Pliny has Serri and not Seraci, "... post eos Serri, 
Cephalotomi"], However, we learn from Strabo, XI, xiv, 14 [L, V, 334/5-336/7], 
that their real name was Saraparas and that part of them moved to the Median border, 
" ,,. $acrt §€ Kal ®pq,Kcbv rtvas, rovs irpQcrayopsvonipovs Saparrapas, olov /ce^aAoTo/iovs, 
oudjertu virkp rijs * Apytstfias » irfajcriw ... Mrfiwv, OypiwBeis avdpc&Trovs ... aTTOK^aAiGrds* 
tovto yap BijXovmv ol Sapairdpai ". If we accept the hypothesis that the Saraparas 
settled in Armenia together with their neighbours the Sirakeni, a new light is cast on 
the gentilicial name of the Kamsarakan princes: Sara-para has the literal sense of 
KG^aAo-Toixos, according to MX, II, Ixxxvii, Mm Sump had the same meaning, " .,, 
tqwlfuiu qnjnil pninpnLpfiLh qwqwpkwhh^ whm.whkijWL MmSump mjhp 
wqwtpuL ", i.e. the name was composed of Kam and sar. On this basis we can presume 
that the Kamsarakan were descended from Saraparian immigrants into Sirak. Sub- 
sequently, with the transfer of Sirak to the branch of the Arsacids which had settled 
in Ajaarunik 5 , it also received the name of the Kamsarakans. According to 
the Armenian tradition transmitted hj MX [II, v, viii], Sirak < Saray, and Gugark' < 
Gusar, It is clear that tjwpwj and %nLywp are merely eponyms, &wpwj has the 
same relation to (^fjpiij^ as fyntyuip to %nLq.wpg+ The correct form of the latter 
may be fynL^wp ef modern Gujareti, It is, of course possible that Sirak means " field " 
(cf. the Sirak plain along the Alazan), x$oi Bra, seJira, sirak. [On the Gargarians, see also, 
Trever, Albania, pp. 31, 46, 48-50, 58, 66, 140. On the Gusarids and the vita^ate of 
Gogarene, cf, Toumanoff, Studies, 183 sqq,, 467-473 and notes. On Sirak, Manandian, 
Problems, pp. 61 sqq.. On the Kamsarakan house, Ibid., 110 n. 173, 130 n. 229, 
171 n. 90, 193 n. 207, 206-207]. (424, 3) 

77 Eerod., VII, 68 [L. HI, 380/1], the Mvkoi lived next to the Qfcioi and were armed 
like the IlaKTVGs, i.e. the inhabitants of Bohtan, This last fact suggests the possibility 
of a link between the Mvkoi, and the Mok-s-ians (Jfmfm^j). It is interesting that .MX 
[II, viii] believed that the inhabitants of Kstunik*, the district next to Mokk' had come 
from Siwnik', a territory near Mughan, "^tf^r anymmnthfiu h ai^nqphk^Jiu 
qinfi ufiumrfhwL t 1 "fwutfywhih wpqiupk ^wmrihwb. n£ t^jimhSy pi jwhrnSi wpmhqh 
qjj.wLwnh rnhnmuhkhy k pi jwhrnSi qwnimmgh qhuifuwpwp rn.pfjn.huh I^nshajkin^ ", 
[Of. Toumanoff, Studies, p. 68 n, 65, see above nn. 19, 58] (425, 1) 

78 Strabo, XI, vii, 1 [L, V, 248/9], According to the geographer, there existed, 
" Ilappamwv nvas ... ovs /caAacrfleu vvv Ilapviovs ... ". More correctly, the Parsioi 
and the Parrasioi were separate but related peoples (Pliny, NH, VT, xviii (48) [L. II, 
372/3] also asserts that " Gaeli quos Graeci Gadusios appellavere ... although we know 
that Gaeli and the Cadusii were separate peoples). According to Mov. Kalanh\ I, xiii, 



NOTES : CHAPTER XIV 



499 



" m^fumpi iffj IXpbwfumj ufiupuwliiuh whm.h " exists, the name is given as tywpu- 
l(U]hij in the heating of the chapter. In IMd., HI, xx, p, 264, the same name is given 
as ^a/^— ^«/&— g while the -4rm, Geogr., [p, 33/44] has tywbliuihg* The other district 
^nLuin—fj-^iuipn^u is found in .4so&&, III, xvii, p. 198; xxx, p. 256; xlviii, p. 283 
in the form <pwnfiu <nu>. (425, 2) 

78a [Of. above pp. 317 sgg. and n. 38a.] 

78l > [#ee next chapter.] 



500 



NOTES : CHAPTER XV 



CHAPTER XV 



a [The bibliographic indications given at the beginning of Chapters IX and XIII are 
also relevant for this chapter. For works relating to the problem of Armenian feudalism, 
see below n. 15d.] 

1 The Tigranid dynasty goes back to the period of Cyrus and Astyages, whose contem- 
porary was an Armenian king having two sons: Tigranes and Sabaris (see above, Chapter 
XIV and n. 1). Erom them were descended the following: 

1. Artaxias [Artases] in 190 B.C. 

2. Tigranes [Tigran] I. According to Appian, Syr,, XI, viii, 48 [L. II, 196/7] Tigran 
the Great was the son of a Tigran, *' fiamXevs y Ap}x^vias Tvypdvqs 6 Tiypdvovs ... ". 

3. Artavazdes [Artawazd] I. 94 B.C. 

4. Tigranes II the Great (94-56 B.C.). His son also named Tigran was married to 
the daughter of Kong Phraates of Parthia. The daughter of Tigran was married 
to King Mithradates of Atropatene, [Media], Cass, Dio., XXXVI, xiv [L. in, 20/1], 
" 6 Midpibdrss 6 tripos 6 €K Myblas yafxppos rod Tiypdvov ... "]. MX [II, xi, xiv- 
xvi, xviii], confuses Mithradates of Pontus with Mithradates of Media. Tigran II 
was married to Cleopatra, the daughter of the former, whereas his own daughter 
was married to the latter. 

5. Artavazdes II (56-30 B.C.), Taken prisoner in 33 B.C. Killed by Cleopatra VII 
of Egypt in 30 B.C. 

6. Artaxias II, son of Artavazdes I. Killed " ... per dolum propinqxiorum ". Tac. 
Ann., II, iii [L. II, 388/9]. 

7. Tigranes LEI. Brother of Artaxias and Erato, [" Nee Tigrani diutumum imperium 
fuit neque liberis eius, quamquam societatis more externo in matrimonium reg- 
numque ".] Idem, 

8. Artavazdes III. Deposed in A.D. 1. Ibid., II, iv [L. II, 388/9], [" Dein iussu 
Augusti inpositus Artavasdes et non sine clade nostra deiectus. Turn Gaius 
Caesar componendae Armeniae deligitur ".] 

9. Ariobarzanes of Media. [" Is Ariobarzanen, origine Medum, ob insignem corporis 
formam et praeclarum, animnm volentibus Armeniis praefecit, Ariobarzane morte 
fortuita absumpto stirpem eius haud toleravere; ...".] Idem, 

10. Erato, sister and wife of Tigranes III (for the second time), Idem, [" temptatoque 
feminae imperio, cui nomen Erato, eaque brevi pulsa ... ".] 

11. Vonones of Parthia, [" „. incerti solutique et magis sine domino quam in libertate 
profugum Vononen in regnum aeeipiunt. Sed ubi minitari Artabanus et parum 
subsidii in Armeniis, vel si nostra vi defenderetur, bellum adversus Parthos sumen- 
dum erat, rector Syriae Creticus Silanus exeitum custodia circumdat, manente 
luxu et regio nomine ".] Idem. 

12. Artaxias III = Zeno, son of King Polemon of Pontus, Ibid., II, lvi [L, II, 472/3- 
474/5], [" (Armenii) Ambigua gens ea antiquitus hominum ingeniis et situ terrarum, 
... maximisque imperils interiecti et saepius discordes sunt, adversus Romanos 
odio et in Parthum invidia. Regem ilia tempestate non habebant, amoto Vonone: 
sed favor nationis inclinabat in Zenonem Polemonis regis Pontici filium, quod is 
prima ab infantia instituta et cultum Armeniorum aemulatus. ... proceres ple- 
blemque iuxta devinxerat. Igitur Germanicus in urbe Artaxata adprobantibus 
nobilibus, circumfusa multitudine, insigne regium capiti eius imposuit, Ceteri 



NOTES : CHAPTER XV 



501 



13, 



14, 



15. 
16, 

17, 



The 
1. 



venerantes regem Artaxiam consalutavere, quod illi voeabulum indiderant ex 
nomine urbis ".] 

Arsaces, son of Artabanus III of Parthia, Ibid., VI, xxxi [L, III, 206/7], [" (Arta- 
banus) avidusque Armeniae, eui defuncto rege Artaxia Arsacen liberorum suorum 
veterrimum inposuit, ... ".] 

Tigranes IV [grandson of Herod the Great], executed in A.D, 36, Ibid., VI, xl [L, III, 
224/5], [" Ne Tigranes quidem, Armenia quondam potitus ac tunc retis, nomine 
regio supplicia eivium effugit.] 

Mithradates of Iberia, Ibid., XII, xliv-xlvii [L, III, 376/7-384/5], 
Rhadamistes of Iberia, opposed to Tigran V from A,D. 50, [Ibid,, XII, xliY-li ; 
XIII, vi, xxxvii,[L, III, 376/7-390/1; IV, 10/1, 60/1], 

Tigranes V of Cappadocia, the Roman candidate and rival of Tiridates I [" (Veru- 
lanus) .,, quosque nobis a versos animis cognoverat, caedibus et incendiis perpopula- 
tus possessionem Armeniae usurpabat, cum advenit Tigranes a Nerone ad capessen- 
dum imperium delectus, Cappadocum ex nobilitate, regis Archelai nepos, sed quod 
diu obses apud urbem fuerat, usque ad servilem patientiam demissus, Nee consensu 
acceptus, durante apud quosdam favore Arsacidarum. At plerique superbiam 
Partborum perosi datum a Romania regem malebant ". Ibid,, XIV, xxvi [L. IV, 
Ibid., XV, vi, xxiv [L. IV, 224/5, 252/3], 
Arsacid Dynasty: 

Tiridates [Trdat] I, from A.D. 50, permanently A.D. 66-80, Ibid., XII, 1-li; XHI, 
xxxiv, xxxvii-xli ; XIV, xxvi ; XV, i-xvii, xxiv-xxxi ; XVI, xxii, xxiii-xxiv [L, III, 
388/9; IV, 56/7, 60/1-72/3; 148/9; 216/7-242/3, 252/3-262/3; 372/3], For the 
reception of Trdat I in Rome see below n, 7, 

Axidares, A.D. 110, son of Paeorus and brother of Tiridates [see next entry], 
Parthamasiris, A.D, 111-114, brother of Axidares, Cass, Dio., LXVIII, xvii, [L, VIII, 
392/3-400/1], 

[Vologaesus] — 117, SEA, " Hadrianus ", xxi [L.I, 66/7], [" Armeniis regem 
habere permisit, cum sub Traiano legatum habuissent ... ".] 

(A king appointed by Antoninus Pius, A,D. 138-161). [" Sous le regne d'Antonin 
le Pieux, une monnaie des annees 140-144, avec la legende ** Bex Armeniis datus ", 
nous montre cet empereur posant la tiare sur la t§te d'un prince ... ". Grousset, 
Arminie, p. 111.] 

Sohaemus (159-162). [Jamblichus, in Photius, Bibliomane, xeiv, II, p. 40, " ... 
aKfjid&w em £oo.ip,ov rov ' Axaip>zvi§ov rov Mpcra/aSou, os /JacrtAevs fy €/c iraripwv 
/JacriAecov, yeyove Se op,ws Kal rrjs ovyKArp-ov fiovXrjs tt}s iv *PwfM7), Kal vfraros 8e, etra 
Kal fiaoiAevs TrdAw ttjs /xeyaA^s" *Apfj,svias "«] 

Paeorus (162-164), [Fronto, Ad Verum Imp, L, II, p. 144/5] CIG 6559 " 'Avp&ios 
IlaKoopos fiaoiXevs ixeyakys ^ApjjLzvias ".] Sohaemus, bis (164-169). [See above, entry 
No, 6 and Cass, Dio., LXXI, 2, [L. IX, 4/5], 

Sanatruces [Sanatruk] (190-197). [The problem of Sanatruk is particularly compli- 
cated, see Toumanoff, Studies, p. 284, Debevoise, Parthia, p, 235, Maricq, Sanatrmq, 
et at, also next entry,] 

Vologaesus [Valarsak] (197-216). [Cass, Dio., LXXV, ix, L, VIII, 418/9], " *Qti r$ 
OvoXoyaiatp tw HavarpovKOv ttcuSi avrnrapara^apLeva) tqis Trnpl SsovijpQv, ... p,4pos ti 
rijs 'Apfievias hrl rfj zlprpsr) ixapioaro ". Ibid., LXXVII (LXXVUI), xii, 1, [L, IX, 
304/5], " Tqv Se tcuv y Ap}j,Gvicov jSaatAea ... e/caAe<7e p,kv <f>ihiKois ypap,p,auw ... IB/jacre 
Se Kal 7repl rov Avyapov "• Cf, Maricq, Chronologie']. 



502 



NOTES : CHAPTER XV 



10, Tiridates II (217). The kings of the Illth century are not known, and the existing 
evidence is very confused. We have mentions of: 

11, Tiridates III in 253 

12, Artavazdes in 269 

13, Tiridates IV, the first Christian ruler contemporary with Gregory the nominator. 

14, Chosroes [Xosrov II] 

15- Tiran [Tigranes VH], deposed in 344. 

16, Arsaces [Arsak II], (346-349), 

17, Pap (369-374). . 

18, Varazdates (374-378), 
[Manuel Mamikonean, 378-385], 

19, Arsaces and Valarsaces [Arsak and Valarsak], 

20, Vramsapuh, 

21, Artases, 

[The Armenian chronology remains riddled -with problems despite the considerable 
-work done on it since the p\ablication of Adontz's book. Only the briefest indications 
can be given here of subsequent developments. The main areas of transformation have 
been: 1, the identification of the Orontid dynasty in Hellenistic Armenia and the ad- 
ditional material furnished by the Nemrut dag inscriptions and the Aramaic inscriptions 
of Artasdas I' found in Armenia. 2. The attempts to clarify the confusion created by 
Armenian sources, particularly Movses Xorenaei, who confuse the Artaxiad and Arsacid 
dynasties in Armenia, confuse rulers and totally misplace a number of them such as 
Abgar of Edessa, Tigran the Great, Sanatruk, etc. 3. The attempt to bring greater 
precision in the chronology of Romano-Parthian relations, particularly in the periods 
of Nero, Trajan, Marcus Aurelius and Septimius Severus. 4. The publication of the 
Sasanian inscriptions of Paikuli, the " Kaabah of Zoroaster " the " Res Gestae " of 
§ahpuh I, etc, as well as of the Greek inscription at Garni, which brought considerable 
new evidence on Roman-Persian relations in the nith century A,D. 5. Pinallythe 
chronology of the Christian Armenian Arsacids tied to the still controversial problem 
of the date of Christianization of Armenia, has received no final solution. 

The chronology of the entire period discussed here was given by Asdourian, Bezie- 
hungen, who already noted a number of errors in Armenian sources such as the mis- 
placing of Sanatruk. Magie's Roman Rule, covers the entire Roman period to the end 
of the IHth century, see also, Toumanoff, Studies, pp. 73-77, 83 n, 105, 111, 120 n. 207, 
166-167, 213 n. 241, 283-286, 291-294, 314, etc, 1. On the Artaxiad dynasty and its 
accession, see Manandian, Trade, pp. 32 sqq., Trever, Armenia, Toumanoff, Studies, 
pp. 73-77, 83 n, 105, 111 nn, 174, 176, 283-286, 291-294, Pebevoise, Parthia, Manan- 
dian, Tigrane II, etc, 2, On the Romano-Parthian relations from the accession of 
the Armenian Arsacids to the Sasanian revolution in Persia, see Schur, Orientpolitih, 
XIX, XX; Henderson, Chronology; Egli, Feldzuge; Kudriavtsev, TBI (1948-1949), 
all of which deal with the period of Nero and of the Peace of Rhandeia; Lepper, Parthian 
War, for the chronological problems connected with Trajan's campaign. Maricq, 
Sanatrouq, and Glironologie, Toumanoff, Studies, pp. 83 n. 105, 166, 213 n. 241, 284, 
for the reigns of Sohaemus, Sanatruk, etc. and the emprisonement of the Armenian 
king by Caracalla; and in general, Manandian, Critical History, II, 1 ; Debevoise, Parthia; 
and Magie, Moman Mule. 3. Por the IHth century Sasanian period, see, 
Trever, Garni and Armenia; Maricq and Honigman, PGDS, Sprengling, Third 



-t.^^'^*^: 



NOTES : CHAPTER XV 



503 



Century Irani Herzfeld, Paihuli; Seston, Diocletien; CJiristensen; Manandian, Critical 
History, II, I ; Gage, Sassanides, etc. 4, The problem of the IVth century chronology 
and the date of Christianization of Armenia seemed solved by Adontz's suggestion 
of A.D, 288 for this event, Vestige, and Baynes xe- working of the chronology found in 
Armenian sources, particularly Ifaustus, Home and Armenia, Unfortunately, Adontz's 
date has been challenged by numerous scholars, and Baynes' chronology, though very 
tempting, is unacceptable; cf. Ananian, La Data; see also, Manandian, Critical History, 
II, 1; Garitte, Narratio; Peeters, Persecution and Intervention; and my forthcoming 
study on Armenia in the IVth century,] (427, 1) 

2 Tac, Ann., XIV, xxvi [L, IV, 150/1], [See preceding note for the text,] (429, 1) 

3 Ibid,, XV, i [L, IV, 218/9], " Tiridates quoque regni profugus per silentium aut 
modice querendo gravior erat: — ISTon enim ignavia magna imperia contineri; virorum 
armorumque faciendum certamen ; id in summa fortuna aequius quod validius, et sua 
retinere privatae domus, de alienis certare regiam laudem esse ", (429, 2) 

4 Ibid., XV, ii [L. IV, 218/9], [" 'Hunc ego eodem mecum patre genitum, cum mihi 
per aetatem summo nomine concessisset, in possessionem Armeniae deduxi, qui tertius 
potentiae gradus habetur: nam Medos Pacorus ante ceperat. Videbarque contra 
vetera fratrum odia et eertamina familiae nostrae penatis rite composuisse, Prohibent 
Bomani et paeem numquam ipsis prospere lacessitam nunc quoque in exitium suum 
abrumpunt. Non ibo iofitias: aequitate quam sanguine, causa quam armis, retinere 
parta maioribus malueram. Si eunctatione deliqui, virtute corrigam ♦.,' "J (429, 3) 

5 Ibid., XV, vi [L, IV, 224/5], [" .., Dilata prorsus arma, ut Vologeses cum alio quam 
cum Corbulone certaret, Corbulo meritae tot per annos gloriae non ultra periculum 
faeeret. Nam, ut rettuh, proprium ducem tuendae Armeniae poposcerat, et adventare 
Caesennius Paetus audiebatur. "] (430, 1) 

6 Ibid,, XV, vi-xvii; xxiv-xxix [L. IV, 224/5-242/3 j 252/3-260/1], [" Turn plaeuit 
Tiridaten ponere apud effigiem Caesaris insigne regium nee nisi manu JSTeronisresumere; 
..."] (430,2) 

ea [Adontz's text has "the former" city, but Cass. JDio., LXHI,iii, 4 [L. VHI, 140/1, 
sets the coronation ceremony in, Rome, and the same seems to be implied in Tacitus' 
shorter account, Ann., XV, xxxi; XVI, xxiii-xxiv, L. IV, 262/3, 372/3. Bee next 
note for the text,] 

7 Ibid., XVI, xxiii-xxiv, [L, IV, 372/3], Cass. Dio., LXIII, i-vii, [L, VIII, 
138/9-146/7, " ... o TipiS&Trjs is t-^v "Pwp.ijv, ... avTJx® 7 )* # a & iyivsro avrwv Tropxrr) 
Bid irdoris tt}s cwto rov Ev<j>pdrov yfjs cSoTrejo £v imviKiois. avros tc yap o TipiBdrrjs Kal TjAiKia 
Kal /caAAet Kal yivsi Kal ^povrjixan tjvBsi, Kal 7} 0epcw7€ta 7) t€ irapauKsm} i} fiaaikiKTj Traaa 
avrcp qvv7)kqAovgi, .„ /cat avrovs oX T? ttoAcis AafMTpWS K^kOafiTf^ivai Kal Ot BljflOl TToXXa Kal 
Xapi€vra dvafiotovres vireSixovTO* „, koI rovro iir* iwia ixyvas, ots (hBomoprjaaVj 6p>o(ws 
iyivsro. (Wevcre Bk Travraxf) ^XP l r V s *ItoM&S, ko>1 avrtp Kal yvvij av/wrajoiWeue, Kpdvos 
Xpvfjovv dvrl KaAvTTTpas %x ovaa > <5or€ fx.7) opauOai wrapa rd irdrpia. iv 8e rf\ 'iraAia ^vysm 
7reju^^€tcrtv vttq rov Nipcavos iKQjjttodi], Kal Bid HiKGvrcov is Niav ttoXw irpds avrov a^t/cero, 
ov ixivroh Kal tov aKivaKTjVj, ot€ TTjoocrget avrto, KaradiuBai Kamsp KsAsvodsls TjOiATjusv* dAA* 
tJXois avrov to) KoAstp TTpoaiirr]^, Kairoi Kal is yqv to yovv KaBsls Kal rds X^P a $ ifra>^d£a$, 
BzwTTQTrjv ts avrov ovofxddas Kal iTpQGKwrjaas. o ofiv Niptov Kal iirl tovtoj avTov Savfxdaas 
tqis Te aAAois eSe|jcucraTo Kal fiovo^axias ev HqvtgqXois e^eTO, ,., . 

Msrd 8e tovto Is tc ttjv *Pa>jj,7]v avrov 6 Nipcov dvrjyays Kal to BidSypLa avrtp hriBrjKZ. koi 
Tiaaa fih> tj noAis iKGKOGtirjro Kal <j>cavl Kal pr^avu>fia(Jtv, ol Te av$pamot ttoAAoi travraxov 



504 



NOTES s CHAPTER XV 



iwpcovro, ju-dAicrra hk ij dyopd i'rr^TrX'^ptoro' to juev yap fxiaov avTjjs 6 StJ/xos Aeu^ajtovcov /cat 
Ba<f>v7}<j>Qpcbv Kara T&A77 et^e, ^d 8* dAAa ot arpanwrai Xaprnporara oVAttrjueVot, wcttc /cat T(i 
oirAa avrtov /cat ra oij^Ta dorpaTTTetv, ot Te /cepajuot /cat avrol iravrcov r€>v t^Ss olKohop/T}}xdrwv 

iKpVTTTQVTQ VTTO TCOV dva/fe/fy/COTCOV, TQVTC0V 8' OVTOJS €/C W/CTOS 7TpQlTapa(7KGVaU$4vTWV iafjXBev 

is Trp> dyopdv 6 N4pcov dfia rij 17/xepa, ttjv iaS'fJTa ttjv emvt/aov evSeSti/ctus, ovv tc T77 fiovXi} /cat 
qvv tqis Sopu<j&dpotsv Kal em Te to jS^jLta dve^ij koX em Btypov dp^i/coi? e/ca#e£eTO /cat /Aerd TOVTO 
o Te TipiScmjs Kal ot ju.eT* auTou Std Te orot^cov ootAztcov e/caTeptoa'ev 7rapaTeTayp,eVcov BiijABov 
/cat Trpos t<3 fiypan TTjoocrcTTavres' TrpoaeKVvyaav avTov, coof^rep /cat irporepov ... o §e iVeptuv 
^jueit^aTO avTOV eBSe* 'dAA* e£ Tot hroitjuas avros Seupo eAflcuv, tva /cat /cat iraptbv irapovros 
p,ov d7roXavu7)s m a yap aot ovVe d ttglttjp /caTeAwrev ovVe ot dSeA<£ot SdvTes irijpTjvav, ravra iyay 
Xapt£ojuat /cat jSamAea ttjs 1 > App.m>ias ttoiw 3 tva /cat av /cat e'/cetvot fxdBwaiv on /cat d$atpetcr#ai 
jSacriAetas' /cat 8a>peta0at Svvap-at. 'tovt' e«TaV dveA0etv Te avTov /caTa t^v dvoSov ttjv eV 
awcp Tovrai GfXTrpoaBsv top ^fxaros 7TG7ron]p,4vT)v e/ceAeuae, /cat KaBifyqBevn avra> vttq tov 
iroBa to StdS^jua eWfoj/ce, /foat Te /cat em rovrtp 77oAAat /cat TravToSamxt eyevovro. eyeveTO §e 
/caTa ipr}(j>iQ[Aa /cat mzvijyvpts dearpwrj, /cat Biarpov, ovx on 17 cr/ajvi) dAAa /cat 17 7rept$epeia 
auTou TraVa IvSo^ev iKGXpvocoro, /cat Ta"AAa oca ecnjei ^pvcrw e/ce/cdajUijTO* d^* ot? /cat t^v 
Tjfiipav avrijv ^pvcnjv €7rcovo/xacrav, Ta y€ juev 77apa77€Tao"ju.aTa rd Std tov depos SiaTa0£Fra, 
owais tov ^Atov aTrepv/cot, dAoupyd ^v, /cat eV jxictp avrcov dpfMa iXavvcov 6 N4pcov eWart/CTO, 
7TGpl£ 8e daripGS X9 vao ^ eireAa/*wov. 

Tavra p,h> ovrtos iy4vsro f /cat S^Aov oti /cat uvpmouim iroAuTeAsT ixp^joavro^ ,.. eV Se S17 toij 
aAAois: e/coAd/cevaev awov /cat virihpap.^ Set^OTaTa, /cat Std tovto 8copd T€ wavToSaTTa 'Trevra/ctcr- 
^tAtcov jxvpidtmv a^ta, cuj ^acrtv, eAajSe, /cat "Aprd^ara dvoiKoBofjurjaai eVreTpd-777/* ,,. ". 
(7/. below n. 8] (430, 3) 

7a [^fX» II, i, "••- *hpkii gkq^ tujdS ♦•♦ IfiupqwL, np jih^ hqkm^ mum qnpb 
^m^nLpkmu k mpni,phmu, fjSmiimfiij k fympquig ilpnj iSfinj nLpnLg ji ungiuh^ 
npg jwpgmjlu tympufjij 1Xp£Uifywj k ^wijwpmp^wlimj hi^popl unpm, qnp 
i/kpnj wjffljiu pwq.wuipkijnjt]* k fiu£ np jkm hnpm knkh prnqmuipg w^fump^liu 
ifkpnj fi unpfiu ukpdmh^ n p r l[ } j 1 £°pt mnukjni[ i£m£pnLpfiLUU* mumrnhh^mh 
wpyml^nLufi^ jUpywJjiuj* mLkjnpn.gh jwqg L [j pw^SnLpfiLU ui£kijkw]j>y h 
ilfiuiljh pum l^mpqjj £kmkjLwhmj ji pmrpiiuipriLpjidj ". (7/, I, ix, II, iii, xxviii, 
lxTiii, and tlae " Primary History ", Sebeos, p. 13. Also Toumanoff, Studies, pp. Ill 
n. 174, 197 n. 222, 220-221 nn. 257 and 262, 295 n. 75, 313 n. 41, 331.] 

8 Cass, Dio. y LXm, ii, v [L. VHI, 142/3], [" ... Kpavyrjs ts im tovtoj woAA^s 
Qvp,pdo7)s i^eirXdyj) ts o TtptSaT^s, Arat a$covos XP^ V0V Tlv ^ ^ Kai diroXov^^vos iyevsro. 
iTreiTa mcoTTTJs KrfpvxB^iu7]S GTreBappTjas Te, /cat e/cj8tao , dju.evos' to tppovr/jxa rta rk /catpw /cat ?$ 
^peta e'SovAevcre, ft^Sev (f>povriaas et Tt TaTretvov <f>B£y£(UTO, irpos ri}v eAmSa &v TevfotTo. 
etTre yap ovtos' *eyw, SeaTTOTa, '^pad/cov juev l/cyovoy, OvoAoyatcrov 8e /cat 77a/copot> tcov 
jSacrtAecuv dSeA^os 1 , cros 1 8e SovAds et/zt. /cat ^A^dv t€ Trpds" ere tov e/AOv 0edv, irpouKVvrjawv ere cos 
/cat tov MiBpav 3 Kal iaofiai touto o Tt dv crv iiriKXojuTjs' crv yap juot /cat juotpa e? /cat tv^tj ". 
C?/, above n. 7 for the episode of Tiridates' sword or dagger, which had been stipulated 
for in advance, according to Tacitus, Ann., XV, xsxi [L. IV, 262/3], " ... Vologesen 
Ecbatanis repperit, non incuriosum fratris: qnippe et propriis nuntiis a Corbulone 
petierat, ne qxiam imaginem servitii Tiridates perferret nen ferrum traderet ant complexu 
provincias optinentinm areeretur foribnsve eornm adsisteret, tantnsqne ei Romae 
qnantns considibns honor esset, Scilicet externae sxiperbiae sneto non inerat notita 
nostri, apnd qnos vis imperii valet, inania tramittnntxir ".] It is interesting to compare 
this episode with that of Mnsel Mamikonean as related by JSebeos, iii, p. 42. (432, 1) 



m 



NOTES : CHAPTER XV 



505 



9 Vologaeses corresponds to either Arm, ^wpq — ^u or Arm, ^wnjfn^ [Cf above 
n. 7a,] (432, 2) 

10 MX, I, Tiii, " ••• Xhpiml{ && W P£ W J yiwpupg •** PuiqmLJiph^nLgmh^ 
qkr^piujp fnp tj^uiqmp^ml^ *** li uw^Swhu {,uimwhi hSm • ♦♦ np^uiiji itying gn 
L giv^m.piii.1/ £winiulibh. r^fi uwCSwhg ^w^wy^ miff", qihh fiLpkmhg y npgwh 
CwmwhF uijhgwh nLbji ". [Cf. above n, 3], (432, 3) 

ioa [MX, II, iv-v.] 

iob [jjfX, II, i, cf. above chapter YII and nn. 37-37a and chapter XI n. 16.] 

ii Joseph., Bell Jud„ VII (iv), 244-251 [L. in, 574/5-576/7], 

*' To Se rcDv 'AAavcov Wvos on p,iv eicrt UkvBcu, .,, Kara rovrovs Se tous* XP^ V0VS Biavorf^ 
Bivrss els tt)v Mrfiiav koX TTpoacoripU) tclvttjs 'in KaB* dpiray^v ifx^aAeiv rq> jSaaiAe? rwv f Tp/ca- 
vo>v BiaAsyovrar rjjs napohov yap odros BeaiTor'qs ivrtv, fy 6 fiaoiAevs *AA4£avBpos TrvAais 
Gidnqpais KAeiarrjv iTTOt-qas. KaKzlvov ttjv eiaoBov avrois 7rapaux6vros dBpooi koX p/jjSev TTpoihroir- 
revoam rots Mrjbois hmreuovres xa>pav troAvdvBpamov Kal Travroiwv dvdfjiearov fioGKTjpidrwv 
Birjira^ov fMTjBsvds avrois roAfxcovros avBiaraaBai. ... p>erd ttqAAtjs ovv paarwvTjs d/za^et ttoiov- 
jievoi rds dpirayas pixP 1 r VS *App,Gvias TrpoTjABov Trdvra Ae^AaToiWes. Tip&drrjs S* avrijs 
ifiaoiAzvsv, 6s v-navnaoas avrois Kal Tronjcrdp.evos' p>dxt)V napd jxiKpov tJA0€v eV auras' (a>os 
dXwvai rrjs rrapard^ecos' fipoxov yap avrcp 7TGpifiaAa>v ns iroppcoBev 1/zeAAev i'TTiuirdueiv, ei 
p.7} ran gifei Barrov £k€ivos rov rovov xoiffas £$>Bt} 8ia</n>ye?v. ol Bk Kal Sid rrjv p>dyrp> en jLtaAAov 
dypioiBivres rrjv juev ^a>pav eAvp,vdvro, ttoAv Be 'TtAtjBos dvBpuiTrwv /cat rrjs oAAtjs Aeias dyovres 
e£ ajj.tf>polv rcov fiamAeicbv irdAiv els rrjv oiKGiav dveKop,ioB7)aav *\ (433, 1) 

12 Cass. Dio., LXXIII, iii [L. VIII, 140/1], " Kai eSet yap rto IJarpopiw rifji-jv riva Bid 
ravra ysvioBai, iro^vBsv 6 Tipihdr7)s aVa>0ev 4k rrjs zBpas B-qpia, Kal Bvo ye ravpovs fiia d/za 
floA?}, eiydrcpTTiarov, Sterpcoore Kal dW/cTetve ". Of. iJfX, II, lxxix, " ••• tyiuinut qliui- 
Cwwwlim.pkw'Uijh Spqiumiaj ••• ««/ qkplftiLij $lP L j$ i[uijpkbhmg iffwi^ &hnwfip 
l^in^kuij_ qjsij^hpi^y Pwifikiiiij ^mhqhp^ phij.i^qkmj^ ^wfu^wliijiwilp ", 

The comparison found in " Agafangehs " [xi, p, 100, " •♦• ^uiuh wjunpfilf iqium^— 
w£kijwh puihgu mju fi pmhu Jjinp^fi wmuljwi], pi t*pP^ L 1 U £$ Spi}tv^t np 
ujnj.uijpifh wLkpkivg qjpnLSpu qhuwij, h tjwdwgktjnjt} jiulf ji ujiqui^ fiLpmil 
yjnp&iuhu bni^nLij " probably refers to the first of the Armenian Arsacids, Pliny, 
NE, XXX, vi (16-17), [L. VIII, 288/9], says that Trdat, as a Magian, did not wish to 
Kome by sea, so as not to pollute water, " Magus „. Tiridates venerat Armeniacum de 
se triumphum adferens et ideo provinciis gravis, navigare nolnerat, qxioniam expuere 
in maria aliisqne mortalinm necessitatibxis violare naturam earn fas non putant, Magos 
secnm adduxerat, magicis etiam eenis enm initiaverat, non tamen, cum regniim ei daret, 
hanc ab eo artem accipere valnit ". The Armenian saying, which is undoubtedly taken 
from a legendary account, must also be an allusion to Trdat's journey, and should be 
taken in the sense that the king travelled over the sea as though it were over land, 
[Cf. above n, 7 for the passage of Cassius Dio where he notes that Trdat's journey took 
nine month and that the king rode on horseback all the way to Italy. Also Herzfeld, 
Archaeological History, pp. 63-67, for the possible influence of Trdat's journey on the 
story of the Magi.] (433, 2) 

12a [£ee above Chapter XIV n, 3.] 

12D [Pliny, N.H., V, xx (83) [L, II, 284/5], "[Euphrates] oritur in praefeetura Armenia© 
Maioris Garanitide, ... ". See also Chapter XIV n. 3.] 

1 3 Pliny, NH, x (27) [L. II, 356/7], " ... universae [Armeniae] magnitudinem Aufidius 
quinquagiens centena milia prodidit, Claudius Caesar longitudinem a Dascusa ad confi- 



506 



NOTES : CHAPTEK XV 



nium Caspii maris /xiii/ p., latitudinem dimidium eius a Tigranocerta ad Hiberiam ", 

(434, 1) 

14 Tac, Aim,, II, Ivi [L. II, 474/5], according to Tacitus king Zeno changed his name 
to Artaxias, attracted the sympathies of the " proceres and the plebs ", and was crowned 
at Artaxata " adprobantibus nobilibus ", [Of, ahoye n. 1 for the complete text]. Ibid,, 
XII, xliv [L. Ill, 378/9], " „. Erat Pharasmani filins nomine Radamistus, decora proceri- 
tate, vi corporis insignis et patrias artes edoctxis, elaraque inter aceolas fama. ... 
primorum Armeniorum ad res novas inlicit ... ". Ibid,, XV, xxvii [L. IV, 258/9], [Cor- 
bulo] consilio terrorem adicere, et megistanes Armenios .,, pellit sedibus, castella eorum 
exeindit, ... '\ Ibid,, XV, i [L. IV, 216/7], the Parthian nobles were likewise called 
" primores gentium ", and Ibid,, II, lviii [L. II, 476/7], " proceres gentium ", See also. 
Ibid,, II, ii ; VI, xxxi ; VI, xlii, etc. [L. II, 384/5 ; III, 206/7 ; 226/7-228/9]. (435, 1) 

14a [# ee aoove Chapter XIV.] 

15 Manu-h, Arm. 8wiim.li < Sw$ip y Rus. Meni^-miH, Lat. min-or, MX [II, Ixv], 
" ZwwnLiub qhiuykw^ ^wpq^u SwhnLljh [i $m£iuzj qjuuimib, q*jpwuwji 
qkmm[, kl^hmi humhw^ Jppi£ FLP 111 !"* ^Jip^^t 1 } & an L W 'P WL * tpfiwuwq qhuin^ 
l{nh^ fynijikjL qgnuih kpnuuhquij wpgwjji »♦♦ qppnj *ywyp& fyfo uinha^ 
^lupq^uf}^ ♦ ♦♦ ", The word manulc is used here in the sense of " princeling ". [Of, 
Hubschmann, Grammaiih, p. 472, and above Chapter XIV, pp. 311, nn. 22a, 24a, 
26 etc.] (435, 2) 

isa [Of. Toumanoff, Studies, p. 115 n. 186,] 

™ [See above Chapter XIV n. 3b,] 

15e [See Eremyan, Slavery, pp.21, 25; Sukiasian, Armenia, pp. 120-122 and below 
nn. 89-89a,] 

15d The entire section on feudalism should be reviewed in the light of the more recent 
studies of this institution, which have dated some of Adontz's concepts, though he 
reviewed and developed them himself in his later work, Aspect, The interpretation 
of feudalism and the understanding of the term have developed extensively since the 
beginning of the century, and the problems have been complicated by the disagreement 
of Western and Marxist scholars on some of the fundamental aspects. The most, recent 
studies of Armenian " feudalism " are to be found in the various works of Manandian, 
particularly his Feudalism, and in Kherumian, Feodaliie. The application of the 
Marxist thesis to Armenia is given in Sukiasian, Armenia, The most important work 
to be consulted is of necessity Toumanoff, Studies, in which he elaborates the crucial 
distinction between Armenian " feudalism and dynasticism " (p. 110). Without this 
distinction, the question of the correctness with which the Armenian na-^arar system 
may be called "feudal " in the Western sense cannot receive a valid answer. On the 
question of the extension of the term " feudalism " to various institutions, see also 
Coulbora, ed. Feudalism, which is, however, unfortunately poor for Armenia. Specific 
aspects will be dealt with farther in the relevant notes. [Of* above Chapter XIII for 
the comparison with Iranian 1 " feudal " institutions, also below n. 44c] 

16 The first scholar to raise the question of Arsacid feudalism was Edward Gibbon, 
but the formulation of the problem belongs to St. Martin, Discours, p. 293, 

St. Martin goes on to assert that if Europe did not invent feudalism, neither should 
this be held as a creation of the Parthians; " Qu'est-ce que le gouvernement feudal, 
c'est tout simplement l'oecupation militaire d'un vaste territoire, partage entre tous 
les soldats; les rangs y sont distribue comme les grades dans une armee ; c'est la eonse- 



NOTES : CHAPTER XV 



507 



quenee inevitable d'un gouvernement militaire on d'une eonqu§te, Les Arsaeides ne 
furent pas les inventeurs de ce mode de gouvernement, puisqu'ils ne furent pas les 
premiers conquerants de l'Asie ". Ibid., p, 296. Ifrom this definition it, is evident 
that the distinguished scholar belonged to the German school and suseribed to its point 
of view, which is now considered to be ont of date, [See the preceding note and the 
Bibliographical note on Feudalism.] (436, 1) 

16a [See above Chapter XTV, also Eht^eham, Iran, pp. 110 sqq., andTonmanoff, Studies, 

i.J 

17 Xen., Gyr&p., VIII, vi, 4-6 [L, II, 410/1], ** Bqkgi Bi pot. Kal rmv hBdBs. p,&>6vrwv 
vp,wv, ots av iyd> rrpdy para wapixco iripm&v irpd^ovrds n htl ravra ra £Bvt), xwpo.s 
ysviuBai Kal oikovs e/ca, oitws BaGfxoij>op7Jral ts avrots BsvpO; orav re itomv e/ceiae, as 
ot/ceia e^axrt KardyeaBai. 

Tavra €?7re koI I8awe iroXXots t&v <j>(Xwv Kara <7rdoas ras Karaarpafeiaas tfoXgis oikovs 
Kal wttjkoovs* Kal vvv Giaiv It* rots diroyovois rwv totc Xafiovrcov al x&p&t Kara\xivovuai a/\Acu 
£v aXXj) yfj' avrol Be otKOvcn ^rapd'jSacrtAe? ". : (438, 1) 

w Ibid., VIII, i, 9-10 [L. H, 308/9-310/1], " Kvpos 8* im p& r$XAa KaBlvnj SXXovs 
imjxsk'qTds, Kal rjaav avrtp Kal irpovoBcov o.'TtoBskttjpzs Kal BaiTavT)p,drwv Borrjpes koI %pywv 
iiriordrai Kal KTTjpdrcov cj>vXaKZS Kal raw sis rijv Biatrav imTrjBsiwv smjueA^rai* koI farmav 
Se Kal kvvwv impsXaqras KaBianj ovs ivop,il$ Kal Tavra ra Pocxic/jfiara fiiXrwr av irapexsw 
avra> xP*) a ® ai "• ••• " * a * Ta&dpxovs 8e Kal irefwv Kal famicbv iyiyvcovKGv e/c tovtcuv /caracr- 
rariov eTvai", Ibid,, VIII, viii, 20 [L, H, 450/1], " vvv Se rovs T€ Bvpaipovs Kal rovs 
uiroTTOiovs Kal rovs qi/jottqiovs Kal olvoxdovs Kal Xovrpoxoovs Kal irapanBivTas, Kal 
dvaipovvras koi KaraKoip,i^,ovras Kal avivrdvras, Kal rovs KoafMqrds, ... ". [Cf* Ehte- 
cham, L'lran ; Frye, Persia, pp. 90 sqq,] (438, 2) 

19 Xen,, Gyrop., VIII, vi, 10 [L, II, 415/6], " ,„ 077000* 8* avyijvKal dpxsia Xdpcomv... " 
dpx&a has the litteral sense of " adniinistrative institution '-, but it is used here in the 
sense of " power, rule ", [Miller, L. H, p. 416, gives the translation ■" lands and palaces ", 

(439,1) 
i»a [Eht6cham, LUran, pp. 114-116, also above Chapter XIII nn,14c-d.] 
ao Tac, Ann., XV, ii [L. IV, 218/9], see above n, 4 for the text. (439, 2) 

31 Justin,, XLI, ii, see above Chapter XII n, 19 for the text, (439, 3) 

aa Theoph, Sim., Ill, xviii, 7-10, " BtJjaois yap irapa rots MtjSois eiTrd twv -jrpdfecuv ra 
avxivod re Kal rifuwrara Biawop,4vois, v6pq> irpsofivrT) kXtjpoBotoviagvqiS; P#} aXXws gx&v 
ra TTpdyfiara I^acr/cc^, Kal </>am rov p,h> ^ApcfaKiBfjv eiriAeyoftevov Bfjpov rijv fiamXziav 
Karixsw, Kal rovrov iinTiB^aQai to) ftamAet to StaS^a, STspov ttjs ttqXgimktjs TTpozordvai 
uvvrd^cos, aAAov Bk ras TroXiriKas nspiKstoBai <j>pQvriBas, rov 8* irepov ras Bia(f>opas 
BiaXvGiv ra>v 7TGpi ri Karaarama^ovrcov Kal Biairrjrov BeofjuevcDv, rov Bs 'nipmrov TjysivBai 
rrjs ittttov, rov Be p,zra rovrov (j>opoXoystv ro vtt'qkoov Kal rwv fiamXiKcov rap,siw>v £<f>opov 
ir^vKivait rov 8* IjSSojuov KyBsfiova rwv onXcov Kal tjjs 'ttoXs^iktjs ioByfros zmGrarstv, 
Aapziov rov ^YurduTTOv rovrovl tov vofiov iv rots fiaaiXsiois ivrep,evtaavros ". 
Theophylakt traces the origin of the families invested with these duties to the period 
of Darius I, obviously under the influence of the tale of the seven satraps, The Arab 
sources likewise speak of the seven satrapal houses, Noldeke, Taban, p, 437. We 
even find the legend that Arsak, like Darius, had been placed on the throne by seven 
satraps. It is possible that certain functions were considered particularly honourable 
under the Arsacids, as they had been under the Achaemenids, so that there was a corres- 
ponding number of satrapal houses, but it is altogether unlikely that the Arsacid families 



508 



NOTES : CHAPTER XV 



should have a genetic link with the satraps of Darius I. Tabari links the Suren, Karin 
and Spandiyadh families, which are known to have been powerful Parthian houses, 
with the satraps of Darius I. According to MX, II, xxviii, Ixviii, the sister of Karin 
and Suren was called Kosma, and her descendents were named Aspahapetuni from the 
name of her husband (Cf. Sebeos, iii, p. 36, IXuinwpwinhtn = Procopius, Pers., I, 
xxiii, 6 ; I, si, 5 [L. 1, 210/1 ; 82/3], *AairGp4&i}s 9 Pers. aspabadh < aspa bed > asparapet). 
Kosma seems to bear the same relationship to the district of Ko^cnjvrj as the Karin 
to the district of Kdpiva. Even in the Arab period, the Karins lived in Nihavend 
near the ancient Karin, cf. JNoldeke, Tabari, p. 437. The powerful house of Mihran 
also belonged among the great Parthian families, Theoph. $w&., Ill, xviii, 10 considers 
it to be one of the Median (i.e. Parthian) families, " „. tov Bk Bapdfx rrjs tov Mippdvov 
olKapxtas ysvofxzvov, 8^/i.ou S' 'ApuaKihov, ... ". The name Vsnasp was current in 
this family: LP\ lxiv, p. 366, " 8%wtttt[£uuwui^ npqji Vqmmmmj [i Ifft^pmu 
mwhl*^ " he is considered to have been the qutjkwlinpjj.jt of Peroz, Ibid., lx, p. 346, 
and consequently his father JXyuiww was Peroz's tutor, but in ISMse, p, 197 the name 
of Peroz's tutor is given as Eaham, '* Ifpinukp npr^mju Qwql^kpwp Tj.mjhmljh^ 
fj*ut£iuij wli nth ji IffiCpmh mn^ih ". We know that the father of Vahram 
Choben was named Vahram Gusnasp, Noldeke, Tabari, p, 270; Theoph. Sim., V, 
xiii, 4, " ... Bapdp., vlov Bap[ap,y]ovavas. ". The son of the all-powerfull dignitary 
Mihr-Nerseh of the house of Mihran was named Mah-gusnasp, according to Tabari, 
p, 110, Vsnasp is the title of the holy fire whose temple was found at Ganjak-§iz of 
Atropatene, Ibid., p. 100, cf. Sebeos, xxvi, p, 92, " ••• ft %w utility ... ^putqfjuu 
Zpuiwhh ilkbh nptiLii ^l^hiuuiq l^n%kjjh ", Near Ganjak is found the district 
of Xu}pQ-p,iTpT)VT}, Ptot, VI, xxvi, which probably was the residence of the Mihranian 
house, It is possible that the Vsnasp was their family cult; fire was wor- 
shiped as the visible symbol of the great god Mithra or Mihr, from which 
the name Mihran was derived. Horses were sacrified to Mihr* just as cows 
were dedicated to Ahura-mazda, Xen., Gyrop., VIII, iii, 11 [L. II, 354/5] also VIII, in, 
24 [L, II, 360/1], (t ... iBvaav t3> Ad kcli coXoKavryaav rovs ravpovs* e^retra tG> 'HAitp kcli 
ihXoKavrrjuav rovs lttttovs* ... ". [Cf. Xen., Anab., IV, v, 35 [L, II, 56/7], see above 
Chapter XIV n, 2 for the text]. Eor this reason the god was called Mi Bra or A Bra 
vrsnaspa, [i.e. the Mihr, or Eire, demanding good horses (< wsni, Pers. gusn, " male " 
and aspa, " horse "), The Moon was worshipped as well as the Sun, but were the Surens 
her devotees ? Cf. Suraena and i7eAijv^. 2ik is also a family name of unquestionable 
Parthian origin. The Persian ambassador to Justinian in 562 was named 'UaB^yovavd^ 
and had the title of Zix> Men. Proi., p. 346, "... ijcnipm^Tai avroBi ko.1 IJepvcov 
7Tpecr/?ei>T^s, & Bijra vtttjpxg ftev aiicopa to Zlx, piyicrrov n rovro irapa rots IHpaais ydpas, 
TTpotnyyopia Se avrov VecrSeyovcTva^. o£tos Se TrapswaoTrjp tov kclt clvtov f$aaiA4co$ vTrr/pxe **. 
Historians took the names Surena and Mihran to be similar titles: Amm. Marc, XXIV, 
ii, 4 [L. II, 410/1], " Surena post regem apud Persas promeritae dignitatis .„ "; Zosim., 
Ill, xv, 5, " o yap aovprjvas (dpx'rjs Be tovto irapa Ilipuais ovop,a) ", Proc, Pers., I, 
xiii, 16 [L, I, 106/7], " GTpaTrjyos 8e els diramv e^eicr-r^/cet, IUpcrqs dvrjp, fjuppdvrjs ftev 
to dgtwfJLa (ovtco yap tt^v dpx^v KoXovm Ilipuai), iJepd^s 8c ovop,a ", JPB, IV, lv, 
" Hiqiu wpawljhwrj wpgwjh tywpufiij £Tu/i^i7i.^ ft t[hpuJj w^fuwp^fih Zwjntj 
hplpnui nSmhu jfoju tubing uthtnp jiLpnzj, djinufii §^f utumu h ifftnufit 1jwp£u ", 
When the king of Persia appointed the Zik as tutor to the Armenian king Xosrov III, 
he probably had in mind the relationship of the regent to the young king, Ibid., VI, i, 



NOTES : CHAPTBE XV 



509 



" •♦• qhmqjtu mn pwqwLnph tf)ujpu[jij *** h jvhqpkqfjh fi hSmhl pmqwuip 
wp^wlj^nihji ♦ •♦ IXiqm qmwh£p fi hdjih mn£i)l Smhrnlj i/Jj rnhndi }\}mipnq\ h. 
Ijimuhw^ pwq. j] qtni-fij hnpm h km hfiw Ifjiu qgnjp jup ♦ «♦ k Sfilf umfipmlj 
qwumjuupinlj km wpgwjjiu Jjnupni^mj ". [From the text the relationship seems 
to he rather between the two rulers]. If we find that there really were seven great 
families among the Parthians, we should perhaps add the house of the ^Ti^orakan, 
from Deh-iJNixorakan on the east shore of Lake Unniah, to the Suren, Karin, Span- 
diyadh, Aspabadh-Kosma, Mihran, and Zik. [Of, Ohrislensen, pp. 103-110; Frye, 
Persia, pp. 184 sqq.; Toumanoff, Studies, pp. 40 n, 14, 83-84 and n. 105, 103 n. 159, 
149, 187 sqq., 206, 208 n^236, 218, 225 n. 270, 253, 317 and n. 58a, 325-326 and n, 91, 
335, 473 sqq., etc, and below n. 36]. (440, 1) 

22a [Tonmanoff, Studies, pp. 119, 267.] 

22b [Ibid., pp. 40 n. 14, 111 n. 176, etc.] 

22c [On the Hadjiabadh inscription, see, Ohrislensen, p. 52, and 100, n. 1 ; Herzfeld, 
Paihuli, I ; Nyberg, Eaffiabad, The date of Sahpnhr I's accession is given as 240 hj 
Frye, Persia, p, 283 and as 241 by Sprengling, Third Century, p, 2 and Christensen, p. 179, 
who also sets the date of the coronation as 242, Ibid., pp. 180-181, this date has been 
controversial, see If rye, Persia, p. 200, The king's death is dated 271 by Sprengling, 
Third Century, p, 2, and 272 by Ohrislensen, p. 226 and 'Frye, Persia, p. 283, with some 
hesitation.] 

23 WZKM, VI, p, 72 [Of Herzfeld, Paikuli, I, pp. 87-89. Nyberg, Hajjmbad, pp. 67- 
68, gives the following transcriptions: 1. sadriBaran u vispuhrdn u vazurlcan u azatan. 
2. x.sa0n8&m& vispuhrdn vazurhan u azatan,'] (442, 1) 

24 Noldeke, Tabari, p. 71 and note 1. [Of. Ohrislensen, pp. 98-113 and notes, Frye, 
Persia, pp. 206-208 see below above Chapter XIII n, a.] (442, 2) 

25 TSToldeke, Tabari, p. 8 and n. The Arab authors give different interpretations 
of the four classes, evidently basing themselves on later conditions. [Of, preceding note], 

(442,3) 
25a [Of Ohrislensen, pp. 100 n, 1, 101-103; Toumanoff, Studies, p. 40 n. 14.] 

26 Leist, Graeco-ltalische Bechtsgeschichte, p. 123 n, c, (443, 1) 

27 Darmesteter, Etudes iraniennes, I, pp. 140 sqq. [See also next note] (443, 2) 
27a [On the vaspuhran, see also: Ohrislensen, pp. 103-110, Benveniste, Les Classes 

sociales, p. 131 ; Frye, Persia, pp.94, 206; Toumanoff, Studies, pp, 40 n, 14, 115 n. 186.] 
271j [On the vuzurgan, see also: Ohrislensen, pp. 110-111; Frye, Persia, pp. 184, 206; 

Toumanoff, Studies, p. 40 n. 14.] 

27c [On the dzadhdn, see also: Ohrislensen, pp. 111-113; IPrye, Pem&, pp, 111, 184-185, 

206; Toumanoff, Studies, p. 40 n, 14, 94 and nn. 137, 138, 95 n. 140, 124-127 and nn. 215- 

216, 220, 235, 238-239, etc.; Kherumian, Peodaliie, pp, 11-12, 19-21; Manandian, 

Feudalism, pp. 90 sqq. ; Sukiasian, Armenia, p. 106, et al,] 
27d [M.g„ PB, III, viii; IV, ii, etc.] 

28 See above Chapter X, pp, 185 and n, la. (444, 1) 
2sa [MX, n, iii, vii-vni.] 

29 Tac, Ann,, VI, xlii [L. Ill, 228/9], " Surena patrio more Tiridatem insigni regio 
evinxit ". On the Bagratids; PB, V, xliv, " ••• mmjp hSm qij.ni.ump wuiqkmjiu 
pwqpmmnLunj fj JJ"(kp rpummk, np Pwq.WL.npg Pmqwfymiugu jkwj^ £fiu Ji ph££ 
wqqfih pwqwLnpnLphwuu wp^wl^rnhnj ", This title was not known to the 
Achaemenids, and was apparently abolished after the Arsaeids, Amm. Marc, XXIV, 
ii, 4 [L, II, 410/1] mentions the high position of the Surena under the Sasanians, " .,. 



510 



NOTES ; CHAPTER XV 



potestatis seeundae post regem ..." but he does not mention their office of eoronants. 
[Cf. Toumanoff, Studies, pp. 112 n. 176, 132, 160, 162, 202, 325 n. 88, 326, 342, and 
above n. 36, also Christensen, p. 107, who cites without comment Theophylakt's claim 
that the Artabids were the hereditary eoronants under the Sasanians,] (445, 1) 

29a [On the office of sparapet = Erdn-spdhbadh, see: Christensen, pp, 99, 107 n. 3, 
109, 130-132, 263, 265, 370, 520 sqq,, for the Sasanian office; Manandian, Feudalism, 
73-75, etc., and Toumanoff, Studies, pp, 97 n. 144, 112 n, 176, 132, 141 n. 253, 160, 162, 
201 n. 228, 209-211 and n. 238, 325-326 and n, 89, for the Armenian office. See also 
above n, 22, and Chapter X, n, 70.] 

S9tl [See above n. 22 for the text of Theophylakt, Eor the hazarapet = hazdrapati = 
vuzurg-framadhdr, see: Christensen, pp, 113-116, 319, 395, for the Sasanian office; 
Manandian, Feudalism, pp, 69-73, etc. Toumanoff, Studies, pp. 112 n, 176, 205-206 
and n. 238, for the Armenian office. Also below nn. 29e-33. 

29c ]?& jy, ij ? « fc L ulffjqpu ij.npbu]ljU]jnLpkwhh {uiqinpuiuibinnLpbiuhh wsjvwpC— 
win bub fuhmSmljmjrnLphmhh^ w^juuipCw^h w^jump^mmmb qb^lfiuJLwLpbwh, 
^hhml^mhw^h wqTj.h ^hniJibwij* {,mqwpmiqbmh mSbhuijh bpljpjih : \jnjhiqlu 
uuipimnhjUjinnLphiulj uiqmpmtqbmnLpbmhh qopmiluipnipbiuhh fiwpwji l^nmJsjpj^ 
qnpbnj pninqrijilj £inljmmu]SnLq hjitjuil^jih jipiutfuunh wquihuiqpo^ wpbrn.}; — 
li^whg, ljwpdhm^tuhji'^ mtikpl^jiLii^ ^lu^wufipin^ bw^winwlfg gw^inhiiLlig^ 
pmpbh^iuhu pwpbCwdpuiLg pwpbtf.npbg jui^nquilfg fi qnpb upuwbpiuiijiuirj ji 
pnhh ft hwfuhkiuijh Ijwpquiij fi IfmSpl^nhbwh mqtj.ph, ft ^bpinj jr^JvwhnLpbwhh 
pm^whrpulf ji *[bpwj wilbhuijh qopwg qppwi]wpm.pbwhh pwq^SnLpbmhh Z m j n g 
ribbing, ^wlnvuimqjwribhuijh duiS jmqPnfi wqqfih iqwpqkbin^jbplihjig putpfjuthmh 
jipwgw^ fi ilbbjj hwCwuibmni.pbwh upuimbpmipfjp ", [Cf, nn. 29b, 31-33] 

8 ° Xen., Cyrop., VIII, a, 14-15 [L. II, 312/3], " Ovtw St? ukoitwv, ottos av rd rs 
oIkovoimko, KaX&s ix 01 Ka ^ V ^X°^V yivoiro, KarzvoTjcri ttws tt)v QrpariwriKrjv cruvra^w. ... 
wottzp o$v ravr lx et * ovrco xal 6 Kvpos uvv^K^aXaiwaaro ras oikovo^ikcls irpd^is' ... *'. 
It is interesting to observe that the New Testament uses the term ^mqmpmiqbm to 
translate the Greek oI K ov6}jlos, I Cor,, iv, 1, " ji ribb ^mqmpmiqbm " = " h rois 
oiKovofMois ". The same term is sometimes used to translate ^iAmp^os, Marie, VI, 21, 
" phppfiu mmjp ibpni^qlu • •• itujjuwpiupwg fit-png k £wqiupwiqbmujij ki/bbuiribbu/n 
£puiuiuujuip *\ " Hp(x>Bi]s ,., Bgiwpqv iTrolrjvsv rots fieyiarduiv avrov Ktxl rots ^tAiap^oiy 
/cat rots TTpcorois ... ". [Cf, above n. 29b], also Toumanoff, Studies, p. 206 n. 234]. (446, 1) 

31 Noldeke, Tabari, p. Ill, fmmadhdr = CpwSwwwp in Elise, ii, p. 24, FB, III,xiv, 
to which he attributes the meaning of " Vorankommer ". In Tabari, the title of Mihr 
Nerseh is given as hazdrabanda, Ibid,, p. 76, which Noldeke connects with the definition 
of Theod,, EH, V, xxxix, " ZovpTJvrjv xiAtW owcercDv Seo-n-d-np ... ". Since the Ar- 
menians refer to the Suren as Hazarapet, we should perhaps read Jtj < \ \& pro J^i \ I \& , 
and the indication of Theodoret probably results from the incorrect interpretation of 
the same word. [The confusion seems to stem from Lazar P'arpeci. Elise distinguishes 
between the Armenian hazarapet dismissed by the Persians, " Cuiqmpwiqbwh £p 
m^juwpCjih, frppk 1 - H^ m JP i[bpuiIfwt]nL ^m^ujpbmj_ £p iu}fvwp£wl[iuuujn 
gpfiuwnhlliii **♦ £whbiu^ qput fi q.npbnjb x ijinjvujhiulj hnpiu iqwpuji^ tub 
jw^juwpCh ", p, 23, and the hramatar and great hazarapet of Erdn, Mihr-Nerseh, who 
styles himself, " IfhZDjbPUbZ "IkDhP^ CpwSwwiup bpmh k W>kpwb", 
p, 24, and is addressed by the Armenian bishops as " •*» IfjiCphbpubCfi ribbfi 
{uiqiupwiqbwji \Xpbmg k IXhwpbiag", p. 28; there is no mention of the Suren 
in Elise. ZP\ on the other hand, gives the title of hazarapet to both the Suren and 



NOTES : CHAPTER XV 



511 



Mihr-Nerseh, whom we know to have been a member of the Spandiyadh house, " ,,. 
\JnLplhh mw^iwL lp fj duiSwhmljph jwjluljj^ ^mampmmhm fj qprnh wpgniSifi ♦ ♦ ♦ ". 
xiv, p. 70, and " \Fji£,phkpuk£ Cfliqmpmmhmii JXpkwij ♦ *♦ ", Ibid,, xx, p, 116. 
On the functions of the great Iranian houses, see above n, 29b, and the next notes. Manan- 
dian, Feudalism, pp. 28, 70], (446, 2) 

32 Noldeke, Tabari, p, 110. (446, 3) 

33 An earlier hazarapet was Valars, prince of Anjit 1 , FB, III, xii, " qni^iuphm^ 
phn hdm pmawmph Sppwfo ••• ^^^ /^wqiupimqkwfi fj ^wq^ujpiinqkmnLpkiuIift 
[litlf fj wn^fif; wSkhiujb Z m J n $ tfkbwtj, tj^wqwp^ np \p fefuuth Hh&miuj 
The last incumbents of this office were Afawan and Vahan Amatuni, Koriwn, U, 1, 
p, 11, " \Jfw£wntf\ l{wghw^ jmp^mhml^mh qJiLiuhJih* j^jihk^uwuwiiJLnp iupgwj- 
mmntp ^putSiuhwghy ma £wqwpwwhumLpkwSp$i wyfump^jju Ztujnn JXnutiwhrnj 
nLpnufu ", Ibid., XIX, 4, p. 61, " ^w^iua mhnLli fyn^fthy jwtjtfj-h JX^mmniuhm^^ 
np ip ^mqmpmhm Zwjng XThbuig^** ". After them Persian officials were appointed, 
Flise, p, 42 and LP', lxvi, p. 383 [Gf. However, Akinian, Koriwn, p. 73, who is of the 
opinion that Afawan should be identified with the 2ik, and presumably, therefore, 
be a Persian, though see on this point Ghrisiensen, p. 105 n, 3. Also, Toumanoff, Studies, 
pp. 171 n, 88 and 205 and n. 234, in whose opinion the office of hazarapet passed in 363 
to the Gnuni, " „, succeeding in this the House of Anzitene *\ See also, above nn. 29b- 
32. Manandian, Feudalism, pp. 281-282] 

34 FB, HI, xiv. In the Qahnamah, " ftwuhhnj nmmmutph " may perhaps be a 
reference to episcopal authority. [On the administration of justice under the Sasanians 
and its relation to the clergy, see, Ghristensen, pp. 299 sqq,, cf. pp, 116 sqq. On the 
situation in Armenia, see Manandian, Feudalism, pp. 76-78, 140, 282, and Sukiasian, 
Armenia, pp. 224-225, 238-248, also next note.] (447,1) 

35 Mlise, p. 23, after the abolition of the Armenian kingship, the Persians began to 
appoint, " ilhm liu Snqiqkm^ nutwiump wyfuutpCphy qji qkl{hnhtjuijli ijiwrdw 
wnwLiunkutjhh ", Ndldeke, Talari, p. 110. Under the Sasanians the chief JBerbadh = 
aeBrapaiii, who was almost the equal in rank of the mobadMn mobadh — magupati, 
supervised " die Kirche und die Rechtgelehrsamkeit ". \Gf. Ghrisiensen, pp. 116 sqq., 
particularly pp, 119-120, 519 sqq,, and the preceding note.] (447,2) 

36 In the Iranian Fpic,. Spandiyadh, the son of Vistasp, is famous for his victories 
over many nations. . To him are attributed the taking of the Gates of the Alans and 
the fortress of the Iron City, Markwarb, Slreifzuge, p, 116. He is undoubtedly the 
'* £iu$ <^>& Uwutbnjiimn <iuj>, *{pp$h wuhi i pwppwpnugh hpt Zwuhwi 
uw iqimnkpuia^SutL iljihik guiju t[tvjp tjtjkw^ ahftqmli jiLp fj qkmhfi : *\ Sebeos, 
ii, p. 30. Mov. Kalanh., II, xl identifies him with a god named T'angri ytm worshipped 
by the Huns " Using horses as burnt offerings ", *' £u1fwjji nLifkSh i[uijpiuq[i 
wli^nnhmj &jiw[unpml qnty.Lg- iqmymoh fiwmnLijujblfih* Ifwpqmjpii[ npm 
Pwbnpli jvwb wumnLiub, nnp tywpujilig JXuu^mlinlmm tyn^hh ♦ ♦ ♦ ". p, 273 = Dowsett, 

Mov. Dasx., p. 156, " IXmqwhn.t [wwf]mj imlhiB i[uJJp £u!jiiij[ih ^jini^g mjph^hjn^ 
ijpi£ Swwm.tjwhlg * * * ", Mov. Kalanh., II, xl, p. 278 =s Dowsett, Mov. basx*, pp. 158- 
159. The victims offered to Spandiyadh link him with the cult of the iFire a dr-vinasp 
[Adhur-GusJmasp] in Atropatene which was the family cult of the Mihranids [Cf. above 
n. 22, Ghrisiensen, pp. n45-146 andn. 3, 158-159, 166-167; Duchesne- Guillemin,.i?e%^o^, 
pp, 87 sqq., 99 sqq.], Spandiyadh was of Arsacid origin, and Sebeos refers to him in 
connexion with the raid of the Arsacid Vahram Ghoben [ii, p. 30, cf. Macler, Sebeos, 



512 



NOTES : CHAPTER XV 



p, 11 n, 5, Chrisiensen, pp, 105, 443-444], Might the horse sacrifices he a symbol of 
the office of the Spandiyadhs as commanders of the eavalry ? The name in its older 
form, Spandarat, spanda-data, occurs in the Kamsarakan house; are we entitled to 
conclude on this basis that this family was related through blood and office with the 
Spandiyadhs ? [On the office of Aspet and the Spandiyadh house, see above Chapter 
XIV, pp. 311 and notes, and Chapter XV n. 22, Also, Christensen, pp, 103, 107- 
109 and notes; Manandian, Feudalism, pp. 61-64; Toumanoff, Studies, pp. 220-221, 
and nn, 259, 264, 324-326 and notes.] (447, 3) 

36a [Qj % above n, 22. On the office of the Hayr- Mardpet, and its mistaken interpreta- 
tions and identifications, see above, Chapter XI, pp. 249-250 and notes ; Manandian, 
Feudalism, pp. 64-67; Toumanoff, Studies, pp. 131, 168-170 and n, 81,' 200, 248, 314- 
315, and next notes,] 

37 MX, II, vii, " \fW k ukp^fjufiu ^pmSwjl ji hnjh iunql fal^l* &• hui^wuihm 
hntjw nZwjp fejuuihu Swufih jlXinpupuinuj^iijIit llfih ^k ^f^ntwy A X^wju^wLwi : 
pwjg npii{£u I k m.p unpm ppg iw&i/^k mhjfr^ujmmJjk^^ ku n£ ^wkd ; ", 

(448, 1) 

38 FB, V, vi [The 1894 Venice edition erroneously gives the Mardpefs name as 
9»jp^ throughout,] (448, 2) 

3S > Ibid,, IV, xiv, [See above n. 36a.] (448, 3) 

40 Ibid., V, vii, " IXju qnpbwl^ujipLpjiLh [sc, ^wLmmwpftd qwh^m^ L 
pkpn.wrj\ k fiwpqwkujm.piji.ti, npm.il £wjph hndjjh hkpgjibhwtj qnpb j^^l \} 
phi dwdwhwlfwrj ", the edition has a mistake in punctuation, the correct reading 
should be, hkpgfchkwij qnpb fkuife and not £wjpb hkpgJj$jkwt}[SeeaboveTi,Z§Q>,~\(4£&,4:) 

41 The Military List gives correctly two of the duties of the Mardpet: " ^ m P 1 }' 
wkwwl^wh, np I hhpgfihji ww^wwwh fi jfhpwj Pwqm^Lnjh k qwh^mtj M [Of, 
Appendix III], Nerses, ii, p. 15, attributes the guard of the royal treasure to the senehapet, 
" \,kpuj;u ifrbip uhhkfjwwbw fi jflptuj wi/hhwjb qwh&mn hnpw*** ". In the 
same context, FB, IV, iii, says only that Nerses, as senehapet held the royal sword on a 
pearl studded belt, " JXjl ft duiSmbml^ph jwjhfijjl{ %p fj t[tupjJ q{jhmnpmphwii 
jj tpipbwliwjjiLphmh^ ufjpkjji ukhklfwrnkw wpgwjfjh llp^wlpuj, £wLwwwpfjii 
fi *[kpwj wfikuwjh Ipwpqwq l^khing p m qw tnpm.p hwhh ji hhpgnj h wpmwgnji 
♦ ♦♦ tywjp ft uwwuijl pwqwuipfjb fiow jj uhwpu wpgwjfih, nrnp^ntUmipuh 
untuhph qiqn^wmfil^j qnul^wwwhwhh £whqbp& wlpuhwl^ww Swpqwpwwqwpq 
l^wfiwpwLU fj utqwum Itnpw pwp&bui]^ m.hl;p, • •« ". (Nerses, iii, p. 19, has at 
the same place, " qwp^mhwljwh dwinwhjih pwp&hw^ mh£p ", which is obvi- 
ously a distortion of " nmp^ntiimliwh < umukph nul^l > wwmbwhh * ♦ * ", 

FB, [loc cit.~\. In view of the words of the Mardpet, FB, TV, xiv, "qinkqjni* wpgm- 
hfi ukhhwl] IjwqtJbijfi^ ♦«♦ ", it is possible that the Mardpet was connected 
with the royal ukhkwlj* Ubbblfwwbw may be the title of the official entrusted with 
the guard of the royal treasure, especially since FB, IV, iii takes the " qnpbwl^w^ 
riLpjiLU ubhbljwu^bw " in a broad sense to mean, " £,ujLwmmp^3 ft t[kpwj wilkhmjh 
lfu]pn.wij Ifhhwg pwij.wi.npm.pkujih fi hkp^nj h wpmwgnj ". [Cf. above nn. 36a-40.] 

(448, 5) 

42 ^^ y^ y^ Cf, also the request of the eunuch Drastamat, Ibid., V, vii, that he be 
allowed to see King Arsak, " **» jnLwhiu^ qif^pLlu hnpm k obmhk^ h wqriLijiuIiki 
hSui mwwtini£iuh '»• h uijjh". (449,1) 



NOTES : CHAPTER XV 



513 



43 Paustus is acquainted with several representatives of the Mardpet house; Under 
King Tiran he knows, " uijp lift uihop^h... np niSi£p quiwwjiiJi Jhbfi Smpqi^kmnL- 
pkwhb, uijp hbpgfjbfi* npnid Z m )p ^ n ?i^ *** "* -^> HI, xviii. The same 
personage is found in the train of St. ISTerses, on his journey to Caesarea, " ^Z^jp && 
jj^Juwbb SwpqiqhuinLphuih ", Ibid,, IV, iv, and is killed by Sawasp Areruni, Ibid., 
IV, xiv. Another Mardpet, entered the fortress of Artagers, after the emprisonment of 
King At sakH, and insulted his queen, " hljh kfini.w tj-utqp fj bkpu fj pkpqh Zwjp 
Swpqu^hui hhpgjihfj kp^hiuifkujij qtnfj^hh iJhdiuu[£u jjpph ijjini^ifp : ". Ibid; IV, Ixii 
He was killed by Musel Mamikonean, and "ft wktlji £wjpnLphuih Smpr^u^kmnLpkiahh 
Ijiupql i£< f h>qquil[ ndh mhnth^ npjiumLpu Hp^uilfinj piuquiLnpfib IjiuS Sppwhwj 
Cop hnpw quidm. iffm^ lP m l I 1 hnjh qnpb SwpifiqkinnLptuib •♦• ". Ibid.,Y, 
iii. This Glak was likewise executed under Pap for his attempt to shift to the Persian 
side, Ibid., V, vi ; and he is the " Zwjp Swpquihm " mentioned as Musel's collaborator 
after the death of St. Nerses, Ibid., V, xxxiii. Finally, in the account of Arsak IPs 
tragic death, we have a mention of still one more Mardpet, named Drastamat, Ibid., V, 
vii. [On this occasion, the Mardpet is identified as being " qji^juwhi qUJiqkq^ mmhh ".] 

One authentic historical figure is found among these personages, namely Glak, who 
is well known to Ammianus Marcellinus, according to whom Glak [Cylaces] was killed 
together with his collaborator Arrabanes (Paustus' Musel) at the instigation of the 
Persian king, Amm. Marc, XXVII, xii, 5-14 [L. Ill, 78/9-84/5], " [Sapor] Cylaci spadoni 
et -Ajrabanni, quos olim susceperat perfugas, commisit Armeniam (horum alter ante 
gentis praefectus, magister alter fuisse dicebatur armorum) [5] ... Sapor ... Papam ut 
incuriosum sui per latentes nuntios increpabat, quod maiestatis regiae velamento, 
Cylaci serviret et Arrabanni, quos ille praeceps blanditiarum illecebris interfecit, capi- 
taque caesorum ad Saporem ut ei morigerus misit ". [14]. According to Paustus, 
Musel outlived Glak and plotted to shift to the imperial side together with the Zwjp 
Smpqu^km^ who is in fact one and the same Glak [FB, V, xxxiii]. According to Amm, 
Marc, XXVII, xii, 9 [L. Ill, 82/3], Cylax and Arrabannes had also'turaed to the Emperor 
Valens for help and so sealed their own doom " Qua humanitate Cyclaces et Arrabannes 
illecti, missis oratoribus ad Valentem, auxilium eundemque Papam sibi regem tribui 
poposcerunt ". Moreover, Ammianus says that the negociations with Queen P'afanjem, 
who was shut in the fortress of Artagers, were carried on by the same Cylaces and Arra- 
bannes, who were acting for the Persian king, but changed side during the negociations 
through pity of a defenceless woman, Amm. Marc, XXVII, xii, 5-6 [L. Ill, 78/9-80/1], 
" [Sapor] eisdemque mandarat, ut Artogerassam intentiore cura exscinderent, oppidum 
muris et viribus validum, quod thesauros, et uxorem cum filio tuebatur Arsacis. .., 
eunuchus Cylades, aptusque ad muliebria palpamenta, Arrabanne, ascito prope moenia 
ipsa, ... propere venit, et cum socio ad interiora susceptus ut postulavit, suadebat 
minaeiter defensoribus et reginae .... Multis post haec ultro citroque dictitatis, heiulan- 
teque muliere truces mariti fortunas, proditionis acerrimi compulsores in misericordiam 
flexi, mutavere consilium, ... ", Consequently, the Mardpet who insulted the Queen 
according to Paustus must be the same Cylaces- Glak. FB, V, iii, asserts that Glak 
had already held the office of Mardpet under Arsak II, or even under his father Tiran ; 
this statement is supported by Amm. Marc, XXVII, xii, 5 [L. Ill, 78/9] according to 
whom Cylaces and Arrabanes had fled from Armenia to Sapor, who had received them 
as refugees and entrusted Armenia to them. This fact leads us to identify Glak with 
the first mardpet mentioned by Paustus as living under King Tiran [FB, III, xviii], 



/ 



514 



NOTES ; CHAPTER XV 



Neither Sawasp Arcrmii nor Musel Mamikonean, in fact, killed a Mardpet, these episodes 
are purely didactic passages in Paustus' work intended as condemnations of crimes: 
the Mardpet falls for his massacre of the Arerunis, at the hand of Sawasp, the last survivor 
of this house ; whereas, Musel Mamikonean avenges the insult to the Armenian queen. 

(Paustus also says of Drastamat, that he had served under Tiran \J?B, V, vii] it may 
he that Drastamat had replaced Glak after the latter's flight to Persia, But Ammianus 
is not acquainted with this Drastamat, and in his account Arsak II dies at the hand of 
the executioner. Judging from his name; durusi-amat t( he who comes in time " 
[" schnell-gekommen ", Hubschmann, GrammaWk, p. 38], Drastamat is merely a ficti- 
tious character created to enhance the episode of Arsak's tragic end.) 

Paustus refers to the first two Mardpeis as Z m Jp> which can he taken as either a 
proper name or a title. Our identification of these officials with Glak resolves the 
problem in favour of the title. In connexion with this same identification, we doubt 
the characterization of the Mardpets as eunuchs. Glak himself, was indeed a eunuch, 
Ammianus calls him spado, [Amm, Marc, XXVII, xii, 5 ; L, III, 78/9], but with the 
division of this personage into four different figures, this incidental detail was transferred 
to the Mardpets in general. [The figures of Cylaces and Arrabannes remain enigmatic. 
Markwart suggested that they be identified with the Zlk and the Karin, but this identi- 
fication is rejected by Christensen, p. 105 n. 2,, see also Toumanoff, Studies, pp. 177-178 
and n. 118, and above nn. 36a-42], (449, 2) 

44 All attempts have remained unsuccessful so far: Inci5ean, Antiquities, II, p, 77 
compared hwjuwpwp with the Lat. prae-fectus and < Iiuifu — wpiiip* Emin, 
Mov. Xor., Supp., p, 298 accepted this explanation with a slight correction. Marr, 
Etymologie, suggested the etymologie hmju or hui{, connected with hw^iuL^ " pro- 
vince ", having the same origin as the Arab.-Iran. term nahie (A^S) "province", 
while the last syllable —pwp was an alteration oirad, Zend, ratu " lord ". The famous 
Iranist, Andreas, supposed that hwjvwpwp is related to Nohodares, the name of one 
of the Persian generals given by Ammianus Marcellinus [Amm. Marc, XIV, iii, 1-2, 
L. I, 24/5, etc], Hubschmann, Grammatik, pp. 514-515. The second half of the word 
unquestionably < raB, and not from ratu, the difficulty lies in the first syllable. Its 
explanation must take account of: hwfu " first ", hw^—wuiw^ huih — tultq, Pers* 
Ttayust, naMe, possibly X,Ji4np^uj^mh, etc. Historical considerations lead to the 
linking ofnaxarar with sa Brdar, consequently Marr's suggestion that there is a connexion 
between nah, nahr, and s&h, sahr, cf. bw£—wwwli and £iu£-—tuwiul{, deserves serious 
consideration. It is interesting to note that EUse, iii, p. 84, divides the nobles at the 
Persian court into " iftjjini^ k u^wmmwlimh hwjvwpuipg " which corresponds 
to burzugan and sahrd%ran. 

Since the etymology of fjwfumpwp has not yet moved from the realm of guesses and 
hypotheses, it is perhaps not superfluous to note that the mysterious word nah- may 
be a contraction of da-nhiu, with the loss of the initial da-, with nahoBar < da-nhudar, 
as nahapet < da-nhiipati. [After an additional half- century of scholarship, the problem 
of the term na-xarar, remains without a final solution. Adontz, himself considered the 
probem again in his Aspect, and some of the most distinguished specialists in the field 
have concerned themselves with it, see, Meillet, Mots parthes and Benveniste, Titres* 
Por a review of the scholarship and evidence, see Toumanoff, Studies, pp. 114 sqq. 
and n. 188. See also next note.] (451, 1) 



NOTES : CHAPTER XV 



515 



44a Adontz's recognition of the double aspect of the Armenian social system: feudal 
and dynastic has received considerable development and elaboration in ToumanofFs 
Studies, pp. 34-144, 154, 188, et passim,'] 

44t> [See preceeding note, particularly, Ibid., pp. 113-119 and n. 181, Also, Manandian, 
Feudalism, pp. 42-89 and Critical History, II, 1, pp. 314-334; Sukiasian, Armenia, 
pp. 97 sqq,, 184 sqq,, especially 193-195 et sqq., where he supports Adontz's conclusions 
against Manandian's, 226-227, etc.] 

440 [The thesis of the similarity between the Armenian naxarar system and West- 
european feudalism has found favour among Soviet scholars, cf, Manandian, Feudalism ; 
Sukiasian, Armenia, pp. 97, 191 sqq., et al. There are, however, important dissimilarities 
observed by Toumanoff, Studies, some are derived from the general survival of what 
Toumanoff has called " dynasticism " in the lt feudal " pattern see above n. 73, others 
deal with specific if related manifestations such as the apparent absence of the crucial 
act ofhommage as distinct from the oath of fealty, see, Toumanoff, Studies, p. 117 n. 191, 
but also pp. 40 n. 14, 144 n, 262, etc. The entire question of the comparison must, 
evidently, hinge ultimately on the crucial and debated definition of the term 
*' feudalism " and the legitimacy of applying it to different societies, see, Coulborn, 
Feudalism, Kosminski, Basic Problems, and below n, 45a,] 

44d [See preceding note. The proliferation of studies on both western and eastern 
" feudalism " in the last fifty years have of necessity rendered part of the discussion 
in this section out of date and in need of serious revision. In addition to works concerned 
directly with Armenian (S feudalism " for which see above nn. 44a-b, the Bibliographical 
Note should be consulted for the problem in general.] 

45 Luchaire, Manuel, pp. 235-236. [See preceding note.] (454, 1) 
45a [The definition of "feudalism" as "an inevitable stage of development " has, 

of necessity, commended itself to Marxist scholars, while finding less appreciation 
in the West. See, Coulborn, Feudalism; Kosminski, Basic Problems; IXth Congres, 
I, 417-471 ; BSJB, I, et at. The problem was set with his usual felicity by Bloeh, Societe 
feodale, II, 241 sqq., who incidentally noted its appearance in the XVIIIth century, 
almost simultaneously with the term " feudal " itself, " Aux yeux de Montesquieu, 
Tetablissement des 'lois feodales' en Europe etait un phenomene unique en son genre, 
'un ev&aement arrive une fois dans le monde et qui n'arrivera peut-§tre jamais'. Moins 
rompu, sans doute, a la precision des definitions juridiques, mais curieux d'horizons 
plus larges, Voltaire protesta : 'La feodalite n'est point un tenement; c'est une forme 
tres ancienne qui sxibsiste dans les trois quarts de notre hemisphere, avec des adminis- 
trations diff&rentes' ", Cf, Ibid., I, pp. 1 sqq.] 

46 Luchaire, Manuel, p. 155. 
4 ? Ibid,, p. 156. 
47 a [See above n. 44a.] 
48 Blise, hi, p. 85, " ... kp^p^, iftl ^mjphhfi^ hfit 

bpt tpuwliwqfjhg) k Ctubkwj^ riLpnLg jigj;, ^piuSmjk^w^ t^ji 
next note], 

48a [Cf, Manandian, Feudalism, pp. 122-123, 310 sqq.; Toumanoff, Studies, pp. 118- 
119.] 

48D [cy, Manandian, Feudalism, pp. 24 sqq,, 32 sqq. ; Toumanoff, Shidies, pp, 112 sqq. 
See above n. 27d.] 

48c [See above nn. 44a-d, 47d-48a.] 



(455, 1) 
(456, 1) 

tmupt^hml^wli^ k 
qwp&rjjih : *\ [See 
(457, 1) 



516 



NOTES : CHAPTER XV 



49 Waitz, VerfassungsgescMcMe, IV, p, 273, " Das Verhaltniss zum Herrn als ein 
Diener (service), Dienst (servitium) bezeichnet ; der Wassail heisst -urn des Willen audi 
wohl selber Diener (famulus, puer) ". FB, III, ix, " Qmjunt. dmSmumljmL mmuinwii~ 
pbmtj juipgwjih Zwjnif iffi fi bmnmjfin unpm ifkb fi^fumh IXq&ukmij, np rnhnLiuhbiu^ 
l[n^tp pqbm^fuh. np £p dfi fi ^npjt^^ qm^bpig pmphbptg mm£mpfiu mp^nLufi : • ♦ ♦ 
IXium umuigip pmLj.mi.npu Z m jn$ a ,p m P^ J ^l t P n p UiJlLU J u f* L P<> q^^fumuu ^npqnmjn 
t^Qnuu, h t^fi^fumuu ilkbft tymfimij qJTwp^ • •♦ fcz. frpP^ W JL n £ n t fy L ^ UUJ S^ m i 
jwqi^Iii km lupgwj yjuq^filjhfi l{ unL Pkmu uftphiytju mpgmjfi ^rnqfilibl^wj 
ufuStuij^ k duwluu JXh&ubmt]. k mpmp qpm pqbm^fu k mmjmnmm mmu unprnu : * * • 
l[wjp pnkmvfuu ^mnjiuuil^ fi bmnmjnL.pfiLU mp^mjfiu £,mummmLi ". Ibid., V, 
xxxviii, " ***^/' f} ^m aknu imnjrjbh h ohm mw^mfitfhu, k udm iffimSmntpbrnSp 
bmnmjhunhhy k mmjgbu ufim qmyfump^u Zwjng * ♦** A m£p Ljmbmj^ fupbmun 
mupgmju tflmpufirj, k l[nijfih fi bmnmjnL.pbmu unpin ". LP\ xxv, p. 140, 
" * * *Pt ^% ^fr^P 1 } P nL P^ I 1 bnnwjfiij tujqmfiufi Biuhplifiui £mdmp&mlinL,pbmLip 
qpb^ mn fiLpbmLg mj;pu". Ibid., xxvi, p. 148, " ***qifmumml^ k uCmmmmlpL- 
Pfii-u % nnp bmnmjfin inmpm £ mn fiLpbmuij mbmnuu k mn Pmn.mL.npu mnhbjj\ 

Ibid., xxviii, p. 168, " ♦•♦ fi dmSmhm^ jnpSl £bmi l^mpiibm^ bfig fi &bp 
bmnmjm.pfiL.un.". Ibid., xlv, p. 263, " * + *bwnmju &bp *lwpn.mU"+". Ibid., 
Ixxy, p, 442, " k jwpgwjfy mpgmjfi ft bwnmjni,pbul iffi bjmu^p^ ". Ibid., 
lxxv, pp.447: 450, " ♦•♦ k ^bij Upbmu mbmnuu mppmub^nLpbrnh mpdmhfi bn%, 
[i 3m fi bq.fi mnm^fi Hummbnj k 3mpql[wu l mpn.mp Smog $bwnu k ^bq 
bmnmjnL.pfiLU mnub^ [447] **+ifbp pufil[ wbmpg £g k ^mpnu Spun k 
hmfuubmij, k ifbg bmnmjnLpfiLpfiLLi mnhb^*** [450]". Ibid., xcii, p, 548, 
" •♦♦ k Jbp phfili mbwpg lg y kifbg &bp pkfilj bmnwjg bdg". Elise, " ^wpgu 
ilhp .♦♦ fast 1 " f} & umLU J n '-Pbmu tfmummlpi **♦ k dbg nhnjh bmnmjnL.pfnu 
bmnmjbgmg ". [Cf. Toiimanoff, Studies, p, 117], {459, 1) 

50 I/uchaire, Manuel, p. 159, " il faut distkiguer \es fiefs qui sont des seigneuries et 
les fiefs qui n'en sont pas ". In his investigation of Russian feudalism, N,P. SilvansMi, 
Feudalism, p. 76, calls the first type "fief - benefice " and the second " fief-allod ", 
{Cf, Manandian, Feudalism, pp. 104 sqq, on the ^osto&s which are not discussed by 
Adontz]. (460, 1) 

51 Mortet, " Feodalite " in La Grande Fncychpidie, XVIII, p, 206, (461, 1) 
53 Eaustus commonly uses the terms umfumpmp and um£mmbm. Smumwlp occurs 

only seven times: three times together with hmCmmbw, FB, IV, xvi, p. 134; V, xxxvii 
pp. 242, 243, and four times alone, Ibid., Ill, xx, p. 62; IV, xvii, p. 136; IV, 1, p. 167 
V, xxxviii, p. 249. The terms m£p and wlpnLpfiLu, in their technical sense are rare; 
Ibid., xliii, p, 253; V, xliv, p, 256; IV, xix, p. 137; IV, vi, p. 208; IV, xxxii, p, 23a 
IV, xxxvii, p. 246 ; V, xxxviii, p, 249. Lazar P'arpeci on the contrary avoids um£w- 
lubm and uses mmhntmlp^ w£p y umfumpmp. Elise, prefers umfumpmp and 
never uses either um^miubw or mmunLm^p. {Cf. above n. 48c, and Toumanoff, Studies, 
pp. 117 n. 190 and 130 n. 229]. (463, 1) 

53 In the period under discussion, Valarsapat was the ostan of the Arsacids, Erwanda- 
sat, that of the Kamsarakans; Hadamakert, that of the Arcruni, Dariwnk', that of the 
Bagratids; Ostan, that of the ftstuni, etc. [Gf. Toumanoff, Studies, pp. 114, 193 
n. 209, 199, 202, 206, 213, 322-323 and n, 77, etc. Manandian, Feudalism, pp. 56-58, 
etc.] (463, 2) 

53a [q^ Manandian, Feudalism, pp. 97 sqq.] 



NOTES : CHAPTER XV 



517 



54 EUse, v, p, 99 indicates the order of succession, " -^ifjufuwhwlj jbwu IpuLjbinijh* 
np bjjih n£,bm ji^fiiLuhjih Ufu-hbiug, nbnjpiupu Ipiid nnprifiu l^wS nbq^pmpnh 
nqfiuu fi uihqji hnrjm SwwnLtjwhlp*** ". 

Succession according to seniority can be traced in the Mamikonean house, FB, 
III, iv, gives the fwifuwpiiip of this house under .Xosrov II of Armenia, as Vaee, son of 
Artawazd, " ♦♦. tuphml^btug pmiiwLnph ji ifbpujj un S m n^w^l npij.fi IXpimmiuq^- 
qwj nhw^uiuihm Xfrndfj^nhhuih wn£ilfih y jwtffil uiqiupiuuikinnLphwb Zwjntj, 
qifbb qopwuup nop tug fiLpnn ". The power passed to has only son Artawazd, 
Ibid,, III, xi, " Uwliiuju n^w^lfi qopuiLUipli ifwgpfil} Sink Iffy ilfi nhnpmh 
qnpilli p ^wjpbhfi pwp& qw^nth LiiuwnLijwhlJih, npnLi) whrnh lp jwhmSi ^wliuSj 
jiLpnj JlpmwLiun^n. : l£nin£fi PutqiuLnpfih q£oph iqwwfiLU fi qjnLfuh qhlfjh^ h 
nuiutupwwbmnLpbLuh hjipmu uibiifi. qfi fywpfi n p I }f l ifiuuuiLulfLuuipfi lp^ k fi 
ijmummlj^wuip luljilI, qji mj^jmnjifih n£ lluilul ^odljUL^ ljJi fi ifbb wuiinbpwnjJfih 
ifbnmh ". Under Tiran, the representatives of the family were Artawazd and Vasak, 
Ibid,, III, xviii, " J)l whq. n.fiiqbw^ IXpuiuiLtuntj. k *£u/uu/^, wpg fi Swilfiljnhhwh 
inn£fit&) npg Ifih qopwLwpg Luilbhwjh qoprng Z*uj n g ♦♦• "• Arsak II, " • «♦ 
Ipuzjnjij «•• q^wpqiuh qbpln bqpwjph fi hw£w m b m ni.ph mhh wqqfih fiLpbiuhn^ 
k ijfl^muml^ qiffi^fih bqpwjph qjiLp fiwjbiuljh fi uu^mpunubuinLpftL.hh nopwtf- 
wprnphwh jfipu uLLuwbpuinpLuijh ",' Ibid,, IV", ii. Consequently, Vardan is called, 
" dbb hw^Luiubm SmSfil^nhbmh uin^ilfi " [as agains his brother, ts • •* tfbbfi 
uinpiubjujinfih Zwjnj npnLii ^Luuwljh tyn^lp "], Ibid,, IV, xi, and IV, xv, as well 
as " uiwhnnnlph hm^uiLubmh liwtlfilfnhbLuh wn^ilfth^ bpln bqpwjp ^wuwIjwj 
uuiLupLULubrnfih ", Ibid,, IV, xvi, Vardan was succeeded by his next brother Vasak, 
Ibid,, IV, 1, " .«. SwmnLijiuhhp hu mSpmummh *♦♦ qjiLpSl mLuhnmibmnhlh 
^muml^mj ♦♦♦ ". Vasak was simultaneously sparapet. After him, his son Muse!, 
assumed the function of sparapet, " ••• h lp hntjw nopuiLjipLfu IfnL^bq npqfih 
^wuiulpjuj uwuipLULubinfih ". Ibid,, IV, Iv. King Varazdat had Musei killed and 
transferred the office of sparapet to Bat Saharuni, " IAujw fywrjnjfj PwipiiLnpb 
*luipwqrpmn fi ipipb uu^mpmu^bmnLpbmhh npwm hmCua^bmh IJwCwnnLhbwt} 
mn£dfih ♦♦» fiLp qiajbtulf, ", but " mntj.fih fiwdfilinhbwu mn^ilfih mwhnnnlp 
bw£wu[bw Ipurjnjt] pwqwuip* ^w^i uihniij jj hilfih rnn^S^ ", Ibid,, V, xxxvii. 
At the return of Manuel Mamikonean from captivity, " ♦♦* frppk bwbu qhut ^w^h* 
np jumw^u ip jbtuj^ hm^wu^bmh^ ujiu£ ^A bl^w^p hm x bm rjhw nuj^wmfiL fejuwunL- 
pbwuh, nnp uinbmj^ lp fi pinn-iump^u t lwpwnjn.wmwj, giuhnji hm lp bplg 
jui^a-fin. k JfmhnLljh m.ubp nJiw^wuibwnLpbinu wrp^jiu wwununl;pni.pbwh 
iuwuwlJju, k ^w^u ifibkp kplipnpq : frulj jnpdwdwd b£uw blpnq ]Jmuni^^ 
fi ijrnmu mlpm.pbmu pLpnj, umfu umwhtj {.pwSutufi puin.iULnpj)u ^mpmnr^mmtuj 
j/jn^u jwtji^mwljbmij ntp2pwtjwpni.pbuiuh rjui^wpm unburn tPpL.hu, njp wjh ftui 
np jiLpnij uwjvubwiju jj pu^u Ipa^bm^ lp fi ul^njpmhl^ nnp wpgwju ^wpwn- 
fiium >unp£ wnSilp jjLpmiS qwjlm}jfih Pwmwj, nmjh Jfwunilj^ Jpugu ^whbmj^ 
mulp nji^jumhnLp fiL.hu ", Idem, Manuel was succeeded by his son Artasir, Ibid,, V, 
xliv, " ♦♦• ^fjLwhqwtjiUL uLuwpmujbmh nopmijmph Zwjng XTiuhnLl^ nmfum £fu.whjinL- 
pbiuh lIw&il. L Jfnibwzj appTiJ] fjLp nJXpuiujyfip^ k bm hSm nmlpnLpjuJ} fiLp h 
Lj^uupupwu^bLnnLpfjLh nopmifrnpULpbtuhh fiLpnj ". The son of Artasir was probably 
the Hamazasp, who was the son-in-law of the Kat'oHkos Sahak I, and who in turn had 
three sons: Vardan, Hmayeak, and Hamazaspean, LP\ xviii, p. Ill, " [y«/^iu^ziy] 
fet tjwuh qfi n£ qnjp h$w mpuL qwLuili pwjg iffiLujh qnLuuip iffi^ nnp lp mnh- 



518 



NOTES ; CHAPTER XV 



kutj_ IfhriLpkuih ZmSmqmuiumj mkmnh Ifwtffilinuijin k uiqwpwiqkwfi Z*njnn, 
np bhwi. [j ZujSint^iuuujUjj hpjiu npqfju^ qumppu ^uipquihy li qZ^nijhml^ k 
nkpajukifih Zm^uiniuuinkiuh ". Vardan was the next tanuier, and after him his 
nephew Vahan, because of the premature death of his brothers, ZP\ xxiii, xxv, 
pp.134, 144, etc, " • •• ^w^uih [mu^p] qwkpm.pfcrf} lTuji/jilinuj;ljn k quiqw— 
pwuihwnLpjiLh ". Ibid,, lv, p. 393, [Of. Toumanoff, Studies, pp. 120-122, 209], 

(464, 1) 

55 FB, IV, ii, Arsak II appointed Vardan, nahapet of the Mamikonean, and Varazdat 
appointed Vage, Ibid., V, xxxvii [see preceding note for the texts]. Manuel seized 
the sovereign power, Ibid,, V, xlii, " \,njuiu£u jwilhiujh qiununu l^mnnL^uihlp 
uufuipimifhinh IfwItnL^ uw£iuiukwu k wkmpu quiLinnuin quiLumwij ", LP\ 
xlii, p, 236, has the marzpan reassure the Armenian nobles that, " n£ qui£pnLpfn.uq 
li qiqwwfiL ^whkdgjnLilhgi ". After the fall of Vasak of Siwnik, the Persians appoint- 
ed his enemy, the apostate Varazvalan, " "<ui£p ji ifkpwj w^fuuiptyh \Jfii.hhwij'\ 
Ibid., xlvi, p. 270. The Persian king assured the Armenian nobles that, if they 
accepted Mazdeism, " ••• qJiLpuigiuu^fiLp nLpnLg nwwhnunkpnLpfiLU mwS L 
qq.uj/1 k qiqwuspL ". Ibid,, lv, p. 311, also, Ibid,, lxii, p. 357, " fi tyuipuljuin 
qwujunLUi}pnLpfiLUU qhhiuj^ nLpwnnL.pkutdp, ♦ *• ", and Ibid,, lxxv, p. 443, " L 
mpn g n jg [*&& m JP *$ / u, _/ z,1 Y/' ii /' uiyJumpCji* np pi wnmLh^uiL pmqwuipnLpkuiSp 
wpdutliwLnpwujtu £wuhw^ J) inwunLin£pnLpiiLU. wjj_ nphmp uiLinnwl^ k whinjiuiuiU^ 
l,in rjl^m mm l^p L ifiuinLnn£iifil[, np qutu nwjuiuj-iuft i/hb pwn.WL.npnLpJiLU jumphh y 
li phn. l^pml^fi l/iiunu qhhh h w£pnLpjiLU ". Vahan made the condition that " • •• 
nLi/hg jmnmij.u SnqnLpbwh quit, L luwmjiL swuijg ", Ibid., lxxxix, p. 523, and he 
interceded before the king, " ••• }unp£kj^ &hq qwiuuni.mi;pm.pfiLU ^wSuwpwlpuujih^, 
which was granted, " ♦ ♦• kp[jnfi i[iuuu gn nindsm^ qui£pnLp[iLU ^wSumpm^ 
wlijih ". Ibid,, xcvi, p. 572. Services were taken into consideration on appointment 
to a tanuteriitHwn. Rejecting a similar request of Vahan in favour of Prince 
Arcruni, the king pointed out that " • •• jmqmqu m^pnLpfihh IXp&pnriiLnjh, Pnq 
Jujjp rift* q[i qbwwunbh 3iupqfil[u p^ np ft uw(,iij;u ^ u ^ uiqwu fiu£ wpdwhwL- 
npwu^u nnLnmhh^ tun ilkq, k tfuiu intuitu fiu£ joqnLin Qpbwn w^fuwptyu 
ijujumwl^kjj £,uijkunnLg wiqui k ilkg jiupdtuhu ". Idem. The king dismissed Prince 
Garegin, " k £wukw^ ft pmn qwipnLpjiLUU Ji hdiuhl; ♦•• ", Elise, i,p, 13. Ibid., 
v, p. 115, " ♦ •* k uiilkhwju Cwjpkuji quiLumg £iumujh fi hngmht ••♦ ", c/. Ibid., 
vii, p. 164, Bishop Sahak of flstuni blames the king that, (< \}mnLg qSm nm^pnt 
Pjithh UfiLnLukwn ". Ibid., vi, p. 135. The king " ••* funuiniuniuL unuj^ unnw- 
qjiLpm^mhsjiLp pyfvwunLpfjLh pum l^wpqjt ^mjpkhfi tquiuiLnjh^" ", Ibid., p. 197, 
According to Sebeos, xvii, p. 65, .Xusrd, " wwj hSm \JSpmuiwf\ *»* ninwhnLinkpm.— 
Pfjkh •♦♦ ". Kavadh granted to his son Varaztiroe, t( •♦• njt^uuihnLpfiih 
wwhnLw};pnLpi;ujuh ". Ibid., xxviii, p. 97, Finally, " nnpq.fi unpin nwun-pwufili 
npni-d wumSih £p \JSpuim % l^wpqhuin wpgwj ji upiiwjjL CnJiLp fiLpnj^ mnL- 
km^ hSm qfi^fuwhnLpjiLhh pnfj^ inwunLinipnLinkujhu uwiqkwni.pkwu y k mpwp 
qhm qpnLUqtnp nuiLpmn pLpnij ". Ibid., xxxii, p. 116. (464, 2) 

55a [Cf. Toumanoff, Studies, p. 119, and Sukiasian, Armenia, pp. 204-205, 307.] 

56 FB, III, xi [#ee above n, 54 for the text], Ibid., IV, ii, " ••• umw^Ji wpgutjfih 
tnmmjiL ji qifiLfu pmqSifih ". Tov. Arc, iv, p, 272, " ♦ • • wuij nfiuiuiwufi fi 
&knu IXynwnj k uifipkijnLnuju£ uhiji^mljujli dumwun.ni.pkwuu »** £nfunL- 
pkmSp ". (466, 1) 



< *s^ 



NOTES ; CHAPTER XV 



519 



57 FB, IV, iii, " Ui^h/ ft i/p dnipn^ Ij^nLmh^mh ma wpgwjh Hp^mli dhbwifbbg 
hw£uiuikwg rnqr^mtj wipfwij, mn£fiwij mnCSw^^ qbnfj^ b qpo^nLij mhmpu *\ 

Each princely house had its own standard, on which were represented, among other 
devices; doves, wqwhwnpo^ wijwhnwL npo> turtle-doves , ifwpdwhwiiwhfe < 
jjuipmdwhuil^ — wh^, eagles, wpbuM^tuh^ (466,2) 

58 See above Chapter X, p. 215 and n, 50, [Also the g rant of insignia to the 
Armenian satraps in the IVth century, of. Chapter V, pp, 87-88 and n. 27. Toumanoff, 
Studies, pp, 117 and n. 193, 134, n. 235,] (466, 3) 

59 Luchaire, Manuel, pp. 250 sqq. [Toumanoff, Studies, p. 118.] (466, 4) 
80 There is no word meaning " law " in Armenian so that an Iranian loan word has 

to be used uiL—p£& < daena. [Cf, Sukiasian, Armenia, pp, 238-254, 381 sqq,] (467, 1) 

61 IneiSean, Antiquities, U, p. 87, observes correctly that we have no case of royal 
interference in the internal affairs of the principalities, and consequently deduces their 
autonomy, [Of, the preceding note awe? the status of the Armenian satrapies vis-a-vis 
the Koman Empire as civitates foederatae liberae et immunes, above Chapter V, pp. 75-101], 

(467, 2) 

62 Of, FB, III, viii for the execution for treason of Prince Databey 
Bznuni [and not Rstuni], " ^wmmpmj hm^mwhmjj PqfimSikuig ♦•• funp^nipn 
ifjjmpmhriLphmh nh£p phq fejuwhuh qopwgft tywpujiq. ljinifhqwL Smmhkt h 
&knu hngw qwpgwjh Zwjni) qmipb fiLp. ♦♦* Qkppmlpnj^ mpmpkmj^ ifl+mmwpihh 
^w^t umwpmw}mh A giug ^w^wli w^winnLhjjy mb£fih qhw wnwgft ilhbfi 
JdwqujLnpfih JoaupnijpL^ A gwpljntj mnhhjih tjhw gwpwfig frpph 1 W JP* n P 
m^juuipifi li qhqji A qopwtj wkwnh fri-pjij nwhwiwh ]kw^ jig} : fei qniqq hnpm 
A qflfih A qnpqfiu awwhf-p jwdpnrjfi mhn fefiiwhfih £n£inniSi£wij juihnLwhhuijfi 
jlAqPnifiwp' liqijLn^. jj huiL kjjhm^ ^w^j- uiqwpujufkmh, mh^kw^ fj l[qnJiSi % wn 
iwuiuptulj n£ q£q pnipijp A n£ qwpni. ; &l mjuiqiu pwp&WL wqtpasmnljS im^m— 
pmpnLpkmhh mjhnpji^ fci. qmnii hngw jwpgntStjiu \w)wh ", (467, 3) 

63 According to FB, IV, ii, nine hundred persons sat at the royal table, not counting 
the ones who remained standing on their feet. We learn from Persian sources that, 
" one of the Persian -Xusros ordered to assemble all kinds of men from all parts of the 
realm and to seat them in a definite order for a feast, At the same time, a certain part 
of the officials was separated from the group during this distribution, in order to conform 
to the traditional practice, " K. Inostrantsev, Materials, p. 177. The reference is 
to the creation of the Throne List or Gahnamah with which we are already familiar, 
The officials supervising the organization were probably the very ones to remain standing 
in Eaustus 5 account. See above, Chapter X n, 2 for the text. [On the gorcalcals, see, 
Manandian, Feudalism, pp. 60-65, 67-69, 71-79, 81-83, 121-122. Toumanoff, Studies, 
pp. 204, 220-222, 230 n, 279, etc., cf, above Chapter X, pp. 185 sqq and notes.] (468, 1) 

64 Luchaire, Manuel, p, 262, " ... dans la plupart des etats feodaux, le plus important 
des hauts fonctionnaires est le sinechal, Aprcs le due ou le eomte, il est le souverain 
en second de la seigneurie; il dirige la justice et l'adrninistration locale, commande 
Varmee ... , Le cTiancelier est un personnage non moins considerable ... il est percepteur 
et collecteur perpetuel de ious les revenus du comte ". [Cf, Lot and Eawtier, Institutions'], 

64 a See above pp, 339-341 and nn, 22, 29a-43, 63, 

65 Luchaire, Manuel, p. 264, " ^wuwwkmg ", is found in FB, IV, li, [" ifkbui- 
ilhbg hwfuwpwpg l^ntuwljmi^ k n T* w k w L£ qwLwnwuikwpg qnpbiulfwjj) A 
nmum inking ^[jhuj^iuhw^ = nkqgwmiq '% ■&£» p, 41 ~ " qjtLqwtqkm ", Mov* 



520 



NOTES j CHAPTER XV 



Kalanh\ p, 124, [Cf. Manandian, Feudalism, pp, 82 and n. 1, 174, 187-188, 232, and 
below n. 00, 

66 Sebeos, xvii, p, 62 speaks of a member of the Dimak'sean house who was an official 
of the Bagratids in the Vllth century, " • • • fympakw^ tg fj q[kpmj mmu £wlw- 
mmppS k IjiuSiul^mmmp ", [Cf, Sxikiasian, Armenia, pp, 226-231.] (469, 2) 

67 FB, IV, xxiii, Prince Meruzan Arcruni, having gone over from Arsak II to the 
Persian king, " nhfcp pun htiiu ulJuui bpndmiip, qp jWLpmkwh buinwj j[iq[i 
u^m ". Ibid,, IV, liii, Arsak II asked for an oath as safe-conduct before his hourney 
to the Persian court, " Up^aty fuhrj.pl p iusLwmmpjtS bpqmifu ULfumfiL fi hSwhlu^ 
qji JujjuS £kin£ mhljmuJjmb bpPfitj£ um uw : fei. hm km pbph£ pum opjiuwqu 
CmLwmwpJiS hpnSwuij pwnwLnpnLpbwuu tyuipujig* wq y Ifogk^ l \_ m P mt l ulpnp- 
iun[ip 8iumwubwL, k jqbmu ". [Curiously, Adontz says here that the Persian king's 
ring bore the effigy of a s * leopard ", or " panther " rather than of a wild-boar, which 
is the description given by Paustus, and is indeed, the correct device for the official 
seal of the Sasanians, cf. Christensen, p, 394.] Sebeos, iii, p. 43, " km mmhbj^ mq 
kpij.ni.ifu Ifhgkw]^ ". According to Proc. Pers., I, iv, 9 [L. I, 22/3, the Persian king 
Peroz had sworn an oath to the Hephthalite Huns over salt, and the ruler of the 
Huns, " tovs* Se aAas atcpov oTjfxeiov rov /JacriAetov am€Kp4p,ac?€.v is ov$ rov opKOv Hzpotfls 
dJftocre vrpoTGpov, ov hj] aAoyrjvas eiTa im Ovwovs iarpdr^vaev ". Mlise, iii, p. 86, the 
Persian king, " Jvuqpip fi ungmhk i^IftujnLpfiLu CuiimmuipSnLpbmh jbplffjpu 
Zwjnij) k bpnfiwiip julJuw Smmh^p umw^fi unguj*** ". [While these oaths are 
duly attested, not all have the character of " homage ", also, as observed by 
Toumanoff, Studies, p. 117 n. 192, "The concept of homage as a separate act from 
the oath of fealty does not appear to have existed in Armenia ".] (469, 3) 

68 ItP\ xxvi, p. 146, " • •♦ ilkp wn. tibq mlpm.pjiiSi k fuhm^mpljnLpjiLU^ k 
&bp wn i/bq wpnwpwSmni.pkwi}p bwnwjm.ppLU k ^urnqmuqutppLu ". (470, 1) 

69 Luchaire, Manuel, p. 195, (470, 2) 

70 Ibid,, p. 196. [Cf. Toumanoff, Studies, pp, 117-118, and next note,] (470, 3) 

71 The Armenian cavalry was also drawn from peasants, as was that of the Parthian 
custom, hence the expressions, *' nwdjify ibbbwj^ nbn^nii^ £bbbwj_ ", See above, 
Chapter XIII, pp. 299-300. [See also Chapter XIII n, 21a for the critique of this thesis. 
On the size of the Armenian feudal contingents and a comparison with western ones, 
see, Toumanoff, Studies, pp, 234-241, particularly p. 235 n. 300, and Table V], (470, 4) 

72 jpg t y^ yj^ Drastamat was, "fi^fuwu mmh nwLwnfiu k £wLiuwwp]i£ nwu— 
&nLtj JXunbq pbpnjiu^ k wuguwju pbpqmau wpgniJip up ft l^nnpmhu jmjhu, 
unjuw£u k jkplfpjjh tTntptug ft Pumpbq pbpnpu nwukgh jbiuj^ Ifiu pun 
uni[wL ", [Cf. above n. 36a.] (470, 5) 

73 Noldeke, Tabari, pp.5, 111. " dpyairer^s " is found in Palmyrene inscriptions, 
cf. Levy, ZDMG, XVIII, p. 89. The word is derived from Pers, " ij* | fortress " = 
Lat. arx, and means " commander, keeper, of a fortress = pkpnwlfw^. ] Cf, Christen- 
sen, p. 107.] (470, 6) 

74 Elise, iii, p. 68, " jwp&wljktjwu ji i[kpwj pkpnjjnu k wlwuwij, anp 
mutjiu fywpujiljgu fi mknfiu mkqjiu^ jwiipnnu w^fuwp^fju. *♦♦ Unw^fiu 
qilkbh Upmwymm {,wunkp& wlwuoj> jiLpnifg. k mnhnLph uwhSwmnjn mJUmpuu* 
q^mnufi gwqwgu, qJXufiu, oJXpmmnkpu k uuilwuu fiLpkwug, qppl^mjunpquh 
k aUpjvbjiu k owlwuu fupkwun. nfiwp&pwpnnu, afynpwujiuwu^ qp'wpjwupumh^ 
nwulimuliwbkjji finuil^mhh^ k pun hnum k nmimhu fiLpkutuij, nUpifiwukwjjty 



NOTES : CHAPTER XV 



521 



q^wuh wlwu, puq uuJju h qwLwbuh jiLp. qQnbwfi h tftwiqnjwh, qflpnwu 
£ If q^w^wlfw^wwu J \ (471, 1) 

75 FB, III, xiv, " Hp^ i^ftwdnqpil bqbwj^ dbbwdbbg hwjuwpwpwiju, 
li fj iffi fywjp kljbmj^^ [vnp^nLpq^ funp£brjwu ". Ibid,, xxi, " dnqmjbtjwh 
ji lIJi dnqm[ iSpmpmhnLphwhh Swpqfilj w^JuwpCfiu Zwjwumwh bpfypfiu. tjw— 
Juwpwpg ilbbwi/bbg, WLwqg, IjnLumi^wjD^ w^wpiwIjwjD, wqwwg, qopwq[iiL(vD, 
jj.mwwi.npg, iqbwg, fefuwug. P- m JS I 1 qopwijwpwiju, wj^ h [i ^hhwljwhwq 
whqwS nwiI^Jj dwpqljwhu ". Ibid,, IV, iii, *' JXwui jt dfc dnqnij^ IjtiLwk^wh 
wn wpgwjh IXp^wIf ifbbg uw^wtqbwg •♦•", c/. Ibid,, IV, iv, Hj V, Iv, " ifkb 
Iiwfuwpwpgh Z w J n $ ••• uljuwh wjum/lbinbL puw IjwSwq fiLpbwhq Juipbi 
qPwq.WL.nph ", etc. Many cases of the princes meeting in council are also found in 
LF\ (471, 2) 

76 FB, III, viii, after the treason of Prince Databey, King Xosrov, " quip opluii* 
q[i dbbwifbb WLwqwhfih % hwfuwpwpgu w^jvwp^wlfWj^gu w^fvwp^wwbwpgu 
np ijjh pJjLpwmpgh h iwqwpwwpgu, Ijwjqbft wn wpgwjfih, k phq u£w ^p£~ 
buqfiu^ k i/jj ng bpPfiqi jj hnqwh£ puq qopu wpgnLUp ", [CJ. above p. 335 
and n. 19 for the regulations of Cyrus]. (471, 3) 

7 ea [Of. Toumanoff, Studies, pp. 119-126, particularly p. 120 n. 207.] 

77 Lazar P'arpeci calls the following relatives sepuhs of the iaymUn 

1. his brother:"*** tfbbji ukiq£fiu Jfrnilfilinhlfiq bqpoph ^wpqwhwj • ♦• = ". 
ZdwjkwLj, LP\ xxvii, p. 159, xli, p. 234. Vardan had another brother named 
Hamazaspean, I bid,, xviii, p. Ill, but Hmayeak was the older of Vardan's two 
brothers, and is consequently called " ifhb ubiqni.£ ". When Vahan Mami- 
konean became taimler of the house, his brother Vasak was called " uhiqm^u 
ITwufilfnhllig ", Ibid,, lxix, p. 401. 

2. his son: " • • ♦ ukiqm.£ i/b nLpbwgfj . . * n P t }{ 1 ^nqpblfwj fefuwufju f)i.pbwj ", 
Ibid,, Ixvi, p. 391, 

3. his nephew: Vahan, " wp&wlf£p qfiLp kqpopnpqfi i/Jj, qnpq[i uw£wwwlifju 
^wuwljwj Qpfjqnp : ••♦ ubiqm^u lXwifJjlinii£[iij Qpfiqnp ♦*♦ ". Ibid., xciv, 
pp. 557-558. 

4. his kinsman: Sewuk, prince of the Anjewaei, had an wqqwlfprj named John, to 
whom the historian refers as " /?#& jlXu&WLWtjfj hwfuwpwpwpwtjh* ubwntC 
ilji npnL.8 wuiilu £p Qn^wh ♦ •♦ J?t jnibw^ wwwwufuwhjj jwqqwljqlu 
Qn^whwf jfefuwhiu H£&kw£jkw£j Vidjnj ••• ". Ibid., lxx, pp. 412-413. Va- 
razyalan is also called sepuh, as " wjp iffi fi wn^S^h JJftLubwq ". Ibid.* 
xx, p. 115, " ^p^p ukiqnL^u UjiLubwzj q^wpwqijwqwh ", p. 118. He was 
the son in law of Vasak of Siwnik, "♦♦♦ ijtbuwjwqbwj^ ji^fuwufih UJjLubwq 
*lwuwlfwj, ubwnL^ JJjiLubwij ", p. 115-116, and it is not clear whether he was 
entitled sepuh as a member of the house of Siwnik', or as Vasak's son-in-law. 

5. Sebeos refers to three individuals from the same family as sepuhs " ubwmXgh 
^w£kni.u]ig JJwSnLlih qnp wuwqfi k JJwpqJiu k *t\wpwq \>bpub£ k 
Xrbpuiu ••♦", vii, p. 50. Their tawnier was, " hnupni^ ^wifimhkmg w£p". 
Ibid., xi-xii, pp. 56, 58. 

Other historians rarely use the term septih. Twice in FB, IV, xv, ** qtyhbj^ ithb 
ubiqnLjC wp^wfynLufi *** " and " Zwfiwqwuwbwh ubiqnii lip fi Jfwiljjl^nhbwh 
wn4Si "• Ibid,, V, xxxvii. In Eliiz, likewise twice: " *• ubiqnt^ nSwug 
jwiuSl 1 wn^Sl ", iii, p, 74 = (£P\, xxxvi, p. 200), and " ubiqm.£ tffr wqq£h 



522 



NOTES : CHAPTER XV 



JXSwuin diking ". Ibid., v, p. 105 — (£P\, xxxvii, p. 215), In Sebeos, only the 
passage given above, [Cf. Toumanoff, Studies, p, 130], (472, 1) 

77a [On the sephahan, see below n. 82.] 

78 The answer to the Persian king, LP', xxiii, p. 135, was composed ** ♦♦♦ uiLiua 
mwhnmihwpgy £uihqkp& WLwn. ukiuCog ". The king summoned, Ibid., xxv, p. 141 
" wifhhwjh wuihnLuihiupgh k uiLtun. uhinm,£,gh kpknnt jwyfuwpCuinu ". 

Again, Ibid,, xxvii, p. 154, " + **nwju miShhwjh ndmuiL uiufuwpuipqu wifkhbnnLU 
k ukinCwyh ♦ ♦* ", Cf. Ibid., xxvii, pp. 157, 160, etc, Ibid., xxxii, p, 190, 
" wilbhuiju inuihnLinhpwiipg k ubiuCog ^mumiuinlhh nkpiwufu. k npnn Swtniu- 
tihiiiL £k ku lp Ifugkw^ q\Xi.kinwpwhb lihgifih ". Ibid., xxxii, p. 194, " Ijug- 
hmj^ hwfu dmmuihhwL fi^jutuufih Dfidihusg i^uwwlfuij, L wiuw Swmwhhog 
uiilkuwju mwhnLUi^pwnh Ztujnn It iulwu uku^^mqh ". Ibid., xli, p. 231, " uiuhi- 
uipwpg k uhinni£g ", etc. {474, 1) 

79 MX, II, xxii, " Zwjng pmn.uiLnpi typinuiLwgq "P^f 1 Sfiqpiuhiuj : \}ui 
dumwun.knm.nwul; nknjiwpu fnp L ggnpu f? n.wLwnu JXqJw£injj k JXnpkpwunj, 
Pninpi[ [i unuw qpwuu wpgnLufi np [i -^hu wjuS qiuLumwg^ £wfiqkp& umwutyu 
Suifizj L nntyuiij, puin opfiunj^ji wnj^ml^mhrnnh^ np fi l^nijpmuu Zwymbujiq. 
npiu£u qfi ifoh^ hngm luwwnnul^uihiuqnjh L mnmih^ pwn.wLnpuinJiu gwu 
uwjhu uip^w^nLUfiu. ifjiwju opfifiuiqpt n£ \km^ jjjjpuipwm fi phml^nLp^d] 
wp^wjfi : ", Ibid., II, lxii, 8" fci bljkm^ mn urn l^uiujuuignjh fiLpnj uipywIfnLukuiq, 
np £ fib fi l^nqpuihu Z^^khfig^ uiubu, Qliquip^uj^km ifkq qdwnwunnLpfiLuu, 
qji ukq t* «£"'%/' pmnpuigui^ : fcz. bin ^pwSwj^ nfiwbn fi bngwbl; hpPwi h 
quiLuinb Hiifinliwfi L IXnpbpwbnj : Jiufy ungui wnnii^ Lu pnqng lfni[kiu^ um 
uip^wjfi, ph umuiLbiiun.nju bkq £ ifkn^ n£ jib^ nLhljbqjip ifjbfi Sfipuib. uiji 
£wu in minting ij^fin n£ mj^ dmnmbgnLpjiLb inuij_ hngui, pmjg nnp mJihu £uiLwuwp 
mpni,hj_ jfiujphuiuu : Qnp pwduihkw^ puin dwpqwpnLfi, qmuiL u^ml^mu dwnuju— 
q.ni.pfiLU phwl^nquinh Zw£wkbjig. i\wuu npnj pwnni-dg fi hnijwuj; hljhui^ fi 
quiLumu lXn_jini^jmi L IXnpkpuihnj ", (474,2) 

80 One of the canons of this Council begins; "jnpduiLl uiqwmg gjupkiubn qkpq 
h afiwpn. pwdwhhh ". The king of Albania, driven from his throne by the Persian 
king, demanded from him his own sepuh possessions, " fuhqphwn nSmhl^nLphwh 
uhiu^wljwb'u^ ijnp £oph {jLpnj £unp£kwj^ lp hiiiii fi wniujnLpkwuu Cwqwp 
bpq. ". JSlise, p. 199. This passage is found in Book VIII, which is undoubtedly of 
a different and relatively later origin than the others. [Cf, Manandian's objection, 
Feudalism, p. 192.] (475, 1) 

81 MX, I, iii, " •♦• qfiuifitj k quiLtunuin ku k fiLpiu^iub^liLp inwun wnwu&<— 
uwlpnunLpkwhij^ k £uiunLpn Cml^mnmlpnhpkuiun k nuiywun uijdi inn dhn 
n-inuiujiu infipuuL gprnnmn fiwwkuiug, SmhuiLwhq np fi ukiji^wkwh wnwwnL.pkwh 
iumjuin^iumnLp[iLu ". The word umwu&hinljiuunLpbwun does not have here the 
opposite sense to ^uihnLpn, as might seem likely (and it was understood by Marr in his 
review of Java^isvili's Polity, which we have cited earlier). It is evident from the 
context that litigations arose primarily at the time of division of the sepuh inheritance. 
The sepuh having received his portion left the family, he became an uminu&hwkuiu, 
and the quarrels arose from just such wnwu&uiulfuiunLpkuiun. [Cf. Manandian, 
Feudalism, pp. 171-172 and next note.] {475, 2) 

82 According to Xosrov, the patriarch, " ukin£wlfuiu uinji ukpmh k wu£uikuj~ 
nwtjkjiu ", Arm, Diet., II, p, 706. [Cf. Manandian, Feudalism, pp. 77, 93-94, 172, 



NOTES : CHAPTER XV 



523 



192-195. See also Ibid., 127-130, etc., for zarangufiwn, as used in connexion with the 
clergy. Also Toumanoff, Studies, pp. 120 n. 206, 124.] (475, 3) 

83 MX, III, xx, " fci. 1 qkplfnLUU qnjjunujil^ jwqquin uwjuwpwpwpwqu pumhwj* 
ilfi x qifhpkwLnpuin juuuiSni-PJiLU, qnp ijmuh wqw^kjnj ukiji^ml^wh wquiwnL.pkujuh 
wnhijtu ". The passage concerns the forbiding of such practices by Nerses I. At the 
same point in FB, IV, iv, " ki qjt ifihjihnfih jwSnLuhnLpkwh opjjuwuipp y ilh 
uinh^ k Jji qmi pkphj^ jiLphmun wSnLuuni-PJiLupuqui^ Ifnqfiiuuij. k ihmjyyhi 
unkj^ji ji dkp&wtnp h jwqqfiu mnCSml^jin jumnhwl^nLpkwh wiinLuuni.pkuih*** ". 

Faustus gives no reason for such consanguinous marriages. Movses, while borrowing 
the information from Faustus, adds his own commentary in which he presumes that 
these marriages were meant to preserve the inheritance of the " sephahan - freedom ". 
And, in reality, a brother marrying his sister-in-law, inherited his brother's sepuh share, 
Movses' interpretation is characteristic for his period and not for that of Faustus, in 
whose History, the injunction of Nerses I is taken primarily as a reference to the marriage 
of King Arsak II with the wife of his nephew Gnel. (475, 4) 

84 Tov. Arc, III, ii, p. 131, " ^wCpuiS ukiqnL^ Pjilfuwwiu£ fejuwujiu ... k 
W JL>P t 1 u ^ wnL ^ tqw^uiouljin ". Ibid., p, 134, he says of the same Vahram, " ... 
np k pjiljliini^wi, ji^jmuhjiu^ ... ". It is interesting that Thomas seems to use the 
term sepuh as the indication of a particular rank, and without any indication of the 
family to which the particular sepuh belonged, as this would have been done by earlier 
authors. So also, Ibid., II, vi, p, 109, " ... k wjlg fi hwjuwpwpwtju*" ukiuntOi 
%wlJiP"» uhiqniX tffj Uw4iul[ ♦•• ". In AL, x, p, 60, " UtiqnL^ qnibqu wqw~ 
inma kjkui^ ji ^tujpkhkujij ... finuiku mp k kh*** ". (476,1) 

85 MiSe, iv, p. 92, " pjutjpLd k mjj^ mquiin Smpqjilj^ qnp^ nuwwujilpiu want— 
mhku juipgniSiji uiiuh£ " = ZP\ xxvi, p, 209, " k ujjl,p jnuwuihl{uiij, k uktumCg 
nfiwhg jfiLpwgiuh ^jup mn£5£ ". Elise. i, p. 10, " Qm.uq kmnd^p h Zmjnn 
Ifkbmn qmqmm k qwqwwnpjiq, k jwpgnLufi wuihf; qnumrnhjil^ Swpqpl^, ". 
Ibid., iii, p. 74. " ... imfkuiujh huijuuipwpgu qopogb jitpkwutj jjiLpwgiuuijiLp 
wwlif; ... piuqniS k wj^uijpnL&fi, np jwpgnLuji uiujui uihmji tp"> Idem, " nopu 
jwpgnLufi wwhlh " Ibid., p, 194, " fc/. pinqni-S ku wjj_ tuqmm fiwpqfik, kit 
np jiupgnLufi mwhi, ... ". LP\, xli, p. 231, " hwfuwpwpg k nkmntCp^ 
nummhji^ k nwiljitj^ ". Ibid., xciii, p. 554, " uuifuuipwpg k wqmmp k uijip 
jnumiuhfjlf Swpqlfwhl; ". MX, I, xxx, [See below n. 86 for the text. On the 
disputed problem of the osianik, see, Manandian, Feudalism, pp. 56-58, 90-104, 116-117, 
123-124; Sukiasian, Armenia, p, 101; Toumanoff, Studies, pp.114, 125-126 n, 215, 
Perikhanian, Ostaniks.] {476, 2) 

85 a [See above, n. 53, and preceding note,] 

86 MX, I, xxx, " fci. qnummhu uihnLinhkwj^ l^nqSmb^h wjuntjjik maw inn l— 
Ppiif * [i qmpdjpa unpm tuu^ ikutju fipp prnqmLnpiul^mh qiupifu " Ibid., II, vii, " fez. 
qmSiqu Al upu^wiqwuu qpwu wpgnLufi tquimpmuml qsnpuu, qi/f] jiLpwguiuihLp 
ljuinhipi^g [j hnjh ifiu qwpiffiij PwqwLnpuinh np fi Zwjgwj', np phq diufiwuwlpi 
dwSwhwlpj dumwhqni.pkwu Ji Cwpwhy umm^km^ qkqu k qmumml^kpmu : 
foul} wuui nLpkilli fyuipufiii pwquiLnpnLpkwu, npiq^u [uh£ y J w JlpS jwpnujkwi 
qnLuqu k nuwuiu mbntwuky ni qjiinki) pi ijwuu uiqumkwj wnqhh mnw9unj^ 
Pi ifiuuh phqq]?i)m.pkwu fipfrg ifrukipj ^lpumkwj_ k ji ping phlfhykw^ qwqqu y 
qtujju ft uikqfiu hntjw jutpnujjih qmuqu uihnLwSp wp^mufi : JXjl mnw9fihh 
Cm u in win ji quipi/jig piuqimnpuiiju umw^Iinij, nptqiu k wjdii ji *lpwzf uiyjump^jiUy 
np IJkijibnLqb l[n^ji ", 



524 



NOTES : CHAPTER XV 



Sep'cul < bQcgQ " sepuh " and ^£)C?o " son, child ", Marr, Etymologies, p. 286. 
[Cf. Toumanoff, Studies, pp, 92 n, 132, 130 n, 229, 408,] (477, 1) 

86a [O/. above n, 85.] 

87 Movses' concept of the bdesxs is extremely unclear. He mentions the bde$x& 
of Atjnik' and Gugark' [MX, II, viii ; HI, vi, lx], hut gives no information about the 
others. He does not indicate that there were four of them, or that they occupied a 
pre-eminent position at the Arsacid court. [On the bdesxs, see above, Chapter X, 
pp. 222-224 and rat. 69-73, and Chapter XIV, nn. 39-40.] (478, 1) 

88 MX, II, vii, " ♦ ♦• jumw£qlii)nLphh£ uiqwinwtjkww, npuf£u puuiwufjli 
Pwquiuipwij ". ■ . (478,2) 

8Sa [See above n. 85.] 

89 FB, III, xiii, " pwqSni.pfiLU Swpql^mh dnqp^jpqn^ hwjvwpwpwijh k ^mS 
^jiuutlfuiunLpkuibu ", Ibid,, III, xxi, " Jliqw umutLkj^ dnqnu^kguiu * * ♦ uuijuwpwpg 
ifk&wtfhbg. WLwqg, * * ♦ uijj^ k ji ^jihwl^muuig mhqmS nuu/fify Swpql^whh ". Ibid,, 
IV, iv, " q£nqknp ufipnju bnmuqh wb£p um wtfhhhufih, um ilbbuiiShbu k um 
Swul^muu^ um u^mmnLwl^muu k um whuipqu, um iwpnutm k um umgwwh, um 
mqmm k um ^ftuwliwhu ". Ibid,, V, xxiii, " k ufipljiu qhui um £uwwpwli ilbbw- 
i/kbg k ifingnLUg, ufwimiLUiliwhg k whwpqg, wqwrng L ^Jihml^mlimu^ 
djimuquiSmju ". Ibid,, V, xxx, LF\ xxxvi, p. 211, " ^muml} n£ kppkg quiqw- 
p£p qpki uwSuiljwufi um fj^juwhu k ^Jjhw^utuu L gw^wuwju w^jvwp^fiu 
Zwjntf*** ". Ibid,, xli, p, 231, " hmfumpuip^ k uknmi£g y nummhfi^ k nut lifting 
♦ • • ", Ibid,, lxxi, p. 416, " fei umhw^ puq fup qhwjuuipwpuwu k quiqwmu li 
wji_ wi/buujju niuifjjlfu *** ". Ibid,, lxxix, p. 472, " ♦ •• qpuiqiuiiu [i nmiljtlj 
fiwpqljwulu Zwjng ". Ibid., xct, p. 5^5, " kP u ^P zfowtywug Smpq^ w^fuwp — 
4fib Zwjnij ♦ *♦ ", Ibid., xviii, p. 109, " ♦•• h ifjiw^wfirim dnqni[pqntju • ♦• ". 

Ibid., Ixix, p. 408, " mSkhwjh hwfuwpwpmijh Zwjntj h tuui^wgfi ", etc, 
MUse, ii, p. 52, st pLuuikp^ k qumhpg wquimwij k ^fihm^mum^***^. Ibid., iii, 
p. 57, " dnqnilk^pu qpwqdm.pfii.u uipwuy k l^mhmh^^ ^jium^muuig h wqwmwtj 
qgui£uiuu3j]iij k qdhhwljkcjwij ", Ibid., iii, p, 67, " ♦♦» wpg k l^mhwj^ It wdh- 
hwju nuiujili pwqSnLppLUU *♦* li n£ mqmm u\mu\]^m^hm^ gum qqhq^ml^ 
ihmwijkw^ ♦•* ", etc, [On the extensive work done on the famih and the sinalcan, 
and the distinction of the two terms by Manandian, see the bibliography in Touma- 
noff, Studies, p. 127 and n. 222, also Eremyan, Slavery; and Sukiasian, Armenia, 
pp. 120-143, 149-156, as well as the Bibliographical Note.] (479, 1) 

89a [qf, Manandian, Feudalism, pp. 69-73, and the preceding note.] 
89iD ppor a detailed investigation of the various aspects, see, e.g. Buby, L'Fconomie 
rurale.'] 

90 FUse, vi, p. 131, " ♦♦♦ kp£ jwqwmmij fitjbu) hp£ fi ^jthm^uihuj^^ kp£ 
hl^hqh^tnj^ qfiu* Ifkuibu h. pnqkmj^ fighu^ kljhutjkh k Ifiujjlsh qJupuigwh^JiLp 
qmpmpu ♦♦♦ prnqphUg fiul^ kljfiu k dnqmjktjuiu, L Ipu^mu qjiLpwgwu^fiLp Ipu— 
WLwbu ", [Of. Manandian, Feudalism, pp, 27, 182, 197-201, 204.] (480, 1) 

91 FB, V, viii-xix. In each of the twelve chapters devoted to the victories of Musel 
Mamikonean, we find repeated the stereotype phrase; " qpuwtjnpquh hmw^lp k 
CuipJjJt IfwtjnLtfwulp " or " ft ^plffi bumwjm.pkwh l^m^nL^mh^p " ; in two 
eases the formula is " £wj}\ t} u lp ffitutijnpquiiju " [V, x-xi], Eut the expression 
" p £wpl{fi liwtjriLijwuk^ " does not mean " to levy a tribute " ; it has the more 
general sense of " bring into subjection, subject ". The complete formula is: " ft 



'^mw?zm 



NOTES i CHAPTER XV 



525 



£wpl[li < bmnmjnLphwh > IfwijnLijwui^ ", " to force into subjection ", (of I 
Mace, 4, " qtfuwtjnpn.uu ji implf/i fymtjnLijfiu mmj^ hnrjw Cjuplpj mfijmil}; "). 
MoreoTer the authenticity of this section of Faustus' History is known to he doubtful. 
[Of Manandian, Feudalism, pp* 85-89, 154-159, 166, 173, 233-234, and following notes.'] 

(480, 2) 

92 l*P\ xlv, p. 264, " ♦ ♦ ♦ Cwpligu wifhuwja W££utp£[iu Zwjnij urn fiu kh, 
li qnpbiulfw^ mdkhwjh jpSntS khnfiu^ hu^ zVi uij^ Lu pmqntS fympmufi 1 ' qnp fi 
tumpuplj qnpbml[miiuiju £mu[i np mum Jj £mju £fiu ", Of the king's words, 
Ibid,, xlvi, p, 269, " fei nj(mpljkp qjiSnj m^jump^fiu ZwjnjJ & ^ioj? tj^impul^mgu y 
npp uiqmhhph^ h qlfmpmufiu mjusmfu* qnp Zmuhp^ ^pm^mjbgm^ juuqphj^ ft 
mmul gnuifil h jnpqxng gng* iffiu^h. ij^mphugjj milbhmjuh ", Of Mise, vi, 
p. 133, " • ♦ ♦ qnq ku qmmuf;p ^wpfyji myjuwpifih ". After the Armenian 
rebellion, the Persians tried to pacify the country, LP\ xlii, p. 236, " k ji £mpfym— 
qpmphmh l^m^m^rnhk^ ". Of Mlise, vi, p. 131, " • •« pniipij^CpmSmjlp a^mpl^u 
mifvwpCph V (481, 1) 

93 JSUse, ii, pp. 22-23, " ♦ ♦♦ nfjapl^ m^juwp^jiu mpmLkj_ brnhprnynjij : ♦•♦ fez. qmju 
wilhhwju mnulp ♦ •♦ a^uwlimuiih ijmmhhu^^ *** k ns ng puqqjifimijmi. ufim y 
pkiqkm k ip bmupniPfjLU ^wplfmiju t 'fimhqji nmmji mpdmu ip mnunt^ ^mpjup 
nwCfslpuumh smiji^ \p\\} u mnunLJiu. hnjuiqiu k kuifjulfntumimtj It kpfigwuij 
qutjju, ti£ rijimjh ^jiumg, wj^ k mihpmlpu^ ; %m k puwi. n^ fiulf fywpi iqwmi/kj^ 
ifrnuu bmupnLpkmu 8mjiij li umlffig, pmdjig It £mufii3 jkpmhq A qm-smm^ h 
Smjpkmij ; f}^ pum mpgndtji wpdmumLtipm.phmuIt wnjjni.jiu, mj^ £jtuwpwp 

jmijj^mmlihjn^ i/jiuili jiughmhg jiulf ilkbrnuffcu nmpduihmjjiu, pi mumji wju 
wifhumju n.mti<k kjmhl; x qjtm°pri. p£h I[Wjgi wpjvmp^u ", 

[On sah, haz, wM, and Imrk, see Hnbsehmann, Grammatih, pp, 114-115, 234; Avdal- 
begyan, I AN A (1926, 1); Manandian, Imdalism, 205-208, 316-318, also above n, 91 
and Mow nn, 96, 97.] (481, 2) 

93a [Noldeke, Tabari, p. 241 n. 1, "Desname der Gxnndstener ist axabisch cJiardfj, 
wofor in Pers, etwa chardg anzusetsen ist, da das Wort im Talmud j^*75 lantet ; auffallen- 
derweise steht dies im Talmud abe rgrade fur die ELopfsteuer ,„ wahrend Grundsteuer da 
}^j?0O (arab, tasq) ist ... . Die pers, Kopfsteuer heisst in unseren Texten ijizja; 

das ist das afamaische gezUh(&) 9 welches in's Pers, als gezit gekommen ist Beide 

Ausdrucke, ohard§ und gizja sind in die Terminologie des muslimischen Eeehts auf- 
genommen ". Bee also MI, hJmrddj, mukasama, wazifa; OTiristensen, p. 124 and n. 3; 
Henning, Orientalia, IV, pp, 291-293; Prye, Persia, pp, 108-109, 218-219. Of Ehte- 
cham, I?Iran, p. 97 for the Achaemenian antecedents, and Jones, LBB, I, pp. 61-65, 
De*leage, Oapitation, for the reform of Diocletian.] 

94 Slrabo, XV, iii, 19 [L, VII, 182/3], st ErpaTsvovrai Se kclI apxovmv diro gikoviv 
ir&v £ws TTwrfjKQVTa ", (482, 1) 

95 Pers. ^ o^AJ = ^ Q ^* *bahrah, Syr, bahrag. This word is still used in Trans- 
caucasia in the sense of okamfart — malzTiegaL Muqasama4 is also found in the Tatar 
version, Joasimet. [See above n, 93a.] (482, 2) 

96 MatiJcdn-i-caimng, I, 12. Sahname, I, p. 247, [#ee above n. 93.] (482, 3) 

97 Baji < root baga 9 Skr. bJiaga, Zend, baz. In our opinion hah/rah also > it. Iranists 
link bahrah with Ay. ba%hra, meaning "tribute, food", Horn, Worterbuch, p. 56 > 
Av. Jiii-baBra, " fortunate ", Both forms presuppose the root bag with the suffix -Bra, 
and like bai, and the past part. ba-$a, pmjum, can have the sense of both " tribute " 



526 



NOTES : CHAPTER XV 



and " fate, fortune ". Baxra or bahra and barx f-j* < baffia, [See Meillet-Benveniste, 
Grammaire, p. 49, and above n. 93.] (483, 1) 

98 Uwlfiui,, " little ", lit. " parcimoniously, ealculatedly " and uwlpj^ " share, 
rank " {cf. Rus. Hac^eT'B) should also he related to it. (483, 2) 

98a [Cf. Lagarde, Arm. Siudien, No. 479, p. 35; Hubschmann, GrammatiJc, No. 19, 
p. 302. Manandian, Feudalism, p. 132, for the Armenian equivalents of the Sasanian 
hharaj and jizya, in his opinion, also above n. 93a.] 

99 Noldeke, Tabari, pp. 241-246 [See above n. 93a]. Luchaire, Manuel, p. 337, 
in southern France, the tax in kind, or champart, is called tasca. (484, 1) 

ioo Dwin Canons, viii, p. 190, " ijl[£u ^hnp/tjih Ifinjjli jwqwqii &kn.bwqpnLphwh, 
wnmm fi £uiplf£ h ji pklpup£ ". [Cf, Manandian, Feudalism, pp. 167, 208-211.] 

(484, 2) 

100a [Noldeke, Tabari, pp. 241-245, C( Die Konige von Persien pflegten vor der Regierung 
des Chosrau Anosarwan von den einzelnen Kreisen als Grundsteuer ein Drittel oder ein 
Viertel oder ein Eunftel oder ein Sechstel des Ertrags zu erheben, je nach Maassgahe 
der Bewasserung tind Bodeneultur im Kreise, ... Nun hatte Kawadh, Sohn des Peroz, 
gegen Ende seiner Regierung angeordnet, dass das Land, Ebene wie Gebirg, vermessen 
werde, um danach die Grundsteuer richtig zu bestimmen. Diese Vermessung war 
geschehn ; doch war die Sache bei ICawadh's Tode noch nicht vollig beendet, Als nun 
aber sein Sohn Chosran zur Regierung kam, Hess er die Vermessung zu Ende fiihren, 
... . Auf jeden Garib Land, der mit Waizen oder Gerste besat war, legten sie 1 Dirham 
Grundsteuer, auf den GarSb Weinland 8, auf den Garib Luzerne 7, auf je 4 persische 
Dattelpalmen 1 Dirham, auf je 6 gemeine Dattelpalmen eben so viel, desgleichen auf 
je 6 Olivenbaume ". [Cf. Chrisiensen, pp, 366-367 and Pigulevskaja, TBI I (1937) 
for the fiscal reform of Xusro I, The dimension of the garib is given by him as 2400 
square meters, Ibid,, p. 366 n. 2, whereas Manandian, Feudalism, p. 206 n. 2, following 
Decourdemanche, gives it as 1470.871/2 meters square, and Kremer, CulturgeschicMe, 
I, pp. 98-99, takes it at 1169.64 square meters.] 

101 Noldeke, Tabari, p. 246, [" Die Kopfsteuer legten sie Allen auf mit Ausnahme 
der Adlicheu, der Grossen, der Soldaten, der Priester, der Schreiber, und die (sonst) 
im koniglichen Dienst Besehafbigten, Sie richteten mehrere Classen ein zu 12, 8, 6 und 
4 Dirham, je nach dem grosseren oder kleineren Vermogen des Mannes. Die, welche 
noch nicht 20 oder mehr als 50 Jahr alt waren, befreiten sie von der Kopfsteuer, Diese 
von ihnen festgestellten Satze legten sie dem Chosrau vor. Derselbe gemehmigte sie, 
befahl, sie einzufuhren und danach die Steuern jahrlich in 3 Raten, jede Rate fur 4 
Monate, zu erheben ", Despite Tabari's specific indication, Chrisiensen, p. 367 
asserts that, " les impots furent payes par termes tous les trois mois "]. Simarrah is 
probably composed of si, se, " three " and hmar, CwSwp ** sum, count ". The present 
liquid measure, the unSwp^ or umiiiup — 2, or 3, puds is probably derived from this 
measure, probably as a result of the fact that a certain quantity of grain went into each 
simarrah, (485, 1) 

ioia [According to Deleage, Capitation, pp. 244-245, basing himself on the Theodosian 
Code, lt La perception des impots annuels 6tait repartie en trois termes .... Les percep- 
tions commencaient pour chaque annee fiscale le l er mars. Les possesseurs de biens 
prives devaient avant le l er juillet se lib6rer du tiers de leur prestation, tandis que les 
possesseurs de biens emphytSotiques ... avaient la faculte de ne se liberer pendant ce 
premier terme que du sixieme de leur imposition ... , Les prestations en nature etaient 



NOTES : CHAPTER XV 



527 



seules acquit6es par termes §chelonnes et par contre la taxe en especes etait reclamees, 
au moms aux possesseurs prives, en totality entre le 22 novembre et le 31 decembre ... ". 
Cod. Th., XI, i, 16, " Provinciates nostri tributa fiscalia per anni curriculum tripertita 
satiafactione restituant ". Ibid., XI, vii, 19, " Tertiam partem eanonis fundorum 
privatorum vel sextam eiifyteuticorum ex kalendis martiis, sieut consuetudo deposcit, 
reliquam vero eanonis summam ex kalendis iuliis par erit postulari ". Ibid., XI, 
xix, 3, " Ab enfyteuticariis possessoribus annonariam quidem solutionem per quattuor 
menses ita statuimus procurari ut circa ultimos anni terminos paria concludantur, 
aurum vero non ex die X kalendas deeembres in pridie kalendarum ianuariarum, sed 
per annum solidum, prout quisque pendere potuerit, inferatur ", Cf. however Appendix 
IC] 

102 EUSe, ii, p. 46, u \,w b ji um.pp. bl^bnjsnLnjh^ np £p mqmm ji *fipjiuwnu y 
pum fywpqfi huijuhhwn ifbpnn ji ulffyiliiiiit, ••• ", Dwin Canons, ix, pp. 193-194, 
" nji hpmhhjLnjh Qpjin.npji b uppnjh Spqwuituj qumpp hlfbnhijLnj dwhl[m.hu 
pbqmnmmuin uin£rfjiu bu l[inpn.buij^ b mnmm £jib uppnj bl^bnbnuij £nq b 2 nL P 
hmb ji fyiupujin hub uilpnLpkwuu jwjw mpmpbmi h *£ juiunubwi ji ffiimiljuiu, 
fojuwum.phwSp qbpjinnLubwn uwluuu ji qjiLinhjiu bqbin^ bu^ pwjg lfjiuijh 
qimuu inut^jiupgnLup ", [Cf, next note]. (485, 2) 

103 Ibid., xi, pp. 196-197, " f)Siuug jimjwmwtj ji fopwj mlpupmunnmn 
jipjuwunLpjiLU mnhbu L npiufcu ji l^buquihbuin £uipfyu tuui^uju^bh k ji IfbpuifynLpu 
b jpSinbj^jiu qjiLplwuij qnpbwljwiu SwwnLijwubh ", instead of the proper beha- 
viour: " £nqin^ b juuimfbj^ jjiLpiuginuijiLp ijwuuiwlpin ji fyu>[pj b ji £h&wh£ b 
juiifbuwju wpSuibuiij wiu^ Swuh b puidjiu *** ", Cf. Satiate Canons, in MelhV 
Tangean, Canon Law, p. 557, " jtul^ ji l(uiipj b ji {Jj&whj; b juij^n^ mpSuibwn 
uiutuiuunpqu £uiuijbu ". [Cf. Manandian, Feudalism, pp. 126 sqq., particularly 
129-132, where still more extensive use is made of the material found in Armenian 
Canons.] (486,1) 

104 Mov. Kaianh, I, xxvi, p. 99, " ♦ • • tywjiijy? ^r dnninjjin.bul; bpjignili* Iftupqu uiju 
ijigji* f^mjwhjiljh^ >npu qpjiL ijnpbwu^ ijbn qpjiL *{ m pj} ^ ijb^uiiuuuiu ijunu giui^ani.. 
b np uiuinwiqbwjh £* b nXmajiq Ijluh mwgl b qjihji x np£iuiji b ifuipnnh jin^ fwl{ 
npnj ijwp b uijqfi n^ ^ i/ji umnbu • * • b njp n^jump ^ ji mmh n^juwp iffi b hphg 
q^P pmpq b S^lj uj^rnhjip^ b njp &ji t p ji mmu jitjkh* ifjib jmjmhwl^ b uijp iquijup^ 
jiijbu* npP ilji " = Dowsett, Mov. Dasx-> p. 51 [Adontz's translation varies slightly 
in the case of the poor man who gives "half as much bread" (i.e. 2 griws of wheat 
and 8 [sic] of barley)]. In the Armen. Diet, p. 588, the word pntutujilj appears 
in the form n.pm.wufil{, the diminutive of qptiLWu, but this word is the Pers. AlaJ 
Pehl. iuvanih, " powerful, possessor, wealthy ", The canon with which we ar© con- 
cerned is found in one collection with the following reading, '* * * * ji priLuihjiljh np £ 

£^Z<? h UI lPJ' i *l wt 11 - S n P^ w ^ n P i l" 1 ^' H. W 7 /* 2 " ¥ z// / 7 /' ^ ^1 l j} usn ^ tP^IB* 11 - "' 

Dashian, Catalogue, p. 275. This interpretation is only a supposition of the scribe, 

perhaps suggested hj the confusion of pnuuhjilj with phn_, pkqiuujilf. Here too, 
ijiuju is replaced by ijiwnu " a jug, a liquid measure " = Pers. «A.^j> whereas 
ijiuiu = Pers. j^rU? * s watch, guard, part of the foot " and " en general partie d'une 
chose ". It corresponds to O.P. pah, pahr, Arm, upm^ iipu£p > pas, as puhr > pus 
" son ". The Arm. qpjiL = garib, a land measure, and subsequently a measure of the 
grain needed to sow such an area of land. Prom it is derived the Arm. qpnuuhy 
qprnwuguij, a measure of weight, now one pound. [Both the Arm. Diet, he. cit. 



528 



NOTES : CHAPTER XV 



and Hiibsehmann, GmmmaiiJc, p. 131 give the form qpnmihfjty "with the translation 
" modius ", and Hiibsehmann derives this term from griw. See above n, 100a.] 

(486, 2) 

105 MX, II, lix, " mjj^ k n* jwrikumju mkqfw hptypmqnpbnLpfiJi^^ p m JS 
nnjtju nLpkg* pum opfihmtyfi {,fjmfiumtymumg tynqSmhg ^mqqmlikpntpkmSp k 
w JL n i*P mjuwfiukog Ifkgkiu^g ". Emin attributes the words qnujh mqkf to the 
preceding phrase in his translation [as does Le Vaillant de Florival in the 1841 Venice 
bi-lingual edition, p, 287]. This is evidently a mistake, since it is impossible to believe 
tha£ even in Artases* time, the land was not cultivated " ... en tons lienx, mais sur quel- 
qnes points settlement " [de Elorival translation loc. tit.]. [NB. Dowsett, Mov. Dasx*, 
p. 50 n, 1, dates the Council " during the period AJD. 484/5-487/8], (487, 1) 

106 Myjtfw Gos, II, i = Karst, SempadscTier Kodex, I, p. 22-23, " />«// q^mptyu 
Ej.WLiijn.ujtj k mqnwij pmzj.wi.npg h fefuwug mpnmpnLpkwSp urngkh. tlfi fiu£ wt-k^ji 
gwh amnwQhnijh umfnpm.p fn.hu tym^qfih^*^ fez. uipq mjum£u ]_f}$f} * Quihqw— 
umwuwiju tpCfiuqkpnpq. Smuh wngku, *** JJnjuw^u k mjdS ifrgfi* qji mpbw— 
Pmqfih mhnwummh It mjnkuwwh It pnLpmummh ufi ifitffi phq. CmptymL ^hqbtyfj^ 
hdwuwu^lu k ppwqwtj k mnih k fuwhnip : JXjl Pi phmtyfjyph phn mpmbumji 
k IpuS phn ifm£mnfj {mplffinfih, qfi n£ \\ £mplj mfunj gpthumnh£fizj, f m JS m JlT~ 
mqahmq jnpdmS pnhft pun. £wpfywL mpmuqbh : puijn mhqmummhg Ppmppfi^ 
Lpgfiu put} iuqkfyuiL, fluty qprnfumg mmuhwhnpnhuijfih. qfi £nq ilfimjh £ Pwn.wi.npfj 
k fofuwufiy h J7S pntp* unjhm£u numfjhj), m J t lf 1 ^ °- bwnwumwhg : Hjuw£u 
h jmiMLpu ^wpuipnL jkophj; lJfih ipipbfitjfi fi^fumhfjh k w£pm.ufi. tujj^ WLbjJi 
m^fuwmbjh qhbnmdp mhtybwjh wufjpwLnLpftLU ilhb £ : kqfjh umwhkfiuu ujj^ 
Cjupty lift ifyf*) ♦•• ^fnifm ifiuip hnh iffiwju ifyfi * Upowfi wj^£wplj iffi [Jigfi^ •♦• 
fllfwupg ftnmnfjhuh wwuwhnpnhutjfih. * * ♦ Qfmj A ^npLnj k fi^nj ufi ifitjft 4wptyi 
♦ ♦* J) mwphSmmu pum l[U3pfj bmnwjkugf; h jmuitpu uioufjij ". Taxes on 
animals also existed in the West, " bovagram, vaceaginm, mnltonaginm, eaprhrum) 
etc. ", Lnchaire, Manuel, p. 388. (487, 2, 

10? Alishan, Sirah 9 p. 91 [Gf. also Marr, Ani}. (487, 3) 

io?a [j or SOIQe f Movses' etymologies, see, Tonmanoff, Studies, pp. 201, n, 22S, 300, 

303, 327. Eor his dating and appreciation of Movses' History, pp, 108, 306 sqq,, 330- 

334, et passim, also, The Date of Pseudo-Moses. On the problem of Movses in general, 

see above, Chapter X n. b.] 

108 MX, I, ix, " *** funp^nipq. fi Smfi hq.fi q.fjmk[y Pi njg nSwhg jmnm^ 
gwh ajid f*gku mfjphw^ m^fump^fju Zwjng** °. nmmfi uwfuuipiupnLpfiLugu* np 
mum ipuu : Qfi n^ Ifmpq.g fiu* fkutl mum jmjwufj*** L n£ qjjuminpm^ m^fump^fiu 
mnwpfiuu jmjmufj £ It n£ iflp^fiuu, ♦•• mj^ fumnu [j fum.au fi funmu milhumju 
k ifwjpkhfi"* Ibid.,ll,m, iit ^mnmp^ml}***tympa.u tykutjwmutywuuy npxmtfi Smppip^ 
uwCSmukwtj m^fumpifiu, k umfuwpwpm.pfiLuu, k unrjfiu uw£mpmpni.pkwunu 
hmimmkmnLpfiLUu Cmumwmkmij qmpu wftmrnufiu^ np ft nmhmlpn^ uwfuuuijh 
ifbp Z^jtytuj k jwjipij ". (490, 1) 

109 Ibid,, I, iii, " •*♦ 8kq funjq fuunpnju mnw9fi mpfymuk£ ^pljuip k 
£w£mt.np qnpbnif qmqnftu Jkpnj fympnk^ mummjfim.pfiLU. &>qfiL qpmqmmpmqu 
k uhmfumpmpmtymhmg mqnwij k mn£ilfirj^ p£ mf julM^ k qfib} fupmgwu^fiLp 
ng fi untjmuf; n.npbkmtj, k ntf ng ft gkqfigu npn^kpng pliwmufj k Smpmqhhmj^ 
k njg nSmug bfyg phmwuwijkmjD k dkpmniwijkmij)*** ". (490,2) 

110 Ibid., II, xi and I, xii, the king appointed Varaz: " fi^fumu wpgnLuwgmuwijh 



NOTES : CHAPTER XV 



529 



npuntj k ^kuu mwpn.kkm^ hSw inn Zpwqgwu nkwni^ ". Ibid. II, vii, to, Gabal and 
Abel, he granted, " ^hu wmpqlil ungm, jnpnij k whm.h Ifn^fiu uwin£u umjuwpwpnL- 
PfiLugu JXpkqlh k ^mphq^h ♦♦* ". Hid,, II, Ivii, to the newly come Amatuni, 
" mwmnLpu qfjLqjjL^ k rj.mumwlfkpmo£, • • * ", Ibid,, II, lxxxiv, the Mamiko- 
nean received, " jWLJiwkhwIjwIt fojumunLphmiip ntLkoqu k q^nwuinwfykpwu^ 
••• ", as did the Kamsarakans, Ibid., II, xlii, xc, etc. {491, 1) 

111 Ibid., II, lxxxiv, " • •• mwtj hSw jLULJiwkuwlfwu fojvujunLpkmiip nqkoqu ", 
etc. On the creation of new houses, II, xlvii, lviii, and lxiv. (491, 2) 

112 Ibid,, II, xiv, Tigran took away the possessions of the Vahewrmi, " pulfbunt 
nhnum ji gpilni.flkuf;hknri : knu jmpgnLufiu tunhnt •♦* ". Ibid,, II, li, Under Artases, 
the Mnracans lost their possessions, " n.pmt-1 jff^^u o^huu k nwdkuwju fyfumu- 
ni.pfiLUu untjui ". Ibid., II, liii, the king, " wmpn.kJ; qpmuu mpDULU^ np ji 
>t uu nq^piult k f]i-qinn LuilfnL.hu ••• "to Smbat Bagratuni, etc. (491,3) 

113 Ibid., I, iii, " nLumft umfumpmpnLpfiLUgu, np mum jimh jj pjiL umfumpwp- 
nL.pkwug lfmpq.£ pun. uwfumpwpnLpjiLUU. umfumprnpuLpfitu JXmw^nLukmij^ ", 
etc. Ibid,, II, vii-viii, " uw£wmbmm.pfjLU fl^mnLukmtji \]fiumlfmh^ 1jmn,ilkmu ", 
etc. " IfnLpmbmh wj;p Ibid., II, viii, ", IXp^mS mmhntmlp wij^jih JfnLpmbwu 
= Hpiiwd uwCmmkm IfnLpmbmh ". Ibid,, II, xlvi, xliv, " jwnw^mumijkw^ 
mmunLm^pmlfmu mmmjiL ". Ibid,, II, vii, !t lXuujw£wwkmli mm^jmL jmnuu 
mwuntml;pnLpku£ q^wuntuu npn^kmj^ ". Ibid,, II, xxviii. It is evident from the 
expression " ^[pLfu hmCmmhm unrjm n]Jmnjumif ". Ibid,, II, vii, that the head of 
house was in fact the nakapet. Movses refers in the same way to the patriarchs of the 
the Old Testament, as this is done in the Bible as well, Ibid,, I, iv. According to Mov- 
ses, the term hmfuuiprnp consists of hwfu and mpwp^ with the sense uwfumpmpkm^ 
as can be deduced from the following examples: Ibid., I, v, " qjnm npqkhmpkmh 
hwfumpwpmlfmumg bliuqng jkpfy mumfi npn,Lniju \?njfi ". Ibid., I, ix, " ft 
bhunjiy jkpfy umjuwpwpwlfmuwrju mjungjilf wpwhiju, QpnLrnh^ Sfrmwu, 
Smmkmnupty ". Ibid., I, iii, " qPmn.WL.npmijh k qhmfumpwljmumij wntfrng k 
mnii/fig '\ Gf, LP\ " fi umfuwpmp uiwwL}miimL.npwijuS\ (492, 1) 

ii3a p? or Mananclian's suggestion that some traces of the system survived to the fifteenth 
century, see, Feudalism, pp, 138, 251-260, 304-305.] 



/ 



APPENDICES 

The Appendices are an addition in the present edition and were not 
part of the original publication, although some of the material included 
in them was printed as part of Adontz's text, and much of it was 
referred to in his notes. In some cases, such as the new Greek version 
of the Life of St Gregory, the Appendices contain material which has 
become known since the publication of the Russian edition, 

Each of the documents included is given either in extenso, or, as in 
the case of geographical documents or such administrative documents 
as the various Notitiae, in their relevant portions, The test from which 
a given document has been cited will be indicated in every case, 
but for variant readings or editorial notes, the reader is referred to 
the original edition. 



«■ 



/ 



APPENDICES 

The Appendices are an addition in the present edition and were not 
part of the original publication, although some of the material included 
in them was printed as part of Adontz's text, and much of it was 
referred to in his notes. In some cases, such as the new Greek version 
of the Life of St Gregory, the Appendices contain material which has 
become known since the publication of the Russian edition, 

Each of the documents included is given either in extenso, or, as in 
the case of geographical documents or such administrative documents 
as the various Notitiae, in their relevant portions, The test from which 
a given document has been cited will be indicated in every case, 
but for variant readings or editorial notes, the reader is referred to 
the original edition. 



«■ 



L GKEEK AND LATIN LEGAL DOCUMENTS 

A. Codex Theodosianus * 
Liber XII Titulus XIII De auro coronario 

6. Iidem AAA [Gratianus, Valentinianus et Theodosius] Gaddanae 
Satrapae Sophanenae. Aurum coronarium his reddi restituique de- 
eernimus, quibus illicite videtur ablatum, ut, secundum consuetu- 
dinem moris antiqui, omes satrapae pro devotione, quae Komano 
debetur imperio, coronam ex propriis facultatibus faciant serenitati 
nostras solenniter offerendam. DaL XTIII. KaL Iul. Consianti- 
nopoli, Valentiniano A. Ill, et Eutropio Coss. (387). 

B, Codex Justinianus 3 
Liber I Titulus XXV I III De officio magistri militum 

5. Imp, Iustinianus A, Zetae viro illustri magistro militum per 
Armeniam et Pontem Polemoniacum et gentes. Cum propitia divinitate 
Komanum nobis sit delatum imperium, sollicita cura cauta diligentia 
pertraetantes perspeximus oportere etiam partibus Armeniae et 
Ponto Polemoniaco et gentibus proprium magistrum militum per 
banc legem eonstituere, tuamque magnitudinem, quae nobis ex ante 
gestis optime commendata est, idoneam ad talem fore dignitatem 
confidentes elegimus certasque provincias, id est magnam Armeniam, 
quae interior dieebatur, et gentes (Anzetenam videlicet, Ingilenam, 
Astbianenam, Sophenam, Sopbanenam, in qua est Martyropolis, 
Balabitenam) et primam et secundam Armeniam et Pontum Pole- 
moniacum tuae curae cum suis ducibus comrnisimus, eomite Armeniae 
penitus sublato, certosque subdidimus numeros, non modo quos in 
praesenti novos constituimus, sed etiam de praesentalibus et Orienta- 
libus et aliis agminibus segregatos, non tamen quantitatem eorum 
agminum minuentes: sed quia plures eis addidimus sine rei publicae 
gravamine et sine augmento sumptuum, aliquantos subtraximus, 



i CTh, II, 1, p, 731, 
2 CJC, 6th ed., II, p. 82. 



GREEK AOT) LATE* LEGAL DOCUMENTS 



3* 



ita tamen, nt et post hanc subtxactioneni amplioxes xemansexint, 
quam usque ad nostra felicia fuexant tempoxa. 

C. Codex Jtjstiniantjs 3 
Liber X Tiiulus XVI De annona et tributis 

13, AvTQKp&TCQp * AvaOT&GlQ$ A, ' 'AvBep.lO) 67Tap^O> TVOV TTjpatTCU' 

plcov. 

*Eav /xeV alrrjur) eVappa r) ttoXis /cov^toyxov Aa/Jetv rvx^rjs crwre- 
Aetas* r] 4tt67tt7)v r) 4£iawrr)v TrspLcfcdrjvai, dva</>€p4a9co /xev r) birjais 
avrwv eis* /Jao-tAea, /cat e£ 6-7rtAoy^s aurov e-TTir^Seios Trpds 1 to£to 
TTspmioBw opKov rrporepov SiSovs, /cat lav n p/qvvar) ofiros rots strap- 
X^iSf fJ/rjSzls tvttos BiSoudw iirl rr)v prqvvmv avrov, ct p,r) avrol oi 
eTrap^ot d^aStSd^cDcrt jSacrtAea Trdvra rd Trap* avrov dvaStSaj^eWa 
/cat t^rrjdivra, /cat ovtcos- Setos 1 £K<j>a)vr)dzir) tvttos d$eiAa>v Tracrt 
TpoiTOis rrapa^vXarrsadai, o Se Kadsls avdpamos Szrjasis Trept roiovrwv 
ivvoi&v fjufj €77tSiSoTa>* jU-' ] 7 Te §£ Kov(f>iup,6s r) p-stows St^ve/cats 7} rrpocr- 
Kaipws /x^re irroifsla psqrz iijtowois yiviaBw xwpls jSacrtAt/aJs £yypd(j>ov 
/ceAewews". MAAa /x^tc e/crayds TrotetTojow ot errapxoi fee/) xpovcov 
7rpoXap,f$av6vTtov rrjv avrr)v apxtfy* p^Te airrjpdaid nmv a^opi^irwaav 
rj aAAcos 'ttcos' xP^ll JLara 7} p.77 7rap€^o/xeya T7)v apxTjV* r) Trape^opeva 
flip, Sid Se tt)v eAAeti/ftv rtui' ravTa /coptfopeVa>p uwp,drwv dpyrjvavra 
r] Sid to rravdfjvai rr)v air lav o^oAdcrat, St* 77V Trapsixovro rrjv dpxv v * 
dAAd ravra Trdvra <?/c fiacriXiKfjs p*6vr)s avBsvrias ywiudvo Kara rdv 
#eiov eyypa<j>ov rvrrov, oi 8k p,r) rovro Trapa^vXarrovrss /cat to StSo- 
p,svov oiKodsv aTTohihorijooav /cat rraaav dXXrjv t^p^iav, rjv dv viroarfj 
rd Sr)p,6aiov* Mryrz Se VTrspdiaeis r) Trpodsoptias sttI tovtois o^€tAo- 
p,4vois 8r}p,oa(ois ivSiSorco ris r] Tas vsvop,iap,ivas TrpoOeap^ias X^P^ 
£yypd<j>ov Bsias /ceAevcrea)^. 6 Se tovto Kara avdsvriav ot/cetav 7rotwv 
otKo^ev /caTajSaAAeVa) Ta Kexp^coaTTjpiiva to) $rjp,ocr(cp. Mrjrz Se 
Ta 7roAtTi/cd xPVl xaTa > oaa ^ T( ? 8r}p,oviw sio<f>ipzrai r] rois ttoXzuiv 
d^cijotcrTat, ets iripas pLzrafapiodw ^pe/as* 77 TrpouvoTTOis rimv d$opi~ 
£4o9w xcupts* betas' /ceAeweeys". if at 7} rd£is Se tcov iTrdpx^y> €t /xt) 
Trdvra rd rr^piexopL^va rfj Siard^i ravra Trapa<f>vXd$;£i /cat StSd^et 
tovs- irTapxovs, /cat ot apxovrss rcov irrapx^v ko\ at Treido^vai 
rd^eis avrois /cat ot AotTrot SrjfjLouisvovreSf et Tats TOta^Tats /ceAevcrecrtv 
wovpy^aatev, ot/co^ev Si8oto>cw rr)v avp,fSaivowav rw S^/xocrta) 



3 CJC, 6th ed„ n, p, 402. 



4* 



APPENDIX I 



/JAd/J^v /cat d>s rov vojjLQV KaTa(f>povrjcravT€S TTGvrrJKovTa xpvcrlov 
Xlrpas TTpooTiixauduioav , Tpipsptos Se Trdvra rd Srjfioaia slofepicrdco, 
rd T€ d'AAa /cat rd Aeydjiteva *ApjxzviaKd, tovtsgti KaXdvSais 'Iavova- 
plais /cat KaXdvSais Maiais /cat rrpos rw reAet rfjs eVtve/xTycrews', i£ 
iacov rpioov fxepwv Siaipovp,iv(x>v tu>v Sypioaicov, /cat ju^Se/xtas' /catvoro- 
jjLias iv rw jJiiaip yivopbivrjs Kara rwv owreAcuv. 'EttziBtj Se rd 'App,~ 
ewa/cd reAecr/xara iv Svo /cara/JoAats awereAetro, e^ecrrt rots* ravra 
avvTzXovoiv, €i fiovXovrai, ttjv Trporipav uvvrjBeiav irpoTipav /cat iv 
Svo KarafioXais dvd tjixiuv /cara/JdAAetv, /cat to erspov tjiiiov iv rw 
27e77Te/zjSpta> rfjs fx^XXovarjs iiriv&p^u^ws /cara/JdAAetv. El Se /cat 
Tpipbepws fiovXovrai rd > Apjj,€ViaKa S^/xdcrta /cara/JdAAetv, i-gir woav 
rov sSsTrrifAfipiov pjr\va rr\s p,$KXovur\s imv^pLrja^oos rrpos viripdsaiv 
avTOis Se8o/xeVoj>. To Se rrpoaireaTaXjMivov /cara ovvBeaiv €lo<j>€p4a9w 
iv ra> TTpovojxiip iKdarrjS iTTiv^pLrju^cos, iTreiSrj tovto $7)Xoi /cat 77 
TTpooTjyopia avrov. 



D. h April. Paulo vc. cons. 



[a, 496] 



D. Novella viii 4 

£/£ indices sine quoquo suffragio fiant 

H AIATA3I2 IIEPI TOY TOYS APXONTAE XQPIE THE 
OIAEOYN AOEEQE riNEE®AL 

c avros jSacrtAeus ^Iwdvvrj iirdp-^cp TTpaircoptwv to 
j3\ drro vnaTvov Kal TrarpiKia), 
<C % IIpooip f iov^> . ^Andaas 'fjpSv rjp,4pas re /cat vvKras ovpbfSaivei 
/x€Ta Trdaiqs dypvTrvlas Te /cat <f>povTi8o$ Sidyeiv del fiovXevo pivots, 
ottws dv -^prfurov Tt /cat dpiuKov #ea> Trap* rjptdv rots' vtttjkoois Sodelrj. 
Kal ov irdpepyov ttjv dypvirvlav Xap,f3dvop,€V, dAA* et's Tot auras* avTrjv 
dvaXiuKopev j3ovXd$ St^/xepeuovreV re /cat vv£lv iv luco rats rjfjiepais 
XpwfAZvoi, a>ST€ tovs ^/-terepous* vTTTjKoovs iv ewa^eta yivsadai Trdarjs 
<f>povTiSo$ dTT7)XXayp,ivovs , rj(Jicov ets* iavTOVS rds* virkp aTrdvTWV 
jueptjuvas d^aSe^o/xeVaJV. Zltd Trdurjs yap ipevvrjs /cat ^r^crea)? d/cptj8ot?s* 
ipXopieda, irpdrr^iv iKeiva £,7)tovvt€S, direp o<j>eXos tols rj^Tipois 
vutt]k6ois sisdyovTa iravTOS avTovs diraXXd^i fidpovs Kal Trdoys 



4 CJC t 6th ed., Ill, pp, 64 sqq. 



GREEK AND LATIN LEGAL DOCUMENTS 



5* 



^rjfjLias €^w8ev €7T€i,$ayofjL€V7)s irapd rrjv SrjjJLOotav diroypa^v Kal 
ttjv hiKaiav re Kal vevopuopLivvjv ovvriXeiav , EvpioKop,ev yap ttoXXtjv 
irreiseXBovoav rois rrpdy\iaoiv dSi/aav, Kal ravrrjv ovk avuiBev, dAA* 
€K rivcov xpovaiv, ^laoafxivjjv rovs rjjxeripovs vtttjkoovs Kal els rreviav 
eXavvovoav, ws els TeXeiordrrjV avrovs drropiav KivSvveveiv eXBelv 
Kal fjL7)8e ra ovvrjBr} Kal vevojjuofjLeva rcov SrjfjLooiaiV Kal rais 
dXrjBeiais evoefiwv (f>6pa>v /card tt)f SrjjJLoolav diToypa^rjv hvvaoBai 
XWpls fJLeydXrjs dvayKTjS riBevai. IJws yap av loyyov ol ovvreXeis, 
r&v re e/c rivos xP^ vov fiejSaoiXevKorwv del ri KepSaiveiv e/c rrjs em 
rats djo^aZs 1 Trpoaycoyrjs f$ovXop,evu)V ? gIkqtu>s re rovrois aKoXovBovvrwv 
Kal rcov ev8o£ordru)v vrrapx^v, e/c re rrjs ivrevBev dSiKias rais re 
etjwBev ^TjfMiaLS rats re vevojj,io}j,evais evoefieoiv eirapKeiv elo<f>opaXs ; 
"Evvoia roivvv rjpZv yeyove, ri rrore av npd^avres airav, ooov iv rais 
rj^erepais i7rap-)(iaL$ eorlv €mj8Aa/?eV, rrpd^ei /zta koivtj rrpos ra 
Kpelrrw iAeTaorr)oaip,ev . rovro Se rrdvrws aTro^rjOov.evov evptoKop,ev, 
el rovs r)yovp,evovs rcov iBvajv, vooi ras ttoXitikoIs dpgds rwv errapxicov 
cloven, Kadapals rrapaoKevdoaifxev -)(prju6ai rats x € P ai Kai ^^vros 
drrexeoBai XrjfjLpLaros, jxovois dpKovjxevovs rois rrapa rov Srjpiooiov 
bioopievois. "Qrrep ovk av aXXcos yevoiro, el pjrj Kal avrol ras dpxds 
dfiloBovs TTapaXafijSdvoiev, ovS* oriovv StS ovres ovoe rrpo<f>doei rcov 
KaXovfjLevojv suffragicov, ovre rois ras dpxds exovoiv ovre erepcp 
rcov irdvrwv ovoevL eoK07rr)oap,ev yap oriirep, el Kal iropos ov fiiKpos 
eXarrovrai rfj fHaaiXela, dAA' ofiv rcov rjfjLerepajv viroreXoov eirihooiv 
fxeydXrjV Xap,j3avoVTWv, e'lrrep d^rjfxioi rrapa rwv dpxovrwv (frvXarrowro, 
7] re jSacrtAeta to re hrjiiouiov evBrjvrjcxei ^pcu/xeVo] vtttjkools sviropois, 
p,ias r€ ravTTjS eisayoixevrjs rdtjeajs ttoXXtj Kal dfivBjjros eurai rod 
irpdyixaTOs d<j>dovia. rj ov iraoiv ion <j>avepov s oriirep 6 x? vu>l0V 
oihovs Kal ovrw rrjv dpxrjv tovovfj,€VO$ ovk avro StSwai p,6vov, ooov 
7rpO(f>do€i rcDv KaXov}X€va)v iirevorfii) suffragicuv, dAAd Kal zrepov 
€^w6ev TTpos^TTiBrjoei rrXeiov 7Tpo<f>doei rrjs r(ov dAAcov rcov rrjv dp^rjv 
7) oi86vra)v ?j fjLV7jOT€v6vra>v depaTrelas ; Kal p.ias dpxTJs droirov oodziorjs 
TToAAd? avdyKT) ^etpas* TTSpivooTziv rov rrjs ooozcos dp-go jAevov, Kal 
rovro ok to xP v ®*> 0V °vk oiKodsv lows TTape^ctv, dAAd o€oav€iop,4vov, 
Kal Xva oaveioaodai SwrjOeirj, £,rjjMovjj,€Vov, Kal ovA^oyl^soB ai /car' 
avrov, or 1 rrposrJKov ion rooovrov e/c rrjs irrapx^s Xafielv, ottooov 
oiaXvoei fjiev avrcp ra 6(f>Xrjfiara ? K€<f>dXaid re Kal tokov 9 Kal ras 
vvep avrov rod SaveioaoOai Zprjixias, Scuaet Se Kal rrjv iv fxioa) oaTrdvrjV 
Sai/jiXeoripav re 7]Srj Kal dpxovri Kal rois dpu^ avrov Trpirrovoav, 
Kai riva iavrw Kal rrposarrodijoeTai rropov Kara rov i£fjs y^p6vav> 



6* 



APPENDIX I 



kcl9* ov luws ovk ap£ei m costs tov Trap* avTov $i8op,4vov TpinXdaiov, 
jjl&XAov Se, si Set TaXrjdsuTSpov sIttsiv, bzKairXdoiov to Trapa tvqv 
r)iiSTSpwv vrtoTsXwv stsTTpaTTOixsvov earat. *Evtsv9sv tg Kat to 
SrjfjLouiov sXaTTw9iusTai' a yap sxpfjv e ^ T ° 8r)ixomov slsax9r)vai > 
tov ttjv o>pxT)v e^ovTos Kadapals xpojjiteVov Ta ^ X € P U ^ ^avTa sis 
ttjv olKslav 9spa7rs(av Aaficbv 6 ttjv dpxyv sxwv amopov ts a7ro<f>r)vas 
Tjfuv tov avvTsXsQTTjv , ttjv diroplav sksivov ttjv St' avVdv ysvofxsvrjv 
rjfMV viroXoyl^STai, Iloua 8s dcrsfir] /cat aAAa ytverat sis rr)v twv 
kXottwv toitwv sIkotws dva^spopisva Trpo<j>auiv ; ol yap 817 Tas dpxds 

GXOVTSS TCLS STT^WplOVS sis TO A^/X/Xa TTpOS^XOVTSS TOVTO TToXXoVS }XSV 

tcov virsvBvvwv a<j>iam 3 ttwXovvtss avTois to TrA^^eAi^a, TroAAovs 
8s twv dvsv9vvwv KaTaKpivovaiv , Iva tois vttsv9vvois xaptcrcu^Tar 
Kat tovto ovk sttI tois xP 7 ]P' a ' TlKa ^ p>dvov TrpaTTOvaiv amai^ aAAd 
/cav Tot? ivKArjfjLamv, kv9a Trspl $vx i rjs eWtv 6 kivSwos* <j>vyal ts sk 
Ttov iirapx^v yivovTai> Kat uvppiovmv evTavftx irdvTss 68vp6pisvoi i 
Ispsis ts i<al fiovXsvTal /cat Ta^eaVat Kat KT^TOpes 1 /cat S^oTat /cat 
yswpyol, Tais twv dpxdvTwv kXottoIs ts sIkotws /cat dStKtatS' p,sp,~ 
tj>6ixsvoi, .Kat ov Tavra 807 ytWrat jiova > aAAa /cat at t<&f TroAewv 
crTaaets /cat ot 8tjjmw8sis dopvfioi to. ttoAAo. xP r )l J ' ( ^ TWV ywovrai ts /cat 
mtvovTat. /cat oAa>s p,ia tis sqtIv avTrj irdvTwv d<f>op}j,r) twv /ca/ccuv, 
/cat to ys dpyvpoXoysiv Tas dpxds ttvutjs eort Trovrjptas TrpooijMov 
T€ /cat wspas* /cat Icftw dpa Kat tovto twv 9slwv Xoylwv davjxaoTov 
T€ /cat aXr)9soTaTOV to r^v (f>iXapyvplav irdvTwv etvat psqTspa twv 
KaKwv, Kat jU-aAicrTa 6Vav ^ Tat? tcuv tSta)TCov, aAAd Tats tcDv apxovTwv 
iyysvrjTai ^v^ats*. TVy ydp ov/c a^ aKivSvvws kXstttoi, tis Se ovk dv 
Xr)UT€vvzi€v dvsvdvva, sis ttjv dpx^)v dTTofSXiirwv KdKstvrjv opwv 
ovrravTa ^pvcrtov TrmpdoKovoav, /cat 9appwv ws, oirsp dv irpd^sisv 
cltottov, tovto xPVH ,aTa Sovs i^wvrjasTa ; svtsvBsv dv8potf>oviai ts 
/cat /zot^etat Kat s(f>o8oi Kai TrA^yat Kat dpTrayat irapdivwv Kat Travrj- 
yvpswv avyxvusis Kat KaTa<f>povr)osis twv ts vo\xwv Kat toV dpxwv, 
irdvTwv avTas wviovs trpoKsiudai voimIovtwv, wsnsp Tt toV KaKiaTwv 
dvSpaTroSwv* Kal ovk dv dpKiaaifxsv irpossvvosiv ts Kat dcfrrjysiadal, 
oTTOcra Kat 6K t^s kXotttjs tcuv iirixwpiwv apxovTwv ylvsTai ^aAeTra, 
ovScvos 1 avTots dappovvTos p,STa Trapprjulas siriTip.dv, sksivwv svdvs 
to Tas dp^ds wvr)aaadai Trpoiuxofxsvwv, 



CAPUT I. 



TavYa a7ravTa Ka0* iavTovs f$ovXsvudp,$voi Ka^Tav^a Kotvan^dv 
tov povXsvjjLaTos TrapaXafiovTSs tt)v £k dsov SeSojLte^v r)pZv svosfisa- 



GREEK AND LATIN LEGAL DOCUMENTS 



7# 



rari)v gvvoikov, /cat rrj urj ye vrrepoxjj to rrpayfxa Koivwaafxevoi /cat 
rt /cat jrapa Trjs orjs AafiovTes fiovAfjs, €7rl tovSg tov delov A iArjAvdapiev 
vqjjlov' St' oS Oecrrrtlopiev, /x^re avQvnareiav jU/qSe^ttav /^T€ tt^v ^%pi 
vvv KaAovfJL€V7)V jiiKapiav firjre tov kojjltjtcl Trjs iu>as p,rjT€ d/\Arjv 
olavovv apyj]v t pirjTe VTraTiKrjv \iryre rjyepioviKrjv, as 8r) KovaovAapias 

KOI KOpp€KTOpiaS KaAoVGlV (&V TIVWV pTjTCQS lX€jJLV7]Tai r) vrroKeipievr} 

rcbSe rjjjLcov tw Beiw vofxcp airoypa^rj, as 8rj /cat jxovas vtto tovSg tov 
vop,ov ayofiev), StobVat ri suffragiov jLt^Ss vrrkp rrjs dp^rjs ttjv 
olavovv 86aiv p^ryre ap-^ovn p,r)8evl ju.tJt€ twv irepl Tas dp-gas Ttvt 
\x,t\tg. STepco TrposcoTTO) Kara 7rpv</>acnv TrpouTaulas* aAAd TTpdiKa p,iv 
KOfAL^sudai rd$ ap-^ds, pieTpia ok 7rap4)(€iv irpo<j>do€i rtov vrrkp eKacrTrjs 
SiSofjLtvwv avpifioAoov t<= /cat ^apTcD^, ifat yap Sr) /cat vrredf^Kapiev 
diroypa<f>r)v to>§€ to> deiw rjpLWV vojjlw 8rjAovcrav, rl rrposrJKOv iariv 
iKaarrjv dpxrjv Trape^etv rj els to Beiov rjpiajv latereulov rj els to 
SiKauTrjpiov Trjs crrjs vrrepoxrjs 7Tpo<j>duei t€>v kcoSiksXAcov rj ovpifioAwv 
rj TTposTaypLaTUiV costs eKeivo [tz] uvvecrTaAdai /cat jj,r) Trape^etv 
avToo pieydArjv atodrjcriv. 



CAPUT II. 

B *Ek€ivo jxevTOi Stopt^o/xeVj to xprjvai tov fiiKapiov Trjs *Acnavrjs, 
ovTa 8k /cat ap^ovTa Trjs TIaKariavrjs &pvylas, jj,t)k4ti p,€V ovtw 
rrposayopeveadai, aAAd tov Aoittov Kop/qTa 0pvytas IJaKaTiavrjs 
oVo/xa£ecr#at, /cat /co/zi^ea&xt e/c tov Srjpiocriov, drrep Kal vvv rrpo^doei 
dvvovwv T€ /cat KarnTaTiuyvoyv vrrkp iKaripas dpx^js eAaju/Javev, ovBsvos 
iAaTTOVjxivov tovtwv Kal fj.rj 8vo Tafjeai ^p^cr^at, aAAd dVa/xtyetcrav 
SKaTepav, Tfjv t€ tov dp^ovTOS Trjv t€ tov fiiKapiov, fxlav yevevBai, 
KopiiTiavrjV ovadv T€ /cat ovopLa^ojjLsvrjv, rod kiv8vvov tcuv SrjfJLoalaiv 
tj>6pcov aifTw ts Kal rrauw 6p,olcos sttovtos — ■ ota ju-tas* rd^ews Ka9saT- 
worjs, p,r) 8ir)pr}fj,4vr}s avTrjs TravTsAws, aAAd /caTa /xtav drrdvTwv 
aTpaTSvofjLsvojv crwe^etav — , KOfjLi^ofjLSvrjv }j,4vtoi Kal avTrjv 8id to 
SittAovv tov kivSvvov Tas dvvovas Kal KamTaTiwvas, drrzp ZKaripa 
TTpoorjV £kojjli£,€to Tafi?. p,T} jir)v iripas tivqs apx^iv tov irpa%r\v pkv 
fiiKapiov, vvv Se TrepifSAsTTTOv KopurjTa Trjs IJaKaTiavrjs &pvytas, 
ovk k*xovTa rravTsAws ov8s}xiav fxsTovu lav iv Tats aAAats tois Trjs 
'Aaiavrjs Stot/c7]aea)s eTrap^tats", dAA* l^ovTa p,kv Trjv tov TTZpifiAeTTTOV 
KOjjirjTos Trjs IJaKaTiavrjs 0pvyias irrvowiiiav, apKovp,€.vov 8k pLovrj 
tjj TlaKaTiavfj, Kaddrrep sIttovtss s^drffxev, &pvyla. 



8* APPENDIX I 

CAPUT HI, 

r Avro Se rovro /cat irrl daripov tov irp<jbr)v fiiKapiov Stopitopev, 
(j>ap,GV St) tov Kara ttjv JJovtik^v Sioiktjqw cosre p/rj Svo Kadeardvai 
to Xoittqv, dAA' eVcij Kopyra p,kv FaXarias 7rpwT7)$ 6vop,alop€Vov ■, 
/cat ^x oVTa Ka * T V V Kar ^ T & v arparicorcbv i£ovoiav, /ca^dVep e^et 
/cat vvv 9 /cat tcls GKaripas dpxV$ KopLi^opLGVov criTTJveis, ov p,fjv e£a> 
rrjs irpa>TJ}s JaAaTtas. ovSe/xtaF yap avrw iripav TravTzAtos StSo/zev 
l^etv i^ovaiav /car* ov8epn.as IJovtiktjs hrapxias* dAAd Kara p,6vy)v 
JaAcmav rrjv TrpwTrjv. ttjs Ta^ecos t€ 6p,oiws dvapuyvvp,€vrjs /cat 
Kara /xiav, chs etpTjTat, voovp,€V7}s /cat api8p,ovp,4pr]S avviyGiav , /co/i.t- 
Tiavrjs ovu7]s re /cat ovofialopLivrjs* /cat ov&svos TTavTsAcos e| avrcov 
Trpos tovs dAXovs €^ovt6s nva Sicxjtopdv, dAAd jutav etvat rd^iv, vfiivl 
TGTayp,4vr)V apyovri, p,ias eTrap^tas rjyovp,4v(x>* ird(77]s 6p,oia)s rrjs 
rdisws a/xa rat o<j>cov avrcov dp^ovn TTGpl rd 8r[p,Q(jia kiv8vv€.vqvqt)s . 

CAPUT IV, 

<^]> OvSevt §e ap)(QVTt iravreXtas e^tejuev ot>T€ TroAm/ctp ovre 
urpanwTiKw skits pxrew iv rats rroksaiv ttjs hrap-gias, tjs dp^et, 

TQVS KaXoVftivOVS T OTT OTT) QT)T d$ > yiVWUKQVUlV, (X)S €t TOIOVTO Tt TTpd^aiSV, 

avTol iravTeAcbs e/CTreo-owTat rrjs dpxTJs °* BappTjaavTSS iripovs sis 
ttjv iavrcov rd^iv e/^8t/3d£eti> \ 

CAPUT V, 

JE Avro 8z rovro <f>ap,kv koX £ttI tov Aa/xTrpoTdVov Koprjros rrjs 
icpas /cat tov Aa/ZTrpoTdVov apxovTos, /cd/cetcre yap p<iav dpxty dpj>o- 
Tspas TroiovpLsBa, e^ovTos ftev /cat to tov TrepijSAiiTTov ko^tos ttjs 
*Etpas 6vop,a > Tarsals Se puds ap-govros Kop,iTiavT}s ovorjs T€ /cat 
6vop,a^opL€V7)s> /cat ttjs' TrpwTTjs fiovTjs Svpias /cat to>v KvpprjuTtKWV 
rjyovpiivov^ /cat rds ixarepas dpx^js e^ovros crtTTjcrcts. e^ taa) yap 
rots jSt/captots* /cd/c€t^ov rWep.ev, wsts a/xa /cat avrov r'g TTGidopevy) 
avT(p Ta^€i KwSvvGvsiv vrrep T6 ttjs tcov St] pLOultov etsTTpd^ecos virip T€ 
rrjs ttoAitik7)S /cat 87]p.oaias /caTaoTacreaJS'. 



CAPUT YI, 



S* BovXo^dd ye pbrjv airacri tois ap^ovcrt tcov ^/xerepwv e^ap^tajv 
TravTas 1 t577o/cetcTc9at, tous ju-€v tStcoras /caTa to t^s dp^yjs tSiov iirl 
Trdaais airiais Kat Tracrats ^p^/xaTt/cats' T6 /cat ey/cA^/i.aTt/catS' TTpo^dascn, 



GREEK AHD LATIN LEGAL DOCUMENTS 



9* 



rovs Se ye iv arpareiais ovras Kal vtto ISikqis dp-govras rerayp.evovs 
Kal rovrovs qvSgv r)rrov 7rpotj>daei 8rjp,oGicov re Kat iyKArjp,drcov 
VTTQKeiudat Tract rpo- <lZ^> ttois avrtils* MAAd Kat rovs ivrevdev 
Kanovras e£ olov8rJ7Tore StKaar^ptov Kal rds olasovv p-era^etpt^o- 
p,evovs ^rjt^ovs i^elvah rovs rcov eVap^tdV dpyovras p/rj avy^wpeiv 
rrXelov ri rwv rrj 6e(a rjpLcbv Stard^et Sirjyopevixevcov Xap,f$dveiv spoxtul- 
cov > yivcouKovras cbs, ei rovrov padvp/fjuemv, irauav t^p.iav ivrevdev 
rois rjfjierepois vrrorekemv hrayojiirnqv avrol Karadyabvai* Ai8op,€V 
Se avrots aSetav /cat yvcopi£,eiv rd rrepl rovrov purj p,6vov els rds ap^ds, 
e£ wv elalv oi areAAdjuevotj dAAd Kat ets ^juas avrovs, cosre r)p,as 
ravra yivwaKovras tw rrpdyiiari rrposrjKovrcos eVe^tevat, El Se 
Kat airoi rivas evpoiev 8id rrjv eK rrjs dftas r) rrjs fcov^s vrreporfsiav 
rovs rjpierepovs vVoreAets dStKovvras* dSetav avrots StSo^tev /cat 
e£era£etv rd dStK^jLtara /cat rovs VTrevdvvovs evpiuKOfievovs dtj>aipeiodai 
rrjs Icovrjs /cat r^v rjp.erepav rd^iv iv reus eVapj(tats trT^qpovv, rovro 
virep Kal rois dp^atots Sfqydpevrat rcov vop,cov. cosrrep yap avrovs 
iravros dStKov KepSovs eipyojxev, ovra> /cat KaBapcos rats dp^ats 
Ke^prj^evovs Trdarfs rip.rjs re Kat atSovs Kat aep,v6rr}ros cbroAavetv 
Becmi^oiiev* 



CAPUT Yll 

i? Ovrco roivvv r)juv rcov dpycbv 8iaKeKpip,evcov rrpoarJKei rov 
ivravBa rrapaAafiovra rr)v dp^v jxerd rrjs rov Beov p,vr}p,T)s evavrlov 
r)p f cbv > r) etrrep rjfuv ovk ew? a^oA^ ivavriov rrjs Te urjs vrrepoyrjs * 
Kat rcov del rov gov KaraKoufMrjaovrwv Bpovov, rov re aet ivSo^ordrov 
Kop/r^ros rcov Beicov r)p,cbv laTgitioncup rov re iv8o£ordrov quaestoros 
rov Beiov r)fj,cbv rraXariov rov re ev8o£ordrov Kopsqros rcov drravra^pv 
Beicov rjfMcov privata>v, rrapovros 8r) Kal rov Kara Kaipov p.eyaAo7r- 
perceurdrov -gaprovkapiov rcov Beicov rjp,cov KOircovcov rov ro%s ovjj,J36Xoi$ 
rovrois rots* 'Trap* ypSv virTjperovfxdpov, opKov StSdvat, pwjSevl Travr^Atos 
/x^Se oriovv Trape^ety ft^Te irpo^doei Sdcrea>s p-^Te TTpoaraalas, ft^Se 
eirayyetAaa^at^ p.7jSe e*K r^s e7rap^tas" djuoAoy^crat creAAetv, ft^Te 
Tots e^So^oTaTOts' irrdp^ois prffG rois aAAots* rots' rds dp%ds e^owt 
ju-T^re rots rrzpl avrovs Kadsorcoat, ft^re erepa> rtvt Kara 7rp6(f>acnv 
rrpoaraaias* aAA* aJs^rep dpLia9ov Aa/x/?dVet r^v dp^p, 7rpoakap,f$dv€i 
re rrapd rov S^ftocrtov rds o"tr^orets ( ravras yap 8rj koI p,6vas Aa/xjSd^etv 
avrdv e^te/xe^), ovra>s avrrjv KaOapais ^vAa^et rats yspvi, dew re 
Kal r\pXv rov virep avrrjs v<f>e^cov Adyo^* Vara) ydp r) or) virepoxr) Kal 
oi fxerd ae rov avrov em^rjo , 6p f evoi Bpovov > cos* elre avrol dapprjvemv 



10* 



APPENDIX I 



Aa/Jetv ri irapa tcov els ra$ elpjjpevas apxds irapiovTcov evre ol 7rapa- 

SwatJTGVOVTGS ai>TOlS €tT€ /Cat Tj T&£lS 7} U7] TTepaiTepCO TCOV 7TpO<f>d(JGl 

uvv7]8eicov Trap r^icov avTols wpiufxevcov (arrep Srj Kal dp/cetv /xdva 
vofit^ovTes StSoa&xt 8icoplaap,ev), cos ovk iv fjLiKpols rj ttoivt) yevr\aeTav 
dAA' ol p,ev jxiyiuroi apxovres ol Aa/Jetv n dapprjcravTes Trapd tcov 
iirl ras dp^as napiovTcov rj Kal avyxtoprjaavTes r V olKela rdtjei toiovto 
ti TTpdrreiv, Kal irposayyeABev ov BepairevovTes , <hs ov \xovov TGrparr- 
Adcriov aTTohwuovui irav oaov elX^aaiv, dAAd /cat fxeydXrjv dyavd/c- 
T7)mv VTTOGTTjcrovTai Kal tov irrl t# dpxfj kivSwov evAafiyjOrjcrovTai,. 
Kal 01 ye d/x<£ J avrovs ovTes Kal r) TreiBojxevTi rdtjis avTols, el irXeiov 
ri tcov Trap 7 rjfxcov 8e8ofievcov eTTix^pTjoaiev Aa/Jetv, avrol re vTTOKei- 
uovrai tois it > 7]p l icop,evois rfj els to rerparrXovv drroSouei eKTreuovvrai 
re Kal ovcrlas Kal ^covrjs 9 rrpos tco /cat ripLtoplais VTrofidXAeadai Trpe- 
TTovuais tois TrXrjfXjjLeXrJiJLaui rots avTcov, 



CAPUT VIII. 

& Tovs 8e ovtcos djxiadovs Trapa\ap,f$dvovTas ras dp^ds TrpcoTov 
drrdvTCov O7rov8aop,a 4'^etv xp 7 ) to rois 8rjfj,ocrlois dypvirvcos Trpocre^etv, 
/cat tovs* p<hs dyvcop,ovovvTas Kal 8eop,evovs avdyKrjs jiera rrdarjs 
elsTTpdrreiv rrjs v^oSporijTos, [MTjSev VTTOKaraKkivop.evovs p^8e vrrep 
avrov tovtov KepSos ti iravrdiraoiv evvoovvras, rols 8e evyvcbp,oai 
TrarpiKws 7rpos<f>epop,evovs* iireiTa to tovs rjfxeTepovs vtttjkoovs 
<f>vXdTTeiv rravTaxodev aveirripedoTovs > ov8* oriovv Trap 5 ovSevos 
avTcov Kopulop,evovs* dAA' taot p>ev iv Tats St/cats, ^croi 8e iv Tais 
8rjfxw8eui KaTaordaeaw evTtouav, iirei-iovTes Te tois dp,apT7)p,acn Kal 
tovs p*ev dvevdvvovs TravTaxddev <f>v\a*TTOVTes KaBapovs, tois VTtevBvvois 
8e eTTiTidevTes irpos tov vojjlov tt)v TroivrjV, Kal ovtcos apxovres twv 
vtttjkocov cos aV TraTepes vlwv, dyairwvTes p*ev avrovs dvevdvvovs 
ovTas* virevOvvovs 8e <f>aivop,evovs uw^povi^ovTes Te /cat Tip,w povp,evoi> 
Kal irauav SiKaiouvvrjv ev Te tois 8tj fjLoatois ev Te tois l8iois ovfifioAalois 
avrois 8iaT7)povvTes* Kal ovk avrol p,6voi tovto irpaTTOVTes, dAAa 
/cat tov dei 7rape8pevovTa toiovtov Xap,fidvovTes /cat tovs Trepl avToi>s 
airavTaSy tbs p>t) So/cetv eKelvovs p>ev 87]6ev dvevdvvovs e&at, St* eTepwv 
8e TrArjiXpteAeiv Te Kal KXeiTTeiv, tovto orrep eVt p,aAAov alaxpoTepov 
iuTi to Kal koivwvovs tcov d8iK7)jjLdTU)v Aa/xjSdvetv. "Qsre e^eorat tjj 
orfj vnepoxj} twv uep,voTepwv Tivds irrl Tas dpx&S irepmew Kal tcov 
eTTiaTajxevcov to. STjjudcria, fiovAevTcov Te <j>ap,ev Kal erepcov TrposcoTrwv, 
Trelpav eavTcbv 8e8u>KOTcov dyadrjv /cat Trpos tcls dpxds e7TiTTi8eiwv, 
tis ydp av ovk ayaTryveis Kal aevvoTTjTos ep/neTrX^aBai p,ey 1 dXrjs 



GREEK AND LATIN LEGAL DOCUMENTS 



IP 



vojjLiaeiev, etVep rjjxsripa i/rrjcfxp Kal /cptVet rrjs vrjs VTTspoxrjs irn ttjv 
dpxty TrapiXdoi, }j,€p,apTvp7]iJ,€VQs p,kv ws sir) xpyjuros, rrpotKa Se 
avrrjv Se^ojuevos*, ovk ivrjaxoXrjfxivos Se TravrzXws ov8<=vl <f>avXw Kara rr)v 
-)((i>paVy ovSk ottws to 8odev a9pol(j€i€v, ov8k O0€V cruXXiijzis -ftpvuiov, 
aXX* tva Si) tovto jmovov l^ot CTrouSacr/xa to too 9zto re /cat rjfuv iavrov 
avuTTJucu, /cat 86^rjs drroXavaai XP 7 ) CJT V S '> Ka ^ dpoifids iXTTiaai peydXas ; 
El 84 ns irapa ravra n irpa^sizv, terra? /cat cols' iirl ttjs dpxys, *4>* 
fjs ion SiKaorrjs, kXotttjs alriav vTroarrjaopLevos* Kal etye <j>av€ivrj 
Sovs x ov(J t° v virkp rod Aa/Jeti; rrjv dpx^)v V Xaficbv e/c ttjs dpx'rjs 
fe/cdVepov yap ofxoiws vrrzvdvvov), on /cat Brj^evaiv zeal i^oplav 
vTToorrjosTai Kal rrjv sis ro ucbfia fiauavov re Kal rijj,u>piav, /cat 
avTOF 8i) rov Xafiovra Trap 9 avrov, Kadamzp shrovrss s^dr^xsv, 
KaKois vTroOrjasi jueydAots 1 . KaBapds yap dirairovixsv stvai Tats* 
hnxwpiois apxais ras ^€t/>as , > tva rovs dpxopisvovs dfyfjitovs rs /cat 
svdrjvovfxivovs <f>vXd^aiiisv '. iTat ai?Tat }xkv €/c ts twv vofJLeov e/c rs 
rtov apx&v iiriKsivovrai rroival rots iv rals slpTjjAsvais dp^ats" odaiv, 
€t n toi ovro irpd^aisv. AtSofjLGV 8k Kal rots iirapx^torais d8siav, 
el Tt Kara ttjs iirapx^s dSt/cov 6 rrjv apxh v ^X wv 8iaTrpd^i)rai 
Kal Ztfi&iais nalv r) iTrrjpsiais TrcptjSdAAot rovs r)p,sr4povs VTrorsXsiSy 
c5st€ rov 6so(f>iX4ararov iirluKOTrov /cat rovs iv rfj x^P a TTpwrsvovras 
hsiqusis sis rjfJLas dvarrsp/Trsiv , KaraXiyovras rov rrjv apxh v sxovros 
ra 7rXr}p,iJ,izXrjij,aTa* rjjxsis yap ravra fjuavddvovrss ursXovfxsv iv rrj 
X<*>pa rov ravra i^srdoovra, £<j>* <S T€ avTov, zvda rjSiKTjazv, e/cetcre 
/cat tcxs Troivas wocr^etv rtov TrX^fZfx^XTjfxdrcov costs jit^Se zrepov riva 
roiovro ti irpa^ai dappijuai irpos ro mtpdSety/za jSAeVovTa. 



CAPUT IX, 

* AvdyKr\v %xovros rov rrfv apx^v 8i4itovto$ Kara ras efiTrpoudev 
8iard£eis, iTrziSdv Karddoiro rrjv ^wvtjv, ras TTSvrrjKOvra rjfj,€pas iv 
rfj GTTapxia 8iarpif$€iv S^/zoata ^atvd/zevov, /cat ras irapa wavrwv 
§€^dju-€Foy ivaywyds* €t jjlsvtoi, irplv TrXrjpwusis ras TrevrrJKovra 
rjfjLGpas, drroSiSpdaKWV dXoiT) Kaddirzp ri r€>v drifxoraTWV dv8pa f TT68o>v > 
§t8o/xev aSetav Tots woTeAecrt Karix^iv avrov iv rfj x^P a [v T V ^^X^J 
Kal irav et ri §€§cu/cacrtv avTai rrpo^duei kXotttjs tovto GlsTrparrziv, 
wapovros fxivroi rod dso^iXeurdrov irnuKoirov /cat to irpayjia iyypd(f>ws 
8iauK07TOVVTOs» €cos av aTToSoir] Trav oirsp /ce/cAo<£a>s tf>aveii)* ^AAa. 
/cat avTOiis' rovs iTrapx^coras ■, etTrep ai'crSotVTo ttjs twv dpxovr ojv 
kXotttjS) aSetav %x €lv > p>&XXov fj,£v odv /cat dvdyKTjv, ravra psqvvsiv 
els 07/*as* wstg rjpLas p,av9dvovras> on wep oXcos XP VU ^ 0V 7Ti7rpduK€i 



12* 



APPENDIX I 



to SiKaiov, Tats elp7)p,evais avTov vTro/JdAAetv 7roivats > irpos too Kal 
rats' e£ ovpavov Tipicoplais evoxov ehai, irapafidvTa rovs opKovs £</>* 
ots eAajSe ttjv apyrps* El Se /cat layyozis KaB oiavovv cunav p,r) ttgttXtjp- 
cokcos rds irevTTjKOVTa rjpepas £k rrjs eVap^tas (f>vyetv, T7]viKavra 
uvAArjcfrdeis, evda av 8iarplj3wv ^avetT}, eiravaxBrjaeTai p,ev els rrjv 
eirapxiav tfs J r\PX* v > ^av Se, oaov av evpeBeif) Aaficov, aTroSaJcret rerpaTT- 
AdcrtoT>» 



CAPUT X. 

^Ekzivqv S^AaS^ <£vAarrojueVot> rov psqhepiav etvat Tots rjpeTepois 
vtttjkoois dSetav £<fy iripto nvl ttXtjv tj kAotttj ravra rrepl rovs dpxovras 
TTp&TTSiv* Ov yap el <f>aveir) o<f>oSpoTepos Tots ayvcop,om Std ttjv tcov 
8rjpLomcov etsirpa^tv 77 Std rrjv tcov TrXrj^eAripbdTCov eire^eXevmv $ 

8cbaop,eV Tots V7T7]KQOlS TTpOTTSlV Th KaT aVTOV* TOVVaVTlOV pkv Q$V 

Kal TTQivais avrovs rats iraucov iriKpordrais vTrd^ofAev, et rovs KaBapats 
XOpaap,evovs rats ^epatv /cat rfj tcov §rjp,oalcov elsirpd^ei pueTa irdays 
TrposevexBivTas d/cpt/Jetas, etra /cara^e/xevovs rrjv apx^v vfipiuai 
Bapprjuaiev, dAAd pur] avv ev<j>7]p,ia irdarj jxera tov vevopuap,evov %povov 
rds eVa/^tas aTToXip/rrdvovTas . diroirepAJjahev , Aet yap rovs jtterd 
roVSe rjpbwv rov vop,ov yivop^evovs AapmporaTovs tcov vrroTeTayp,evcov 
eVap^ttov dpxovras eVTeBvpur]u6ai > ttoutjs p>iv diroXavnovm Sogjjs 
toiovtoi jtaivoptsvoi) TToaais Se TrepiireaovvTai Svs/coAtats tov vop,ov 
tovtov TrapaXvcrai to ye £<$* eavTots BapprjaavTes * git) yap av tcov 
cltottcov, el tovs p>ev eir evreXeatv dAoVras kXottois avrot KoAd£,oiev, 
Kal fiaadvois avrovs VTrofidXAoiev, Kal ov rrpoTepov avyx^potev ecos 
av awoSotev rd <j>cbpia, avrot Se dvevBvvoi puevoiev eVt peydAwv yevop,evoi 
kKott&V) /cat ovSe to Trpos rovs vtttjkoovs epvdpicovTes irapaSeiyp^a* 
Sv e^euTiv avTois xmepihovui uepwois re /cat eXevBepois Kal 7ravTO)(69ev 
eirawovp,evohs <j>avrjvai /cat rrjv e£ rjp.cov l^ etv KaArjv I A 1 p,apTVpiav 
Te /cat eAmSa, Ov uvyxcopovjxev Se ovTe rots* TrepijSAeTTTOis Sov^lv 
oi)Te erepco rtvl ttjv olavovv avTots irAeove^iav rj dSt/ctav eirayayeiv, tj 
ttoAitikois SAcos TTpdypuaai Koivcoveiv s tva /cat r}p,€is avTOisrrjv aep,v6TrjTa 
tf>vAaTTOip,ev KaKeivQt ttjv KaOaporrjTa Te TjpXv /cat evvoiav dvrtStSotev. 
"Jarct* yap aitav to VTTrjKoov^ cos Std ttjv ovtcov chtfieAeiav Kal to iravTax- 
69ev az>Tcbv d^/xtov Kat to Std TrduTjs avTovs dyeiv eviraBeias /cat 
pWj KaTavayKa^eadai Tas ^cupas a7roAtju7rdvetv pwfie ev £ev7) raAat- 
7Tcopeiodai 3 Std tovto tov irapovTa vop,ov eypdi^ap^ev, 6eco Te avTOV 
dvar hdivTes /cat rats Trapovoats crejSaer^twrdrats ttjs pbeydArjs avTov 
Kal KOhVordTT]s iopT^s r)p,epais* tva waaiv e£f) iraTepas Se^ea#at 



GKEEEl AND LATIN LEGAL DOCUMENTS 



IS* 



/xaA/W apxovras ?? KXiirrovrds re Kat dv8parro8co8^is Kat Tats avrcbv 
ovuicus i<f>z8pzvovTas* Azi 8k /cat vpas rovs ^pzripovs woTeAets 
slSoras, TToarjv vpcbv idepsda wpovoiav, psra irdayjs GvyvcopoavvTjs 
rovs St) p,oaiovs aVeAAtTrcus' j>6pov$ slsaysiv, Kat p/rj8£ rrjs irapa rcov 
apxovrcov dvdyK7]S Setcrflat, aAA* ovrcos svyvcbpovas iavrovs irapd^iv, 
costs rjpiv ££ avrcbv iv8<zi£aadai rcbv spycoV; on /cat avTot rrjs roaavT7]s 
^iKavBpcorrias rrjv otK€tav rjfuv $vyvcop*ouvvT)V dVn8t§0T€, /cat gIkotcos 
l^€T€ rrduav rrapd rcbv apxovrcov 7TpQ<j>ausi rfjs svyvcopoavvTjs GTT0v8 r t)v 
re Kat irpovoiav, £k$wq yivcouKovrzs chs* e77€t&7 Tots' apxovaw iTTiKzirai 
7ravra^o0ev d t&V S^p^oaicov kiv8wos koX rcbv dvcopoXoyrjp4vcov 
iariv, cos iirl rep adcbv avrwv kiv8vvco rds dpxds VTT^isipxovrai^ /cat 
vpds tqvtq yivcouKovras e/c TpoTrov rravros ei)Aa/?€tcr#at rr)v dyvco~ 
IxoavvifV} /cat p/r) rds iavrcbv yvcopas ovrco Trapiysiv dirsidzis, cos 
Kat rrjs ££ avrcbv Ssiodai u^oSporTjTos^ rjv dvayKoiov avrois £cm 
TTpos^apfiavsiv 8ta rrjv dirapair^rov rcbv 87)p,ocricQV sts7rpa£w* si86rcov 
vp,cbv rcbv rjperspcov vttj^kocov^ cos at vrparicortKal Sairdvai Kat rj 
rcov troAspicov Starts ttoAA^s 1 SstTat rfjs iinpsXsias, /cat ov/c lort 
Xp7)p,&Tcov x&pls ravra TrpaxBrjvai^ rod rrpdy pharos p^rjSspias avafSoXrjs 
8zop*ivov > ov8k 7jp,cQV alpovp,4vcov rrspiopav rrjv *Pcop,aicov yrjv iXarrco- 
Qzioav s dAAd Aifivyv T€ Traaav dvaKrr]aap*£vcov /cat Bav§iXov$ /ca- 
raSovAwaavTCjov /cat 7roAAd ye IVt Kat pzl^ova rovrcov iXirilovrcov 
wapd rov Beov Aa/Jetv re Kat rrpd^ai^ sis a TrposrJKov iari rovs Sypocrlovs 
<f>6povs aveXXmcos Kat euy^co/xoya)^ Kat Kara ras 1 chpwp,ivas Gisirpdr- 
Tecr^at rrpodzaptias* costs eiirzp vpsis ftev evyvco/zovcus' aTrai'T'^aotTe 
rots djo^ovcrtv^ ot §€ paStav T€ Kat €K irpo^ipov rrjv rcov 8rjp,oa(a)v 
zl$KQp,i8r)V ets 1 ^jaaV Trototvro, Kat tou? ap-^ovras iiraiviuopLsv rrjs 
gttqvStjs Kat vp,ds aTToSel o^tte 0a r^s yvcbp/r)s* koI rravraxpB^v pia 
ns eWat KaAa^ re Kat avp,<j>covQS rcov re dp^dvrcuv Kat rcov dpxop,<htoV 
dppLOvla, 



CAPUT XL 

To) ju,€ydAa> roivvv dsca Kat acorTJpi rjpcov ^It\(jov Xpiarw irdvrzs 
opticas dvaiTspmircauav vpwovs virsp rovrov Stj rov vopov, os avrolls 
8coo€i Kat rds rrarpi8as oikgiv da<f>aXcos Kat rds oiK^tas Trepiovuias 
€X €IV fisfiaicos Kat ttjs rcov dp^ovrcov wrroXav^iv 8iKaioavvrjs* -Kat 
yap 8r) Kal 7}p,Gis 8id tqvtq avrdv i94pi^9a > orrcos av eK rfjs iv rep 
vopco 8iKaioavv7]s laxvacopLGV rep Szuttott) dscp otKetcSaat iavrovs 
Kat rrjv rjfXGrdpav avarrjaai jSaatAetav, tva pur) 86i;cop>GV irtpiopav 
dvBpcorrovs dSiKovpdvovs, ovs rjpuv 7rap48coK€V o dsos* qttcos av avrcbv 



14* 



APPENDIX I 



Std Trdvrwv <f>€i8wp,iz8a, rfj avrov KaraKoAovBovvr^s dyaBoryn. 
"Q$T€ to ye itf> 9 rjjjuv d<f>ooiova6w rw Bsw, Stdrt jxr]8kv rwv €ts vovv 
rjfjuv ipxofjizvcov dyaBwv virkp K7)8€}j,ovias rwv vtttjkowv TrapaAipLTrdvoiisv . 
jSouAd^tevot yap ras dvsAevBipovs ravras Kal avSpaTro&coSeis /cAo7rds 
dveAetv /cat rovs rjiierzpovs vttqtgAzis £v ewafleta Trapd rwv ras imx- 
wpiovs dp%as £)(6vtwv tf>vAd£ai, Std tovto £u7rsvcra}j,sv irpoiKa ras dpgds 
avrots Sowat, ottws dv /xtjSc avrots 1 i£r) TrArjfAfieAetv re Kal apTrd^iv 
ro vtttjkoov* odnep !v€/ca iravra aipovp,eBa irovov, ovk d^iovvres 
lMp,€iodai rovs irpd 7}{awv fSefiauiAevKoras > omep xpTipsdrwv irpov- 
fidAAovro ras dp-gas* iavrots dvaipovvres rrjv dSetav rov ye rois ev 
rats dpgxus dSiKovaiv eVtTt/xaV St/cata, dAA* avrol re ols eAdp,f3avov 
ey KaAvtrreaB ai Si/catot KaBeurwres, rovs re oiKelovs viroreAets St' 
avro 8rj tovto e^apird^eiv rwv KaKws dpgovrwv ov Svvdfievoi ovbe 
avrots rots apgovutv eTnrt/Aav aw<f>povetv Trpo(f>daei rrjs elprjpivrjs 
air las* r)jj,ets 8e apKovvra rfj fSaaiAeia iropov etvat vop,i£ > ojjLev ro 
jjlovovs evreAets rovs SypLOoiovs KOfil^eaBai $>6povs, dAAd pjr\ ri Kal 
eijwBev TrposeTritpqretVy OTrep Tots* vtt7]k6ois Trdvra Karaoel8ei rov 



CAPUT XII. 

IB To 8e €v TOty ep/rrpouBev rjfJLiv elpi)jxevov en pLel^ovi xpijvai 
Kal aKpifiearepa irepiAafietv wrjB^jxev vop,o8eala, wsre rov r)p,erepov 
ukottov aVacrt yeveoBai cf>avep6v. ©euTrilofxev yap rovs AapLTrpordrovs 
rwv VTrorefaypiivwv iirapxitov dpxovras, x w P 1 ^ dirduTjs yivopievovs 
Xpypbdrcov 86crews Kal rwv Trap* avrwv 8i8op,€vwv opKwv fj,ep,vr)p,ivovs, 
eyew /cat ravrrjv Trap* rjfxwv rrjv Trapprjuiav rov p,7)8ep,iav etvat prrj8evl 
irpos avrovs iravreAws tj>6pov [r)roi Kpiriqpiov] irapaypaff>r)v pxyre 
ev rats dp,apravop,evais Trapd rivwv filais p/qre eVt tois 1 £yKXf)fxaai Kal 
rats £vt€v9gv dSiKiais fxrjrz £ttl rats urduecn rats Srji^oaiais pbrjre 
£tti rats rwv SrjfjLoaiwv (f>6pwv elaTrpdijzuiv, dAAd 7rdvTas 6p,ota>s 
VTTOKstadai rfj rovrcov SiKaioBoala, ovk dva\x$v6vrtov ovnz irposra^is 
AajSetv £k rwv apyovTwv ots VTroKGivrai, ovtg jxtjvvziv ets auTOtfe, 
dAA' dpKztaBai twSg r)p,wv rw vdfxw, St' o5 irauav avrots i^ovaiav 
7rap€^o/zev, ovSevos* dSetav egovros TravrsAws £ttI rwv ^Iprji^ivwv 
alriwv ovrz Trpovofxtw xprjvBal nvi ovts e/C€t#ev iavrw KaropBovv to 
7rA^/z/xeAetv dvsvBvvws* ov yap av ol Trdo7)s aTre^dju-evot Arjiftews 
dpxovres tr^pov ri rov Beov /cat to£ vojxov /cat tov rj^repov TTpoBrjaovcn 
Seovs, dAA* €t? £kzivo jSAerrovTes $>vkdt;ovm rots vtttjkoois to St/catoF, 
Trdvra Kara rovs ^/xe 1 ripovs Kpivovris ts /cat Trpdrrovr^s vdfiovs* 



GREEK AND LATIN LEGAL DOCUMENTS 



15* 



'EttI yap rois roiovrots Kal rovs arpariwras rovs iv rais irrapx^ais 
ovras avrois VTrordrToixzv, ov8k e/ceure Seo/xeVots rrposrd^ws rivos 
I8tas rj nap' rjpiwv rj rrapa rwv ^/xereptov dpxdvrcov, dXXa rep rrapovri 
vop,cp xpwpLevois Kal rovrov avrois Sgikvvovctiv, c5st€ irrapivveiv avrois 
Xpcopievois rfj rrjs dpx'rjs rrappTjala, yivwuKovras cos, et pur) rovro 
rrpdiaiev, Kal uittjuzcov Kal avrrjs ttjs arpareias vrropizvovcriv ektttwuiv 
Kal rov eis ucopia kivSvvov vrro arrj govt tu . "Qsre rjpuv p/^Sevos iripov 
rravreXcos 8ziv apyovros, Kal XrjuroSiwKras r) rovs KaXovpievovs 
/3ioKcoXvras> piaXAov 8i XcoTro8vras, rj d<f>07rXiaras e/CTre/xTmv, rrpo^dozoi 
ixkv StJ#€V zvXoyois ^pa>/xevovs , > avrovs 8k ra irdvrcov ^e/ptcrra rrpdr- 
rovras* rcov yap apxovrcov rcov irrapx^v rrjv iKaarov rcov fjieylarcov 
dpx&v rrX^povvrcov rdtjiv, Kal dvrl 7rdv7]S iripas §e dpxrjs rats irrapxtais 
dpKovvrwv, Kal ra €/c rcov rjpLtrepcov vop,cov to ye eV avrois $7}<f>i- 
£o/xeVtov, Tis dv dapprjazizv 77 <$>6pov 7rapaypa(/>7J rj roiovrco nvl rrpos 
avrovs xprjcrauBai ; 



CAPUT XIII. 

<iT'> * Array opevopiev 8e Kal rtp iv8o£ordrcp urparryytp rrjs 
v Eco Kal iram rois rjpieripois dpxovuiv, r) XrjarooicQKras rj fiioKcoXvras 
r) d(f>o7rXiaras rj rivas roiovrovs iv rais iirapxlais iKTripirreiv. "Iorcouav 
yap ot T€ yzviodai p,€rd rovSe r)p,cov rbv vopiov dapprjoavres ', d>s avXX7]tf>- 
B£vt€s rrapa rcov apxovrcov rcov irrapx^cov Kal 8€up,wrr)piov oIktjuovui 
Kal zls rjpids tov rrpdypiaros p,7]VVop,evov rov kax aTOV VTroarrjaovrai 
kIv8vvov ol re ras roiavras avrois 7rapa8i86vr€s rrposTa^is triginta 
libramm auri virourrjaovrai rroivrjV 3 Kal piGi^ovos 8e in Kal cr<j>o8- 
poripas rjpicbv dyavaKrr)- I A crzeos Treipadrjvovrai . Azi roivvv rovs 
dpxovras rcov irrapx^v roaavrrjs a^itodivras Trap 3 rj}j,wv itjovatas 
ovrw rw TTpdypbari xpfj a @ al > **>$ 8iKala>s Kal vop,ip,oj$ airaaiv stvai 
tftofiepovs, eiSoras* ojs, €t rfj 8^8opiivrj Trap* rj/jicbv apxfj KaKtos Kal 
dvatjlws TTJs imrpaTreloris avrois Trap t)p,(Jl)v rrapp7]crias xP r ) aoVTai > 
vrroKelaovrai rais ripiwpiais ats zpnrpoodzv €i7rop,sv> /cat ktos rrjv 
apxfy €x ovai Tavras vrrop,£vovrzs , Kal irr€i8dv avrrjv KarddoivrOy 
pi^i^ovwv €Ti 7T€ipcbp,€Voi KivSvvwv, 0v8k yap 8l8ofji<zV avrois aSeiav, 
rrplv rov vevopiicrpievov rwv TrevrrjKovra r)7T€pwv rrArjpcoaovcri xp°* vov > 
rG>v irrapxi&v &v rjp^av dvaxcopetv 7} /card rrpo^auiv revocatorias 
[jjroi dvaKXrjcr€a>s] 7) Kara rrpo^amv <f>vyrjs fj Kara aXArjv oiavovv 
alrlav* yivuiUKovuiv ws, Kaddrrep ipirrpoud^v zlrrovrzs £<f>97]p,izV, eire 
errl ravrrjs yevoivro rrjs €v8atp,ovos 7roX^ws eire iv olahrjrrore x&pa, 



16* 



APPENDIX I 



rrpos rrjv irrapxiav aWis irrava^ivr^s rjs rjp^av rroivas v<j>£i;ovuiv > as 

epmpOuBeV 617TOVT6S €(f>67)fJ,€V. 

CAPUT XIV, 

Tov 8e opKov hdxjovuiv ivravda fxev /card to dvcu- IE r4p<x> p7)94v. 
El 84 tiuiv iv reus iirapxiais odcri 7r4p,7roiTO ra rfjs dp^yjs vvp*fio\a, 
irri Te tov BeofaAecFTaTov imaKOTTov ttjs fxrjTpoTroXecos Kal twv iv 

aVTT) 7rpC0T€VOVTC0V TOV OpKOV V<f)4£oVOl, Kal OVTW TCOV TTJS dp^TJS 

dvTiXrjiffovTai irpd^eoiv S^AaS^ ttjs orjs virepox'yjs irpovoovarjs tov, 

€IT€ iirl TaVTTjS TTJS lieyohrjS TToXeWS TTOpaX&fioi ThS dp)(r}V, €tT€ KaTCL 

^copav avT(p to, crv/zjSoAa TavTrjs rre^d eij] irapa ttjs crrjs vTrepoxTJs, 
avTOV tov Aa/x/JdvovTa to du(f>aXes rrepiTTOieiv tw 8r] jjlog iw ire pi ttjs 
twv <f>6pwv djj,4fj,7TTOv elsrrpd^ews, KaBdrrep av avTos KaBapws So/a- 
pdueias* Kelu9w 8e 6 vop,os ?}fHV oStos i<f>' arraai tois Tas 7rap' 
rjfiwv prjTws 6vop,aa9rjvop,4vas dp-gas 4 k tov irapovTos xpovov dp,la9ovs 
TrapaArjifjopevois* tcl yap 8rj TrpoeiArjfioTa tois eii7Tpoo9ev Keip,4vois 
VTTOKslada) vofAois, ov8ep,ias TTOivrjs twv iv Tw8e tjijlwv tw vo/xa> Siwpio- 
\x4vwv imK€ifj,€V7]s tois /J-^pt vvv to\s dp-gas egovai, ttAtjv el p,7) Kal 
avTol p.€Ta Tjjv ip,<j>dviaiv Tov8e tov vojiov kA4tttovt€s dAoZev, 

<* EirtXoyos^ * TavTa toivvv rj utj virepoxj] ixdvTa p,av9dvovua 
iv irauh tois e9vem toXs V7TOTeTayp,4vois $>avepa TrapaaKevaaaTw 
yev4o9ai, /card to vevofMicrpL4vov irposTayixaui xpwfj,4v7] Trpos irdvTas 
tovs twv errap^tcuv r}yovfi4vovs' wsTe avTovs ywwoKovTas ttjv rjp,eT4pav 

TTZpl TOVS VTTTjKOOVS OTTOv8t)v Kal TjV egOjieV TTepl T7JV TCOV dpgOVTWV 

yshpOTOvlav yvwp,T)v, el84vai, ttouwv avTois dyaBwv /x€TaSe§t6/<a/x€V, 
ov8e ttjs ftaaiAiKTJs 9epaTretas 8id ttjv avTcov ev8ai}j,oviav 9etadp,evoi. 
Dat, xvii. k, Mai, CP. Belisario v. c, cons. [a. 535], 



"I8iktov ypatf>kv tois diravTaxov yrjs ^o^tAeaTaTots imoKOTrois Kal 
oaitoTaTois TraTpidpxais . 

Trjs 7rapa8oB^iu7]s r\pXv e/c deov rroXiTelas /c^So/xevoi Kal iv drrdorj 

8lKaiOUVV7} tfiv TOVS TJfAZT4pOVS VTTTjKOOVS UTT €v8 OVT €S y TOV VTTOTGTay- 

p,4vov vojxov iypdiftap,izv f ov 8rj ttj afj oaioTrjTi, Kal St 5 avTrjs diTaui 
Tots ttjs eTrap^/as ttjs orjs woirjaai <f>avepov KaXcos 4x €iv evo/xttra/xev. 
ttjs odv cftjs 0€O(f>iXtas Kal tojv XoiTTcbv imuKOTTOJV 4otoo TavTa Trapa- 
TTipziv, Kal el Tt TrapafSaivoiTo rrapa tcuv ap^oVTCUV, els r)p>as p,i)vveiv, 
O7rcos av jirj ti rrapopadeiT) tcov omws Te Kal St/caicos v<f>* rjfJLcbv vop,o- 
9eT7j94vTO)V . El yap r^xels p,ev tovs 7}p,eT4povs vtttjkoovs iAeovvTes, 



GBEEK AND LATIN LEGAL DOCUMENTS 



17* 



on TTpos rrj tcuv §77 jxouiwv <f>6pwv e/mcret /cat p,sydXas vWp-evov 
€K ttjs twv apxQVTtov kXotttjs d8iKias Std ra$ yivofxevas twv eTrap^taV 
Trp&QGis, TavVas dveAetv Std tov woreray^vov iuTTGvaapLZV v6[jlov, 
vjMzis 8k paBvjiovvTes p,r) TrposayyeiArjTZ, rjfXiv iikv d<f>omovo8w TO 

TTpOS TOV SsaTTOTTjV 0€OV 9 VfJt,€lS 8k aTToAoyiqUSoBe TTpOS aVTOV VTTGp 

rfjs twv dAAtov a8iKia$, et ri Trap a to [it) jxaBeiv rjfjLas fiAdpos tois 
trap* Vjjfiv dvdpwTTOis iirdyoiTO. dAAd Set TrapovTas vpas ttj %wpa 
/cat VTrkp avrtov /cat twv Xoittwv dywviwvTas <f>avepovs rjfuv KaBiGTav 
/cat Toi>$ opBws apxovTas /cat tovs TrapaftaivovTas tovSg rjpwv tov 
vofxov, ottws aV e/caTepovs yivwuKovTSS tovs p*kv /coAa^co/^ev^ tovs 
8k dpeificofjLeBa. 9 ETT€i8dv 8k 6 vopos 8r)p,oala TTpOT^Beiy] Kal dVacrt 
yivoiTO <j>av€poS) TrjviKavTa A^Scts* eVSov aTTOKSiadco iv ttj dy iWTaTT) 
e/c/cA^crta ju-erd twv izpwv cr/cevaV, old /cat avTos dvaTeBGifiivos Bgw 
/cat ?rpos uwTTjplav twv vtt* avTov yevof^ivwv dvBpWTrwv yzypaptiivos, 
TTOirjaaiTe 8k dv /cdAAtov /cat tois avToBi tt&giv dvBpwTrois avp,<f>opwTepov ? 
€t7T€/> avVdV iyKoXdtfsavTGS rj uaviuiv r) XlBois iv Tais oroats ttjs dyta>- 
Tares' iKKXymas dvaypdijjaiTZ) TTpo^ipov Trap^yp^voi rrdui tt\v twv 

VO\hoB$TT}B4vTWV dvdyVWVlV T€ /Cat KTTJUIV* 



CAPUT I. 

El 8k ttjs twv dpxdvTWV KaBapoT7]TOS TovavTTjV iBifxeBa TTpovoiav, 

TTp68l]XoV WS TToXXw jJL&XXoV TOtS €K8lKOlS OV/C i^TjaopLeV Ov8* OTIOVV 

ovre Aa/z/Jd^etv ovVe StSdrat. Sdicrovcrt p,kv yap virkp twv TTap€Xo^ivwv 
avTols TTposTayp.aTWV iv to> 8iKauTT]plw twv iv8o£joTaTwv esrapxa)^ 
et p,kv fjL€i lovzs at TToXeis etev, solidos quattuor, et 8k twv iAaTTovwv, 
solidos tres, /cat Tripa tovtwv ov8iv* A^ovrat 8k oi8* otwvv Trap 9 
ov8svos, 7tXt)v et psj tis €/c tov 8rjp.ouiov vsvofiiujiivos avTols Trpos IT} 
TTopos* r) ewrep ixrjSkv €/c tov S^juoatov Aaftj8dvot€^, p&)8ev TrepatTepw 
tcuv ttj Bsia 7}p,wv 8iT)yop^vixivwv StaTa^et /co/zt£ea#at. iTrehoiy^ et 
ti AafiovTzs dAotev t) avTol rj ol Kakovjisvoi avTcov yapTOvKdpioi t) 
zTspos tis toV 7T€pt a^Tovs^ €/c€t^o TeTpaTrAdcrtov aTToSwaovcnv OTTSp 
lAajSov, /cat tov tfrpovTiafiaTos dTTsXaBrj aovTar Kal TTpos ye Kal i^opla 
8i7)V€K€i ItjiiiwB4vt€s Kal sis vwjia awcj>povia94vT€S 8wvovai x^P av 
dv8pdmv dyaBois ovtI KaKwv tov ^povTia^aTos avTiXap,fidvGuBai* 
"EusuBs 8k /cat tovtov <j>vAaKss vp,s$s» /cat kwAvvtss Ta rrapa TavTa 
yiv6p*eva /cat psqvvovTGS , wsrz p,rj 8iaXaBeiv Tt tcDx' dp,apTavop,€Vwv 
jX7]8k €/c tov XaBziv aTifjt,wp7]T0V stvai, dAAd irauav IvoTrjTa T€ /cat 
St/catocrvp^v Tots rjfLZTspois vtttjkoois iiravBrjaai* El 8k Kal ol /xe^pt 
vvv apxovTGs fMrj jLt€Ta tt)v ifjL<f>dviaiv tov8€ TOV VOjlOV Trdui)s aTTOGXWVTa* 



18* 



APPENDIX I 



kXotttjs, lUTwaav /cat avrol reus e/c rovSe tov vo/jlov itoivcus VTro/cet/xevou 

[ TOVTO TO 'ISlKTOV 7Tp6$ TOV$ 6771(7 KOTTOVS ] . 

Dat, sv, k. Mai, CP. Belisario v. c. cons, [a. 535]. 

9 Eypd(f>7) to laov tov ISiktov KwvaTavTivovTToAiTais ', %X 0V °vtws* 

"Outjv dirdvTwv tG>v vtttjkowv id4p*s8a irpovoiav, SeiKwaiv 6 Trap* 
rjp,wv svayxos re#€ts vojxos, ov St) irpos tovs iv8o£oTa,Tovs rjfxwv 
iirdpxovs iypdiftapLGV. dAAd TrposrJKov ecrri /cat vpuas clvtovs etSevat 
tt)v rjfieT^pav irpovoiav, rjv Trspi TrdvTas dvdpwirovs l^o/xev. /cat Sid 

TOVTO TOV VOjlOV CLVTOV /Cat €V ISlKTOV 7TpOT€9eiKafX€V CTJ^U-aTt* WST€ 

vfxds tw SzairoTT) dzw /cat awTrjpi rjfiwv > Irjaov XpiUTW St/catcus* 
dva7T€fj,7T€iv vfxvovs, /cat Tjj 7}jj,€T€pa ftamAeia, oti irdvTa Std to vp,€Tepov 
uvji<j>ipov alpovfJLtda ttovov* 

Tvwuis Trjs 7rap s iKauTrjS twv VTTOT€Tayp,£vwv dpxwv 6tf>ei/\ovarjs 
Trape^ecr^at Aoyw avvrjd^iwv ttouottjtos , [irapa] twv rets" dpxds ixovTWV 
ovSsvos ToXp,wvTO$ TTapd tcl irpoy^ypapipidva ovt€ Aa/3etv ovtg hovvat 

Tl 7t\£oV* 



1 AtTO TOV TTZpifiMlTTOV KOp/TjTOS TTjS * AvaToXrjS OVTWS' 

iv tw Bslw rjjjiwv kovJUovkAziw vo/x. %y 

tw irpifjuKrjpitp twv AapmpOTaTWV Tpifiovvwv voTa- 
piwv p.erd tcov Tzoudpwv UKpivlwv tov delov Aarep- 

KOvXoV VOfA* V 

tw avTOV fiorjdw vop,. y* 

T7J Ta^Gi twv ivSoijoTaTwv ivrdpxwv virkp irposTay- 



ftaros* 



VOfJL* 



^Atto tov dv9v7rdTov 'Aulas ovtws* iv tw 8elw rjpiwv 

KOVJ3oVk\<zIW VOfJ,* £)/ 

TW TTpllllKTjplop TWV XafJUTpOTaTWV TpifioVVWV VOTa- 

piwv /xerd twv Tsoodpwv uKpiviwv tov delov Aarep- 

KOTjXoV VOjl. [A 

tw avTov f$OT)dw vojjb. y 

Tjj raf et rcov ivSo^oTaTWv iirdpx^v vrrkp irposTay- 

fAtlTOS VOfA* 77*' 

*Atto tov TT^pifiXsTTTOV Kop/T^Tos &pvyia$ UaKaTiavrjs 

OVTWS* 

toXs <.7TGpi(3X€TrT0i$> ^apTouAapiots' Tpiul TOV 
dziov rjfxwv kovJ3ovkAgiov vo/z. 5 



GREEK AND LATEST LEGAL DOCUMENTS 19* 

TW 7TpipLlK7JplW TCOV XapiTTpOrdTWV Tplf$OVVWV VOTCL~ 

plWV VOfJL, K& 

rw avrov fiorjdw VOjl, y* 

rfj rd^€i rwv ivSo^ordrwv irrdp-^wv virkp rrpos- 

rdy p,aros vo/z. v 

4 'Aito rod TTzpifiAiiTTov kojxtjtos PaXaTias irpwrrjs 

OUT CDS" 

rois irepifiXirrTQis gaprovXaplois rpicrl rod dslov 
kovJSovkXglov vop,, #' 

TW TTpllXlKTjpiW TWV XapLTTpOTaTWV TplfioVVWV VOTCL- 

piWV VQjJL* /C§ 9 

rw avrov fiorjdw vop,. y* 

rfi rd^€t rwv ivSo^ordrwv Irrdpywv virep irposray- 
fxaros vop,* v' 

5 Aito rod fiiKaplov rod MaKpov Tsi-govs ovrws* rots 

Trepi^Xinrois gaprovXapiois rpiul rod delov kov- 
fiovKXziov vop,' B* 

tw TTpifxiKTjptw rwv XapLTTpordrwv rpifiorvwv vora- 
piwv vop,, ko 

tw avrov fioT}dw vop,. y 

rj) ra^et rwv ivSo^ordrwv irrdpxwv virep rrposrdy- 
fjbaros vojx* p? 



Kal ouai dp-gal vrrariKal jjroi KovuovXdpiai* 

6 *Att6 rod apgovros naXaioriVTjS Trpwrrjs ovrws* 

rois TTzpifiAdrrrois gaprovXapiois rpiul rod dzlov 
kov^ovkAziov vop, 0* 

rw TTpipLiKTjpiw rwv Xap,7Tpordrwv rpijSovvwv vora- 

plWV VOpL. fcS* 

rw avrov fiorjdw vop. y 5 

rfj rdtjsi rwv ivSotjordrwv iirdp^wv vrrep irposrdy- 
pharos vop. /x' 

7 *Att6 rod ap-govros IlaXaiorivrjs bzvripas ovrws' 

rots TTspifiXirrrois gaprovXaplois rpicrl rod Betov 
kov^ovkX^Iov vop,. 6* 

rw irpipLiKTipiw rwv XapTrpordrwv rpijSovvwv vora- 
plwv vop,. k8* 

vop. y 



rw avrov ftorfiw 



20* APPENDIX I 

rjj rd^€i rwv ivBo^ordrwv iirdpywv virkp rrposrdy- 
fxaros vqjx* /x* 

5 *Atto rod apxovros &oivUv)s TrapdXov ovrws* 

rots 7T€pi^X4rrrois %aprovXa plots rpial rov dsiov 

kovJ3ovkXgiov VOfl. B* 

rw 7Tph}AiKj)piw rwv Xap,7Tpordrwv rpifiovvwv vora- 

piwv vop,. /cS' 

rw avrov j3or]9w vop,. y* 

rfj rd^€i rwv ivbo^ordrwv irrdp^wv virkp irposrdy- 

p,ar os vop,* p? 

9 > Atto rov ap)(QVTQS Svpias osvripas ovrws* 

rots Tr^pi^Xirrrois ^apTovAap tots' rpial rov dstov 

KOvfioVxXsiOV VO/X. B* 

rep TTpifMKTjpta) rwv Xap/npordrwv rpifiovvwv vara- 

piWV VOfJL. /cS 

tcu avrov f$OT)9w vop,. y 

rfj rd£ei rwv iv8o£ordrwv iirapx^v virkp rrposrdy- 
jxaros vofx. fi 

10 *Atto tov ap^ovros ©soSwpidoos ovrws* 

rots 7T€pif$A<z7rrois -^aprovXaplois rpial rov dsiov 
/cov/Jov/cAetou VOjl. 6' 

rw TTpifjUKTjpitp rwv Xapmpordrwv rpifiovvwv vora- 

ptWV VOfJL, k8* 

rw avrov J3ot]9w vop,, y* 

rfj ra^et rwv ivSo^ordrwv irrdp^wv virkp irpoardy- 

jjiaros vop,, p, 

11 * Arro rov apyovros *0apo7)V7Js ovrws* 

rots irspifSXirrrois -)(aprovXaptois rpial rov Beiov 

kovJSovkXgiov vo/x. 9* 

rw 7Tpip,iKrjplw rwv XafiTTpordrwv rpifiovvwv vora- 

piwv vo/x, /co° 

t<S avrov fiorjdw vo/x. y 

rfj rdtjsi rwv ivoo^ordrwv iirdp^wv virep irposrdy- 

jxar os vo[jl, /x 

12 9 Atto rov dp-^ovros KiXiKias Trpwrys ovrws ' 

rots TTZpifiAiTrrois %aprovXapiois rpial rov Bslov 
kovJSovkXgIov vop,. 9 

rw TTpipLiKTjpiw rwv Xafxnpordrwv rpifiovvwv vora- 
piwv vop . kS* 

rw avrov fio7)9w vop, . y* 



GKEEK AND LATIN LEGAL DOCUMENTS 



21* 



rfj Ta£ei twv ivSoiordrwv iirdp-^wv virkp irposray- 
fJLaros vop,, p? 

13 Atto tov apxovros Kvirpov ovtws* 

TOIS TT€plfiM7TTOlS %apTOvXaplOlS TpiVl TOV BeloV 
KOV^OVkX^IOV VOp, 6* 

TW 7TplfMK7]piU) TWV XapLTTpOTaTWV TplfioVVWV VOTCl- 

piwv vo/x, /cS* 

tw avrov fio7)9u) vop, y* 

rfj rdijsi Twv ivbo^ordrcoy iirdp^wv virkp irposTay- 
p,aTos vop, ji 

14 *Atto tov apyovTos IIap$vXias ovtws' 

TO?$ TTSpifSXilTTOlS ftapTOvXapioiS Tpi&l TOV BziOV 

KovfiovKiXsiov vop* 6* 

TW TTpipiKTjpiw TWV XapLTTpOTaWV Tplf$OVVWV VOTO,- 

piwv vop, /cS* 

Tip avTOV fio7)9w vop,. y' 

ttj ra£et twv ivSogoTaTwv iirdp-^wv vrrip irposTay- 

JJLCITOS VO/X, /X* 

15 *Atto to£ dpxovTos BWvvias ovtws* 

TOIS 7T€plfSX4lTTOlS XapTOvXapioiS Tpivl TOV 0SIOV 

kov/3ovkX<zIov vo/x» B* 

toj TrpipiKTjplw twv Xap/TTpoTaTWV Tpifiovvwv yora- 

piwv V0 H" K &* 

TW CLVTOV f$07)9w vop. y* 

ttj ra^€i twv ivSo^oTaTwv iirdp^wv VTrsp TrposTay 
paTos vop* /x* 

16 *Att6 tov &p)(ovtos ' EXXtjsttovtov ovtws* 

tois TrepijSXzTTTOis ^apTovXaplois Tpiul tov dziov 
kov/HovkXgiov vop. 9* 

TW TTpipiKTjplW TCOV XapTTpOTaTWV TpifioVVWV VOTCL- 

piwv vo/x, /cS' 

tw avTov j3o7)9w vop,* y* 

ttj Ta^et tcuv ivSo^oTaTwv iirdp^wv virkp Tr/wrdy- 



p,CLTOS 

17 'Atto tov ap-govTos Avbias ovtws* 



vop,. p 



TOIS TTSpifiXiTTTOlS ^apTOvXapiOlS Tpiul TOV dsiOV 

KovfiovKXeiov vop,, 9* 

TW 7TpipiK7]piw TWV XapTTpOTaTWV TpifioVVWV VQTCt- 

piwv vop,, /cS* 



22* 



APPENDIX I 



rw avrov fioTjdtp vop,, y 

rfj rd^et rwv iv8o£ordrwv e^rap^cuv virkp irposrdy- 
fjuaros vop,. p! 

IS *Att6 rov dp)(OVTO$ 0pvytas aaXovrapias ovrws' 

rots irepi^Xirrrois ^aprovXapiois rpiul rod 9eiov 
kov^ovkX^lov vop,. 9 

rw TTpipiKTjpiw rwv Xaparporarwv rpifiovvwv vora- 

piCOV VOJA. kS' 

rw avrov fiov9w vop,, y 

rfj rdfjei rwv iv8o£ordrwv iirdp-^wv irnkp irposrdy- 
fxaros vop,, p,* 

19 *Atto tov dp-^ovros IliuiSlas ovrws ' 

roXs rr^pi^Xirrrois -^aprovXaplois rpiul rov 9eiov 

KOV^OVkX^IOV VOp. d y 

rw npipuKTjplw rwv Xapmpordrwv rpifiovvwv vora- 
plwv vop,. k8 

rw avrov fiorjdw vop,, y 

rfj rd^ei rwv ivSo^ordrwv iirdp^wv vrrkp TTposrdy- 
p,aros vop,. ju' 

20 y Arro rov dp^ovros AvKaovlas ovrws* 

rols TTZpifSXirrrois xapTovAapiots' rpiul rov 9elov 
kov^ovkXeiov vop,. 9 

rw TTpipbiKTjpiw rwv Xapnrpordrwv rpifiovvwv vora- 
plwv vop,. /c§* 

rw avrov fiorqdw vop,. y 

rfj rd^€i rwv iv8o£ordrwv Irrdp-^wv virkp rrposrdy- 
pharos vop,, p, 

21 *Att6 rov dpxovros N4as ' 'lovunviavrjs ovrws 

rdls rr^pi^Xinrois ^aprovXaptois rpiul rov dziov 
KovfiovKXelov vop,. 9 

rw 7Tpip,iKr)piw rwv Xaparpordrwv rpifiovvwv vvra- 
plwv vop,. /cS* 

rw avrov fiorjOw vop,. y 

rfj rdtjzi rwv ivSo^ordrwv krtdpywv vrrep Trposrdy- 
pharos vop,. p,' 

22 *Att6 rov dpxovros *App,€vlas 8evr4pas ovrws* 

rols TTispifiXeTTTois -^aprovXapiois rpiul rov dslov 
kovJ3ovkXsiov vop,, 9 



GREEK AND LATIN LEGAL DOCUMENTS 

tw TrpipiKrjpiw rwv Xapirpordrwv rpifiovvwv vora- 
piwv vop . 

tw avrov j3o7)dw vop,, 

rfj rd^ei rcbv ivoo^ordrwv irrdp^wv virkp npoardy- 
paros vop. 

23 'Atto rod dp^ovros 'Appevtas peydXrjs ovrws* 

roXs TTtzpifiAiiTTOis -^aprovXapiois rpial rod dsiov 
KovfSovKXziov vop. 

rw TrpipiKrjpiw rwv Xaprrpordrwv rpifiovvwv vora- 
plwv vop. 

rw avrov fiorjdw vop. 

rfj rdijei rcbv ivSo^ordrwv iirdpxcov virkp irposrdy- 
paros vop, 

24 *Atto rov dp^ovros KaTTTraooKias Trpwrrjs ovrws* 

roXs TTZpifSXirrrois ^aprovXaplois rpial rod dziov 
kov/3ovkMiov vop, 

rw Trpip,iKrjpiw rwv Xapirpordrwv rpifiovvwv vora- 
piwv vop, 

rco avrov fiorjOw V0 H" 

rfj rd^€t rwv ivSo^ordrwv indp-^wv virkp rrposrdy 
par os vop, 

25 'Atto rod dp)(ovro$ Kamra^oKias osvripas ovrws* 

rois TTZpifiXirrrois ^aprovXapiois rpial rov 0€iov 
kovJSovkXziov vop, 

rw TrpipiKrjpiw rwv Xapirpordrwv rpifiovvwv vora- 
plwv vop, 

rw avrov fiorjdw vop, 

rfj Tafet rwv ivooijordrwv krrdp^wv virkp rrposrdy- 
paros vop 

26 > Atto rov dpxovros ' EXevorrovrov ovrws* 

rots TTZpifSXiirrois ^apro vAap tots' rpial rod Belov 
kov/SovkXgiov vop 

rw TrpipiKrjpiw rwv Xaprrpordrwv rpifiovvwv vora- 
piwv vop 

rw avrov fSovjdw vop 

rfj rd^€i rwv ivSo^ordrwv iirdpxojv virkp irposrdy- 
paros vop, 

27 j Atto rod dp^ovros Evpwirrjs ovrws* 

rots TTSpifiXeTTrois ^aprouAaptots rpial rod Bstov 
KovfiovKXelov vop 



28* 



*8* 

y 



0' 

y 



0' 

k8' 

7 



8' 

7 



k8' 

r 



24* APPEISDIX I 

rto TTpipiKTjptcp rwv ] XapTTpordrwv rpifiovvwv vora- 
pitov vop. kS' 

tw avrov fioTjdw VOp, k8* 

rfj Ttz^ei rcov ivSo^ordrcov iirdp^wv virkp Trposrdy- 
paros vop, p* 

28 *Air6 TOV ap-ftOVTOS @p<XK7)S QVTWS* 

rois 7reptj8Ae77TotS' %aprovAap(ois rpujl rov 9siov 
kovJSovkKsiov vop, 6* 

TW 7TpipLlK7)piW TCOV XapTTpOrdrWV TpljSoVVWV VOTd- 

piWV VOp. K§* 

rw avrov fioTjdto vop. y 

rfj rdijsi rwv ivSo^ordrwv iirdp-^wv virep Trposrdy- 

pLGLTOS VOp, fl 

29 Atto rov ap-)(ovTos ^PoSotttjs ovrws* 

rois TTZpifSXiiTTois -^aprovXapiois rpial rov delov 

KQvf$QVKA€lQV VOp, 9* 

rw TTpipiKTjpiw rwv XapTTpordrwv rpifiovvwv vora- 
plwv vop. k8 > 

rw avrov. J3qt)9w vop,, y* 

rfj rd^i rwv ivSo^ordrwv iirdpx^v virep TTposrdy- 
paros vop, p* 

30 Atto rov ap^ovros Aipipovrov oiirws* 

rots 7T€pif$Ai£7TTOi$ %aprovAapiois rpiol rov 9siov 
kovJSovkAgiov vop. 9 y 

rw TrpipiKTjplw rwv Aaprrpordrwv rpifiovvwv vora- 
ptcov vop, /cS* 

tw avrov j3o7)9w vop, y 

rfj ra|€t rwv ivSo^ordrwv iirdp-^ajv virkp irposray- 
pharos vop, p* 

31 " "Atto rov dp-)(ovros Kapias ovrws % 

rois Trepi^XeTTrois ^apTOvAapiot? rpiol rov 9siov 

KovfiovKAzlov vop, 9' 

rw TTpipiK7)ploj rcov Xapirpordrwv rpifiovvwv vora- 

plcov vop,. kS' 

rw avrov f}orj9w vop, y > 

rfj rd£zi rwv iv8o§ordrwv iTrdp-^wv VTrkp TTposrdy- 

par os vop, p 9 

32 > Atto rov apxovros AvKias ovrws* 

rois TTepifiAdTTrois ^aprovXapiois rpiol rov 9eiov 

KOvfioVKXsiOV VOp,. 9* 



GKEEK AND LATIN LEGAL DOCTJMENTS 25* 

rw 7rpijj,iK7]pia) rwv Aapmpordrwv rpifiovvwv vora- 

pltOV VQp* kS* 

rw avrov f$Q7)$a> vop,, y 

rfj rd^i rwv ivSo^ordrwv irrdp^wv virep rrposrdy- 
paros vop,* p* 

33 'Aito rov ap^ovros Avyovurapvwrjs irpwrys ovrws* 
rots irepifiAeTTTQis xaprovAaplois rpiul rov Belov 
kovJ3ovkAgiov vop,, 6* 

rw rrpipwripiw rwv Aaprrpordrwr rpifiovvwv vora- 
piwv vop, k$* 

rw avrov j3o7)9w vop,, y* 

rfj rdi~zi rwv ivSo^ordrwv irrdpx^v vtrkp Trpoardy 
paros vop,, p? 



Kal ouai dpx ai rjyzpoviKal ijroi correctoricu* 

34 'Atto rov dp^ovros Aif$vr}S rrjs dvw ovrws* 

rots 7T€pij3At£7Trois xaprovAaplois rpial rov dsiov 

/cou/Jov/cAstov vop, &* 

rw TTpipuKTjpiw rwv Aapirpordrwv rpifiovvwv vora- 

piwv vop,, U 

rw avrov f$07)9w vop, y* 

rfj rd£si rwv ivSo^ordrwv irrdp^wv virzp irposray- 

par os vop,. As* 

35 *Atto rov iLpyovros Aiyvirrov rrpwrr^s ovrws* 

rots TrepifiXirrrois yaprovAapiots rpiui rov Bslov 
KovfiovKhsiov vop, 9 

rw 7rpip,iK7jpiw rwv Aapmpordrwv rpifiovvwv vara- 
piwv vop,, ii 

rw avrov [$07)9w vop,* y 

rfj rd^ei rwv ivSoijordrwv irrdpxwv virsp rrposrdy- 
paros vop. As* 

36 9 Att6 rov apxovros Aiyvirrov Sevrdpas ovrws* 

rots / 77€/?i J 8Aen , T0is' xaprovAapiois rpiai rov Osiov 
KOvfiovKAsiov vop,* 9* 

rw TrpiptiKrjplw rwv Aapmpordrwv rpifiovvwv vora- 
piwv vop, t€* 

rw avrov J3ot)9w vop,, y* 



26* APPENDIX I 

rfj rafet rwv ivoo^ordrwv iirdp-^wv vrrkp trposray- 
/xaros vojit. As 1 ' 

37 ^Atto rod dp^ovros AvyvarapLViKrjs Szvrepas ovrws* 

rots irepi^Xdrrrois ^aprovXapiois rpiul rov dsiov 
kov^ovkX^lov vo/x. d 9 

rw TrpifUK7]plw rwv Xap.irpordrwv rpifiovvwv vora- 
piwv vo/x. te' 

rw avrov fiorjdw vo/x. y* 

rfj raf ei rwv ivBo^ordrwv iirdp^wv vrrsp rrposray- 
ptaros vo/x. As' 

38 'And rod dp^ovros IJaXaiorivr]s rpirrjs ovrcos * 

roT$ TrepifiAdTTrois ^aprovXapiois rpiul rod deiov 
KovfiovKAziov vo/x. 6' 

rw 7Tpip,iK7]plcp rcov Xaparpordrwv rpifiovvwv vora- 
plwv vo/jL, te* 

rai avrov fioydw vo/x. y 

rfj ra^et rwv ivSo^ordrwv Irrdpywv virkp irposrdy- 
jxaros vo/x. As' 

39 *Att6 rov ap-^ovros 'Apafiias ovrws* 

rots 7T€/h/3A iirrois ^aprovXapiois rpiul rov Beiov 

KOvfSoVKXziOV VO/X. 6* 

rw irpipLiKTipiw rwv Aa/X77 pordrwv rpifiovvwv vora- 

piWV VO/X, It 

rw avrov fiorjdw vo/x. y* 

rfj rdfjzi rwv ivSo^ordrwv lirdp-^wv vrrkp trposray- 
piaros vo/x. A?' 

40 'Atto rov dpxovros Ev^parrjuias ovrws ' 

rots 77€pij8Ae7TTOts ^ajorovAaptot? rpiul rov Bstov 

KOvfioVKXzloV VO/X, 6* 

rw TTpipbiKTjpiw tojv XapLirpordrwv rpifiovvwv vora- 

piwV VO/X. t€ 9 

rw avrov fiorjBw vo/x, y 

rfj rdfjei rwv ivSo^ordrwv Irrdp^wv virkp irposrdy- 
piaros vo/x. A?' 

41 *Ait6 rov dpxovros Meuonorapiias ovrws* 

rols 7r6ptj8Ae7TTOiS' xairrovXapiois rpiul rov Bsiov 
KovfiovKXeiov vo/x. 6 9 

rw 7rpipUK7]piw rwv XapiTTpordrwv rpiftovvwv vora- 
riwv vo/x. te 5 

rw avrol j3o7]6w vop,. y* 



GKEEK AND LATIN LEGAL DOCUMENTS 27* 

rfj rd£ei rcbv ivSofjordrwv iirdpxwv we/? irposrdy- 
fjuaros vo/x. As* 

42 > Atto rod dpxovros KiAiKias Szvrzpas ovrws* 

TOIS 7T€ptj8Ae77TOtS XCLTTTQvAaploiS TpiUl TOV 9siOV 

kovJSovkAsiov vop,, 9* 

rw 77pi}UKripl(x> rwv Xapurpordrwv rpifiovvwv vora- 

piWV VOJJL, l€ 

tw avrov fiorjBw vop,* y 

rfj Tct^ei rcbv iv8o£ordrwv iirdpxcov vwep Trposrdy- 
p,aros vofx. As' 

43 'Attq rod dpxovros 'Appzvias TTpwrrjS ovrws* 

TOIS TTSpifiXiTTTQIS -)(apTQv\aploi$ TpiUl TOV 9zloV 

KovfiovKAelov VOjA* 9 

ru) irpifJUKTjpiw rcbv Aapmpordrwv rpifiovvwv vora~ 

pliOV VOjA. i€ ? 

rw avrov f3or)9w vop,* y 

rfj rdijei rcbv ivSo^ordrwv iirdpxcov virkp Trposrdy- 
p,aros vo/x. As 

44 *Atto rov dpxovros PaAartas 8svr4pas ovrws* 

rots iTzpifiXiiTTois xaprovXapiois rpiul rov Betov 

KOV^OVkX^IOV V0JJ,* 9' 

TCp TTplfJLlKTjplCp TCjOV XajlTTpOTaTWV TplfioVVWV VOTO*~ 

picov vop,. l€* 

rw avrov j3orj9w vo/x. y 9 

rfj rd^i rcbv ivSo^ordrcov i^rdpxcov virkp Trposrdy- 
fiaros vofx, As** 

45 9 Atto rod dpxovros ' Ovwpidhos ovrcos* 

rots 7r€pif$M7TT0is ^apTOvAap/ot? rpial rov 9etov 
kovJUovkKsIov vojj,* 9 

tw rrpi}AiK7)picp rcbv Xafj,7Tpordrwv rpifiovvwv vora- 
picov vop,, he 

rw avrov f$o7)9w vop,, y 

rfj rd£<zi rcbv ivBo^ordrwv zTrdpxwv virkp irposrdy- 
jxaros vop,. As* 

46 9 Att6 rov dpxovros rcbv Ntjuwv ovrws* 

rols TTepifiAeTrrois ^aprouAapiots' rpiul rov 9ziov 

KOvfSoVKAsloV VOjJL. 9 

rw 7rpi[MiKrjpicp rcbv XaiiiTpordrwv rpifiovvwv vora- 

piwv VOfl* l€ 

rep avrov f$07)9w vop,, y 



28* 



APPENDIX I 



rfj rd^i rmv iv8o£oT&TWV zirdpxoiv wrcp rrposrdy- 
fMaros vo/x. As' 

47 > Atto tov dpyovros Mvalas Sevripas ovrcos* 

rots 7rspif3\i£7rT0is ^apTOvAaptots' TpiM tov Bsiov 
KovfiovKAtiov ' VOpb. d > 

rtp irpifJUKypiw rcov XafMrrpordrcov rpifiovvcov vora- 
pio>v VOjl* te* 

rip avrov fto7)9(p vojx, y* 

rrj rd^et rcov AvSot; ordroov e-Trdp^coy virkp irposrdy- 

jJLCLTOS VOjU, As' 

48 *Att6 tov dpxovros Sicvdlas ovrws* 

rois Tr^pi^Xhrrois xapfovAaptots rpial rod Bsiov 

KOV^OVkX^IOV VOfJU* B* 

rw 7rpifJUK7]p(<p rcov Aap/irpordrvov rpifiovvcov vora- 

piCOV VOJU-. t€* 

tw avrov fioydcp vop,, y 

rfj rdi~zi rwv ivSo^ordroov errdp^cuF vrrkp Trposrdy- 
pharos vop,. As' 

49 iTapd 8k e/cdoT^s 1 ttoMcqs £k8ikov, €t ftzv git) p,y)rpo7roMr7)$, 
vTTGp Trposrdyfiaros ets* ra rcov iv8o£ordroov iirdpxwv 8l8ou9at, vop,la~ 
juara S% ei 8e aAA^s ttoAgcos, vop^y** kqX iripa rovrcov psr)84v* ov8s 
yap rov$ £k8ikqvs owe StSdvat rots dp-^ovutv ov8k iripoo nvi oi're 
Aafj,f$dvzw jSovAojtteffa, 'jtA^v et /Z77 rt^es €ialv avrois e/c tov 8tj jjLoaiov 
mtp^djitevat ow^etar etSoVtov avrcuv a>s, et p,r}vvd<-i7) rw ^/xerepa* 
Kpdrzi irzpl rivos avrcbv, cos TrapafSalvoi rd Trap* rjj^cov dsuTTiudivra, 
Kot oTrep dv Ad/Jot quadruplum d7roSt6cret, Kal rfjs <j>povrl8o$ 7rapa- 
At»0€ts i£oplav oIktjusi Swpe/aj.' oVdre /cat oi rcov iirapxicbv dpxovT€s> 
et rovrov 7rapap f eXrjCFeiav Kal rov$ it<8iKov$ 7repu8oi€v KkiirrovraSy 
oiK iXdrrpva /cat avrot rroivrjv vTrocmjaovrai* 

Dat xvii k, Mai, CP. Belisario <v. c> cons. [a. 535], 



GREEK AND LATIN LEGAL DOCUMENTS 



29* 



E, Novella XX * 

De administrantibus qfficiis in sacris appelaiionibus 

IIEPI TQN YIIHPETOYMENQN O00IKIQN EN TOIE 
SAKPOIS TQN EKKAHTQN, 

c O avros fiacriXevs *Iwdvvr] rw ev8o^ordrw irrdpyw rwv lepwv 
praetoricuv to /?', and vrrdrwv /cat TrarpiKiw, 

«IlpootiAi,ov). 3f H8r) }iev dziov iTTOirjodfAzda vop,ov irepl rwv itfiiaewv 
8iaXeyop,evov > rival XPV Trapa^vXdrreaBai rporrov eir* avrais, /cat 
oBev els rlvas <f>epeaBai rds ck/cA^tous*' ov rrpos re rr)v otjv v7repoyr}v 
rrpos re rov evSofjorarov rjp,wv Karen epLifsapLev Koiaiurwpa, Eirei8r) 
8e 7roAA^ yeyovev dp,(f>ia^T7juis rrepl rwv V7T7jperovp.evwv Tavrais 
6<f>tj>iKiwv , twv puev €K rov Beiov rwv emaroXwv aKpiviov rds rwv 
spectabilicuv St/caoTtov olKeiovjievwv iv rals eKKXrjrois vrr^pealas, 
rwv 8e €/c rov Bpovov rrjs vrjs V7repo)(rjs jxiyiura 8t]Xovvtwv Tj8iKrjuBai, 
el fj,eraf3el3Xr}pLevov rod or^/xaros' ovkzti jxovoi rats eKKXrjrois VTr^perrj- 
uovai rals aV6 rwv Xafjurpovdrwv rwv iirapxi&v dp^ovTwv ip^ofievais 
els {xovov to uov SiKaurrjpiov, KaBdwep irpoTepov r\v rjviKa iv Belw 
fxev /cat avros rjKpow SiKaorrjplw, vnypereiTO 8e r) rd^is rj ur\ t dXXd 
8id to rwv UTTeKrafSiXlwv cr^jua iv Ta^ei Beiov aKpoarypiov rrjs 

VTToBeueWS KlVQVlxivTjS) KCLI UWaKpOWfJLeVOV rfj ufj V7T€pO-)(f) KCU TOV 

ivSo^ordrov tjjxwv KOiaiurwpos, /cat eKaTepov jxepovs ro irav olKeiov- 
fxevov, /cat uvva^Bevrwv Trapd re rfj ufj virepoxfj xal rwv ivSo^ordrw 
rjfxwv Koiatarwpi TroXXaKis rwv re €/c rwv Beiwv uKpivlwv, otrrep rals 
e<f>euemv VTrTjperovvrai, rwv re eK ttjs rdfjews rod Bpovov rov uov* 
TeXos els nva rvrrov ro irpayp,a irepiearT), ov dypdtf>ws els rjfJL&s rjydyere. 
to TTpaypia 8e /cat rjjjuv ovk 0.776 rpoirov yeyovos e§o^€. Kal rews> 
i7rei8rj7Tep JJa(f>Xayovia /cat *Ovwpids, 8i7]pr}p,evai irporepov els dp^ovras 
8vo y els eva /cat rov avrov irepieuTT^uav to rod irpatrwpos ovojxa 
TrposAafiovra, rovro dvafM^iu^rjr^rws e8ofje ro cr^jua rfj afj rrposrJKeiv 
dpxfj. Tavro Se /cat irrl rwv wore 8vo IJovrwv, rovreariv EXevorrovrov 
re /cat Ilovrov IIoXep i wviaKov t KaKelae yap 8vo Kadeurwrwv ep.irpoaBev 
dpxovrwv, vvv 8e evos rov p,o8epdrwpos yeyovoros, KeKoa[JL7}fj,evov 
/cat avToO rfj rwv TTepifiAeTrrwv d^ta, 7?aAtv ravra uvvefiaive /cat els 
to uov jxovov 8iKaurrjpiov <f>epec?8ai rds a7r6 twv eKKXrjrwv St/cas 
ixprjv, Kara fievTot tovs ottovs rfjs irepl twv iKKXrjTWv 8iaTa£ews* 

5 CJC, 6th ed. Ill, pp, 140 sqq. 



30* 



APPENDIX I 



CAPUT I 

2vv7Jpeae roivvv djxa /xev tols virovpyovuiv e/carepa ra>v dp-gcov, 
a/m 8e vfjuv dfju^OTepois, /cat irpos ye Kal rjfMV opdcjos e8o£ev iyeiv to 
Trapaardv TjpZv, coore p,0V7)v ttjv rd£iv ttjs utjs VTrepoyrjs rats rotavrais 

V7T7)p€T€lV €/C kXtj TO IS, Kad&lTSp Kal 7TpOT€pOV TjV y el KoX €V C^/XCm 

061OV aKpoarrfpiov XeyoiTO Kal TrapeiT) Kal 6 ivBo^oraros rjfxQv KoiaioTwp 

Kal jLt€T€^Ot TCOV 7TpaTTOfJ,€VWV. 
CAPUT II 

MAAa fX7)V eireiirep 6 ttjs 7Tpa>T7)s Ka7nra8oKias TjyovfjLGVos TrpoTepov 
els ttjv (jtjv apx?) v ecopa fiovTjv KaKelue to tvov e(f>eaewv i<f>4p€TO, vvv 
8e els to tov TrepifiAeTTTOv avdviraTOV ju-eTa/Se/JA^Tat cn^/m, ov8ev 
t)ttov TrposrJKov eaTi, Kal tt)s OLp^^S eKelvrjS e<f>ecriv 8exojxevT)s Kal 
ava7TGfjL7TOfjt,4v7)s ttjs 8iktjs ivTavda, Kara, ttjv detav tjjjlwv StdVa^tv 
ev Tatjei deiov aKpoaT7)piov avTrjv aywvi^euBai, ovvovtos Kal tov 
ivSo^oTaTov rjfjLcov KoiatcrTcapos /cat avvaKpocop,evov ttjs VTrodeaems 3 

fAOVTjS §€ TTJS Ta£J€Q>S VTTTjpeTOVfJLeVTJS TTJS CF7JS, eWtSTJ Kal TTpOTGpOV 

tovto £vev6fJuoT0. El yap Kal 6 7Tepif3Ae7TTOS kojxtjs tcov oIkiwv avva- 

V€fll)(9r) VVV T7] dpxjjy aAA* oSv 0VT€ 7TpOT€pOV TToAAdl TW€$ GKIVOVVTO 

St/cat trap* avTtp otrre eK tov 8t.KaoTT)plov tov Kar* ai>Tov i(f>epeTo 
ris cr^eSov e<f>eais evTavda. vvv 8e 8tj Kal rd Trepl rds TapuaKas Stot- 
KTjuets Kal eTepois Tial irape8u>Ka}xev 9 Kal ov 8ei irapa tovto iXaT- 
TwOrjvai tov uov dpovov, dAA* opLolws ttjv ut)v VTrTjpeTelaOai Ta£tv 
fj,6v7)v Tats ivTavda cjtepofxevais virodeaeai, 

CAPUT III 

TavTo 8e tovto /cat eirl tov ttjs *App,evlas avOvwaTov, iwGiSi) 
TTpOTepov apxrjv avTTjv 6p8ivapiav [epb^adpLov] 7TOirjcravTZs vvv ov8ev 
avTjj TrposdevTZS els to ttjs avdviraTeias ixeT7jyayop,ev a^fta. ifat 
yap 8tj Kal Tals eKeWev St/cats rj ttjs crrjs vTrepoyrjs VTnjpeTTJcreTab 
Ta£is, ttjs 8iK7)s p*€v iv Tafei #€tov aKpoaTTjpiov* Kadamep elirovTes 
s^drjpev, Kivovjxevrjs > Trap* ap,<j>OTipoi$ 8k vpZv e^eTa^o/xeV^s ov8kv 
8k tJttov ttjs Tatjecos uTJs vTTTjpeTovp.ivrjS tw TTpayjxaThy Kaddirep Kal 
TTpOTepov rjV) rjviKa p,6vov to ttjs dp^rjs ttjs KaXovp,4v7)s 6p8ivapias 
el)(e ayrjpta jxei^ova Tafjiv ov Trpos^afiovva. 

CAPUT IV 

'ETTeiSi) 8e Kal AvKaoviav Kal iJtcrtStav Kat *Iuavpiav W dpxovvi 
iTpoTzpov T€Tayp,evas Kal Tas €K/cAtJtovs dvaTreparovuas els tov dpovov 



GREEK AND LATIN LEGAL DOCUMENTS 



31* 



rov aov vvvl Koap,r)9rjvai rfj rcov rrpaircopwv dpxfj ovp,f$efir)Kev (el 
/cat SoKei rrcos ovvavap,ep,lx9ai ris avrfj /cat arparicoriKr) Ta^is* erreibr) 
rrporepov Kal Sov£ e<f>* eKaarrjs rovrcov errapxias rjv), dvayKaicos 
fjjuv ex^iv eoo£e Sid rov KaiviapLOV rovrov povco 8r) rc7> 9povco aco /cat 
rep evSo^ordrco Koialarcopi 7rapaSovvat rrjv rcov e<f>eaewv e^eraaiv, 
Sowat ok <f>iAav9pco7Torepov rfj rd^ei rfj afj /cat rois iirl rovrcp rrpar~ 

TOjxivOlS V7T7JpGT€T(j9ai* t5(7T€, €? Tt yeyOVe TOIOVTOV ep,7TpOo9eV Tj Kal 

varepov yevrjrai 3 rrjv avrr)v rep rrpdy pan rd^iv vrreivai 9eo7ril > op,ev . 



CAPUT V 

> Eirei8r) 8e 8vo Ka9apcos r)aav ctp^at rov re Kop/qros rrjs t Ecoas rod 
re dpxovros avrrjs rrjs rrpcorrjs 2vplas, /cat at p,ev rrjs rroXiriKrjs 
ravrrjs dpxrjs e<f>eaeis els rov aov i(f>epovro 9povov > rrjs rd^ecos vrrrjper- 
ovarjs p,6vrjs rrjs orjs, at 8e rov Kop/rjros rrjs *Ecoas, ola arreKrafSikiov, 
Kara ro rcov 9eicov eKpoarrjpicov o)(fjp,a els re rov 9povov rov aov 
eis re rov ivoo^orarov Koalarcopa, jjlovwv rcov 9elcov a/cpt (VI) vtcov 
V7T7)perovp,evcQV* dAAa rovro *** ev rovrcp rev p,epei Kakcos rjpXv eSofjev 
eyew errl ravrrjs 8r) rrjs dpxrjs Koivrjv Sovvai rrjv vrrovpylav rois re eK rov 
rcov 9etcov ernaroXcov uKpivtov rois re eK rrjs rdijews rrjs arjs vrrepoxrjs* To 
yap 8r) rwv rrp6a9ev 8vo j8t/capia>v re IIovriKrjs rrjs re > Aaiavrjs rrav- 
rekcos Kaiviu9ev /cat els rr)v dpxrjv p,6vrjs errapx^s ju-tas 1 [leraardv, 
raXarlas *j>ap,ev /cat &pvyias IlaKariavrjs > <f>oirdrco p,ev rrpos re rr)v 
arjv vrrepoxr)v rrpos re rov ivSo^orarov Koialarcopa, piovrjv 8e rr)v 
vrrrjpeaiav 8exe*a9w rrjs rd^ecos rod 9povov rod aov, 

CAPUT YII 

KaKelvo p,evroi 9ea7rl£ ) op,ev > Sure errl rovrcov Sr) rcov dpx&v rcov 
vvv Trap* r)p,cov e£jevpe9eiucov Kal pierafSakovacbv ro dpxaiov cr^/xa, 
etre avr69ev Kara rrjv <j>vaiv rrjs ot/cetas" St/caaatev dpxrjs elre /cat 
€/c TTapairopmrjs r)p,erepas, ravro <f>vAdrrea9 ai o*^/xa* /cat ev9a p,6vr)v 
virrjpereiv rr)v rd^w rrjs crrjs vrrepo (VIII) x 7 )^ e9eaTriaap f ev > 6p,olcos 
[eire eK rrapaTropmrjs] elre eK rrjs rov hiKaurr) plov <f>voecos r) e<j>ecns 
dveASot, rr)v rd^iv rrjs orjs vrrepoxrjs V7Trjpereta9ai rats i<f>eaem 9eoirit s o- 
p,ev, etre eK 7rapa7rop,7rfjs r)p,ere (IX) pas', 6p,olcos rrjs rdtjecos ear ai rrjs 
orjs* *Ej>* ols re koivy)v emojxev rr)v re rwv <crcDv>> rd^ecov rrjv re 
eK rcov 9eicov OKpivicov vrrovpylav^ 6p,oicos rr)v Koivorryra tj>vXdrrop,ev p 
elre eK rraparvopmrjs elre Kara ro rerayp,evov ev rw oiKaurrjplco 
yevoiro rd rrjs e£erdaews. 9 Ett 9 eKelvwv pevroi rcov Bikcov, as ov arreKra- 
jStAtot SiKaaral Kpivovaw, aAAd avvrjyopoi p,6vov, eeff Sv i(j>epero ra 



32* 



APPENDIX I 



rrjs vito94g€cos ets Te tov dpovov tov uov el's re tov ivSotjorarov rjfxcbv 
KoialvTtopa, twv KadwaiwpLevwv Xi^XXt]giwv V7T7)p£T0vp,£vwv avrais, 
i7T€i8r} jjL7]bkv ttclvtgAws inl ravrais KGKaiviorai, to iraXaiov (j>vXdrro(jLev 
oXqiLa* wsTTZp /cat iirl twv d'AAcov dirdvTwv tcuv ov KaiviuBivTwv ra 
rrjs TraXaids virovpy las iizvziv £<f>* iavTwv Siardrro^v, ovogvos vzwt ipov 
y^vofxevov, 6 yap 67rtav/j.jSas > Kaiviapos dAAotov 7tws xprjvai yzvdudai 
Kal to twv virovpyovvTwv weSet^e u^rjfxa, 

< 5 ErriXoyos^ , Xd toIvvv irapauT&vTa rjfjuv Kal 8td tovSc tov 
Bslov SijAovprGva vofjiov 7} GT) vir^po-^j] epyw Kal iripaTi irapaoovvai 

UTTZVUaTW . 

Dat. xv, k, April, Constantinopoli post cons. Belisarii v.c, [a, 536] 



F. Novella XXI 6 
De Armeniis ut ipsi per omnia sequantur romanorum leges 



KA 

IIEPI APMENIQN Q2TE KAI A YTO YS EN IIASI 
TOIZ PQMAIQN AKOAOY&EIN NOMOIS. 

f O avTOS fiacriAevs %4/ca/aa> tw /xeyaAo7r/>€7recrTaTa> dvdviraTw *AppL€vla$* 

<Ci7poot/xtov>. Ttjv *Ap}i€vlwv xwpav TeXelws €vvop,€io9ai /3ovAo- 
juevot /cat fJLTjSzv ttjs aAA^s" rjf^wv StecrrdVat iroAiTZias dp^ais t€ c Pa>/xai*- 
Kats e/cocr/^cra/xev, twv rrpoTspwv avTrjv dVaAAafavres* ovofjidTWV, 
cr^/xacrt re xpTjadai tois f Pa>ju.aiW> owet^tcra/xeVj 9sup,ovs t€ ovk 
aAAovs €tvat Trap' avTois rj ovs *Pwp,aiot vojxi^ovgiv era^a/xev. Kal 
(prjdrjixev xprjvai pjjTW vop,w /ca/cetvo iiravopBcbuai to KaKtos Trap 
avTois afJLapTavojjLGVov, /cat firj /caTa. to fiapfiapiKov e9os dvSpwv jxkv 
€tvat tcls StaSo^d? twv t€ yoviwv twv t€ aSeA(£a>v to£ re aAAou yivovs, 
yvvaiKwv Se oi5/c en, /i^Se ^tupis irpoiKos auras cts* avSpos cf>oiT&v 
/x^Se ayopa£€tx#at irapa rtov crwot/c€tv /xcAAdvra)^ rouro oTrsp j8ap- 
j8apt/ccoT€pov /xe^pi tou vw Trap' auTots 1 ivofjulaBj]' ovk ovtwv fjbovwv 
TavTa aypia>T€pov ooi^auavTWV, aAAd Kat sTtpwv i9vwv ovtws art/xa- 
aavrcov t^v tf>vGiv Kal to OtjXv TrzpivjSpi&dvTWV, ws ov Trapa $€ov 



8 CJC, 6th e&, III, pp. 144 sqq, 



GREEK AOT> LATIN LEGAL DOCUMENTS 



33* 



y€v6jj,€Vov ovSe ovvtgXovv rfj yevsmovpyia, dAA* (h$ svrshis tg /cat 
77TtjU.a074.eV0v /cat Travis e£cQ irposfJKov /cafleardvat Ttp/rjs, 



CAPUT I, 

©GUTdttOjizv roivvv Std TOvSe rov #etou vojiov, costs /cat irapd 
> Apfj,€viois ra avrd /cparetv dVep /cat Trap* 77/xtv Trpo^dcret tt)s tG>v 
Br\ksh€>v Sia§Qxfj$> xal /^Se/xtav sivai 8iaff>opdv appevos Te /cat ^Aetas. 
MAA* tSsTrep eV rots rjfxzTGpois vofxois riraKTai, Kara rrdiov p,sv ayrjixa 
KXypovofiovm yovsiS; yjyovv TraTepa /cat p/^repa, /cat tto/itttov zeal 
liajiptyv, /cat Tew ert iroppwripw^ ^ /cal tows fter* avrovs, rovr4ariv 
vlov /cat dvyaripa y ottws Te a??Tot KXrfpovopLOVvrar ovtws /cat Tnxpd 
*ApfjL€Viois etvai /cat /xTjSev ra ^Apjxsvias vofiifxa rcov ^Pcopaicov Steer- 
ravat. -Et ydp ttJs rj^ripas TroAtTetas etcrt SouAevaucrt Te 77/xtv juerd 
t<w aAAcuv iBvwv /cat tt&vtwv dmoXavovui rcov ^/xerepcov, ov St^tou 
jxovai Trap* awots at #yJAetat ttJs ^ctp* rjfuv laorrjros iKfiAydrjaovrai* 
dAAd Trdcrtv eV tuto rd rcov rjfjLGTipwv eWat vojjlcjov, oaa re e/c rcjov 
rraXaicbv avvrjdpoiaaiM^v /cat eV rots rj^ripois eVd^aftev ivanrovrois 
re /cat Styecrrots 1 oaa Te e/c ttjs j8acrtAt/C7}s' vop,o9eoia$ rcov re ipmpoadzv 
avroKparopwv /cat rjfAwv [re] avrcov amoyiyptxrnrai,. 

CAPUT n. 

Taura roivvv drravTa /cparetv els rov wrravra #ecrTrt£oju-ev xpovov, 
dp^o/xeva e/c TrpooijMiwy rfjs irapovur}$ Tecrcrapes/cajSe/caT'rjs' em v^rjuews 
/cat avT'^S'^ /ca#* 7?i> rovSz ypd^o^v rov vop,ov. to yap /cat ra TraAatorepa 
TrGpiGpydaauOcu /cat irpos tovs dvw xpovovs dveA#etv avyxvusms 
jxaKKov V) voptodzaias iartv* dAA v e/c rcov xpovcov, /ca#dVep zIttovtss 
gtfidrjfjLZV, TTJs 7rapov(j7]s reo-crapes/catSe/caT^s 1 eTrtve^crecos' /cat avrT}? 
/cat /card rd> e^s diravra ^povov at StaSo^at fxepiTtocrav ojuotat, 
tco^ €/c 7rdo7)$ amW efe StaSo^d? ^epo/zeVajv o/xotco^ p-e^ em yvvat/ca)v, 
ofiolcos Bk £tti dvSpcov tov Xoittov <f>vXaTTQ}i£vtov * To 8e e/^7rpocr9ev 
y€v6p,zvov airav /^eVetv e^rt tov trpoTipov uxrjfxaTos icbjAzv, etre €7rt 
y€V€ap)(iK<jbv €tTe eTri rcov dAAa>p yeyoveVj ovS* ot*ow eTriKoivcovovvrcov 
toov OtjAgiwv 7rpo$a>7r(jQV e^rt rots ^'S^ Stave^fletcrt yevecrtap^t/cots 1 
^copiois rj rats yevofievats StaSo^ats* ^XP l r V$ rptS'/catSe/cdr^s iinvefA- 
rjuecos /cat avT^s" dAA* e/c rod p7)94vTos xpovov > tovt4utiv oltto reaaapss- 
/catSe/car^s iinvep/rjuews^ /cparetv rd Trap tjjjllov vofjLo9€TTj94pTa 
#ea"77i£ojU€V, 

<C*^Tr/Aoyos k ]>» Td to&w Trapaurdvra TjfMV /cat §td rovSe rov 



r 



34* 



APPENDIX I 



delov S^AozJjueva vofiov r) ar) p,eyaXoirpeireia /cat ol jxer avrr)s rrjs 

0>PX?)S dvTlX7)lfj6fM€POl 7Tapa<j>vXaTTGW sis TO 8l7)V€KSS uirovhaudrwaav , 

Dat, xv, k, April, CP, post cons, Belisarii v. c, [a, 536] 



Gk Novella XXXI ' 

De disjpositione quattuor administrationum Armeniae 

AA 

IIEPI AIATYnCtSEQH TQN TEZZAPQN APXONTQN 

APMENIA2J. 

*0 avros fiaaiXevs *Iwdvvj) rw ivSo^ordrti) eirdp-)(ip rwv lepwv rijs 
"Eco irpaiTwpitov ro j8*, and virdrwv /cat irarpiKitp. 

<iU pooipiiov^> . Td p,drr}v Kelp,eva Kal eKKeyypbivws el rrpos rr)v 
TrposrjKovuav dtf>iKoiro rd£w koi Siaredeirj /caAtos, erepd re <aV> 
dv9 erepcov rd irpdypLara <f>alvoiro KaXXlco re e/c ^eipovwv ig aKoapLtov 

T€ KeKOUjlTJlxiva 8l7]p9pQ)fA€Va T€ Kal SiaK€KplfJL€Va €K TCQV efiTTpOudeV 

araKrcov re /cat uvyKeyyixevtov* Tovro Kal eirl rrjs ' App,evia>v -gvopas 
dfiapravofxevov evpovres thrj9r)p,ev xpjjvai, Trpos p,iav dpp,oviav rd^ai 
avrrjv, Kal 4k rrjs evra^las loyyv re avrfj Sovvat rr)v TTposrJKOvuav 
rdtjiv re eirideivai ttjv rrperrovuav, 

CAPUT I, 

Toiyapovv reaaapas etvai ireirovfi Kajxev ^ApjjLevias* rrjv p,ev evSordrrjV, 
r)s r) fZTjrpoTroXis rfj rrjs evcrejUovs rjfjrtov TTposrjyopias eiroyvvp^la /caTa/ce- 
Koufxrjrai rrporepov Ba^avls o^TOt /leovTOTroAts* KaXovfievrj, yvTrep Kal 
avBvTrareia reriixrqKap.ev y 97s ^/cd/aos TrpoearrjKev 6 ixeyaXoTrpeTreara- 
ros > VTTSKiafiiXtav re dVo^TpaFTes' t^v dp-)(r)v /cat irdvra Sovres avrfj 
oiToua TrposrJKov ianv avdvirareiav e^eiv aroXfj re yap avrrjv Kare- 
Koop/rjuapev avdvirdrov /cat rrdvra aKoXovOa rovrois e^eiv Sterwdicra- 
/xev. /cat TroAets* avrfj 8eSa>Kap,ev 0eo8ocnov7roXiv re> t}v /cat irporepov 
elye> UdraXdv re Kal NikottoXiv Kal KoXa>veiav e/c rrjs TTpcorjv irpwrys 
*App,evias KaXovp,evrjs Xafiovres, TpaTre^ovvrd re /cat Kepaaovvra 
e/c IJovrov rod irpwijv TIoXe^wviaKov KaXovjievov, ^wpiaavres avrcov 
ras jiev rov XafXTrpordrov ttjs eirap^ias ap-^ovros ras 8e rod TTepifiXerrrov 
/^.oSepdrcupos'j eirird re iroXem rr)v oXrjv enap'xlav Trepiarrjuavres /cat 
OTToaa rrjs TrepioiKiSos earlv avrcov, 

7 CJC, 6th ed,, III, pp. 235 sqq. 



GREEK AND LATIN LEGAL DOCUMENTS 



35* 



1 Aevripav 8k erdf a^tev *Apjisvlav rrjv spmpoadzv TTpcorrjy KaAov- 
pivqv* r)s rjysirai Ezfiavrsia, itoXsis avrfj irposvGip,avr€$ rrps re rcbv 
Z'ejSacrTOTToAtrcDv rjv Kat rrporepov et^e, Kat irpos ye Kojxavd re e/c 
rov KaXovfxevov TTpcorjv IJoXeixwviaKov JJovrov /cat ^Aav e/c tov 
* EXevoirovrov, /cat ju/yp /cat Bpiaav, w$re ev irevre TrdAscrtv etvat 
t^v eirapxiav ravr7]v ? /cat t^v dpyTjV rjyefiovlav ovuav KaraXarovres 
eirl rod irporepov tT^ju-aros' /cat t6v apxovTa avrrjs ov8evl 
/coo-jU'^craFTes' ovo^an ixelt.ovi, dXX o Trporepov et^e tovto avT<£ 

2 /caraAtTTovTes', *^77t tovtois re rplrrjv 'Appevlav Kareor7)odp,e9a 
rrjv Trporepov Sevrepav, fjs yjyeirai MeXiryvrj ttoXis d/^ata, ttoXis 
iTTiG7)fjLOS, ev KaXw re yrjs /cat depos Keip,ev7} /cat ov8k rroppw SteoTcuaa 

TVOV TOV Ev<j>p&TQV peVjxdrWV, ravrTjV Wr}97}fJL€V §€IV /CaTCt TO TTCLpOV 

av^rjaai kcli els to rtjov aireKra^iXiwv p,eraor7jcrai o-^/xa, tw re 
dp^ovra ravrrjs > lovariviavov oVo/xdcrat Kopyra, 8ovvai re avrcp kclI 
vrrkp mrrjaewv solidos septingentos /cat rep ye avrov Trape8pw 
solidos septuaginta duo /cat rfj ye avrov rd£ei solidos sexaginta 
drravrd re e^ew orroaa rwv rotovrwv earlv tSta 9povwv. rovs re 
Trpwrjv 6vojjLat,opivovs ra^ewras rrdvra p,kv rrpdrreiv orroaa Kat 
ep,7rpoa9ev, /cat jxaXiara rrepl rrjv 8rjp,ouiav elsrrpa^iv r)G)(oXr}o9ai, 
els 8k rrjv r€>v Kopiriavcbv Trposvjyoplav p,erafSaXelv , wdvrcov avrols 
ovrco <f>vXarTop4vwv ws tjvikcl TafetoTat KaSeor^Kecrav. JUoXeis 
8k V7T€KAivap,€V avrfj rovro p,ev "ApKciv /cat "ApafihOGOV, rovro 
8k * Apiapd9eiav /cat ifo/xava irepav (KaXovm 8k avrjjv /cat 
Xpvarjv) Kat Kovkovu6v 9 as* /cat rrporepov e%)(ev ev If TrdAeat 

3 aw€crTcuo*a. Svveur7iudjxe9a 8e /cat rerdprrjv 'AppLevlav, 7) 
TTporspov ovk els inap^tas gw4k€ito u^Tjp^a, aAAa rwv T€ £9vcov fy 
/cat 6/c Siat/>6pwv avvetAeKTO fiapfSapiKwv 6vop,drwv y T^o^avyvrj re 
/cat ^Av^ttjvt) tj Tlocf>7)V7) /cat > AodiavrjV7), rj /cat BaAafiiTTjVT) KaXovp,dvr) 
/cat wo uarpdirais o5o*a* a-p^s* §e tovto ovojxa rjv ovSe *Pcop,aXKOV 
ovSk rtbv rjp.eripwv TTpoyovcov, dXX* i£ iripas TroXireias els^vrjveyixdvov. 
KaKaelvrjv roivvv dpxrjs ttoXitiktjs iKoup/fjuaiiev cr^/xaTt, ap^ovra Te 
TToAtTt/cov ey/caTacrT^aavTes /cat TrdAtv T€ avTTj ti^v tcuv Maprvpo- 
ttoXitcov Kat to Khdapit^ov oovtzs <j>povpiov* Kat avT^ Se €V to) twv 
opSivaplcov dpxwv Kariurrj cr^/xaTt KovoovXapla Trap* ^fico^ ysvoptivrj* 
cust€ reoadpwv *Apij,€viajv ovcrtbv 8vo p,£v etvat crTTCKTajStAtas 1 , tq^v 
T€ ToiJ av0V7raTov t^v T€ tov Kop/i)Tos, Kat dv#v7raTov /zev eJ^at TW 
ttjs TTpwTTjS rjyovptGVov *App,€vlas. Koprqra 8k rov rrjs Tpirys, rov 8k 
rrjs 8svr4pa$ Kat rsrdpTTjs 6p8ivapiovs Kadzurdvai* KaX irr€i8i^'7T€p 
rovro rjpCiv SiZGTTOvSaarai, wsrs rd$ &XP h ™ v ^^VTaKoattuv vop,io~ 



36* 



APPENDIX I 



fidrwv iKKXrjrovs ov-)(l Trpds ravrrjv <f>£p$odai rrjv svBaljxova ttoXiv, 
dAA' irrl rovs avvsyyvs utt€ Kraft iXiovs dp^ovras, Kal rovro SiarvrrovfAev, 
wsrz tw ixkv apxovn, rrjs irpwrrjs > App f €Vias, rovriuri rw avdvirdrto, 
rds e/c rrjs Szvnipas *App,€vias iKKXrjrovs (f>ip€adai p rovriuri rds 
Kara Szftdarziav, rep §€ rrjs rplrrjs Apjxzvlas Kopirjri, rw Kara MzXirrjv- 
rjv ^aju-ev, ras e/c rrjs rerdprrjs * App,$vlas iKKXrjrovs p<4xP l ro ^ pTjBivros 
dvrJKeiv 7TOGOV, 



CAPUT II. 

Tovrwv rolvvv ovrws rjpXv 8iar^raypiiva)v /cd/cetvo TrposSiopiaai 
BiKaiov ert vofjul^opLGV, i<j>* w TTpourrjcrai rrjs rpirrjs ' App,€vias avSpa 
aefjivov, VTrovpyrjKora T€ r\pXv rj8rj Kal d^iov rov rrjs dpyrjs oyKov Kal 
7Tpoo-)(T)iJ,aro$* Evpovrss rolvvv @a>/xdV rov fMeyaXorrperriurarov rj8r) 
fxkv ap^ds* irrl rrjs 'App,€via)v dvvaavra ^copas, Kal rdAAa §€ dVSpa 
^prjurov Kal yvrj ericas rjjjuv vrrrjperrjadiJievov re /cat VTrrjp^rov^voVy 
avrov irrl rfj rrjs dp-^rjs ravrrjs irpoftaXXop^eBa SioiKfjasi, cosre recos 
fxiv rrjs iirap^ias ravrrjs rjyeZadai Kara ro prjBkv rjpXv cr^p-a, trpovosiv 
8i /cat rwv aXXcov oiroaa av avrtp [r]] Std deiwv imrpi\jjaip,<zv com- 
monitorial etre iirl rrjs iirap^as rjv avrw TrapaSeSai/cap-ev etre 
/cat €7r' aAAats* onsp /cat rr^TTpd^a^v Seta Trpos avrov TT^rroirjixivoi 
eommonitoria Tiept ttoXXwv Kal 8ia(j>6pwv irpd^zcov, arrsp avrov 
/cat els iripas xwpas irposrJKov 1 iuriv els epyov dyayeiv. Td \iivroi 
rrepl rds Upcoavvas, Kadd noXXaKis eip^/cap-ev, p,4veiv Kara ro 
TTporspov ftovXojxeda cn^p-a, ov8kv ovre Trepl ro fjLrjrpoTroXiriKov 
SiKaiov ovre rrepl rds ^etpoTovtas 1 rov rrpdypiaros dfjusiftopiivov r) 
Kaivi^ofxivoVy dAAd rG>v rrporepov )(€iporovovvrcov Kal vvv iyovrwv 
rrjv rrjs ^iporovlas i^ovuiav, Kal rvbv Trporipwv p/rjrporroXirwv irrl 
rrjs iavrcbv pievovrcov rdtjews, wsre pirj8ev ro ye irr* avrals Kaiviodrjvai. 

CAPUT III, 

*Ekgivo piivrot ra>v dvajpLoXoyrnjiivcov iarlv, ws iir^ihrjir^p rov 
rrjs rpirrjs * Apjxzvtas KOfjLrjra ov rroAiriKov jjlovov, dAAd /cat ur par xa>- 
riKov rrsrroirjKaixsv ap-govra, avayKalws ^etv /cat rovs orpariwras 
avrco rovs /car' avrrjv ihpvjxivovs VTTOKstadai, dSetav l^ovrt, Kadd 
roXs arparicoriKols dp^ovaiv ifatrai, Kal rrpos ovofxa /caAetv avrovs 
Kal i7ri&)T€w Kal Trpovoelv rwv airr)uewv avrwv Kal iir^iivai rois 
Kar avrovs, eiir^p ahiKoZzv, /cat \ir\ n avy^cop^lv rois arparicvrais 
dSiK€iv rovs vtttjkoovs, et Se ri 7rpdfatev u(f>o8por€pov, Kal iyKXj]" 
fjiariKcov aKpoaudai Sikcqv, kov €t arpariwrai KadsorrjKOHzV, /cat 



X 



GBEEK AISTO LATIN LEGAL DOCUMENTS 



37* 



cwravra irpdrr^iv owoua rois urpanwriKots SeSdi/cap.€V ap^ovcjiv. 
Kal wsirep rw T€ *Iaavpia$ Kop/qn rw rijs IJaKariavrjs &pvylas Kal 
npQS ye rots TTpalrcopm AvKaovlas t<= /cat IJiaiSias /cat ®p&K7)$ /cat to 
arparicoriKov w€/cAtVa/x€>% ovrcu /cat awa> ju^ }xovt)v €ti>at t^f Tan> 
TroAtTt/caV wpayfjidrcov rd^iv, dAAd /cat t^v tcov oTpaTtam/cab' i£ovuiav 
re /cat apxTjv, Kal etvat crejujw avrov arpana>rais re /cat tStdmus* 
KeAevo^ra /cat TravTa rrpdrrovra > cos* facts 8^ t^s Q>PXV$ Ka$€arwa7]s* 
Kal jiiav ridsoBai rrpovoiav rov ft^Sev ly/cA^jita /card rrjv hrap-giav 
ajxaprdveadai, dAAd /cat aax^povt ergots* wo/?dAAea#at Tots* Trpos^/covo't, 
ravryjs 8k 8t] rrjs ££ovaias ovk a<j>aipovp,€6a rravr^Aws avrov Itt* 

OvStzvl TTpOSCOTTCp TCQV /Card T7}V irrap^iaV QVTQIV, €tT£ tStO>Tt/C& €tT€ 

urpariwriKw etTe ra/«t eta/car /ztav yap /cat cwe^Tj T7p zlprjvqv eV 
dVacrt rois vtttjkqois rots rj^ripois ^>vXdrreudai /JovAdjuefla, ov ttJ 
8ia<f>opa rcov 7rposd)7rwv rrfV /card ra>v vop.a>v slsayovrGS Kara<f>p6v7)aiv* 
<i* E7riAoyo$7> * Td rotvw Trapaardvra rffxiv rj ut] vTrzpox?} Kara 
rrjv rcov TGaadpwv *Ap}jL€yicbv 8iarv7ra)mv > Kal ^tdAtcrra /card t^v 
rfjs TpiTTjs, fj$ Kara 7Tpo<f>aoiv rov irapovra detov irrovqua^v vop,ov, 
vvv re /cat els rov i£J7)$ cmavra -gpovov <f>v\drrsaBai cnrsvadrco, irdvraiv 
TTparrop,£v(x>v Kal eyypa^oixivwv rats fxepiKats Siarvirwueai rwv 
6p,o8povwv rcov uwv y oVdcra St8ocr#at /ca#' e/cacrrov iros rr poser d^ajiev* 
Dat. xv, k. April. CP, post consul, Belisarii v,c, [a. 536] ( 8 ) 

H. Ebictum III 9 
De Armeniorum successions 

r 

nEPI TH2 TQN APMENIQN AIAA0XH2. 
< c avros fiaaiXevs .♦,>. 

<.npooip,tov> , ifat *ApiA€vtovs f$ovX6p,eBa r^ rrporipas a7raAAd- 
^avres dSt/ctas €77t Tots' T)}j*€r4pQV$ 8td rrdvrayv dyayziv vop,ovs /cat 
Sowat avTots* lororTjra ryv Trpiirovuav* 



CAPUT I, 

Kal eVeiS^ ju-e/xaS^/cajU-ev Ivay^os fiapfiapiKov riva /cat Bpauvv 
etvat Trap 5 a^Tots* vofjLOV oi *Pa>fiaiois ov8k ryj 8bKaiocrvv7) rrjs 7)iizr£pa$ 
rrpsTTQvra TroAtTetas*, oVws* av dppeves ft<=v /cA^povojuotev rcuv yoviwv> 

8 (7/, Chapter II, n. 2, for Adontz's objection to this version of the text which is, 
however, adopted by Honigmann, Osigrenze, pp, 7-9. 

9 CJC, 6th ed., HI, pp. 760-761. 



38* 



APPENDIX I 



BrjX^iai Se /x^/cert, Std tovto dearri^ofiev tw irapovn Beiip ^pcu/^evoi 
vofiw Trpos rrjv ur)v p,€yaAoTrp47T€iav, 6p,oia$ ztvai ras StaSo^ds 1 
/cat Sua rots *Pa)jj,aLCQv StaTera/crat vo^ois iirt re dVSptDy iiri re 
yvvaiK&v, ravra /cat eV 'A previa, Kparziv. Std tovto yap S^ /cat 

TOWS' TjpLeTSpQVS €K€l(7€ K<XTS7T€p.lfjafJ,€V VOfJLOVS, Iva 6tS* OLVTOVS dtf>0~ 
1 pWVT€$ OVTW 7ToAlT€VOlVTO , ' EtTZiSt} Se T(X I7S77 7TpO€lArj<j>OTa 

anavra avatciveZv rwv aroTTCjordrcvv iarl, Std tovto deairi^op^ev toVSc 
tov vofxov Kpar^lv diro tov /caipov ttjs evutfiovs tjjjLwv /JacrtAetas", 
cSare rds rcov i£ e/cetVov TeAevT^cdVTtov ^XP 1 v ^ v Sia-So^ds 1 /card 
tovtof TroAtTevea^at tov rpoiTov, 7tXt]v el jjiTj €tv)(qv StaAwa/xevot 
t) dAAcus* Trpds* dAA^Aovs aTraAAayevTes'. el yap Tt toiovtov yeyove, 
tovto em ttjs ot/cetas fxeveiv 2 luyvos /cat fJLTjbap.cos dva/ct}>etcr#at 
deaiTiC > o\x,ev , Merexeiv Se avVds /cat rtuv KaXovfjuevcov yeveap^t/cdV 
Xcopiwv diro rod elpT]fj,evov %povov j8ouAd/xe#a, et /xeVrot uvfxpalrj 
rivas evpedijvai, olirep /catVot /x^ KaXovj^evas rds dvyaripas els rrjv 
ef aSta#erov StaSo^v eypaifjav oficos kXtjpovojxovs , /xeTetVat /cat Tots* 
e£ avTcuv yevojxevois ttjs tu>v yeveapxiKcbv Trpayp,dVa>F StaSo^S' 

<;' -Ett tAoy os 1 >. Td Tolvvv Trapaardvra rjfMV /cat Std TovSe tov 
Setov S^AoVjitem vop,ov tj ar) virepoxr) /cat 7rapa<f>vXd£cu /cat irepaTi 
mtpaSovvat a7revcraTa>, tusre rovs TjjxeTepovs vopuovs Std TravTtuv 
/cparetv /cat etvat Kvplovs dpxop,evov p,ev rov irapovros vojxov, Kaddirep 
elwovres k'fydrjiASV, e/c ra)v Trpooipulwv ttjs TjjxeTepas jSaatAetas, to> 
77avTt Se av/jLTrapaTadrjGOfjLevov XP°* vt P Kai e ^ T ° Aowrdv a?rac7t ttoAi- 
Tevaojxivov rpoirois /cat irapd ttuvtcov ^vAa^^cro/xevov, 
Dat, X, kal, Aug, Belisario v. c. cons. [a, 535], 



II. GKEEK AND LATIN ADMINISTKATIVE DOCUMENTS 

A, NOTITIA DIGKITATUM 1 

i. Noiitia dignitatum omnium tarn civilium quam militarium, in jpartibits 

Orientis 



2. 


Praefectus praetorio Orientis ... 


5, 


Magistri equitum et peditum in praesenti duo, 


6, 


Equitum ac peditum per Orientem ,,, 


28. 


Comes Orientis .,, 


30. 


Uicarii quatuor : ... 


32. 


Ponticae ... 


38. 


Duces tredecim : ... 


42. 


Per Orientem sex : .„ 


44. 


Eufratensis et Syriae ... 


46, 


Osrhoenae. 


47. 


Mesopotamiae ... 


49, 


Per Ponticam unus : 


50. 


Armeniae ... 


79. 


Praesides XL : ,,, 


86. 


Per Orientem VIII : ... 


90. 


Eufratensis ... 


92, 


Osrhoenae. 


93. 


Mesopotamiae ... 


103. 


Per Ponticam VIII : 


104. 


Honoriados, 


105. 


Cappadociae primae. 


106. 


Cappadociae secundae. 


107. 


Helenoponti. 


108. 


Ponti Polemoniaci. 


109, 


Armeniae primae. 


110. 


Armeniae secundae. 


111. 


Galatiae salutaris ... 


1 Not. dig., pp. 1 sqq. 



40* 



APPENDIX II 



ii. [Praefectus jpraeiorio per Orieniem] 



1. 


Sub dispositione uin illustris praefecti praetorio 
sunt dioceses infrascriptae : 


per Orientem 


2. 


Oriens ... 




5, 


Pontica ... 




7, 


Prouinciae : 




8, 


Orientis qnindecim : 




9. 


Palaestina. 




10. 


Foenice. 




11. 


Syria, 




12. 


Cilicia, 




13. 


Cyprus. 




14. 


Arabia [et dux et comes rei militarisj 




15. 


Isauria, 




16. 


Palaestina salutaris. 




17. 


Palaestina seeunda, 




18. 


Poenice Libani, 




19. 


Eufratensis. 




20. 


Syria salutaris. 




21. 


Osrhoena. 




22, 


Mesopotamia, 




23, 


Cilicia secxinda ,,, 




41. 


Ponticae decern ; 




42, 


Galatia. 




43, 


Bithynia, 




44. 


Honoxias. 




45. 


Cappadocia prima. 




46, 


Cappadocia seeunda. 




47, 


Pontus Polemoniacns, 




48. 


Helenopontus. 




49. 


Armenia prima. 




50. 


Armenia secxinda. 




51. 


67alatia salxitaris ... 





vi* Magisier militum praesentalis II 
26, Sub dispositione uiri illustris magistri militiim praesentalis ; 



GKEEK AND LATIN ADMINISTKATIVE DOCUMENTS 

27. TJexillationes palatinae sex ; ... 

31. Comites sagittarii Armeni .,, 



41* 



vii. Magister militum per Orientem 

23, Sub dispositione uiri illustris magistri nulitum per Orientem 

48, Item [Legiones] pseudocomitatenses XI : 

49, Prima Armeniaca, 

50, Secunda Armeniaca ,,, 

58, Transtigritani „, 

xxii. Comes Orientis 



17, 


Sub 


dispositione "uiri spectabilis comitis Orientis pionineiae 




infrascriptae : 


18. 




Palaestina, 


19. 




Foeniee. 


20. 




Syria, 


21, 




Cyprus. 


22, 




Cilieia. 


23. 




Palaestina secunda. 


24, 




Palaestina salutaris. 


25. 




Foeniee Libani. 


26. 




Eufratensis, 


27. 




Syria salutaris. 


28, 




Osrhoena. 


29. 




Mesopotamia, 


30. 




Cilieia secunda. 


31. 




Isauria, 


32, 




Arabia ... 



xxv, Uicarius dioeeseos Ponticae, 



14, Sub dispositione uiri spectabilis 
prouinciae infrascriptae : 

15, Bithynia, 

16, Galatia, 

17, Paflagonia, 

18, Honorias, 



uicarii dioeeseos Ponticae 



2* 


APPENDIX II 


19. 


Galatia salutaris. 


20. 


Cappadocia prima. 


21. 


Cappadocia secunda. 


22. 


Helenopontus, 


23. 


Pontus Polemoniacus, 


24, 


Armenia prima. 


25, 


Armenia secxmda ... 



xxviii. Gomes limitis Aegypti 

13. Sub dispositione uiri spectabilis comitis rei militaris per Aegyp- 

tum : ... 
22, Ala secunda Armeniorum, Oasi minore. 

xxxviii. Dux Armeniae 



FL 






Auaxa 


INTALL, 








COMORD 
PR. 






Castellum 


Sabbu 


Domana 


Siluanis 


Castellum 


Castellum 


Castellum 


Apollinaris 


Melitena 


Trapezunta 


Castellum 


Castellum 


Castellum 

^^ Colore 

^*^ caeruleo mare 

^s^ indicatur 



10. Sub dispositione uiri spectabilis dueis Armeniae ; 

11. Equites sagittarii, Sabbu. 

12. Equites sagittarii, Domana. 



mm^Mmmmmmlimmr^ 



GKEEK AND LATIN ADMINISTRATIVE DOCUMENTS 



43* 



13. 
14. 
15. 
16. 

17. 

18. 
19. 
20. 
21. 
22. 
23, 
24. 
25. 
26. 
27. 
28. 
29. 
30. 

31. 
32. 
33. 
34. 
35. 
36. 
37. 
38. 
39. 
40. 
41. 
42. 
43, 
44. 
45. 
46. 



Praefectus legionis quintadecimae Apollinaris, Satala. 
Praefectus legionis duodecimae fulminaZae, Melitena, 
In Ponto : 

Praefectus legionis primae Pontieae, Trapezunta, 



Ala Kizena, Aladaleariza. 
Ala Theodosiana, apud Auaxam. 
Ala felix Theodosiana, Siluanis. 
Et quae de ininore latereulo ernituntur : 
Ala prima Augusta Colonorum, Chiaca. 
Ala Auriana, Dascusa. 
Ala prima Ulpia Dacoram, Suissa, 
Ala secunda Gallorum, Aeliana, 
Ala eastello Tablariensi constituta. 
Ala prima praetoria nuper constituta. 

Cohors tertia Ulpia miliaria Petraeorum, Metita, 

Conors quarta Raetorum, Analiba. 

Cohors miliaria Bosporiana, Arauraca. 

Cohors miliaria Germanorum, Sisila. 



Ala prima Ionia felix, Chaszanenica. 
Ala prima felix Theodosiana, Pithiae. 

Conors prima Theodosiana, Ualentia. 

Cohors Apuleia eiuium Komanorum, Ysiporto. 

Cohors prima Lepidiana, Caene-Parembole. 

Cohors prima Claudia equitata, Sebastopolis, 

Cohors secunda Ualentiniana, Ziganne 

Cohors, Mochora, 
Offieium an tern habet ita : 

Principem de seola agentnm in rebus. 

Numerarios et adiutores eorum. 

Commentariensem, 

Adiutorem. 

A hbellis sine subscribendarium. 

Exceptores et eeteros officiales. 
Dux Armeniae VII 2 , 



2 Cf. Mommsen, Verzeichniss, Bury, " The Notitia dignitatmn ", JES, X (1922), 
and Jones, LBE, II, pp. 1417 sqq. 



APPENDIX II 



B. Lateectjltjs tjerootnsis 3 
Nomina 'prouinciarum omnium 



2. 


Dioeensis Orientis habet prouincias numero XVIII 


3. 


Libia superior. : 


4. 


Libia inferior. 


5. 


Thebais. 


6. 


Aegyptus Iouia. 


7. 


Aegyptus Herculea, 


8. 


Arabia. 


9. 


item Arabia Augusta Libanensis. 


10. 


Palestina. 


11. 


Fenioe. 


12. 


Syria Ooele. 


13. 


Augusta Eupfacatensis. 


14. 


Cilicia. 


15. 


Isauria. 


16. 


Cyprus. 


17. 


Mesopotamia. 


18. 


Osroena. 



IL Diocensis Pontica habet prouincias numero VII : 

% Bitinia, 

3, Cappadoeia, 

4, Galatia, 

5, PapMagonia, nunc in duas diuisa, 

6, Diospontus, 

7, Pontus Polemoniacus. 

8, Armenia minor, nunc et maiox addita 

XIII, Gentes barbarae, quae pullulauerunt sub imperatoribus : ,.• 
38, Armeni ,., 4 , 

3 Not, dig., pp, 249 sqq, 

4 Of, Mommsen, Yerzeichmss, and above Chapter IV, n. 31, Bury, Verona List, Jones, 
VeroTia List* 



GBEEK AND LATEST ADMESHGSTBATIVE DOCUMENTS 



45* 



C, Laterctjltjs POLEMII SlLVII 5 
Nomina Prouinciarum 

,., VIIL InOrienteX: 

2, Prima ; Srria Goele, in qua est Antiochia, 

3, Secunda : (Siria) Palestina. 

4, Tertia : Sixia Phoeniee, 

5, Quarta : Isanria, 

6, Quinta : Cilicia, inxta montem Taiirum, 

7, Sexta : Cyprus, 

8, Septima : Mesopotamia, inter Tigrem et Enpliratem, 

9, Decima : Enfratesia, 

10. Octaua : Hosdroene. 

11, Nona : Sophanene. 

IX InPontoVIII: 

2, Prima : Pontus Polemoniaens, 

3, Secunda : Pontus Amasia, 

4, Tertia : Honoriada, 

5, Quarta : Bithinia, 

6, Quinta : Paflagonia, 

7, Sepiima : Armenia minor, 

8, Sexta : Armenia maior, 

9, Octaua : Oappadocia „, 6 , 



D. HlEROKLES' SYMIKDEMOS 7 



IEPQKAEOYZ 2YNEKAHMOS 



631 3 KcovaravTwovTroXis , Elulv ai Traom iirdp^iai kcu iroAzis 
at wo rov /JactAea rwv t Pa)}jt,aia>v SioiKOvfLGvai rov iv 
KwvaravnvovTroX^i, 
iirapxlcu £S, 7roAe^ Ae, w$ viroriraKrai. 

5 j\^, e%, pp. 258-259. 

6 Of, Mommgen, LaUrwhis* 
1 JSieroMes, pp. 12, 33 sqq. 



46* 



APPENDIX II 



690 


3a 


IIONTIKH [Aia rrjs IIovTiKrjs SioiKijaecos] ... 


698 


5 


KAIJIIAAOKIA A As. 

KovuovXdpiov y TToXeis S. 


'jE^rap^ia KaTTTrahoKias a>, vtto 




6 


Kaiadpeia 


Kaiadpsia 


699 


1 


Nvuua 


Nvuaa 




2 


@€pp,d 


Ta &4pp,a 




3 


peyecov IloSavSos 


* PeyGiroSavSos 


700 


1 


KAIHIAAOKIA B A£, 

Tjy€jj,6va, ttoXgIs rj. 


^Errap-^ia KaTnrahoKias j8, vtto 




2 


Tvava 


Tvava 




3 


0avarivo7ToXi£ 


lJ>aVOTlV0VTT0klS 




4 


KvfSiurpa 


KvfSlorpa 




5 


Na^iav^os 


Nav^iav^os 




6 


Udfjifjia 


2daip,a 




7 


Ilapvauuos 


Tlapvauos 




8 


peyecbv Aoapa 


f PeyeSoapa 


701 


1 


pzyscbv Movkujuqs 


*PzysKovKOVo6s 




2 


EAENOnONTOZ _A^, 

KOVUOvXdpiOV } TToXziS £. 


y Eirapj^ia ' EXsvottovtov, vtto 




3 


'Afxdusia 


^Ajxaula 




4 


"Ifiwpa 


Ifiopa 




5 


ZrjXa 


ZrjXa 




6 


UdXrov ZaXiyiov 


SdXrov ZaXl)(OP 




7 


n Avhpaira 


"AvSpaira 


702 


1 


*Ap,io6s 


*Afjuao$ 




2 


SlVWTTT) 


SlVOJTTT) 




3 


nONTOS nOAEMQNIAKOS X9, y ETra Px la IJ6vtov 
IIoAtiAoviaKQV, vtto ^ye/xova, TroAets 1 e. 




4 


NeoKaicrdpzia 


NeoKaicrdpeia 




5 


K6p,ava 


Kop,dva 




6 


IIoXzfJLCQVlOV 


TO AsfJLOVlOV 




7 


Kspaaovs 


Kspaaovs 




8 


Tpair^ovs 


Tpair^ovs 


702 


9 


APMENIA A £t. *E7rap)(ta * Apjisvlas a, vtto T^ye/xora, 






ttoXgis e. 




703 


1 


27ej8acrreta 


2*f$dcma 




2 


NikottoXis 


NlKOTToXlS 




3 


KoXcoveia 


KoXovia 




4 


SdraXa 


UardXa 




5 


2Jt=j3acrTov7ToAis 


HtfiaGTQVTToXlS 



GREEK AND LATIN ADMINISTRATIVE DOCUMENTS 



47* 



6 


APMENIA B pa. 


^Eirap^la ' Apjj,€Via$ /3, vtto rjyeixova, 




ttoXgis s. 






7 


M^XirrjVTj 




MeAiTivrj 


8 


"ApKa 




"ApKa 


9 


* Apafiioaos 




'Apafiioos 


10 


Kovkovoos 




KOKQVGQS 


11 


Kopiava 




Kop,dva 


12 


* ApiapdBsia 




3 Apapadia 


704 la 


ANATOAIKH [Aid 


rfjs 


'AvaroAiKTJs SioiKrjazws'] ... 


712 10 


EY0PATH2IA_ 
7]yzix6va> ttoXsis t/3. 


M- 


'Eirapxta Ev<f>paTTjolas > vtto 


11 


*l€p&7T0XlS 




' IepatToXis 


713 1 


Kvppos 




Kvpos 


2 


2ap,6aara 




Sajioaara 


3 


AoXl)(r) 




A oXtj-^tj 


4 


Zevypia 




Zevyjxa 


5 


rzpp,aviK€ia 




r^pjxaviKia 


6 


uipprj 




IJ4ppr} 


7 


NlKOTToXlS 




NlKOTToXlS 


8 


EKr^vap^la 




27/cevap^ata 


9 


UdXrov * Epayittfvov 


UaXyevoparilsvov 


10 


Ovpijxa 




2vpi[j,a 


11 


EvpVOTTOS 




EvpOJTTOS 


12 


OZPOHNH JIS. 


*E7Tapj(ta ^Poapoovvrjs, vtto r)y€jj,6va, 




TToXziS 9, 






714 1 


"ESzaaa 




"Eosuua 


2 


KtovvravTLva 




Kwvoravriva 


3 


&€o8omov7ro\is 




&€QOO(JlOV7ToXlS 


4 


Kdppai 




Kdppai 


5 


Bdrvai 




Bdrvai 


6 


N4a OvaXzvrta 




Nea OvaXzvrids 


715 1 


AzovToiToXis tj /cat 


AzOVTOTToXlS Tj KOI 




KaXXtviKos 




KaXXlVlKT) 


2 


Blpda 




Blpda 


3 


MESOnOTAMIA 


V, 


^Errap^ia MeaoTTOTa/xeias', vtto 




rjyeiAova, ttoXis 


a. 




4 


"AfitSa 




"Apaoa ...* 



8 On the date of the Synekdemos and its relation to other sources, see Hierolcles, 
pp. 1 sqq,, and above Chapter IV, n. 42b, also Jones, OREP, p; 503, 



APPENDIX II 



E* Basilii notitia 9 
TA3IS nPOKA&EAPIAS TQN OZIQTATQN IIATPIAPXQN 

6 *P(ibjA7)S 

6 KwvaravrivovTToX^cos 

6 > AX^av8p^ias 

6 ^Avrio^las 

6 AlMas ^I^pouoXv^cov 
Td£is TTpoKaO^Sptas ixrqrpoiroXirwv /cat avroK^aXwv /cat £7tmjk6ttq>v 
reXovvrcov wo rov dwoaroXiKov Bpovov t<xvtt)S ttj$ BzQ<f>v\&KTQV /cat 
jSacrtAtSos" ttqXgws* 



a. 


Ewapxia 


\ KaTnraSoKias a- 


d Kaiaapsias* ... 


ta. 


iirapxia 


*App,$vias j8 


d >27ej8a(7T€ia?. 


£. 


iirapxia 


c .EAevo7r(Wov 


d MjLtacretas 1 . 


iy. 


errapxta 


> Apjx€via$ a 


6 MgXitwtjs* 


iS. 


£irap)(la 


ifaTriraSo/ctas' /? 


6 TvaVCOV TjTQl Xpia- 

TQwrdAetos, ... 


is. 


iirapxia 


* Qva>ptaSos\ 


d -KAauStowdAeais. 


^ 


cmxp^ia 


IJovrov iToAe/^ 
a>Fta/cov _ 


d NzoKaiaapGias* ... 


KS. 


iirap^la 


KaTTTraSoKias f$ 


d iWcu/c^cTcrov. 


^. 


i,trapxia 


Aa&KTJS 


d rov ^acrtSos. ... 


M4%pi > 


TOVTWV ot ( 


p/rjTporroXirai /cat Aot7rdv 


ivTGvdzv dp^ovrai oi avro- 


Ke^aAoi 


u ... 








iirapxia 


*JEWo?rdvTov 


d Sv^atTwv, ... 




iirapxia 


*App,zvias j8 


d *-Hpa/cAowdAea)s 
T^rot 0iXaxOo7)s* 




hrapxia, 


MjSaaytas 


6 UzfiaOTQVTToAGWS . 




iirapxla 


IIqvtqv iJoAcjit- 


d Tpa7re£ow'Ta>j>, ... 



covta/cov 
(Td^is /cat Sialpzats rwv ^rjTpOTroXircbv avv rois w aureus errtcr- 

KOTTQIS)* _ 

^4, s ^7ra/)^ta KaTTTraSoKias 

6 Kaujap€ia$ 
a , d t&V SacrtAt/ctDp ®^pjxwv 

9 GWg?. Cty;pr., pp. 1 sqq. 



GREEK AND LATIN ADMINISTRATIVE DOCUMENTS 



49* 



IA, 



IB. 



h 


O NvU7]S 


y* 


6 ®<zo8omov7r6Xsws 'ApfAsvias 


I, 


6 Kap,ovAiavwv 


€, 


6 KlUKlUOV* ,., 




'Ejrapxia 'Apfxsvias B 




6 SsftaaTGias 


a, 


6 U^aaroviToX^ws 


b 


6 NiKoiroAews 


y* 


6 UardXcov 


s. 


6 KoAcoveias 


c. 


6 BrjpMXjfjS' 




9 Eirap^ia * EAgvottovto v 




6 'AfJLavGias 


a. 


6 'Afjuaaov 


£ 


6 2JlVW7T7]S 


y. 


6 s 7j8dpa>v 


5, 


* AvhpdlTWV 


€. 


6 Za\i)(OV TjTOl AsOVTOVTTokzWS 


s. 


6 ZrjXcov. 




'Errapxla 'ApjjLGVias 




6 MeXinvTJs 


a. 


6 "ApKj}$ 


% 


6 KoVKOVUOV 


y. 


6 *Apa/3i(jaov 


8, 


6 *Apiapadr}$ 


e. 


6 KzOjAOrVWV 



IT. 'Errapxla KaTrrraSoKias 

6 Tvdvwv 7]toi XpiarovTroAscos 
a. d Kvfti&Tpcov 

/}. O 0aV(7TlVOV7ToX€U)S 

y. d Uamptav. ,,♦ 

IE* ^Errapxia * OvcopidSos 

6 KAavBiovTroXzcos 
a, d t HpaKAzlas IJovrov 
j8. d npovmdbos 



50* 



APPENDIX II 

O ltov 

6 Kparsias 

6 ' ABpiavovTToAecos 

k, ^Eirap^la IJoA^pLCOviavrj 

6 NzoKaioapelas 
6 Tpair^ovvrcov 
6 Kepaaovvrcov 
6 rod IloX€fj,a>viov 
6 Koixdvwv. ... 



e. 



a, 
t 

5. 



KE, ^Eirap-^la KaTTiraSoKias 

6 MwK7)a<yov 
a, 6 Na^iav^ov 
j8» 6 KoXcovelas 
y, 6 Ilapvauaov 
8. d iloapeov. 

J&. y Errap-^ia Aa^iKrjs 

6 0dai8os 
a. d ^PqBottoAzws 

/?. d r^s ' Af3ia<J7)VtDV 

y. d IJerptov 

§. d Ziyaviwv. ,,. J(? 



F. Georgii Cyprii descriptio orbis romani n 
... JL4 THE ANATOAIKHS AIOIKHSEQS ... 

*E7rap-)(ia ' Oaporjvrjs 

"ESeooa p,7)T poiroAis 

Kdpai 

KcovardvTGia 

®€o8ocriov7ToAi$ 

KoXAlVlKOS T)TOl AzQVTOTToXlS 

Nia OvaAzvrta 

10 On Basil see Honigmann, Basileios, Laurent, Basile, and Hierokles, pp. 49 sqq, 

11 Creorgr, Cypr., pp. 41, 45 sqq, 



"^ 



GREEK AND LATIN ADMINISTRATIVE DOCUMENTS 51* 

BlpBcOV 

MoviBiXXa 

®Tjpilia")(Q)V 

Moviavya 
MaKapra 
MapKovnoXis 
* Avaoraala 
*Hp,£pio$ 

KipK7]Uia 

Me^pi rwv 3>oi iari to TrArjpwpLa MzaoTrorapilas Kal dpx 7 } rfs yfjs 
lit pa loos , 

'E7Tap-)(la M^aoirorap^las avco jjroi A 'AppL€vta$ 

"ApilSa p,7)TpQ7ToAlS 

MaprvpoiroXis 
cl7to te pLiMcov rrjs avrrjs ttoAzws yevvarai 6 Tlypis TTorapios. 

Aapas 
&7TO $ jxihitov rrjs avrrjs rroXews elai ra /xeflopia Kal ol opoi IJepaloos 
Kal Evplas* 

Kaarpov *PiaKrj<f>as 

Kaarpov TovpdvSios 

Kaarpov MdpSrjs 

K&UTpOV AopVTJS 

Kaarpov *Pi<f>dov 
Kaarpov "Ia^pios 
Kaarpov T^avpas 
Kaarpov Avodaaos 
Kaarpov* Af$dpp,rjs 
Kaarpov Tfyvofilas 
Kaarpov \Zv£iertuv 
Kaarpov BavafirjXcov 
Kaarpov XovoScov 
Kaarpov *A'Caov8ovos 
Kaarpov Maa<f>povas 
Kaarpov BaaiXiKov 
Kaarpov UktjXov koi ^OStjAoqv 
Kaarpov Brj'Covfiaidas 
Kaarpov Mavaoodpwv 



52* APPENDIX n 

Kaorpov &ip9axaj3paT)s 
Kaorpov Utrdwv Xi(j>as 
Kaarpov Kakwvos 
Kaorpov Btfiaodpwv 
Kaorpov Tt,avpas 
Kaorpov BipSas 
Kaorpov ' Arra-gas 

KXlfxaros ' Aplavrjvrjs 

Kaorpov *A<f>oviAcov 

Kaorpov ^ApifSdxwv 

Kaorpov 0Xwpiavwv 

Kaorpov Aa<j>vov8w 

Kaorpov BaXovos 

Kaorpov Sajxo-^dprayv 
^jQSe TrArjpovrai, 77 Msoowor apia, /cat zoriv 6 Tavpos Kal rj KXsioovpa 
BaXaXziowv, Kai ap^ercu Kara ro dpKr wov p,4po$ r) M^yaXi) *App,€vla. 
zlol 8k Kai oi oiKovvrss sis to opos rov Tavpov rrhf]oiov rov avrov 
KXifxaros Xaoi /ToVo/xa£dju,evot 6 jxkv sis XoBaprai > 6 8k srspos J£ava~ 
oovvirai, Kal zorw opos viprjXov, iirovofjia^ofji^vov MapaoKev* iv & 
Kal rj Kifiwros rov JVcoe im$€pop,4v7) rois voaoi irpooeKpovoev sis 
rrjv Kopv<f>rjv rov opovs koI eonv rovro yvcoorov irdoi rois rcov eKeiae 
jAzp&v p>€)(pi rrjs otffjiepov* 



y Eirap^ia A > App l evias aXXrqs 

AaSiftwv vvv fXTjrpoTroXis 

> ApoaiMOvodrcov 

ttoXi^vt) Xoldvwv 

Xoooyid^wv 

Kidapi^wv 

Kaorpov MspriKiprov 

Kaorpov BaXovKovos 

<;* Kaorpov iJaAtd?I> 

Kaorpov x Ap8wv 

/cA//*a Uo^>i]V7)s 
%ojpiov vtto ro avro /cAt/xa, Aeyo/xevov ^IaXipifSdvwv, odsv opparai 
6 rrjv rrapovoav tftiXoTTQvrjoas fiifiXov BaoiXeios* 



GKEEK AND LATIN ADMINISTRATIVE DOCUMENTS 

/cAi/m *Avlrjrivfjs 
/cAt/xa AiyqawT\s 
/cAt/xa rapwY)s 

/cAijua IlaMvfjs 
icAtjita 'Qp&avwrjs 
/cAtjua Martaw/ajs' 
/cAt^ua Mov^ovptov ... 



53* 



^Eirap^la 'Appievias MeydAyjs 

^let eiSevcu, on avV^ avTOKi<j>aX6s icrn pur] reAoucxa wo rov dvaro- 
At/cdv dpovov. aAAa Tip,r)dzXua Sia rov ayiov Tprr\y6piov 7 App f €,vias > 
e^oucra TroAets 1 /ecu Kaurpa /ecu /cAt/xara a. ,,. J5 . 



G. Nova tactica 13 

TMTT^ M£iV TA I1AAAIA TAKTIKA 2KOIIEI AE KAI 

TA NEA 



< Ti^lS TCQV pL7)Tp07T0A€C0V TCOV V7TOK€lfJ,€VU)V TO) TTJS BaUlMSoS 8pOVW,> 



a, 


*H Kai crape ta ,,, 


mx. 


77 JSW/Jdoreta 


ijB. 


7) 9 Api>da€ia 


iy. 


t? MeAmv7? ,., 


17}. 


07 iVeo/caicrdpeta . 


/c<9. 


17 Mco/CT^aos ... 


jas. 


77 Kap,a-)(os ... 



i?itn Se /cat dcroi e/cdoTT? p,7)T poiroAei wd/ceivrat Opovoi, 

<A> Tfj KaiuapGia KaTTTraBoKias* 

a. d Nvu7]$ 

j8. d rcuv SaatAt/ctDv ©epjittuv 

y. d ifa/xoi;Atavcuv 

S. d KioKiaov 

12 On the date of George of Cyprus and his relation to other sources, see Hierokles, 
pp. 1 sqq., and 49 sqq. 

13 Georg, Cypr., pp. 57 sqq. 



54* 





APPENDIX II 


i 


o J&vaiawv 

6 IJzvrjpi&Sos 

6 *Apa6elas 

6 rwv AlttoXicov ... 


LA, 


Tfj JSe/Saoreia rijs 'Apjxevlas 


a. 

I- 
v- 
I. 


6 E$fiaOTQV7r6X€WS 
6 NlKOTToAzWS 

6 UardXwv 

6 BeplOQTjS* 


IB. 


Tjj y Afj,ao€ia * EX^vottovtov 


a. 

I- 
y- 

8. 


6 'AfAlOOV 
6 2lVU)7T7)S 

6 'IfSopcov 

6 'AvBpdirwv 

6 ZaXlov yjroi IIoiatttj'CovttoXzws 


IT. 


Tfj MeAirrjvfj rrjs 'AppLZvlas 


a. 
1' 

y- 

[8.] 


6 "ApK7)s 
6 Kovkovuov 
6 'Apafiiaov 
/ecu AvttCov ... 


IH. 


Tfj <lNeo> Kaiaapeia IIovtov 
i7oAeju.a»vai*/co v 


a. 

i 


6 KspaoovvTcov 
6 rod JJoX^/JLCovtov 
6 Kojxdvrwv 



AT* Tfj Tpa7T€t t OVVTl ttjs A <a£i/c7}s'> . 

<a.> 6 Xepidvwv 

</?.> o XafiaTo^ovp 

<r,> 6XdX 

<8.> o UaiTrep 

<€,> o Kepap^iwv 

<S.> o Azplov 

<£.> o Bi^dvwv ... 



GKEEK AND LATIN ADMINISTRATIVE DOCUMENTS 



55* 



MS*. Tjj JTa/xa^o) *App,svias 



a. 

b 

8. 

6, 

NA. 



6 *ApaapaKtov 
6 Bap^aviuarjs 
6 MgXov 

6 MzhoV €T€pO$ ,.. 

Tols Ev-^airais c EAgvottovtqv 

BpOVOS VTTOKGljJLZVOS OVK 6CTTI 



NB , Tfj * Ap,aurpihi rov Tlovrov 

JW. TtVV *A(JfJ,W(jdT(jQV 

NA. AiX&vai. 



H, Epistulae ad Leonem imperatorem 

xxxvi - Armenia Prima 14 

Piissimo et Christianissimo imperatori uictori semper augusto 
Leoni Iohannis Gregorius Auxentius Eustathius Epiphanius episcopi 
primae Armeniae in domino salutem, Dens nerus dominns noster 
Iesns Christus semper optima natnrae humanae dona concedens 
nullnm tempus sine sua prouidentia dereliquit, qua gratia etiam 
nnnc fidei uerae prospieiens, quae nostrae salutis spes est, in te pio 
et Christiano principe quodam secundo Dauid cornu imperii reclinauit. 
quem sibi nouit religiose a cunabulis seruiturum, tunc sua sententia 
in omni orbe terrarum imperare sanciuit, quatenus ex uestro imperio 
profluerent bona subiectis et ubique pietatis praedicatio praeualeret 
uestra utique mansnetndine nihil aliud praeter fidem sceptra regalia 
iudicante, quarum rernm testis est praesens zelus et studium ilia 
firmandi, unde firmitas uestri accedat imperii, a deo namque unctus 
et regem mox ei qui unxit, ipsa principia commendasti optime satis 
cogitationibus simul et uoeibus ei deseruiens et ut haee bene consis- 
terent, prae ceteris omnibus apud cunctos pro fide orthodoxa decer- 
tatus es, omni scilicet mala secta prorsus expulsa atque sublata, ad 



14 AGO, II, v, pp., 69-71, 



56* 



APPENDIX II 



idem eonuenientibus et ad inuieem coneordantibus qui nuper quod 
fieri non oportuerat, uidebantur esse diuisi. quia prospicit uestra 
pietatis intentio, ad unam reduxistis ecclesiam non solum eos qui 
per dissensionem noua passione languebant, sed etiam eos quorum 
erat a priscis temporibus mens corrupta et a recta et regia uia reee- 
dentes ad loca praua et spinosa blaspbemi erroris abierant, ut secundum 
euangelicam uocem ononis ecclesiae unum ouile consisteret et unus 
pastor dominus Christus esset. sed baec quidem omnia proueniant 
in uestro semper imperio; quia uero et meae simul humilitati piis 
litteris estis iubere dignati ut quid sententiam de bis quae Alexandria 
facta sunt, manifestem, licet [et] exaggeratio rerum ibi gestarum 
neque sententiae tribuit facultatem caligine rerum tristium mentibus 
obumbrante, hoc tamen uobis insinuo quia si uera sunt quae in precibus 
religiosissimorum episcoporum et clericorum Aegyptiacae diocesis 
continentur et auctor Timotheus inuenitur tantorum et talium seele- 
rum, quae propter nimietatem, ut arbitror, non creduntur, alienus 
sacerdotio cum bis qui similia perpetrarunt, apud sanctos canones 
iudicabitur. et haec quidem de bis quae Alexandria gesta sunt, 
cum sancto concilio quod mecum est, deliberans religiositati uestrae 
significaui; fidem uero solam trecentorum XVIII sanctorum patrum 
qui dei gratia conuenerunt in Nicaena urbe sub piae memoriae principe 
Constantino, seruamus, qua ab infantia sumus inbuti et in sacerdotio 
alios inbuere nouimus quamque et post baec CL episcopi congregati 
in ciuitate regia firmauerunt et propriam iudicarunt et ipsa sibi 
cooperatrice utentes et doctrina diuinitus inspirata, sancti scilicet 
spiritus, blasphemias et zizania radicitus amputantes quamque 
nihilo minus et def initio a sancto Calchedonensi consilio explanata 
firmauit, praecipue repugnans uesaniae nefandi Nestorii et sanctam 
synodum quae Ephesi est celebrata, confirmans, cuius praesules 
fuerunt deo amabile et sanctissimae memoriae Romanorum et Alexan- 
drinorum episcopi Caelestinus et Cyrillus, qui maxime aduersus 
sceleratam blaspbemiam Nestorii suis responsionibus doctrinisque 
claruerunt, quorum epustulae aduersus eundem impium Nestorium 
et <ad> Orientales uniuersos datae et ab eodem sanctae memoriae 
Cyrillo contra eundem Nestorium anathemata proposita sunt firmata 
atque roborata. Igitur indicamus prolatam definitionem a sancto 
Caleb edonensi concilio non sieut fidei symbolum, sed sicut defini- 
tionem esse positam ad peremptionem Nestorianae uesaniae et ex- 
clusionem eorum qui salutem incarnationis domini nostri Iesu Christi 



GKEEK AJTD LATEST ADMESTISTEATIVE DOCUMENTS 



57* 



denegaxe noseuntux, ut agnoscant omnes qui ob hoc scandalum pa- 
tiuntur, quia neque nos post oxthodoxum symbolum CCOXVIII 
sanetoxum patxum ant augmentum aut deicrinutionem in Ms quae 
sie perfeete et a sancto spixitu sunt definita, suscipimus <et> fidem 
aliam nescimus, quia neque est nee patimux hoc audixe, licet quidam 
esse dicant, si uexo quibusdam uolunt calumniaxi uexbis, etiam 
hoc uestxae sexenitati indieaxe confidimus quoniam ea quae illis 
uidentux esse dubia, ad inteUegentium sic xespicexe noscantux affectum, 
sunt enim quaedam in definitione quae <si> xecte intelligantux, 
oxthodoxa sunt ; si uexo aliquis ea alitex uelit inspicexe, inueniet hanc 
sensus dubios paxientem, multi siquidem et scxiptuxas diuinas non 
intellegentes sicut scxiptae sunt, pxopxiae blasphemiae dogmata 
genuexunt, quos dominus Chxistus sua dementia et saexa pietatis 
uestxae pxudentia diligentiaque conuextat et xectae pxaedicaxe uexbum 
uexitatis edoceat et boo sapexe quod sancta catholiea dictat eeelesia, 
cuius caput quidem est Chxistus., uos autem xobux ac fundamentum 
imitantes immobilem Cbxisti petxam, supex quam omnium cxeatox 
ecclesiam suam aedificans omnibus Cbxistianis pietatis xequiem 
eondonauit, qui etiam uestxam sexenitatem sempex consexuaxe 
dignetux in talibus uigilantem et insidiantem bestiam Cbxisti ouilibus 
a saeptis fidei oxthodoxae xepellentem, exedimus enim quia postquam 
nostxis littexis baee indieauimus, cuncta uexitatis uia omnis modus 
uestxae panditux pietati quatenus sopiantux scandala et ad fidem 
xectam membia discoxdantia xeducantux domino Cbxisto supex quadam 
inspixatione uestxam txanquilitatem ad bona omnia deducente, 

Iohahhis episcopus misexicoxdia dei Sebastiae metxopolis sanctae 
dei ecclesiae manu mea suscxibens secundum sanctionem uestxae 
pietatis pxaedictam epistula mea sententiam nuntiaui, quam babeo 
pxo fide txecentoxum XVIII sanctoxum patxum in Mcaea conuenien- 
tium, quam et sanctum Calcbedonense concilium confixmauit, simul 
et pxo Timotheo, quern pieces ids aduexsaxium ostendexunt, et conse- 
xuaxi uestxam potestatem tempoxibus longis exopto 

Iobannis episcopus Nicopolitanus similitex 

Gxegoxius episcopus Sebastopolis similitex 

Maxentis episcopus Vaxissae similitex 

Eustathius episcopus Coloniae similitex 

Epiphanius episcopus Satalenus similitex. 



maSm 



58* 



APPENDIX II 



xxxvii - Armenia Secunda 15 

Religiossimo et piissimo et Christianissimo imperatori semper 
augusto Leoni Otrius Acacius Iohannis Adelfius Hormisda Longinus 
secundae Armeniae episcopi in domino salutem. Dens qui glorifi- 
eantes se glorificat, secundum cor suum apicem uestrae tranquilitatis 
inueniens inexpugnabilem palmam et honorem fidei consistentem 
placidus praebuit uobis, Christianissimi principum, super omnes 
homines sine prohibition e ahqua potestatem. Insurgentes enim 
inopinabiliter subdidit uictoriis atque inconparabilibus triumphis 
et uestram pietatem excellentissimis honoribus exornauit, immutilata 
et sine litigio et ab aliis indiuisa praebens sceptra uestri imperii, 
ut firmas atque elaras in uobis diuinas seruantes formas optima 
uoluntate seruitis, bonum enim circa dominum deum favorem 
uobis habentibus, mansuetis nutibus ad legalem et mirabilem uitam 
conuersationemqiie deducitur quidquid sub sceptra uestri imperii 
gubernatur; aequam uero sortem salutis uestrae utilitati reliquiorum 
hominum ponentes inconcussam catholicis totius orbis ecclesiis pacem 
sine tumultuatione confertis et dei clementiam imitantes neque 
paruulos humilesque despieitis, quando per commodum condescen- 
sions summitatis culmen ad nos etiam pro fidei causa deponitis 
sociosque nos, qui pro abiectione nostra nihil in terra sumus, inter 
uestras accipitis curas, non egentes conlationem nostram et in hoc 
utique magnitudinem incomparabilis dei clementiae demonstrantes, 
quapropter quoniam iussi sumus, ultra nos quoque praesumimus et 
quid sentiamus, uestrae pietati suggerimus. nos igitur, uenerabilis 
imperator, in ultimo mundi loco degimus multo spatio a regia ciuitate 
distantes, sed uestrae potentiae in nullo diuisi fauore circa fidem 
equidem rectam sententiam possidemus, ad sermon es uero eonten- 
tionum linguas habemus segnes. cohabitamus enim circa Armenios 
barbaros, fideles quidem, sed recte Romano eloquio non utentes, 
breui quodam ab eis spatio, magis autem intercessione Eufratis 
fluminis separati, et propter frequentem barbarorum permixionem 
longos nequiuimus proferre sermon es, uitamus autem etiam doctrinas 
extraneas proferentes, quia eloquentia quidem sancti spiritus rennu- 
unt et propriam doctrinam in euangelicis eruditionibus adferre noscun- 



15 AGO, II, v, pp. 71-75. 



^^M^rt^rr 



GREEK AND LATIN ADMINISTRATIVE DOCUMENTS 59* 

tux, nos etenim secundam inhabitantes Axmeniam una et immobili 
utimur uoluntate et in una fide consistimus, communitex omnes et 
seoxsum singuli pro uestxa maiestate ad deum facientes orationem 
ab omni haexese et lingua blasphema separati unamque doetrinam 
super omnia elaxam a Sanctis txecentis XVIII patxibus percipientis 
patrum fidem inuiolabilem conseruamus, quaestiones uexo de deo 
tamqxiam intitiles et nostra cogitatione supexioxes effugimus et alind 
quidquam ant <sentientes aut> dicentes ualde declinanras a supexfluis 
quaestionibus abstinentes et leetiones impias refatantes ab eorum 
paxauitate inimica eeelesiae sumus extxanei et txiticum fidei a zizaniis 
separatum nobis fidelibns imperantibus consexuamus, in txaditione 
patxum doctrinae sufficientiam possidentes et nihil supexuacaneum 
quemquam loqui sinentes, „. eum itaque adoxemus Chxistum, qui 
et dispensatiue pro nostra salute suscepit carnem passionis et diuinae 
nobis inpassibilitatis iuxa donauit. sic enim concilium sanctorum 
patrum Calcbedone celebratum sapiens et trecentorum XVIII patrum 
inuiolabilem et intemptabilem custodiuit fidem et fortiter ineruditi 
uiri fatuitatibus resist ens eatholicis ecclesiis in toto orbe fundatis 
contulit bona pacis, cum quibus nos quoque in uno coxpoxe congregati 
per fidem uestrum imperium inmutilatum et ad filiorum filios peruenire 
ab omnium domino depxeeamux, si uero quidam decerpentes con- 
gruentias syllabarum conpositionesque uexborum bella et lites mouere 
temptans aduersus eeclesias, deus restitit eis ; nos autem intentionem 
exponentium fidem et mentem probantes nequaquam a uerboxum 
conpositione xecedimus, sed et pxopugnatores dogmatum et pexfectos 
custodes fidei txecentorum XVIII patrum babemus sanctos patres 
Calcbedone collectos et sicut et ipsos txecentos XVIII patxes honoxa- 
mus, nihil enim adicientes illorum symbolo os obnoxium multis 
suppliciis damnauexunt, igitux sanctionem pietatis uestxae suscipiens 
una cum sanctissimis episcopis nostxae pxouinciae xelegensque pxeces 
ab Alexandxinis clexicis uestxae maiestati poxxectas pxioxes atque 
postexiores in pxioxibus quidem inuasionem ouium factam contxa 
pastoxem fleui et contra ipsum sacexdotium seditionem noui Dathan 
et Abixon nimis ingemui, pudor enim cuctus pex axxogantiam est 
expulsus, omnis lex et timox impexialis et iudicium est contemptus 
et sacexdotales sanctiones uexauexunt nefandissimae uoluntates manus 
contra sacerdotes armantes, quas debuerunt optime temperaxe et 
donum pexcepexunt dignum suae salutis pignus. habens etenim 
mentem xatione priuatam, sicut pxecibus sumus edocti, si tamen 



APPENDIX II 



uexae sunt, scelexatus Timotheus pxincipatus amoxem xLtUitati praepo 
nens ad xes nefandas accessit sedibus non sibi eonpetentibus inxuens 
adhuc uiuo ecclesiae saeexdote dispensationemque ecclesiae petulanti 
uoluntate dixipiens et pxincipium sacexdotii faciens sanguinis effusio- 
nem, sed etiam sanetaxum gxegium caedis faetus occasio inpudentex 
custodem constituit semet ipsum, qui neque uoeari dignus est Chris- 
tianus, quando cxuentis manibus uenexabilia mystexia non dubitat 
impia pxaesumptione contingexe et post damnationem illam opexaxi 
quae <neque> eum neque alios agentes sanetoxum patxum xegulae 
uidexe pexrrrittunt, qui [neqxie] non sustinens ut secundum xegulas 
oxdinaxetux eeclesiastieas et ab his qui sinrili eastigationi uidebantux 
esse subiecti, faetus episcopus ab omnibus eeclesiis semet ipsum 
exeomnrunieasse dinoseitnx quasi faciens diuinae gxatiae donis iniu- 
xiam, deinde cum non naleat cuxaxe quae ab eo male pxaesumpta 
sunt, si tamen quae de ipso dicta sunt, eum uexitate coneoxdant, 
patxum eoneiliis obloquitux et eum ei mala patriae non suffieiant, 
omnes eonttixbaxe temptat eeclesias, quasi potestatem habens gexendi 
quaecumque uoluexit, et neque uiuis neque moxtuis pxaesulibus paxeit 
ecclesiae, sed quasi contxa omnes potestatem impietatis adeptus 
pximum CL patxum synodo dexogat, quam spixitu diuino statuit sedes 
Alexandxina, quod mihi fecisse uidetux, ut effugiat liomicidii adul- 
texiique supplicia, illic namque in ipsis pxincipiis contxa homicidas 
excommunicationis decxeuexunt poenam, non suscipit autem sanctam 
et uniuexsale Oalehedonense concilium nesciens quia etiam ante hoc 
a txecentoxum XVIII sanetoxum patxum fide semet ipsum fecit 
extxaneum, quam sanctum utique Calchedonense concilium confix- 
mauit ae xoboxauit, opoxtebat enim eum ascendentem tyxannice 
ad thxonum beatae memoxiae Cyxilli, illius libxis incumbexe et doe- 
txinam illius possidexe, sed uos, pii, tamquam uniuexsos pxincipes 
optima uixtute supexantes fidem defendite tyxannidem sustinentem, 
patxum sanetionibus confexte uixtutem, sacexdotibus donate a pexiculis 
libextatem, pxohibete eos qui inxegulaxitex contxa ecclesias nituntux 
insuxgexe, quatenus sine seditione et lite atque bello uestxis tempoxibus 
sanctae dei ecclesiae constitutae incessabilitex pxo uestxa longaeuitate 
atque salute emittant domino Chxisto suas oxationes, ut longis pacifi- 
cisque tempoxibus uniuexso oxbo texxaxum et omnibus Sanctis et 
catholicis dei eeclesiis condoneiis, pex omnia inuictissime et Chxistianis- 
sime txiumphatox sempex auguste. 

Otxeius misexicoxdia dei episcopus Melitene metxopolis sanctae et 



GR.EEK AND LATEST ADMINISTRATIVE DOCUMENTS 



61* 



eatholieae ecclesiae propria manu suscribens secundum sanctionem 
uestrae pietatis meam sententiam per praefatam epistulam protuli 
quid sentio et in eonfessione sanctorum patrum Nicaea conuenientium, 
quam secuta est et Calcliedonense sanctam et uniuersale concilium, 
simul et quod sapio super Timotheum, quern preces aecusauerunt, 
et custodiri uestram potestatem sancti ecclesiis et omni orbi terrarum 
temporibus longis exoro 

Iobannis episcopus Axeae similiter 

Adelphius episcopus Arabissi similiter 16 , 



16 On the date of these Letters, see above Chapter VIII, n, 1, 



62* 



APPENDIX II 



I. Tables 

i Helenojpontos, Pontos Polemoniakos, Armenia Prima 17 



Hierokles 


Notitiae I, VIII, IX 


Notitiae III, Z, XIII 


y Afj,aala 


I. 1 'Ap,au€ias 


1 > AfjLau€ias 


"Ifiwpa 


4 *Ij3vpvwv 


4 'Iftopcav ijroi 
nijjioXias 


ZijXa 


7 ZtjXcov om, 
viii, ix 


7 ZrjXtov 


SdXrov ZaXi^rjV 


6 ZaXfyov Tyroi 


6 ZaXiypv tjtoi 




AsOVTOVTToAtzWS 


AeovroiToXecos 


"AvSpaTra 


5 'AvSpaTroSwv 


5 s AvSpaTTtov 


"AfJLlOOS 


2 *Afj,iauov 


2 ^Afxivaov 


UlVCJOTTT) 


3 UlVCOTTTjS 


3 UlVtQTTYJS 


— 


II. Ev)(alrwv 


II, EvxaiTtov 


NeoKaicrdpzia 


1 NzQKaiuapiias 


I. 1 NzoKaiuapeias 


Kofiava 


5 Kopdvcov 


4 Kofxdvwv 


IIoX^iiwviov 


4 IIoXzfxajvzLov 


3 IloX^ficovlov 


Kspauovs 


3 Kspaaovvros 


2 Kzpalovvros 


TpaTTS^OVS 


2 TpaTre^ovvros 


II. Tpairz^ovs 


— 


— 


5 'AXvas 


— 


— 


6 f Pt£atou 


— 


— 


7 Kokkov 


— 


_ 


8 Evvikov 


— 


— 


9 'ApaSdorj 


— 


— 


10 MapTVpOTTO- 

Aecos* 


— 


— 


11 6 'y^Ads* 


Ze/?aaT€ta 


1 Hzfiaariias 


1 Uefiaarelas 


NikottoXis 


3 NikottoAzcos 


3 NikottoXzcos 


KoXovia 


5 KoXcovlas 


5 KoXcovGias 


UdraXa 


4 UardXcov 


4 UardXcov 


JSzfSaurovTToXis 


2 ZefSavTovTToXsojs 


2 UzjSaaTov- 

TToXzCQS 




6 Brjpluurjs 


6 Krjpiaarjs' 18 



17 Ramsay, Hist. Georg,, p, 319. 

18 On the Episcopal Lists in general, see Beck, Kvrche, pp. 148 sqq,, and Robert, 
Villes, pp. 428 sqq. Also, Jones, CEJRP, Appendix. 



GREEK AND LATIN ADMINISTRATIVE DOCUMENTS 



63* 



ii. Pontos and Armenia 19 



Justinian 


's Re-organization 


Older Byzantine 
Arrangement 


Armenia Prima 


Bazanis or Leonto- 
polis 


Unknown 




Theodosiopolis 


Cappadocia Prima 




Trapezoids 


Pontos Polemoniakos 




Kerasous 


?j 




Satala 


Armenia Prima 




Nikopolis 


33 




Koloneia 


33 


Armenia Secunda 


Sebasteia 


Armenia Prima 


>? 


Sebastopolis 


33 


33 


Komana 


Pontos Polemoniakos 


33 


Verissa 


Armenia Prima 


33 


Zela 


Helenopontos 


Helenopontos 


Amaseia nrbs 
Amisus urbs [sic] 
Ibora urbs 
Eukhaita urbs 
Andrapa urbs 
Sinope urbs 
Leontopolis urbs 


Helenopontos 




Neokaisareia urbs 


Pontos Polemoniakos 




Polemonion urbs 


33 




Pityous phrourion 


33 




Sebastopolis phrou- 


33 




rion 





19 Ramsay, Hist, Georg., p. 325. 



64* 



APPENDIX H 



J. Xenophow 20 
KYPOY nAIAEIA 

I,ii 

3, ot 8k iJspcrt/cot vojjLOi irpoXafiovres emfteAovrat ottws rrjv dp-gyv 
jj,rj roiovroi eaovrai oi TroAtrat otoi wovypov nvos rj alaxpov epyov 
i^>ieu9ai, emju-eAovrat Sc cSSe. 

"Eanv avrois iXev94pa dyopd /caAov/xep^ kv9a rd re /JacrtAeta /cat 
raAAa dp^eta 7T€7TOi7)Tai. evrev9ev rd jaev dVta /cat oi dyopatot /cat 
at tqvtqov (f>covai /cat d^etpo/caAtat aTreXijAavrai et? aAAov twof, cos 
}xiyvvi)Tai tj rovrwv rvp^T] rfj rwv Tre7Tai8evp,ivo)v evKoajiia* 4, 
SifjpTjrai 8e aiV^ 77 dyopd 77 77€pt rd dp^eta rerrapa pipy* rovrcov 
S'eWtv eV ftev TraiaiVy ev 8e i(f>7)j3Qi$, dAAo re/Wots d^Spdcrt^, dAAo 
rots virep rd arparevaip,a eri) yeyovoai. vdjua) 8* <=ts rd$ eavrcov 
%wpa$ l/cacrot tovtcuv Trdpeiaiv, oi jaev TratSes" d/xa rfj rjixepa /cat ot 
reAetot dv8pes > ot Se yepairepoi tjvik' av eKaarw irpoxcopfj, ttXtjv ev 
rats rerayp,evais rjjjLepais, ev ats avrovs Set irapetvai* ot Se e<j>7)j3oi 
/cat /cotjLicovrat 77€pt rd dp^eta ow rots* yvpwyriKols ottAois ttXtjv rcov 
yeyaprrjKorwv^ ofiroi 8e ovre iirilrjrovvrai, rjv ju.77 iTpo7nrrj9fj 7rapeivai, 
ovre TroXAaKis aTretvat /caAdV, 

5, "Apxovres 8* e^* €/cdora> tovtcuv raV jieptjov elm ScuSe/ca* 
St68€/ca /cat Uepcrwv $vAat 8irjprjvrai» /cat eVt juev rots' TratatF <?/c 
rcDv yepairepwv r}p7)p,evoi elulv ot a^ So/ccuat tovs TratSas 1 fieArfarovs 
diroSeiKVvvcu* hrl 8e rots £<f>i]j3oi$ e/c rwv reAeta>v dvSpcov ot av ad 
rovs e<j>tfj3ovs ^eAriurovs So/ccuct Trapeze w eVt 8e Tots dVSpdo-tv ot av 
So/ctoat rrapixeiv avrovs ftdAtcrra rd reraypeva iroiovvras /cat rd 
7rapayyeAAoju-€va vtto rrjs fzeylarrjs dpxv}$* elal 8e /cat rcov yepairepwv 
TTpourdrai T^p^juevot, ot irpoorarevovaiv, ottcos /cat o^rot rd KadrjKovra 
a770TeAcScrtv, a Se eKaarrj rjXiKia irpouriraKrai rroielv 8i7)y7)aop,e9a > 
ws p,aX\ov S'^Aoi' yev7)rai § e7Tip>e\ovrai (hs av j34Anarot. elev oi ^roAtrat, 

6* 01 pi]v 8r) TraTSes* el$ rd StSacr/caAeta <j>oirtjovres 8idyovai p,av9d- 
vovres St/catoaw^v .,, 

8, ,,. p,av9dvovui /cat ro^evew kol aKovrt^eiv, 



Sen, Cyrop., L. I, pp, 10/1-24/5. 



GREEK AND LATIN ADMINISTRATIVE DOCUMENTS 



65* 



M4%pi p,kv 8r) ii rj eWa/catSe/ca ircov oltto yzvsas ol TratSes* ravra 
TTp&TTovow, e/c tovtov Se ets tovs £(firj[Sov$ e^ep^ovrat, 

9. Ourot 8* at? oi l^jSoi Stdyowtv cBSe, Se/ca en? d(£* ov" dv e/c 
TraiSscov ££4A9wai KoifMwvrai \xkv Trept rd ap^eta^, c5o"77€/> TrpoetpTj/ca/^eVj 
Kal <j>vAaKr}s eve/ca rfjs ttqAgws Kal uwj>pouvv7}$* So/cet yap avrT} 17 
7}At/cta p-dAtcrra em/xeAetas Seta^at* irapi-^ovai Se /cat r7;v 7)p*4pav 
iavrovs tqIs ap^ovai -^prjadai t\v n Setovrat virkp rod koivov, Kal 
orav jikv SeT/, iravrss p,4vovai Trept rd dp^eta* orav 8e i^irj /JacrtAevs 
677* drjpaVf e^dyet Trjv T^ucretav rrjs <j>vAaKrj$* Trotet Se rovro TroAAd/ctS' 
rod pw)v6s* e^etv Se Set tous i^wvras rd£a /cat Trapd r^v j>ap4prav 
ev /coAed) /coTTtSa 7} udyapw, in Se yippov /cat TraArd Suo, toore to ^ev 
a<f>€ivai> tcq 8 ^ edv SeTj, e/c ^etpos xprjudai* 10* Std rovro Se 
Sj]fjLO(jia rod drjpav eVtp-eAovratj /cat fiamAsvs wairep /cat ev 7roAefta> 
rjy€p,(&v ecrrtv avrots Kal avros re frqpa /cat rcov dAAeuv emp-eAerat 
oVcos av dypwmv, on dA^d^ardrrj avrois So/cet etvat avrTj 17 jueAerT; 
r<3v Trpos rdv tto Ae^tov. ,,♦ , 

22, At 8* ad p,4vovoai <f>vXal StarptjSowt p-eAertucrat rd re aAAa 
a TraiSzs qvtgs ijxaBov /cat ro^evetv /cat d/covrt^etv, /cat Staycovt£dp.evot 
ravra irpos dAA^Aovs StareAowtv, etc^t 8k Kal SyjpLoawi rovrtov dycDves 
/cat a#Aa 7rpor0erar £v fj 8* aV ra)v <f>vAcbv TrAetcrrot tScrt SaTi^uoveVrarot 
/cat dvSpt/ctoraTot /cat ewtcrrorarot, eVatvowtv ot TroAtrat /cat rtp.tucrtv 
ov fxovov tqv vvv ap^ovra avrtbv, dAAa /cat ocrrt? aurovs" TratSas oVras 
£7rai8zvG€* -^pwvrai 8k rots pivovai rwv £<j>rjj3wv at dp^at, ijv n ^ 
<f>pQVpr}(jai Se^cr^ 19 /caKovpyovs 1 ipzwrjcrai 7} Arjaras VTroSpafAzXv rj 
Kal aAAo ti oaa loxvos 7} Ta^ous Ipya eart, 

Tavra juev St) ot l^jSot irpdrrovrnv, OTetSdv Se rd 8e/ca erT) Stare- 
Aecrtocrtv, £§4p%ovTai ets 1 roz)s" reAetovs 1 av8pas. 13* aft 0$ 8* av 
££j4A9co<xi xpovov odroi a5 tt4vt€ /cat etKocrtv erT) Stdyovcrtv cSSe. trpwrov 
p,kv ocrTrep ot %</>7)f$oi rrap£)(0vt7iv iavrovs rats ap-gais xpijadai r\v n 
84t) vnkp rod KOhvov } Sua </>povovvTCov re 7^877 epya ecrrt /cat Irt Suva- 
jx4vwv. rjv 84 77ot S4r} oTpaTGVsoBai} ro^a jikv oi ovrw TreTratScvjueVot 
ovk4ti e^ovres ovSe TraArd crrparevovrat^ rd 8 y dy^jxa-^a 6VAa /caAotf- 
jueva^ dwpaKa re Trept rots or4pvois /cat y4ppov iv ttj dptcrrepa, otovirsp 
ypd<j>ovrai ol JJ4puai i^ovr^St £v 8e r$ Se^ta jud^atpav 7} K07rl8a. 
/cat at apyai 8k iraaai rovrwv KaBioravrai ttAtjv oi rcov irai8mv StSdor- 
KaAot, 

'^TretSdv Se rd TreVre Kat et/cocrtv er^ StareAecrcocrtv, et^crav ftev 
av o-Srot TrAetdv rt ysyovorss t) ra Trevr^/covra er^ a7rd yeveas* e^ep- 
^ovrat 8e rrjviKavra ets rovs y€pair4povs ovras re /cat /caAovftevov^. 



66* 



APPENDIX II 



14. 01 8 ad yspairzpoi odroi urparsvovrai p,kv ovk4ti e£tu ttjs 
iavTcbv, ot/cot Se fxivovTSS 8iKa£,ovoi rd re KOivd Kal t<x tSta 7rdVra, 
Kcd davdrov Se ofiroi Kptvovai, /cat rds dp-^ds ofiroi irdcras alpovvrai* 
/cat tjv tis rj iv £<f>7]f3ot,s rj iv reAelois dvSpdcnv iXAiTrrj ri tu>v vop,ip,tov, 
xftaivowi p,kv ot <f>vXap)(oi e/cacrrot /cat rcbv dXXcov 6 ^ovXofx^vos, ol Se 
yzpairepoi aKOVuavres sKKplvovuiv* 6 Se iKKpidzls drifios StaTeAet 

TOV XoiTTOV j8tOV. 

15, "Iva Se aa^iarepov SyjAcodrj Trdaa rj Ilzpudbv TroAtreta, fxiKpov 
€7ravct/xt* vvv yap iv ^pa-^vrdrto dv 8rjXuodziT) Std rd 7rpo€ipT}p,£va. 
Xiyovrai p,kv yap Ilipuai dp,(f>l Tas SaiSe/ca fjivpidbas etVat* tovtcov 
S* ovSets aTreA^Aarat vojjlu) ti/jlcov /cat dpy&v, aAA* Ifean irdui Hipoais 
7T€fM7T€iv tovs iavTWv TratSas ets to. KOivd rrjs SiKaioavvrjs StSaovcaAeta. 
aAA s ol fx^v SvvdfAizVoi rp4(f>€iv tovs 7rai8as dpyovvras iripmovmv, 
ol Se fJLT) 8vvdp,evoi ov 7T€}j,7Tovaiv \ ol 8' dv TraiSevdtocn irapd tois 
8y]}xouiois StSacr/caAotS", e^eortv avrols iv tois i<f>rjfiois veavioTceveo^at, 
rots Se fjL7) StaTratSeu^etcrtv ovtws ovk e£ecrn>\ ot 8 dv ad iv tois 
i<j>rjfiois StareAeoxocrt ra vopLifxa ttoiovvtzs* e^eort tovtois els tovs 
tgXsiovs dv8pas ovvaXl^zodai Kal dp-gcov Kal Tt/xdn> /xeTe^etv, ot 8* dv 
jxj] 8iay£vwvTai iv toZs iifirjfiois, ovk elcrepxovTai zls tovs TeAeious 1 . 
ot 8 dv ad iv tois reAetots StayeVtuvrat dveTTiXrjTTTOi, oStoi twv yepat- 
Tipwv ylyvovTai, ovtw jxev S77 ot yepaiT^pwv Sta irdvTwv twv rraXcov 
iXrjXvdoTGs KadluTavTav /cat tj TroAtreta avTrj, jj otovrat xpwpzvoi 
/JeArtOTOt av etmt. 



III. AEMENIAN ADMINISTKATIVE DOCUMENTS 
A. Throne list - Gahnamak * 

[fifi Um^Cuiljwj jubqjip tupiupkui^ jj qpmu mpgULuji jlXpmm^iup f^mqmLnp^i 
qnp fj $[iumnuu fumLuhnhw^ hpi hu qmufiiulpiihh uJjmSm qnp IXpmm^pji fr 
qjiLinhfih mhufi ji gwnny wSunj V£ : \}l ^nwdwj mpgmjfi k pmpbpmpfi 
L qjjp mpmpfi hu Utu^wty If iup fulfill nu pi npufiu £hp prnphprnpiuPJiLun. {,pwdiuh 
mmgi C^Jjnq mnmmmq L mmhnLmipmhq npujiu jmnw^ tun. ^tujng mqqfiu y 
Iwjumiu L fi &bpnLi) qpLuiuji ^m^Cmdm^jT^fiu mnhhj^ np JuiJuS £hmi &ujn$ 
wnujuiuiij L mrnhntmipmhu q.m£ ji Jiujm ifatp • \,njumiu \,hpuhC CpwiiwhwL 
wpgwjjii] uipgwjji L hu [JJw^wif £wjnij Ifmpnqlilinu fyujikijwg ut^m^hhSmlfjih L 
hquig {jwrnmuji ump^uij^rj wp^iujfi h nilhpu, k i wjuiu^u mpnmp L fymdmpfim 2 : 
\JXn.w2ph ji^juwh L dwfuniunh^ 



w. 


UjiLuhuin mip 


dk. 


IXSuimnLuji mip 


Z 2 * 


IXuiuhmu 


</,. 


^nnjpwh mip 


1- 


IXpbpnLuhwrj mip 


di. 


^unLlihmg mip 


W 


JfwqjuwgnLU^ 


d E . 


Ujq IXhkmLwaji 


t* 


Ifmd^ilnhijiq w£p 


dp. 


Smjng 


h. 


fjwCwuju \jnifiwij mip 


v- 


pmuhbnj qmrnmrnph 


* 


Jfnlfwn mip 


jim. 


fyupnLubmg mip 


f. 


fh^inntuhujn mip 


h- 


*£ mp&mLnLujiu 


E* 


*lm£unLuhmtj mip 


h- 


^mpqSmumj mip 


p. 


ljmumi[itj mip 


P h. 


JJm^mruiLuhu 


d. 


IXu&wLmghwij mip 


h- 


Qwphqihjiu mip 


dm. 


IXmwCnLuhwij mip 


pi- 


Uphqiufiu mip 


dp. 


^wSumpwlfmuu 


pp. 


UjiLukmij kplfpnpqu 


j+ 


Ujq Hupn&Lufi 


P p. 


IXpbpnLuhmg hplfpnpqh 


j q . 


*£ wumunmnhh 


L- 


IXpbpuLuhmu hppnpqh 



1 See above Chapter X, n. 26. 

2 On the problems of the Preface, see above Chapter X, n. 43. On this document 
in general, Toumanoff, Studies, pp. 229 sqq., and tables iv ? xiii. 



68* 







APPENDIX IH 


jm. 


irwifji![nhl;[ii} bplfpnpijSj 


&w. 


U wwb qnLUjjh 


LP- 


f^ntfjubmhh 




b F . 


IXnwihhhwhb 


LI- 


q-—H£iiijkuihh 




b q . 


Spndifih 


LI- 


^fiiiwgukwhh 




& q . 


Jfwiiphpwgjih 


L k. 


Pntfum *}*fiiiwgubujhh 


bk. 


ZuiLUmSjjjh 


LI- 


fy*L ttpbqbwhh 




& 2 . 


pdhmSifi 


Li- 


JXjl ^yfcSwgubwhh 




H. 


*fiw$pbpm.hfih 


LE- 


tflwjniSifih 




b E . 


]Th$nLhfi 


LP- 


JXnimJsijhmhlii 




tp. 


\,wfv£hpfj'ii 


y 


IX^w^dmphwhh 




k- 


^pmqui^muj[buj!j wpgniSjfi 


futu. 


ZwfipnLdkwbh 




Ipil. 


fipummbmh wp^mSiji 


fup. 


*£ wpwuupul^bmhh 




to- 


JXpuiw^bubwbh 


H* 


QfiiStwtywhh 




to- 


^uiimhqm^jjb bpl^pnptj. 


M* 


«#*#& 




to- 


SrJLfi 


fub. 


Quipk^WLiuhhwhh 




ith. 


<lpdmhnL\hfi] 


to- 


Eh&iiijkgfi 




hi- 


JXgwbm, 


A* 


lTwbijiiilim3j[]b 




hi- 


Cjiptulfw f hpfiw t pubwb 


fa. 


JJijliniijjih 




to- 


^wqjipljmhh 


w* 


Swjqpkwbb 




W- 


Jfmpm^kmh in]jkp~\ 


b. 


bp^whpjnLhfjh 




£ 


*£ mmqpmumnt^jih^ 




B, 


Military 


LIST 3 


IXphSwhmh qntnli 




JXpkhkjbiJjh ipihph 




JXhrjhq^ mmh 


# 




y^t^ [ir'^s 




pTjh^jvh Hqb&bwij 


m 




JXSmulj^nbfi 8 




Ppqhm.hiuliwli 


i 




JXLtnbujgjih S 




JTwbwLiaqhwh 


a 




^wp&WLntSjJi 8 




pmtj.mp mm ndip 


n 




SwSpmpw^fi £ 




IdnpfunnmSifi 


n 




Ifm^mq^aj^lj £ 




X?mjiw£iJi 


n 




Snqljbwwh € 




^UJ&lLhfl 


n 




^pj^rnhp b 




Xkmm^nLhjj 


it 




^mahmip b 




^rnhfi 


Z 




Piugwh b 




fiwubhwyfi 


n 




y£pm.hfj < Li b 




tywipLUfi 


J 




%ntl^mhh b 



3 See above, Chapter X, n. 27. 



ARMENIAN ADMINISTRATIVE DOCUMENTS 



69* 



Qubwgfi 


m 


(J)iuwuiqwpnLhfih 


b 


ITmuqml^nLhp 


J 


^mnjip^wu 


b 


UwnJ^ndifi 


J 


^jidwunLufi 


b 


^wpwdunLufi 


J 


Qmuqiunuiu 


b 


"IXjbhhwIfwh 


£ 


Vnqm^ft 


b 


IXaiiLkhbwh 


J 


U^iubhnfj 


b 


<£ wpdhmJikwh 


£ 


JXu£^thmn 


b 


Vjqmu[q]nLh[j 


J 


yfjhwh 


b 


fhiaifjuhuiu 


£ 


$uiifpkiiih 


b 


Zjiuifnujfth qnmu 




ZuipwLwjJih qnLnh 




Pqh^fuh ^nLquipiutj 


r w^ 


^rnqS^m^fi 


w'i{j) 


^wSuwpwhwh 


n 


^npqnLiuijfi 


n 


ywuujkgli 


i 


IXpbpmuji 


n 


flLwtuitjfi 


n 


n*p£tnm.hfi 


n 


l[JwLqiwn[i 


n 


ITnliwzjfi 


n 


Swjhgfj 


n 


<hnflPhh 3 fj 


^ 


Jfiuif{jl[nhkwh$t 


n 


IXh&luijijIi 


i 


^wuwun.wrjjih 


n 


Zivprndhwh 


i 


^wpqSuiuh^jiu 


n 


Spiqmmndiji 


£ 


OpqnLbJj 


I 


Uh&ntifr 


£ 


JXnnLhijkiubgh 


I 


W»3h 


j 


H^ngkwh 


I 


Qwpi&Liuhkiuj 


j 


*}*fjSuiDijkwh 


J 


hp[pL\mhpniSifj 


j 


fyj&^Jiity 


J 


ZmSmumnLuhmh 


£ 


pnfumjk^ 


J 


JXpinw^hubwu 


J 


^utpkqkuih 


J 


ljmn.pwmni.ufi 


£ 


JXphnhmh 


J 


lXppm£mifhmu 


£ 


ZivthnLiip 


J 


SpnLufi 


J 


IJwCumnihJ] 


J 


fini.dmSih 


i? 


Qkuwl^mhwh 


J 


•fiw^pkpnLufi 


£ 


JX^jumqwpkwh 


£ 


finqniStji 


£ 


<£ wpmqtupmjiljkujh 


£ 


Ifntpmnmu 


J 



♦ ♦•A nnfimuu f>jmjj_ m^fump^u qpimhwiii* k ^jjhfi PJjl Smpmjil^ mpmhn np 
pun. wnjfiULj> ty—fig £mnwpmij ^nq qummumLnpu qpmuu wpgniStJfnp i numwuq 
npg p mmmhpmnS h^tuhtfjh pun. w\jf\j>mjfi A Smpqmhmml^mh np i hhp^php 
mm^mmmu fi i[kpwj Prnqmiinju A qmu^nuju A pun. mdkuwjhu Pfu. nopmnh 
£mjntj x np I ^mpfitp A gumu £mqmpmij 4 ; 

4 On the Military List, see Toumanoff, Studies, pp, 229, 234 sqq,, and table v. 



70* 



APPENDIX III 



C. Pseudo - Gahnamak 5 



fe, 1)l fippk &knhw 


nphnuiL ifkbh 


PqhmSijig* 


\,hpu£u 'jt Mhutupjiui, 


( k hlfkwj^ 


'A 


JJwuuihkwhg, 


Zmju hnpnqip qqwCi 


u fi^fuwhw 


B h 


^jiunhp. 


Zwjn$y qnp ip l^uip^kw^ 'ji }\}nupi 


i — 


fcllkqhwhg. 


i[ujj L JWJL pwqujLnpi 


wrjh : Qhnj 


hi, 


^PuiLu^bmnLUp^. 


L hm Ijiupqip Y* dw3 6uj£njb 


'A 


JXh&wkwjg. 


ukquih tup^wjfih \Xp£Uil{iuj, puip 


(ku 


Ukpwuinkiuiip. 


inphguipjiLp, np kh a 


l J un £M- 




Upnuuhkmkuijg, 


Zmjlpmihli^. 






{fjikuiLinnfig. 


tfliupphhwhg. 






HuufiuqlinLbfig. 


Xhjpwp uimhm hg . 






fy^wnLUfi^. 


prnqpuimnLlJ^. 






^ui^whnLh^, 


Spi}uiinnLUfig. 






IXqkhkwhg* 


JXuu^kwnLhfi^. 






"finijkwhg. 


Ifmjv uiqp Llifig ♦ 






^w^prnhfi^p. 


J\)nphjnTwi.Iihp, 






ywjjuwpiuljuilij). 


IXpbpriLhfig. 






Ifnlpu^p, 


Ifwdjilinhhiuhg* 






Uqfindifip. 


UfiLbkgfig. 






-phpniSife. 


l^S win tiding. 






IXuipufuj math fij>. 


IXbqhtjktuhg. 






f^nqphkwjg. 


*lp m Bh> 






^uinphbmhp. 


tynifip. 






Qnqlikpinp, 


^wpwdhndifig. 






Ifwhiwnkmhp, 


ITuipqiqkinnLhji^. 






ITpnLnLhjig. 


^wCluiLhjig* 






thwiifimSifig. 


tyw^ui Ltl Llljlg • 






^wpkqkwhg* 


^liuiipg. 






Dwiupnthji^. 


XJjjuuihhwhg. 






^ui^mifi^, 


Ijiuqilkiuhp^. 






ilpph^nLhjig. 


IFwhuiLwykwhg* 






JJnLphwhj). 


bqhuhwfig* 






^pSiulpikwhg, 


%w8phwhj>. 






Upnthlig. 


5 Nerses, pp. 32-39, 









ARMENIAN ADMINISTRATIVE DOCUMENTS 



71* 



^mppuihnkui hg . 
JXpwqwbkwhp, 

Qhbwjhnijg. 
Zwpghwhg. 
MnpqnLwjg. 

Zui^inkwhg, 
t^phwhg. 

4> m nw£n LUJig , 

pLpbkwjg. 

Ifk^jiwkiuhg* 

^ui^ujljujpfifi^, 

Puiukhg. 

XjuiqlimSifig. 

Ifwifjilinhg hplfpnpq. 

<php£dm.h]ij>. 

pwqnLiuhj). 

JXphqnLhfig. 

UuiCwnnLhtiD* 
JX^Iuwqnphwii^, 
9*&n *.£/»,£ . 

UnnijiunLliJi^. 
DmCnnw taking* 
Xjuijpfi n L-Iifij} . 
Qwtiw££[ipuj inking. 
*£ uipiunwinwhkwhj>+ 



Chbwjkiifa. 
ITqjtnLUJi^. 
IXp^ndiji^. 
*piunbwhg, 
ITwpquifjkwhg. 
l^kl^mhqkui'h^, 
fonpkkhfig. 
Sfithwlfwiig. 
Zwfini-bkwhg. 
Ijnppkwhg. 
tjqiiLbiilig. 
QwpLCwtig. 
SnLpphpwhkwhg* 
fidnhhjig. 
SiptunLlifj^, 
ITktpmSifig. 
•fituqiugiuufbing. 
IjjiijjnLykw'Ug. 
Utqwitrj.nLhp^^ 
U,pwuj£ujuikwli£>. 
flpuwiqkm^, 
IXpgwbnij inkwpj>. 
f}*wihukwljp* 
fiwqpwuuinLiiJig. 
^ujpuiqnLhfj^, 
Uqkqhwqpo^. 

*pwqnLkwhg. 
IXuiqwpvufvwq^ g. 
JXppiniuii/kwIig. 
Zfu-uhutlpiihg. 
pmqmuuilj^uih^. 

J^kwritiuiliwhg. 

Xfwuipwuwdkwhg, 

Zk£Stummhp, 



72* 



appendix in 



^wpnk^kwug* 
f }*pwSwhrj.g* 

hjiuqpkiubg. 
%po£wl[lipg. 
XfwjuuijnLhjj^ 

IXpnLdbiubg, 

Iqwqwpgg. 

Swpkkiuhg. 

QnpwphwJjg* 

Ikpgwjfibg. 

pwpilhwjj). 
Ijwiljiikujjg. 
9*iy Liu aw wkmg : 

fct in j i npg fifth ^ qw£u. 
QiiLppwkwhg. 



tpogwjnLuJij)* 

Ifnmiijkiuhg. 

UuiklinLhfig. 

hjiwLbfig. 

pmjjhniJijj^* 

IXijhkuii&Jjg* 

SfiLpiuhhuihg. 

IjqhnLufig. 

Sm^nLulig. 

JTwqJim.uJig* 
JTwfuwiiniSifig : 

iXju hh nw&jg k ujjj_ ku jnjni[^ 
tjjip bwhp ^uidwpktjwj mpljmhk^ pfoq 
J}pni[, qnp hnpnqkiuij }Xp-£wl{ 
BtuqujLnpIi k l^wpqkmij ^r dmS 
£utfnjh fiLpnj ^jt ukqwu jitp^ pmph] 
%npkDwpkLpu y CpwfiwhuuL ifkbfih 
X^kpupufi : fiulf y ji pfjl^uihij lfnLu£ 
huuiifjh klfkqkrfwliwug, nuwujihh 6 ; 



D. Armenian princes accompanying St. Gregory to 
Caejsarea 

i. Armenian version 7 

£&P* frul[ PwqujLnph tyuiqi[wquilili lfwLpnil k kplfftuifiL ifkbuiL JuhnnLpkwiip 
^ntj.mgbm^ anhSmpkm^ qj^jJumLnpuh hwfuwpiupuitju, ql[nmuil[ut[u m^fuwpijih^ 
wniu^fii fefuiuliu USjqkq inwh, kpfypnpn fefuwhh IXh&ukwrj* np ip pnkw^jvh 
ifkb, kppnpq fefuwuh XfwpnwkwnLpkwli fejvuihnLpkwhu, fnppnpn fefuwhu 
pLunLuljUJiu x JjyjvujhnL.pkuib JXuiqkmnLpbwLh^ ^ftuqkpnpn jr^jurnhh JJwwpwwk—* 
uwLpkiuh, tjopimjuip Z&j n g m^juwpijih^ tftrjkpnpn ji^jvwhh Ijnpqmlimug 
w^juiiipZJiti, kophkpnpn fefuwhh \f mining m^fuuip^jih, nLpkpnpq fefuwhu 



6 On the Fseudo-QaJmaT/mk in the Vita of St, Nerses, see Toixmanoff, Studies, p. 229. 

7 Ago?., pp. 590-591, 



ARMENIAN ADMINISTRATIVE DOCUMENTS 



73* 



fynjLtywpwgLntj w^fiJWpCfih *np iffim mhrnrnhhrn^ pqkw^fuh, fihhbp'npq. fofuwhh 
fh^mmlikuig m-^umpCfihy mmuhhpnpTj. jj^fuwhh ITnlfUJ!} w^Jvwp^Jih, dkmwuwhh^ 
pnpq fejuwhh Ujuhkiug m^jvuip^jih^ kpl^nmmumhhpnp^. fefuujlifc tyiuLqi[ig 
wpfmup^lih, h phjy ww mnhhpnprj. jt^Juwhh flmiliujijLnij m-^uuip^jj^^ inpkgww— 
uiuhkpnpq fj^fumhh yw£wufi Qwpujmjhq. h Zkp qiummji, ifibqkinwuujhkpnpq 
fefuiufch XfmqjuwqpLpkmh mwhh, ifjs^mmuwhkpnpq. fefuwhb XkpbpnLuhwij ; 
JXju fi^fuiubg hh pbuifjpg, fynLiuulpii^g, ^nqphml^wj^p^ iwzpiipWLiipg, pfupw— 
uipg ft di^ £wjuiumwh w^fump^jj wwhh pnpqnSwj, rpnp qnLduipkwij pwtpuLnpb y 
k mnmokwg nhnuw h hnndmhu kiuii[uiqmjk}iiijLnq ji Drntpuok IfkuuipwijLnijy 
qnp pmn £wjbp£h jkqiiLfih Jfmdm^p Ifnikh, qji wwpkw^ ifopjiqnpV' ^wiuihuijimqkm 
Ipnrjnujrjkh fupkwhij tu^fuiup^fib, k CwbTjhpkkijw'ii qim mhk^ £whwujwp£iuijh : 
fez. ipmSmjkmrj ^pnijmpmml^ qpkj_ opphwlj ipuju : 



ii. Arabic version 8 

86, Cum igitux pexvenexunt epistulae, ad eum venexunt tres xeges, 
rex Abchazoxum {'lob's) et rex G-eoxgioxum (gxg'n) et xex Albanoxum 
flTnyn) et cum ipsis catexva pxincipum : pximus pxinceps 'Ihbns. 
Seeundus pxinceps 'xtnwg nomine vitaxa (bytqs) magmas, Textins 
pxinceps 'xmt'n. Quaxtus pxinceps 'sbytywn nomine 'sbyts, qui 
pxaepositus exat custodiae montium qwsywn et mtznywn. Quintus 
pxinceps mqwny'nwn nomine 'sbYb'ts : Lie autem pxaefectus exat 
exexcitui totius Axmeniae, equitum et peditum, nee discedebat a xege 
magnae Axmeniae, at que in bellis omnes quos memoxavimus pxineipes, 
et memoxabimus, sub eius potestate exant, pxaetexqxiam quod pxinceps 
qmxdl non exat sub ems potestate, quae (xegio) est foxtes qxdytn. 
Septimus xegens magnae xegionis swfnys, quae est iuxta xegionem 
fluminnm, Octavus pxinceps xegionis guxgx. Nonus pxinceps xstnwn, 
Deeimus pxinceps mfnswn. Undeeimus pxinceps swnywn et custos, 
Duodecimals pxinceps dsbwd'wn, Deeimus textins pxinceps ? tywn. 
Deeimus quaxtus pxinceps zwxydwn, et ipse iuxta qxdytwn, Deeimus 
quintus pxinceps Ixbwxwnyws, dux ? qui xectox exat rnlhVwxi'h. Deci- 
mus sextus pxinceps (dpxcov) 'xtzxwnyws. Et omnes gentes convene- 
xunt secundum gxadns suos. 

87. Et H sunt sedecim pxincipes quos enumexavit xex et ad quos 



8 Garitte, Agathcmge, pp. 72-73 = Mair, ChrisUanizaiion, p. 114. 



74* 



APPENDIX III 



misit propter honores eorum, Atque deliberavemnt de iis quae 
inter ipsos erant, constituerantqxie nt ad illxim irent, et hoe perfeee- 
rcmt. 



iii. Greek version 9 

135, ... 6 fiamXevs ... KeXevei ofiv rovs dpxrjyovs rtov uarpairvov 
ovyKaXiuaodai Kal roTrdp^as \ Kal rrpcbros dp-^cov AyyeXirwv oIkov 
6 $€VT€po$ ap-fcwv ^AXaevcbv, 6 ttjs Kop,iTarrjalas* 6 rpiros dp-^cov 
em rr)s itjovaias rrarpiKios Xeyofxevos* 6 rirapros dp-^cov 6 to SidSrjfJLa 
i^ovaiav e^cov iiriSrjcrai rep jSacriAet, d Aeyd/xevos ^Aairircov 6 nepLrrros 
dpXWV (jTpaTOTreSdpxrjs urparrjXariKrjs ifjovulas TrJ£_Apjj,€via>v ^cipas* 
6 zktos apxcov 6 tt)s Kop8ovXrwv jtopa^' 6 sfiSopos dpywv 6 erepos 
KopiTarrjuios' 6 oySoos dpxcQV ttjs *Povotivcqv ^cipas* o eWaros 
ap^cov d rfjs Koplrwv xebpas' 6 84ko,tos dp)(cov 6 rr)s Uvvircov j(wpas' 
6 ScuSe/caTos dp-^wv 6 UovSaicov -^wpas* 6 rpiuKaibzKaros dpywv 6 
ttjs Ovrlwv -)(wpas* 6 rsooapioKaihiKaros dp%u)V 6 idvdpxrjs Zapov- 
av8a>v Kal Xepas rrjs rrarplSos* 6 k4}j,7tto$ Kai hzKaros 6 rov MaX^a- 
£ttuv oikov 6 ef/ccuSe/caros' dpywv 6 ' Apaovpovvcbv . oSroi ol dp^ovr^s 
ol e/cAeKTOt roTTOKpdropes ', irarpiap^iKoi, ^iXiap^oi /cat p^vpiap^oi iv 
p,€<J(p rrjs > App,€vlwv ^cupas oikov ©opycop,, ovs ovvadpoioas 6 fiaoiXsys, 
d/TrecrretAev els rd p>ipr) KaTTiraSoKias els ttoXiv Kaiaapaiwv, Kara 
rr)v 'ApfAZViwv SidXzKrov Maad^ Aeyo/xev7?v, ottojs aTr^Xdovrss rov 
rprjyopiov dp^tepea Karaarr]o(jam irdar) rfj ^cupa. 



iv. Greek Life of Saint Gregory 10 

98. ZvvrjXdov Se TTpos rd TTpoypdp*p,ara ol rpeis /JacrtAcis", Aa^tov 
T€ Kal *Ifiipwv Kal > AX^avCbv > Kal pierd rovs fiaaiXeis d irpcoros ev 
dp-){7) 'IvyiXrjvfjs Kal *AvTio*Trjv7]s roTrapyrqs* 6 Se Sevrzpos rrjs ^Ap^ia- 
<.vr}>vr)s TTiTidtjrjs 6 p,4yas' 6 8e rpiros rwv Miphvov roirapx^s^ 
6 Se rdrapros yevedpxqs rcov ^OaTTiriavcov iTrdpxrjs oaris /<aOa> 
rzrayp,€vos vtttjpxzv (f>vXdrreiv rd p>€prj rtov KavKauivov opivov Kal 



9 Ag„ pp. 68-1 

10 Garitte, 



= AASS, Sept, VIII (Antwerp, 1762), pp. 320-400. 
s, pp. 72-73. 



ARMENIAN ADMINISTRATIVE DOCUMENTS 



75* 



T^dvwv 6 Se TrifXTrros 6 rwv MafxaKovviavwv ^ye/xcuv, da7rapa7T€r7]s 3 

TT&VTCLS TWV 'ApfJLGVLWV ITTTTOTCLS T€ /Cat TTZ^OVS €%WV V7TO T7JV OlKZiaV 

urparrjyiav, avvwv re ast rw /JaatAet rrjs MeydXrjs *App,Gvlas 3 /cat 
iv rats wapard^aiv irdvras rovs t<= TrpoTjyovfjidvovs /cat VTrorarro- 
piivovs <...> V7t6 rrjv oiKsiav OTpaTTjyiav zktos Sg 6 roirdp^s rwv 
KopSovavwv %wpas ovrw KaXovjJi4v7]s, ttXtjuiov VTrap^ovurjs rwv 
* AvhpiOKohpirwv* 6 Se €J38ofj,os i^ovaiaurrjs rrjs MsydTvqs So(j>avrjvrjs 
7TX7)uia^ovurjs rrj MeuoTTorajxla^ 6 Sk oySoos jjy€}j,wv rwv Fovyapwv 
%wpas ? Sans €t%€V feat avros d£iav mrta^ov* eWaros Se 6 rwv *Pova- 
rovvlwv TOTTdp-^TjS' 6 Se Se/caros 6 rwv MoKaawv uarpdin}s* d §€ 
eVSe/caros 6 rwv Svviwv <j>vXap^os' ScoSe/caros d rwv T£>av84wv 
(rrpaTTjyos* 6 8e rptcr/catSe/caros' <..* Tecrcjajoecr/catSe/caros^ 6 oar- 
paTTTjs ZavpafiavSwv /cat Xeipwv ^wpas, /cat rovrwv ttXtjuiwv rwv 
Kohpirwv 6 Se 7T€VTe/catSe/caTos y^v^dp^rjS 6 rwv Xovp^opwv ws 
rjyov}X€vos ix wv T V V ^PXV V T V V /caAou/xevqv MaA^a £dj8e* e^/catSe/caTOS 
d rwv 9 Apr^<ip">ovviwv i£;ovcriaarr)s m /cat rd Aot7ra Se yivr\ Kara 
rd^iv avvrj-^Brjaav n . 

E, Aemenian Princes accompanying Trdat III to Eome 

i Armenian version 13 

2?/*S. ••• Xhww fypk L nLUJL *L W J U wdkhwjh ilkb wpgwjh Zwjng, fvnp&Lpq 
wnhwjp lhuuh £whwwwp£wqh mjhnrjji^ £wbqkpkwj^ IfwqStp, wnhnjp phq 
jiLp ifhb IXp^hwjujl^nwnuh %pfiqnp h qhnpnii npqfih qfypumwglu^k qhrnjiul^n^ 
wnuh Hqpfiuibnu : fei j] qjihnujpiul^wh l^nqSwhlh qynphufih qw£hp£guh fiLpnj 
ww£wp]ih, np pqkw^fvgh fyn^frh. qwnw^jih mu^Smhmljuijh fi \,np ^jipml^mh 
lfnrp}whl;h) k nbpljpnpq uw^whwl^wjh jj£unpkuwwhkwjq Ijnqdwhlh, L qkppnpqh 
jJXpriLkuujujfo l^nqpmhlh^ qsnpnpqlt fi tfwugpixg l^nqSmhlh. tplkb fejuiuhh 
JXhqkrj^ wwh, k qPwij.wq.jip wuiqkwh, h quiqwpwufhwh dhb, h qfefuwlih ITn^zz/i/, 
k qji^juwhh VfiLhhwq^ h qjj^fuwhh fh^uMLLkwij, k qfi^juwhlt IfwqJvwqnLpbwh 
mwihy k q^w^wiqh ^w£,wwpijwhji^ L qfofuwlili UwwuliwwkwnLpkwhb : fe*. 
W JL p mi l nL fi iJkbwiJkbog) L kpwhwuriLU £wqwpWL phwfip qopnth £whqkp& 
fuwqwjp qbwjp jlXjpwpwin qwtwn£ ji ^wqwp^wiqwm gwqwgt whqwlih^ ji 
uwCSwhu Qnuhwi}. •*• 



11 On all the versions of the so-called " Agatcmg&os " and their problems, see Garitte, 
AgatTtange, also above Chapter X, n. 89a, Qn this listing of Armenian princes to which 
he refers as " List A ", see also Tonmanoff, Studies, pp, 159 sqq., and notes, also table vi. 

13 Agat\, pp. 643-644. 



76* 



APPENDIX III 



ii, Greek version 13 

164, ... ore odv TjKovcre roiavra 6 fiaoiXevs, i^ovXevoaro jiera 
oTTovBrjs rropevBTJvai TTpos avrov, Kal eroijxos yevopevos, TrapaXa^dvei 
rov fxeyav dpyievlaKorrov Tprf/opiov Kal rov vlov avrov 'ApcooraKTjv 
eTTiuKOTTOV, Kai and rwv TTGpl avrov ivSo^wv tovs reuuapas rijiivordrovs 
rov tStov TTokariov,, rov Trpwrov roTTOKpdropa airo NopcnpaKecov 
piepcov, rov Sevrepov roTTOKpdropa drro rwv rrjs ' Aoavplas pbepthv, 
rov he rpirov duo ra>v ' Apovaurwv p,epwv, ro Be reraprov aVo ra>v 
Maaaj(ovrwv Ovvvojv /xejoaV, /cat rov \ieyiurov dp-^ovra rov oikov 
rwv Xeyofxevcov *AyyeXirwv, /cat rov ure^>ovra rov jSacrtAea ap-^ovra, 

Aairer Xeyofxevov, /cat rov orparoireBapyr^v rov \Leyav y /cat rov dp^ovra 
MaXKat ) irwv > /cat rov ap^ovra Svvirwv, Kal rov dp^ovra x Pearovvirwv, 
Kai arroypd(f)Ovra ap^ovra, Kal rov eBvdpyr\v HaamavGyv , /cat rov 
dp-^ovra HirauKarrerewv y /cat irXelurovs ra>v dXXcov fxeydXa)V Kal 
i^8ofjL7]Kovra ^tAtaSas* e/cAe/crtov urparoTreBcov, 

165, i£eX9d>v odv *Apapdr rrjs warplBos Kal OvaXapaairdr rrjs 
iroXews, eirepaaev els rd pep?) rrjs *Icovias, ,,, 14 , 



F. Armenian Princes summoned to Gtesiphon 
i, Lazar P'arpeci 15 

II* hi* ••• hi- npg fi dwSnLU ji dkbmilhb uiLiuqwuLnju Zwjng* jnLfiji wblfhiu^ 
£wbw iqwp&prikzjuili* wjungfilf kh : h wn£ii£u Ufidikuii)* ^wuiulj Uftdihina 
miph^ np fj dm drub fiiupn^ufinuu £p Zwjng* k p uinCSih JXpbptiLuhwg \>hp£w— 
ufni.C ]} uin^d£h (^np^innLhkw^ Up mm if, fi mn£,Slh J\] n p fun nn ibhwrj fywqfenj, 
fi uin£ii£h Vmifjjlfnhlfig ^mpqmh^ Zwjng wuwmpmmhm k m£ph IfmilfiJjnhlfio, 
ft mn£fi£u Ifnlpug IXpinwIf. ft mn/rftu l£mw£ni.uhmij \Tuiu££, fi wn£,S£u l£ilw— 
mnthkmg ^w£mu. fi wn^S£h ^wClLnLukmij ^fitm. fi inn^diu IXb&kuinkwn 
&fiwi.nu. L jm^fumpQu tf^puii] pqb^fuu JX^hl^w^ k mj^ wiuunLinkwpg wyfuinpCfiu ; 



13 Ag„ p. 83, 

14 This passage does not occur in the Arabic version or in the new Greek version of 
the Idfe of St. Gregory published by Garitte, Agathange, cf. Ibid,, pp. 327-331. On the 
princes accompanying Trdat III on this hypothetical journey, see Toumanoff, Studies, 
" List 3 ", pp. 159, 161-162, and table vi. 

is LP\ pp. 143-144, 236-237. 



\ 



AEMENI&N ADMINISTRATIVE DOCTJMENTS 



77* 



II* hP ••♦ llnwgip mn unum Smpqu^iuhh IXmpnpitfinrjh ji pmh^h wpgwjfih 
QwqJihpmfiy k kpnhnjp hnqm qwuii£ pi n£ i^ng uufmhrnhkilg jj h^h^^ ^ n l 
qmlpnLpjnSiij. k nu^mmjiL {mbhilg jnLifbgt : fet mjum£u hhhqrnl hmm^km^ 
qjSwpqfjIjh^ Ipumih nuiilkhhukmh k mmjp fumnwijnujmkk^ fj rpiLnh JXpkmtj* 
qji^jumhh ilhb IXpbpnthkmg tj%kp^mmnL^ k nji^jvmuh mmhh JXdmmnthkmg^ 
k nji^fumbu mmfah ^mhmhnkmjy k np-sjvmhu wwhh JXpywpniSikwij qJXp^WLfjp, 
k qhtfumhu mmuh IXu&kmtjkmij qpSmmh^ k qjiyjvmbh mmhu JXSmmnLUkmtj 
tf^m^mhh* k qji^jurnhh QhpniStkmgy k Tjjj^jumhh JX^nymtj, k qji^juiahfi Unztfi— 
kTjkhJit} tjfpmijimg^ k njiyfumhh Sm^pmj xplpfyy k ijjj^jumhh lXpbpm.hkwij 
qJXujpuiufi'i k omjjith [i mmhnunlpmtj'li k jmLmn ubm^mgu^ jumnmgntgmhljiii 
jj qnuift pmnmmpjjlt Qmnfybpmfy : ^wdb^bm^ uil mrikb^ $m^ jj numb k 
fihgh hbhnmunp fejumhh XtJiLtibmij ^mum^ pwjJJ umju mmjp juwnmijnujmhb^ 
jmnm^mnnjh gmh njihgk * 



ii. MSe™ 

II* ♦♦♦ fez. whq£h pinqwuipfi fi hbpgu Ifntbijbm^ nrf.mpmmbmh^ ipwSwjip 
qp^i ipmlwpmwli* k n£ ku pum un^npntpkmhla \ wji pmuu amjpmnjihu fipp 
mn wmkj_[tu k mumfimmhu^ uff^ip^ mtlkhkhjih nilbbmifbb ^mummlpj mfjpmuip 
jimpnlfmbh. wji tljimjh fyn^nj ^pwfimh mntbrn^ jmlfmbi jmlimmhi awpu, nnp 
fihgh £whwslp, npnq lubnLwbgh bfc mju ; ■ 

h wntfib 0/riMznj ^lUVltt ifffimfc: 

h wntfih JlpbpitLhkm 9 \,brUUWfrZw&r,Lh : 

f» mnCiiih fl*p£mnLubmy JXPSUM mhmJi : 

ft mn^li hjnpjunnn diking 9»U7*fe£»/)S mhnik : 

^ inntftf Wini^IinHpS WP'hU'b rnhndt : 

h mntfth Ifnlimy UPSIM mhnth : 

ft mn^Slh I£wm£m.hbwg lTVL\>l*i$ mhmh : 

f* mn£ii£h JXdmmm.ftbmg ^UZUh mftm.b : 

f* mnCfiih ^wCknihhwg *hfohS uthnLb : 

ft wn£Slh tyji&kwijkiuij &lTU»hn\r uihniJi : 

Qwju bujjumpwp^u jwl^mhl jwhnuuhl; J^n^kgjib fi numb mp^nLhji^ ♦•♦ 17 * 



is MUSS, pp. 42-48, 

17 Ob all of tlie princely lists in 3Lazar P'axpeci and Eliae, and their reliability, see 
Toiimanoff, Studies, pp» 246 sqq,, and tables yiii-ix. 



78* 



APPENDIX III 



G. Partisans of Vasak of Siwnik' 18 
i. Lazar P'arpeci 

II, iS ••• fet npg funuinnphgiuh qlju[j uwiniuhujj[ih phq nLJvmwhbhqjiL 
^uiumljuif hh ujjungfilf. Jrsluwhti prnqpinmntubw^ Sfipnq, fi^fuwhh ]\}npfurmnL.h— 
busy ^wqfonj^ fofuuihh IXiquiCnLubwij Jfwui6 y fi^jvmbh ^wCkniSikwij fypLin^ 
fefuwhh fywipLbkujij ^wpwq^iuiqnL^li, fi^juwhh IXpbqbhfig [sic] IXprn^h^ fi^jumhh 
flwbuij Xfbpuk/iy b utJl,p jmjwwhl}wi}^ k ubwniXg nfiwug j^iLpw^mh^jiLp 
uw£fi£ * 



ii. EUie 19 

III, ••• huff qqnLuqh bppnpq jt kbnu ^uiuiul^wj fefuwhjih i/^i&^u/i/, np n^ 

ft pwq Ifwybw^ tp jnLfuin£u ^bpwunuwij pum hbpgjiu junp^pqnqh [iLpng : 

Quwpbwg k I urn phq fiLp x qnpng qjiwtp J Z^ /7JL Z /7L /^/' L ^ ZwLUwmy hngiu. 

QJi^jvwuh prnqpuimntiibwi} qopogh JiLpmj^g : 
Qfefuiuhh hjnpjunnnLubwij qopogh JiLpi^g : 
Qfi^fumhb lXupv£riLbbuiij qopogh fiLpni^g : 
Qfefuwhu ^wCknLhbwij qopogh fiLpni^g ; 
Qfefuiuub ^Iwjvihhwg qopoph fjLpmlg : 
Qfefiiujuh Qwpbqbfijiij qopogh jiLpn^ : 
Qfejvwuu flipbwj qopogh fiLpni^g : 

fez. qwjl putqni.fi qopu jiup^rnhfi inwh^h wplj jjihgb, k qubiqm^u nfiuiltu 
jwjjfii wn£fi£ : 

IV. ••• &bnu £wplf k bCjnu njpwqmfiu fi qSiqth ^Ppfiumnufi^ k juwrihbwy fi 
qnthqu qfiLwg : fez. fi pwqni.fi JWJL wbq jju qnqwpmp unqbqmt k bfinun fjppk 
qo& ji fi££ wfipuigbingB. k jvpuifi Cuimbwjjwiji^inwl^bujg k Iwn k b£iuu jiujuium— 
pbuifip qpwqmfiu juiqiuuiwrj k qpwqfiuiqnjhu fi -^jihml^mhmn^ k qwj^ nfiwhu 
jwtwLwhbui^ ^w^whwjhq : flpnrj uiuntuph bh wju* qnpbwl^qmg unpw. 



18 ZP\ p. 209. 

« EUie, pp. 74, 91-92. 



ARMENIAN ADMINISTRATIVE DOCUMENTS 79* 

h^fumhh f}*£iniiL$ikwij y JXpmwl^ rnhndj, 

fyfuwbb hjnpfunnnL&bwtj* ^mqfi^nj uibmb. 

fyfuwbb ^uiCluirfihwij* QfiLin lubmb. 

fi^Jvwhh pwypuiuinLbbwij* Sfipr^l wbrub. 

fiyhjinhh IXwiu^mbbwD^ \J*uih££ inhmSi, 

Ji^fvwbb ^uipkqlhJigKj* IXpui^h wbm.b. 

fyjuiubb IXtyinj {^b^mq^ wbmb. 

fr^juwhb pLpbiuj* \,bpub£ lubmb, 

ft^jvuibb ifhm ku tywjnLbbwzj* ^wpmij^mmnL^ uibntb, 

UbwnL^ i/Ji IXfiinmnrfibuiij* lfiub£b uibmb, 

PwnriLii b mil in a in in dwpnhh^ qnp nuwiubhliub wbriLivbbb* jwppnLuh uiwbl; : 
JjL pni^wliqwli *l\}^p P n l n P ui^fuwpQj wu^uuiwSpbgnjg jrupuignLpjiLb, n£ 

iljtiujb pum m-^umpC.ml^mb pwqpnipbwbb , wjj^ L qpuiifniSu ji umpp mjuwl 

blfbijbrjLnjh 20 . 



H. Partisans of Vardan Mamikonean 
i. Lazar P'arpeci 21 

£9*. fei uwwpwwbwb Zwjntj wtpb Ifwiffiiinbljii} bpmbb^fib ^wpqwb* 
wnbwjL pbq fiLp fi wwbnLinbpwtjb Zwjnzj, npg ^bpifbnwbqb ufjpm[ 6bin£[ib