Hostname: page-component-78c5997874-m6dg7 Total loading time: 0 Render date: 2024-11-10T08:57:24.464Z Has data issue: false hasContentIssue false
  • English
  • Français

The Fifteenth-Century Bagratids and the Institution of Collegial Sovereignty in Georgia

Published online by Cambridge University Press:  04 August 2017

Cyril Toumanoff
Cyril Toumanoff*
Affiliation:
Georgetown University
Get access Rights & Permissions [Opens in a new window]

Extract

The end of the fifteenth century witnessed the dissolution of the medieval kingdom of Georgia. The three kingdoms — of Georgia proper (Iberia= K'art'li), of Imeretia (Abasgia), and of Kakhetia — and the five sovereign principalities, into which that ancient and once powerful realm had found itself divided, were ill prepared now to face the onslaught of the Islamic imperialism of the Osmans and the Safawis or, finally, to prevent the nearest Christian empire, to which they had turned for aid, from absorbing them.

Type
Articles
Information
Traditio , Volume 7 , 1951 , pp. 169 - 221
Copyright
Copyright © 1951 by Fordham University Press 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Tables 1.625, 2.639; Panaretus, 48 (ed. Lampros, S., 4 [1907] 289): David VII's wife was a sister of Duke of Meschia [Samc'xe], son of John I; Tables 2.638-40).Google Scholar

2 Cf. below, chap. II.Google Scholar

3 The History of the Invasions of Timur (ed. T'aqaišvili, , QM Annex II) 855; Tables 1.625. The former source, possibly dating from c. 1424/1450, depends largely on the (Minorsky, V, ‘Transcaucasica,’ Journ. Asiatique 217 [1930] 91–2; Brosset, HG I 1.393 n.7; idem, Additions et éclaircissements à l'HG [St. Petersburg 1851] 386-97) and forms Part 2 of The First Continuation of The Georgian Annals, compiled in the eighteenth century (MGHL 159-61). The other parts of Contin. I., of importance for this study, are (3) the Bridge Chronicle, a series of entries bridging Parts 2 and 4; QM 887; and (4) The History of Alexander I and His Successors, possibly based on an older source: QM 888-91. There is, moreover, Contin. II. of the Annals: QM 892-973, the opening part of which is used in this study. — A fuller treatment of the Georgian sources for the period in question will be found in chap. 1 of the present writer's unpublished doctoral dissertation, bearing the same title as this study, submitted to the Faculty of the Graduate School of Georgetown University on May 14, 1948.Google Scholar

4 Hist. Inv. Tim. 859-64, represents Bagrat's apostasy as having been simulated, under duress, chiefly to save the others with him. The date of his capture is wrong: 1393; but it reappears, correctly, for Timur's second campaign in Georgia; pp. 858, 860.-' , I (Calcutta 1887) 401–3 (sack of Tiflis and capture of Bagrat), 407-8 (apostasy); Panaretus 53.292 (Nov. 21, 13871); Thomas of Mecop' (Mecop'ec'i) (trans. Nève, F, Exposé des guerres de Tamerlan et de Schah-Rukh d'après la chronique arménienne inédite de Thomas de Medzoph, Brussels 1860) 36-7: the King's Islamism is mentioned as but a ruse; Charter 28 Alex. I. 1440 (Z 247-8) mentions the capture of Bagrat and Anne by the ‘impious Timur’; Chronique géorgienne (ed. Société Asiatique, Paris 1829) 1 puts that event at 1385: it is an eighteenth-century compilation (MGHL 179 n.13). — I wish to express my gratitude to Professor V Minorsky for the kind assistance he has given me in connection with the works of and 'Abd al-Razzaq Google Scholar

5 Hist. Inv. Tim. 871 (no date given); also Chron. géorg. 1/2; Tables 1.625; Dates 380.Google Scholar

6 Chron. géorg. 1/2; Dates 380.Google Scholar

7 Vaxušt, , History of Georgia (tr. Brosset, , HG I 2) 663; Dates 378; Tables 1.625. We have no contemporary data on Queen Helen.Google Scholar

8 The practice of marrying according to Ebenbürtigkeit obtained among the royal, princely, and noble houses of Georgia; cf. Vaxušt, , Geographical Description of Georgia (ed. Brosset, , Descr. géogr. de la Géorgie par le tsarévitch Wakhoucht, St. Petersburg 1842) 6/7, 22/23; Karst, J., Le code géorgien de Vakhtang VI : Commentaire historique-comparatif I (Strasbourg 1935) 246. All the marriages of Georgian kings which can be ascertained have been contracted with other Bagratids, with members of other sovereign houses (Georgian and foreign), or members of non-sovereign princely houses of Georgia. Exceptions to this rule are well known, as, e.g., the marriage of the Crown Prince David, the last King's son (in 1800), which resulted in the raising of the bride's family (Abamelik') to princely status, receiving thus a retroactive legitimation. Cf. Tables. — Georgian nobility was composed of princes and knights; the princely class of the t'avad-s had, together with the Crown, evolved out of the original tribal-dynastic aristocracy of the mamasaxlis-es, sep'ecul-s, or mt'avar-s, as it was variously called. That aristocracy was the ‘first class’ of Iberia according to Strabo, , Geogr. 11.3.6; whereas the knights or aznaur-s represented Strabo's ‘third class’; Karst, , op. cit. 203-4, 218, 237-8, 245-6, 248-9, 251-4; Janašia, S., ‘K Kritike Moiseǎ Xorenskogo,’ Masalebi for the History of Georgia and Caucasia 6 (1937) 471-503. Accordingly, all the members of the princely class were by nature ebenbürtig with the royal house. This equality of birth was explicit, after the fifteenth century, in the case of eleven houses of the t'avad-s (or didebul-t'avad-s). They alone had escaped the weakening division of the princely houses which in that century resulted from the Partition of Georgia; they thus retained their original dynastic status as well as the original title of mt'avar, and their position closely resembled that of the immediate princely vassals of the throne of the Holy Roman Empire; cf. Karst, , op. cit. 228, 169 n.2; II (Strasbourg 1937) 160-2; Brosset, , Introduction lxxix. — The four cases adduced by Brosset, , IV Rapp. 6-25, of intermarriages of the royal house with certain princely families which were regarded as unbecoming (uxvedri, lit. ‘impassable’) cannot impair what has been said, in view of the overwhelming number of other similar marriages. These four alliances were objected to on religious grounds (probably questions of kinship, religious vows, etc.) and resulted in amends being made to the Katholikos of Iberia, presumably in order to obtain, post factum, the necessary dispensations (cf. below, note on George VIII's daughter).Google Scholar

9 Panaretus 21.280 (birth), 29.286-7 (marriage), 53.292 (capture by Timur); Hist. Inv. Tim. 859. The date of Anne's death is unknown. Brosset mistook that of her double sister-in-law, the Empress Eudocia, on March 2, 1395, for her own; HG I 2.653 n.3. Anne's mother was Theodora Cantacuzena, a cousin of the Eastern Roman Emperor John VI; Panaretus 16.277-8, 21.280. Anne's brother, the Emperor Manuel III, married on Oct. 6, 1379 Eudocia, formerly Gulk'an-Xatun Bagrat V's sister; Panaretus 48.289-90, 50.291, 54.293.—For the surname of Grand Comnenus, adopted by the imperial house of Trebizond, cf. Vasiliev, A., ‘The Foundation of the Empire of Trebizond,’ Speculum 11 (1936) 36–7; for Anne's own surname, Panaretus 29.286. — The family tradition of the House of Orbeli(ani)-Barat'ašvili claims that a daughter of Bagrat V, Thamar, married Eles Orbeli-Barat'ašvili. Since, however, this tradition is found only in the charter (gujar) of the Orbelian abbey of Bet'ania, copied in 1704, its acceptance must be withheld until further confirmation. Cf. T'aqaišvili, E., ‘EšǍe Odin iz IstoǍnikov Istorii Gruzii CareviǍa Vaxušta: Gudžar Baratovǐx iz Betanii,’ Zapiski of the East. Div. of the Imp. Russ. Archaeological Society 8.113-28. — For other possible children of Bagrat V cf. below, n.28.Google Scholar

10 Hist. Inv. Tim. 866, 867, 868, 871; Thomas of Mecop' 37.Google Scholar

11 Tables 1.625; cf. below, n.19.Google Scholar

12 Cf. below, chap. II.Google Scholar

13 Hist. Inv. Tim. 866.Google Scholar

14 Ibid. 871; Tables 1.625; Dates 380.Google Scholar

15 Of 1393 (résumé, IV Rapp. 9), Jul. 9, 1399 (resume, IV Rapp. 9), and 1401 or, rather, 1395 (résumé, HG II 2.461). The regnal year of the first act is 24, that of the last 26: the date of the latter is incorrect. The year 1401 is the year 89 of the XIV Paschal Cycle (k'oronikon) of the Georgian Era. The numeral 89, expressed by the letters p.t' of the Military Alphabet (mxedruli), can be easily an error for 83, in Georgian: p.g., since the letter g. in an erroneous, horizontal position looks like the letter t' of that alphabet. And the year 83 G.E. indeed corresponds to A.D. 1395 which is the 26th year after George's elevation to the co-kingship in 1369.Google Scholar

16 Cf. note on Constantine I.Google Scholar

17 History (HG I 2) 677; Dates 380.Google Scholar

18 Charter 4 Jul. 3 Const. I. 1408 (AG 3.461) mentions both his father Bagrat and his brother George; Thomas of Mecop' 37, 75-6; Šaraf ad-dīn, II (Calcutta 1888) 512.Google Scholar

19 Cf. above, n.11. George, already a co-king in 1369, could not have been a son of Anne Comnena, who was born in 1357 and who married in 1367. For David, cf. Panaretus 53.292.Google Scholar

20 Cf. Vasiliev, , The Foundation of the Empire of Trebizond 6; Toumanoff, , ‘On the Relationship Between the Founder of the Empire of Trebizond and the Georgian Queen Thamar,’ Speculum 15 (1940) 299ff.Google Scholar

21 The Hist. of the Mongol Invasions [last part of The Georgian Annals; MGHL 176] (ed. T'aqaišvili, , QM) 719, 720, 734; Tables 1.624. This Comnenian princess must have been a daughter of Manuel I of Trebizond by his third, Georgian wife, Rusudan; she married Demetrius II in 1277, according to Vaxušt (HG I 2) 591 n.1, 579 n.3. Brosset suggested that she must have been Theodora, daughter of the same parents, who later became Empress of Trebizond (1285); ibid. His suggestion, however, is based on an erroneous reading of Panaretus 5.268: mistaking the last five words to refer to whence she fled and not, as is correct, to her parentage. She fled from Trebizond; cf. e.g., Miller, W., Trebizond: the Last Greek Empire (London 1926) 30. Since Theodora is called Manuel's ‘first daughter’, there must have been another, or others, who could have married Demetrius II.—The only other Bagratid named Manuel was the father of Smbat IV (†616/617); cf. Marquart, J., Osteuropäische und ostasiatische Streifzüge (Leipzig 1903) 436; cf. Toumanoff, , ‘The Early Bagratids,’ Le Muséon 62 (1949) 1-2, 38; Tables passim. Google Scholar

22 According to the Georgian sources, David IV (cf. note on Alexander I) married a daughter of the Emperor Michael VIII Palaeologus; Hist. Mong. Inv. 703, 721, 747; Tables 4.642. Pachymeres, on the other hand, reports that David married a natural daughter of Michael's sister-in-law, the wife of his brother, the Despot John, and daughter of the Sebastocrator Constantine Tornices; De Mich. Palaeologo 3.21. Two of David's sons were named Constantine and Michael, presumably for his queen's maternal grandfather Tornices and for the Emperor (who may have been her father after all?). Curiously enough, a recent student of the Palaeologan genealogy mistook David's wife for a daughter of the Despot John himself; Papadopulos, A. Th., Versuch einer Genealogie der Palaiologen, 1259–1453 (Munich 1938) 4–5 and Gen. Table. —The Tornicii may have been descended from the Bagratid princes of Taraun; Adontz, N., ‘Les Taronites à Byzance,’ Byzantion 11 (1936) 21-46.Google Scholar

23 Hist. Mong. Inv. 719, qualifies the Trapezuntine father-in-law of Demetrius II as 'Emperor [lit. ‘King of the Byzantines'] Comnenus who is of the House of the great Constantine’ Google Scholar

24 Hist. Inv. Tim. 880; Šaraf ad-dīn, II 379 (the name of George's brother is not given here).Google Scholar

25 Ibid. 512–4; Hist. Inv. Tim. 882.Google Scholar

26 Dates 380; HG I 2.664, 667 n.2; Allen, W. E. D., A History of the Georgian People (London 1932) 126; Gugushvili, A., ‘The Chronological-Genealogical Table of the Kings of Georgia,’ Georgica 1.1-2 (1936) 126-127.Google Scholar

27 Cf. preceding note.Google Scholar

28 Samarqandī, 'Abd al-Razzaq, Matla' al-sa'dayn (Lahore 1360/1941) 243; Mīr-Xāwand, , cited by Minorsky, V, ‘Tiflis,’ The Encyclopaedia of Islam 4 (1934) 757. There is a certain chronological difficulty connected with the death of King Constantine. According to 'Abd al-Razzaq, Qara-Yusuf crossed the Araxes shortly after Ša'ban 12, 815=Nov. 25, 1412, so that the battle in which the King of Georgia fell must have taken place sometime in the Winter of 1412–1413. Yet Alexander, his son, ascended the throne between Febr. 21 and Mar. 21, 1412, as is proved by his charters (cf. below, n.51); and the charter of the Prince-Chamberlain K'uc'na, written between the accession of Alexander and Dec. 7, 1412 (cf. below, n. 47), leaves no doubt as to Constantine's being deceased then (cf. below, n.45, IV). The information of the contemporary documents should carry more weight, in my estimation, than this work which in its final form dates from 1470. — 'Abd al-Razzaq states that Constantine was slain together with his brothers. It is quite admissible that, in addition to David, the King had other younger brothers unmentioned in the Georgian sources; yet the word ‘brothers’ may be used here in a broader sense of ‘relatives,’ or ‘brothers-in-law’ etc., and this information of a foreign author ought not to be accepted unconditionally without a confirmation from other sources.Google Scholar

29 Cf. below, n.51.Google Scholar

30 Constantine I has been occasionally styled ‘the Second’ by the older historians who counted the tenth-century usurping Abasgian ruler of that name as ‘the First’ But, then, the other usurping Abasgians (another Constantine, two Georges, and Demetrius) have never been taken into account in establishing the ordinal numbers of the legitimate Kings of Georgia, which is as it ought to be; and at present Constantine is referred to as ‘the First’ (cf. e.g., Gugushvili, , op. cit. 126–7. For the Abasgian subjection of Iberia (the nucleus of the Georgian State) in the ninth and the tenth century, until 978, cf. J II 378, 383-4, 385-9.Google Scholar

31 Panaretus 53.292; Thomas of Mecop' 37, 75-6; cf; Tables 1.625 and the notes on Bagrat V and Constantine I.Google Scholar

32 Undated fifteenth-century document addressed by a member of the House of Zedginije to the King of Georgia; ed. AG 4; cf. J IV 93, 88-9.Google Scholar

33 The Bridge Ch, 887, relates a wholly imaginary story of David's flight to the Kakhetian province of Didoet'i and of his being proclaimed there, after the death of George VII, ‘King of Iberia and Kakhetia’; of his reign and that of his son, another King George; of the latter's marriage to Nat'ia, daughter of the Prince-Chamberlain K'uc'na; and of the their son, the great Alexander. Apart from other arguments against the historicity of this tale (cf. note on Alexander I), its very chronology is untenable. Bagrat married in 1367 the ten-year-old Anne. Even if David had been their first son (which he was not), he could not have been born before, let us say, 1369. Alexander I, on the other hand, was born in 1390, when his presumed grandfather was at the most about twenty! Google Scholar

34 The approximate date of the document is determined by the combined mention of George VII as ‘King of Kings’, i.e., King-regnant (cf. below, chap. II), 1395–1405, and of the Katholikos of Iberia, Elias, 1399–1419 (e.g., Janin, R., ‘Géorgie,’ DThC 6 [1924] 1271). — Olympias was neither known to Vaxušt nor mentioned in the Hist. Inv. Tim.; therefore Brosset, who could not well dismiss her charter altogether, placed her in Tables (1.625) only as a doubtful member of Bagrat V's family. And yet the evidence of this document can in no way be vitiated by the silence, not unusual in Georgian chronicles, regarding daughters of the royal house; cf. The Founder of the Emp. of Trebizond and the Georg. Queen Thamar 305ff., 309ff.Google Scholar

35 In the Georgian Church the term ağapi meant a commemmorative meal or distribution of victuals, offered to ecclesiastics, the poor, or passers-by, accompanying the funeral service on the anniversary of the departed. The celebration of agapae (which seems to be a pagan survival, though deriving its name from an early Christian practice) was assured in perpetuity by legacies and foundations; Janin, DThC s.v. ‘Géorgie’ 1264; Javaxišvili, , Sak'art'velos Ekonomiuri Istoria I (Tiflis 1903) 93ff.; Brosset, , Introduction cxiv.Google Scholar

36 Ed. Sas. Sigel. 25 Google Scholar

37 C'agareli, A., ‘Pamǎtniki Gruzinskoy Starinǐ v Svǎtoy Zemlě i na Sinaě,’ 'Pravoslavnǐy Palestinskiy Sbornik 6 (1888) 220.Google Scholar

38 Ceret'eli, G., ‘ArxeologiǍeskaǎ Progulka po Kviril'skomu UšǍel'ǔ,’ Materialǐ po Arxeologii Kavkaza 7 (1898) 93ff. — Ceret'eli refers, erroneously, to Olympias' husband as Kaxaber III.Google Scholar

39 For the House of Orbeliani, cf. note on Alexander I.Google Scholar

40 Kakabaje, S., ‘K Voprosu o Ktitorax Mg'vime v Imeretii,’ Bulletin de l'Institut caucasien d'histoire et d'archéologie 4 (1926) 126–7. — In the fifteenth century, the Duchy of RaǍa passed to the House of Č'xe(t)ije; ibid. ; Vaxušt, , Geogr. Descr. 48-9.Google Scholar

41 Spiski 98. The Kakhetian house of ČavǍavaje may be a branch of it.Google Scholar

42 Cf. Tables 1.625. — In his charter, George VIII speaks of the invasion of Timur ‘at the time of Our grandfather's brother, the King of Kings George’=žamsa papis jmisa Ǎ'uenisa, mep'et'-mep'isa giorgissa. Brosset was perplexed by this document, so definitely militating against Vaxušt's genealogical construction; IV Rapp. 20. The responsibility for the erroneous notion that Alexander I was the son of a King George cannot, however, be laid exclusively to Vaxušt's door. It seems to have been a common eighteenth-century error, found in the Continuations of the Annals and in one of Vaxušt's sources, the charter of the Barat'ads from Bethania Abbey, copied in 1704 (cf. above, n.9), and it appears to have arisen from a certain confusion regarding the filiation of Bagrat VI (q.v.).Google Scholar

43 Cf. e.g., J IV 7-20; Gugushvili, , Chron.-Geneal. Table 126–7.Google Scholar

44 Cf. Allen, , History 126. Brosset was somewhat puzzled by the divergence between Vaxušt, the Bridge Chron. (cf. above, n.33), and the documents; cf. Tables 1.625.Google Scholar

45 This is proved by the combined evidence of the following sources: (I) A fifteenth-century MS of agapae has: ‘May God have mercy on the soul of Alexander, illustrious among all the kings and Sovereign Lord of All the East even to the West; may God have mercy on the soul of his mother, the former Nat'ia, now [nun] Nino’; ed. Z 208. (The Bridge Chron. 887, while giving a wrong name to Alexander's father, concurs with this document in naming his mother, a daughter of the Prince-Chamberlain K'uc'na, Nat'ia; above, n.33).—(II) Charter 20 Sept. 1 Alex. I. 1413 (K 3.7-10; Z 220) whereby Alexander founds an agape for his grandmother Rusa, who had directed his education and had undertaken, after the devastations of Timur, the restoration of the primatial cathedral at Mc'xet'a, in the course of which he died. (J IV 14 supposes that she undertook to bring up Alexander after Queen Nat'ia had taken the veil. But this is unlikely, as the latter would hardly have taken that step before the death of Constantine I, in the winter of 1412; and Alexander ascended the throne, aged twenty-two, immediately thereafter; cf. below.) (The Hist. Alex. I is in agreement with this charter: it mentions Rusa, the Prince-Chamberlain K'uc'na's wife, who brought up her grandson, King Alexander; p. 888.).—(III) Charter 19 Alex. I. 1431 [I] (résumé, IV Rapp. 15) whereby the King institutes an agape for the Prince-Chamberlain (though no name is given).— (IV) Charter of 1412 of K'uc'na himself (Z 209) : in connection with the restoration of the Abbey of Ulumba, he founds agapae for his wife Rusa, King Alexander [then reigning, hence mentioned first], King Constantine, Queen Nat'ia [the latter's wife, else she would not preceed the following]., King George [VII], and Queen Rusudan (erstwhile protectress of the family). Cf. J IV 15-6. It is difficult to see why Javaxišvili should so hesitatingly have admitted what is so unmistakably patent in the light of the above combined evidence; J IV 13-6.Google Scholar

46 In his charter K'uc'na mentions his forefather, of the same praenomen, as having been Prince-Constable of Georgia under Queen Rusudan (1223–1245). But, unfortunately, between the constableship of Zacharias II Mxargrjeli, who died in 1212 (The Histories and Eulogies of the Sovereigns [part of The Georgian Annals: MGHL 175-6] ed. T'aqaišvili, QM 525), and that of his nephew Avak-Sargis III Mxargrjeli, who filled that office from 1233/4 to 1250 (Hist. Mong. Inv. 570-668; for the date of his accession, cf. Yovsepyan, G., Xalbakyank' kam [Valaršapad 1928] 98–100), the incumbents of it are not known to us. The praenomen K'urc'ik (Xurc'ik), born by K'uc'na's grandfather (according to his charter), may possibly suggest that he belonged to the Meschian feudal house of Xurc'ikije or Xurc'ije (for which cf. T'aqaisvili, , ‘ArxeologiǍeskiǎ Ěkskursii etc.,’ Sbornik Materialov for the Description of the Peoples and Localities of Caucasia 35 [1905] 1-60). This problem is treated more fully in the dissertation referred to above, n.3 (pp. 80-83).— A fresco portrait of the Prince-Chamberlain K'uc'na is found on one of the three remaining walls of the church at Nebaxt'evi, near the village of Brili, in the Gori district; M. Xaxanov considers it, and the church, as belonging to the sixteenth century; ‘Ěkspedicii na Kavkaz,’ Materialǐ po Arxeologii Kavkaza 7.65. He may possibly have postdated both the building and the painting.Google Scholar

47 The date of this undated document is determined by the fact that Alexander is mentioned as king-regnant (after Febr. 21/Mar. 1412; cf. nn.45 and 51) and K'uc'na's wife Rusa (†Dec. 7, 1412; cf. n.49) as still living. Agapae might, to be sure, be founded in the lifetime of their eventual beneficiaries.Google Scholar

48 This is implied in K'uc'na's charter.Google Scholar

49 Charter 10 Sept. 8 Alex. I. 1420 (K 3.10-12; Z 226) states that Alexander was aged twenty-two when he undertook the restoration of Mc'xet'a, after the destructions of Timur. But from the charter 20 Sept. 1 Alex. I. 1413 (K 3.7-10; Z 220) we learn that it was his grandmother Rusa who had begun that work and died in the course of it (cf. above, n.45 II). It is clear, then, that Alexander merely continued the work. Rusa appears as still living in K'uc'na's charter, written after the accession of Alexander (Febr. 21/Mar. 21, 1412; cf. n.51); she is mentioned as deceased in the latter's charter of Sept. 20, 1413. Since the agape for Rusa's soul, instituted by that act, was to take place on St. Ambrose's day (Dec. 7, according to the Byzantine calendar), we may assume that she died on Dec. 7, 1412. Being aged twenty-two, when the task of restoration was left to him in 1412, Alexander must have been born in 1390; cf. J IV 10-11. The charter 22 Sept. 7 Alex. I. 1419 (Z 224) states that the King was aged twenty-four when he began the restoration, but the document is a seventeenth-century copy; J IV 11. Brosset knew only this act and concluded that Alexander was born in 1389, since he counted his reign from 1413; IV Rapp. 12, 13.Google Scholar

50 Cf. below, chap. II.Google Scholar

51 J IV 7-10 (on the basis of the analysis of 18 charters, from 20 Sept. 1 Alex. I. 1413 [K 3.10; Z 220] to 29 Alex. I. 1441 [K 3.24]).Google Scholar

52 Hist. Alex. I. 889; following note; cf. J IV 35-6.Google Scholar

53 Charter of Mar. 7, 1446 (résumé, IV Rapp. 21), issued by 'George son of the King of Kings and monk, the blessed [sanatrel , i.e., ‘late'] Alexander’; charter of Aug. 26, 1445 (Z 257), issued by ‘King Demetrius, son of the King of Kings Alexander’; had Alexander died before the issuance of this document, he would have been referred to as ‘late’ in it. Hist. Alex. I. 891 records his death after 1445. Thomas of Mecop‘ 146, claims that the King developed an ulcer of the stomach a year after the supposed murder of his father-in-law, Bešk'en II of Siunik’ (cf. below).Google Scholar

54 Aeneas Sylvius (Pope Pius II), Epistolarum lib. I (Basel 1571) 852: George VIII's ambassador is described as Nicolaus orator Georgii Alexandri magni regis. Google Scholar

55 Thomas of Mecop' 144. Bešk'en fled to the Court of Georgia, having been despoiled of his princedom by the Timurid Šāh-Rux, and received from Alexander the fortress-city of Lori. Thomas would have us believe that Alexander caused, in 1437/8, the poisoning of his father-in-law; ibid. 144-146. Nève suggests this may have been due to his fear of a possible revival of the Orbeliani power; ibid. 153-154 (for the Orbelianis, cf. below). — The Queen's name is revealed by the charter 20 Sept. 1 Alex. I. 1413 (K 3.7-10; Z 220), issued by ‘the King of Kings, the Lord Alexander and the Queen of Queens, the Lady Dulanduxt’ These words figure at the end of the document, its initial part having been lost. It is difficult to see why Javaxišvili should have hesitated to accept this text as an evidence for the praenomen of Alexander's first wife only because Dulanduxt is not expressly so called and because we do not know what the lost opening part contained; J IV 19-20. What else could she have been? Her title, Queen of Queens, shows her to be the wife of a king-regnant (cf. below, chap. II n.27). Of all the female members of a sovereign's family, only his consort and occasionally — but then with a specification — his mother were mentioned in royal diplomata. Since Dulanduxt was not Alexander's mother and since her relationship is not further specified, we must assume that she was his wife. —L. Ališan states, citing Thomas of Mecop', that Alexander was a brother-in-law of Bešk'en, ; Sisakan (Venice 1893) 96 and Stemma ad p. 92. The text of Thomas 144, it is true, is vague enough to admit of the interpretation that Bešken II was a grandson and not (as is correct) a great-grandson of Būrt'ēl II of Siunik', nevertheless Thomas makes it quite clear that Bešk'en II's paternal uncle was Būrt'ēl II's grandson; Ališan 96 n.2.Google Scholar

56 Cf. below, n.65.Google Scholar

57 Hence possibly the unhappy relations with his father-in-law, Bešk'en of Siunik'; hence also, presumably (whether or not Alexander was, in addition, responsible for Bešk'en's death), his expiatory act of entering a monastery; and hence, finally, the official silence of the Georgian chroniclers regarding his first wife. On the analogous case of David II of Georgia's repentance after the repudiation of his first wife and the official silence regarding her, cf. Avalishvili, Z., ‘The Cross from Overseas,’ Georgica 1.2-3 (1936) 3–11.Google Scholar

58 ‘Orbeli’ is the territorial epithet derived from the castle of Orbet'i (Samšvilde), in the Duchy of Samšvilde, in Lower Iberia (cf. Vaxušt, , Geogr. Description 166/167). The older form ‘Orbeli’ was later replaced by ‘Orbeliani’, originally the collective name of the gens. The Armenian branch was called ‘Ωrbēlean’ The Liparitids were also known under the name of ‘Bağuaši’ Google Scholar

59 Adonc' (Adontz), N., Armeniǎ v Epoxu Yustiniana (St. Petersburg 1908) 402–4; T'aqaišvili, E., ‘Georgian Chronology and the Beginnings of Bagratid Rule in Georgia,’ Georgica 1.1 (1935) 20-1. Like the Mamikonids, the Liparitids claimed Chinese imperial descent and the gentilitial title of Čen-Bakur (later Jambakur), i.e., ‘Son of Heaven of China’ (=Pers. bağpūr=Ind. devaputra=Chin. t'ien-tzu; cf. Marquart, , Streifzüge 133-4; Justi, F., Iranisches Namenbuch [Marburg 1895] 240). Cf. Scöld, H., ‘L'origine des Mamiconiens,’ Revue des études arméniennes 5.1 (1925) 131-6; Mlaker, K., ‘Die Herkunft der Mamikonier und der Titel Čenbakur,’ Wiener Zeitschrift für die Kunde des Morgenlandes 39 (1932) 133-45; Justi, , op. cit. 424-5; Adonc', op. cit. chaps. 10, 11; Laurent, J., L'Arménie entre Byzance et l'Islam (Paris 1919) 90; Grousset, R., Histoire de l'Arménie (Paris 1947) 290-1, 641. Cf. Ωrbēlean, Stephen, The History of the Orbelians (ed. Saint-Martin, J., Mémoires historiques et géographiques sur l'Arménie II, Paris 1819); Saint-Martin, J., ‘Dissertation sur la famille des Orbélians,’ Mémories II; Brosset, , Additions 213-5, 257-64, 317-29, 334-9, 346-61; Alisan, , Sisakan 92-8. — Exactly as the Mamikonids were the hereditary High Constables of Armenia (Adonc', 282-3, 370), so the Liparitids were enfeoffed of the office of Prince-Constable of Georgia; and the struggle against the Bagratids, carried on by the former in Armenia and by the latter in Georgia, had the character of a family feud. The policy of the early Liparitids was pro-Armenian (cf. Allen, , History 86) and it was to Armenia that the principal branch of the house retired after the disgrace of 1177 (cf. below). It will be recalled that a Mamikonid prince, Artavazd, later imperial Strategus of the Anatolics, fled to Georgia in 771 and held fiefs there (cf. Grousset, , op. cit. 324) and that there have been other Georgian houses, e.g., the T'umanids, deducing their descent from the Mamikonid dynasty (N. Marr in Xristianskiy Vostok [1913] 2.144-145; Chahnazarian, G., Histoire par Ghévond [Paris 1865] 13 n.2). It is true that Stephen Ωrbēlean, Archbishop of Siunik' (†1304), the historian of his house, alleged that his ancestors had come to Georgia directly from China (at a remote epoch before our era) and omitted all mention of their Armenian past. But then, his sources must have been affected by the same nascent nationalism of the Georgia of the eleventh century (and later) which made the family historian of the Bagratids, Sumbat, omit their Armenian origin and trace that house directly from Judaea to Georgia (cf. Streifzüge 391-403; MGHL 154-5; The Early Bagratids 22-3). Actually, the first historical Liparitids appear in Georgia in the ninth century — the epoch of the decline of the Mamikonids in Armenia.Google Scholar

60 Ca. 876, Liparit I ‘took possession of the lands of T'rialet'i and reared the castle of Klde-Karni’; The Chronicle of Iberia [part of The Georgian Annals; cf. MGHL 173-174] (ed. QM 224; ed. QA: QauxǍ'išvili, S., K'art'lis-C'xovreba: Ana Dedop'liseuli Nusxa, Tiflis 1942) 164. These fiefs lay in Lower Iberia; Vaxušt, , Geogr. Description 190/191.Google Scholar

61 The House of Orbeliani ranked, in the seventeenth and eighteenth centuries, as fourth among the ‘undivided’ princes of Iberia (mt'avar-s, cf. above, n.8) and was enfeoffed of the princedom of South Sabarat'iano, comprising the Duchies of Samšvilde and Xunani and the offices of Constable of the Van (Sabarat'iano) and (jointly with the Houses of Muxrani and P'anaskerteli) of Prince-Master of the Palace of Georgia; Vaxtang VI, King of Georgia, Code of Laws §35 (ed. Karst, J., Code géorgien du roi Vakhtang VI, Strasbourg 1934); Vaxušt, , Geogr. Description 40/41, 46/47; Heraclius II, King of Georgia, A Short Description of the Princely and Noble Houses of Georgia (cited by Xaxanov, M., Gruzinskie Dvorǎnskie Aktǐ i Rodoslovnǐǎ Rospisi, Moscow 1893) vi; Aktǐ 2 (Tiflis 1867) 39-43 [List of the last occupants of the great Crown offices of Georgia]; Karst, , Code géorgien: Comment. I 228ff.; V Ivanenko, N., Graždanskoe Upravlenie Zakavkaz'em (vol. 12 of Utverždenie Russkago VladǐǍestva na Kabkazě ed. Potto, V A., St. Petersburg 1901) 9; Peter Dolgorukov, Pr., Rossiyskaǎ Rodoslovnaǎ Kniga III (St. Petersburg 1856) 475-6; II (1855) 62-3; Spiski 36-7, 67-8.Google Scholar

62 Thus in 1045–1046 Liparit IV, Duke of T'rialet'i, Argvet'i, Lower and Upper Iberia, Prince-Constable of Georgia, and Magister and general of the Roman Empire of the East, forced King Bagrat IV of Georgia, through the mediation of the Emperor Constantine IX Monomachus, to cede to himself one half of the Realm (south of the Kur); Chron. Iber. (ed. QauxǍ'išvili=QA, ) 185-91=QM 260-7; Cedrenus (PG 122) 304-5; cf. Allen, , History 85-94, 102-3. (On the pages of CMH 4 [1923] 166, the powerful Liparit IV has become ‘Liparid, King of Georgia’, possibly via Ibn al-Aθīr's ‘ King of the Apxāz’, cf. Saint-Martin, , Mémoires II 214-6.) Google Scholar

63 It is difficult to understand how S. Runciman, who, in chap. 8 of The Emperor Romanus Lecapenus and His Reign (Cambridge 1929), has presented with a great penetration the picture of the Caucasian frontier of the Eastern Empire, could have confused the old house of Siunik' (cf. Grousset, , Hist. de l'Arm. 291, 245; Laurent, , L'Arménie 275-6; Justi, , Iran. Namenbuch 426-7; Ališan, , Sisakan 6-19) with the second Siunian dynasty, the Orbelians, established there only after 1177; cf. Runciman, , op. cit. 153, 160ff. and ‘The Geneal. Trees’ between pp. 262 and 263, referring to the first dynasty as ‘Orbelians of Siounia’ Google Scholar

64 Brosset, , Additions 362–7; Ališan, , Sisakan 92-8, Stemma ad p. 92; Justi, , Iran. Namenbuch 446.— The historian Stephen Ωrbēlean, Archbishop of Siunik' (†1304), was Bešk'en I's great-great-granduncle.Google Scholar

65 Vaxtang IV and Demetrius III (qq.v.) appear as early as 1413 in the charters of their father, King Alexander; this is likewise the year in which Queen Dulanduxt is mentioned n a document (cf. n.55). Queen Thamar appears first in the charter 10 Sept. 8 Alex. I. 420 (K 3.10-12; Z 226) and possibly in the acephalous charter 29 Sept. 5 Alex. I. 1417 Z 223-4); J IV 19. It is in 1417 that Alexander's sons George VIII and David (qq.v.) begin to figure in documents, together with Thamar. Since the two sons by Dulanduxt appear already in 1413, Alexander's first marriage must have taken place c. 1410/1411; and since the first two of Thamar's sons appear in 1417, Alexander must have married for ***he second time c. 1414/1415. According to Vaxušt, that event occurred in 1414; History ***HG II 1) 249; cf. Tables 625; J IV 19-20.Google Scholar

66 Hist. Mong. Inv. 571 calls him ‘son of Ortul’, i.e., of Tūgril(-Šāh). Upon the division ***of the Iconian Realm by his father II, in 1192, Tūğril-Šāh received Abulustayn appanage; but he exchanged it, in 1200–1201, for Erzerum; cf. Zetterstéen, K. V, Tūghril-***hāh b. Kilidj-Arslān,' The Encyclopaedia of Islam Suppl. 5 (1938) 251. A vassal of Georgia, ***Tūğril-Šāh, caused, according to Ibn al-Aθīr, 12.270, his son to embrace Christianity in ***order to marry Queen Rusudan; ibid.; Minorsky. Encycl. Islam s.v. ‘Tiflis’ 756. Rukn ad***in-Jahān-Šāh, who succeeded his father Tūğril-Šāh in Erzerum in 1225, must be distinct ***from Rusudan's prince consort, whose name has not been recorded in the available sources, ***nless we so construe the passage of Fida, Abū'l, ad ann. H 620, as to infer that he was ***named Muğīθ ad-dīn like his father; cf. Brosset, HG I 2.501 n.2.Google Scholar

67 Cf. Allen, , History chaps. 9, 10.Google Scholar

68 For the Imeretian Seljukids and the above stemma, cf. Hist. Mong. Inv.; Vaxu History (HG I 2) 647, 668, (II 1) 245-9; Dates; Tables 4.642. 1.624-5; Gugushvili Chron.-Geneal. Table 123, 125, 128; de Zambaur, E., Manuel de généalogie et de chronologie pour l'histoire de l'Islam (Hanover 1927) 143–4; and above, nn. 22, 66. — Rusudan's daugh-***ter, Thamar married, in 1236, her second cousin, Giyāθ ad-dīn-Kayxusraw II, Sultan o Iconium, who chose the emblem of the lion and the sun to commemorate the occasion Hist. Mong. Inv. 571; art. ‘Kaikhusraw II,’ The Encyclopaedia of Islam 2 (1927) 639-40*** Google Scholar

69 Cf. MGHL 157 n.44. Georgian knows no distinction of genders. Cf. below, II n.27.Google Scholar

70 For Thamar the Great, cf. Allen, , History 103–8.Google Scholar

71 Some scholars at first believed the charter to have been actually issued by Thamar the Great and proposed the emendation of the date 121 of the XIV Paschal Cycle of G.E (1433) to 421 of the XIIIth (1201); T'aqaišvili, AG 2.371-372.Google Scholar

72 Charter 4 Jul. 3 Const. I. 1408 (AG 3.461); charters of Alexander I of 1414 (K 2.3-4) Sept. 22, 1419 (Z 224), Sept. 10, 1420 (K 3.10-12; Z 226), Jan. 6, 1424 (Z 228). Bridge Ch 887 mentions them as Alexander's brothers. Not mentioned by Vaxušt, they were bu doubtfully admitted by Brosset in Tables 1.625. — The Matla' al-sa'dayn 432 records th arrival at the court of Šāh-Rux of Alexander's representative and of the sons of Co*** stantine, at the end of 1420.Google Scholar

73 Cf. below, chap. II.Google Scholar

74 Chalcocondyles, Laonicus, De reb. turc. 9 (PG 159.456).Google Scholar

75 For John IV and his dates, cf. Miller, , Trebizond 81–96; Vasiliev, A., ‘Pero Tafur, *** Spanish Traveler of the XVth Century and His Visit to Constantinople, Trebizond, and Italy,’ Byzantion 7 (1932) 99-101. — Manuel III of Trebizond was the brother of Anne, Bagrat V's queen, and the husband of Bagrat's sister Gulk'an-Eudocia; cf. above, note on Bagrat V Google Scholar

76 Tafur, Pero, Andanças e viajes de Pero Tafur por diversas partes del mundo avidas*** (1435–1439) (ed. de la Espada, Marcos, Collección de libros españoles raros o curiosos 8, Madrid 1874) 159–60.Google Scholar

77 His father's charters from 20 Sept. 1 Alex. I. 1413 (K 3.7-10; Z 220) to 1442 (Z 253), where Vaxtang always precedes Demetrius; Hist. Alex. I. 889–91; Tables 1.625.Google Scholar

78 Cf. below, chap. II.Google Scholar

79 Hist. Alex. I. 890–1; Tables 1.625; cf. J IV 38-9; Chron. géorg. 2/3 puts his death at 1447.Google Scholar

80 Hist. Alex. I. 889; Chron. géorg. 2/3 (‘Sil-Xat'un’); Tables 1 625. Both the Hist. Alex. I. and the Chron. géorg. call her ‘daughter of the P'anaskertel and sister of T'aq*** P'anaskerteli.’ T'aqa appears in the available documents — late in life, it should seem—in the charter of Bagrat III of Imeretia of 1511 (Sas. Sigel. 28), together with his elder brothers Zacharias and C'ic'i. Zacharias P'anaskerteli, on the other hand, is mentioned in Constantine II's letters-patent of 1467 (Sas. Sigel. 30-31), together with his elder brothers*** and his father, Prince Zaza. Sitixat'un, who married in 1442, must have been the oldest*** of the eight known children of Zaza I.Google Scholar

81 P'anaskert, the patrimonial castle of the family, was situated in the Duchy of Tao, on the river of the same name, a tributary of the Čoroxi; Vaxušt, , Geogr. Description 118/119.Google Scholar

82 Hist. and Eul. of the Sov. 445.Google Scholar

83 Dolgorukov, , Rossiyskaǎ Rodoslovnaǎ Kniga II 45–6. P'anaskerteli, T'aqa, Duke of Tao, defeated the Turkomans invading Georgia c. 1302, at T'ort'omi Castle; Hist. Mong.*** Inv. 767.Google Scholar

84 Invested with the princedom of Sac'ic'iano — comprising Xveduret'i, Kareli and other seigniories, and the hereditary offices of Constable of the Royal Banner (Tiflis) and (jointly with the Houses of Muxrani and Orbeliani) of Prince-Master of the Palace of Georgia — the House of P'anaskerteli-C'ic'išvili ranked as fifth among the six ‘undivided’ princes of Iberia (mt'avar-s, cf above, n.8). The division of Sac'ic'iano between the two lines of the house, into Upper and Lower, entailed the loss of its dynastic status in the seventeenth century. Cf. Vaxtang VI, Code §35; Vaxušt, , Geogr. Description 40/41, 46/47, 202/203; II Heraclius, Short Description v; Aktǐ 2.39; Karst, , Comment. I 228; Ivanenko, , Graždanskoe Upravlenie Zakavakaz'em 9; Dolgorukov, , Ross. Rod. Kniga II 45-6, 62-63.—The second surname, C'ic'išvili (and the name Sac'ic'iano), must have been derived from an earlier member of the house called C'ic'i, and not, as has been supposed, from the Queen's brother (Dolgorukov II 45 calls him her nephew), because this patronymic is found already under Alexander I (1412–1442): David C'ic'išvili figures among the witnesses of that King's undated charter (AG 2.28). — Upon the Russian annexation of Georgia, the House of P'anaskerteli-C'ic'išvili was received into the princely nobility of the Empire under the name of Cicianov; Spiski 91-3.Google Scholar

85 Dates 381.Google Scholar

86 His father's charters from 20 Sept. 1 Alex. I. 1413 (K 3.7-10; Z 220) to 1442 (Z 253), where Demetrius always follows Vaxtang; his own charter of Aug. 26, 1445, (Z 257); Hist. Alex. I. 890-1; cf. Tables 1.625; cf. J IV 16ff.Google Scholar

87 Cf. Tables 1.625.Google Scholar

88 Cf. below, chap. II.Google Scholar

89 Vaxušt, , History (HG II 1) 145; Dates 381.Google Scholar

90 Hist. Alex. I. 890: 'And when [Alexander] set up [his] first son Vaxtang in Iberia and Imeretia, together with him he established his [=the latter's] younger brother Demetrius; and he set up in Kakhetia, in the year 1445, as King, Demetrius‘ younger brother George.’ The word mast'an (‘together with him’) precludes an interpretation to the effect that Demetrius was in Imeretia while Vaxtang was in Iberia, and justifies the comma after ***imeret's (‘in Imeretia’). Brosset, nevertheless, translated this passage as follows: ‘Lorsque le roi Alexandré installa son fils Wakhtang dans le Karthli, il plaça en même temps sur le trône d'Iméréth Dimitri, son second fils, et en 1445 George, frère cadet de ce dernier en Cakheth’; HG I 2.682. Javaxišvili seems inclined to accept the passage in this sense; J IV 36. This interpretation was caused, no doubt, by the passage of the text which mentions the appanaging of Alexander's third son in Kakhetia, and by the inability of so many scholars to accept the fact of co-optation (cf. below, chap. II). This text must be regarded as having contributed to the tradition that Alexander I divided his realm among his sons (cf. ibid.). — Apart from the fact that the revolts of the Seljukids served to accentuate the renascent separatism of Western Georgia, the title of the Georgian sovereigns had been, since the eleventh century, ‘of Abasgia (=Imeretia) and Iberia’ (cf. below, chap. II, n.28). — For the date 1445, given in the above text for George's installation in Kakhetia, cf. note on George VIII and below, chap. II, n.46.Google Scholar

91 Cf. below, note on George VIII.Google Scholar

92 In 1453; Hist. Alex. I. 891; in 1452: Vaxušt, , Dates 381. J IV 37 accepts the latter date. Demetrius was killed by a horse while hunting; Dates 381. The former date seems preferable because it brings the King's death closer to the date of Bagrat VI's revolt in Imeretia; cf. below, chap. II.Google Scholar

93 Charters of their son, Constantin II, of 1466 and 1475 (AG 2.13 [for the date cf J IV 92-3] and AG 2.31); Hist. Alex. I. 891. Brosset attempted to identify Queen Gulašar with a Queen Gulk'an, whose death is reported sub anno 1471 in the Chron. géorg. 3/4; HG I 2.684 n.5. He was not aware of the existence of the charter of 1475. Queen Gulk'an must be a different person, perhaps the consort of one of King Alexander's brothers (qq.v.).Google Scholar

94 His father's charters from 29 Sept. 5 Alex. I. 1417 (Z 223-4) to 1442 (Z 253); his own charter of Mar. 7, 1446 (résumé, IV Rapp. 21); Hist. Alex. I. 890-1; cf. Tables 1.625; cf. J IV 16ff. 20, 40.Google Scholar

95 Cf. below, chap. II.Google Scholar

96 Cf. note on Bagrat VI, n.127.Google Scholar

97 Phrantzes; cf. below, n.198.Google Scholar

98 J. IV 41-2 (on the basis of nine charters, from 3 Geo. VIII. 1448 [K 3 29; Z 260] to 2 Sept. 17 Geo. VIII. [AG 2.12]).Google Scholar

99 Cf. note on Bagrat VI. — The Western Dukes, who subsequently became sovereign princes, were those of Abkhazia (of the House of Šarvašije), Guria (of the House of Dadian*** Gurieli), Meschia (of the House of Jaqeli), Mingrelia (of the House of Dadian), and Suania (of the House of Gelovani); cf. Gugushvili, , Chron.-Geneal. Table 132–3; Allen, , History 135-7. It is difficult to agree with Javaxišvili that, at the end of the fifteenth century, Abkhazia and Guria formed parts of Mingrelia; J IV 110, 162. Guria was a fief of the secundogeniture of the Dadianis, separate from Mingrelia, as early as 1352 (Bak'raje, D., ArxeologiǍeskoe Putešestvie po Gurii i AdǍarě [St. Petersburg 1872] 335; a [also: paid his respects to the Emperor of Trebizond in 1372; Panaretus 44.228). After Mamia I Gurieli, younger son of Liparit I of Mingrelia (mentioned in George VIII's charter of 1463 [AG 3.462]), the Dukes and, then, Princes of Guria formed one continuous dynasty down to the Russian annexation of 1829. Mamia, by the way, was a son-in-law of the Emperor David of Trebizond; to him the Empress Helen Cantacuzena fled after the disaster of 1462 (Chalcocondyles, , De reb. turc. 9.488). Furthermore, among the Christian princes of the East ready to take up armes in the proposed crusade of Pope Pius II, in 1459–1460, we find not only Bendian (Bendias) rex Mingreliae (Mangreliae) and Gorgora dux Georgianae, but also Rabia (Fabia) dux Anocasiae (Anogosiae) and Mania marchio Goriae (Sylvius, Aeneas, Epist. 849-50; Wadding, Lucas, Annales Minorum 13 [Rome 1735] 153, 60; cf. note on Bagrat VI). Bendian is, of course, Bediani — a territorial epithet of the Dadianis, derived from the province of Bedia, occasionally used as a praenomen (cf. Hist. Eul. Sov. 411; Hist. Mong. Inv. 583, 605, 656; VII Rapp. [Brosset, , Rapports, 1re livraison 1849] 42-3, 44-5). In 1459 this term designated Liparit I (†1470), but when Barbaro and Contarini used it of the Prince of Mingrelia, they could refer only to his elder son and successor Šamandavle I ( Barbaro, Iosafa, Viaggio alla Tana 10 [ed. Ramusio, , Secondo Volume delle Navigationi et Viaggi, Venice 1559, 96]; Contarini, Ambrogio, Viaggio 2, 5 [ibid. 114, 115, 119]). Anocasia is a further corruption of Auogasia, used by Contarini for Abkhazia (this appears from the context; op. cit. 6.119. Cf. Avegie, Anegie, for Abkhazia: Intinéraire français ed. La société de l'Orient latin, 1882, index i.V.). Gorgora dux Georgianae is Qvarqvare III (also styled ‘the Second’), Duke of Meschia and Atabeg of Georgia (†1466). When Barbaro (Viaggio nella Persia 27, 28 [ed. Ramusio, 109, 110]) and Contarini (op. cit. 6.120) speak of Gorgora, they must refer to his successor Baadur I (1466-1475). Chalcocondyles (op. cit. 9.460) refers to the princedom of Meschia as and to the Dadiani fiefs, somewhat confusedly, as [Dioscurias-Suxum] Google Scholar

100 Cf. J IV 77-101.Google Scholar

101 Cf. ibid. 97.Google Scholar

102 It is in this sense that one must interpret Vaxušt's statements regarding the return of George to Eastern Georgia, replacing Bagrat there in 1466, and the conclusion of peace between them in order to facilitate for the latter the subdual of Constantine, son of Demetrius (III); History (HG I 2) 687, (II 1) 250. Cf. J IV 94-7; Kakabaje, , Sas. Sigel. 31.Google Scholar

103 Contin. II. 894.Google Scholar

104 The charters of Jan. 23 and 24, 1479 (Z 300-1; Z 301).Google Scholar

105 Hist. Alex. I. 890; cf. note on Thamar-Daria and Bagrat VI.Google Scholar

106 So J IV 43-4.Google Scholar

107 Cf. above, n.105.Google Scholar

108 Thus, George's own son Alexander of Kakhetia (q.v.) married a princess of the House of Irubak'ije who was called both Anne and T'inat'in; David VIII of Georgia, son of Constantine II and great-nephew of George, married a princess of the House of Orbeliani who bore three names; Thamar, and Miraingul; Tables 1.626; cf. IV Rapp. 25. King Rostom, David VIII's great-grandson, married a princess of the House of Abašije who was named Catherine (K'et'evan) and Gulduxtar; Tables, loc. cit. This custom had, of course, nothing in common with assuming a new name upon becoming a religious.Google Scholar

109 Dates 382.Google Scholar

110 History (HG II 1) 148.Google Scholar

111 Ibid. 144–8, 11.Google Scholar

112 Cf. above, nn.33 and (for the source) 3. It is rather interesting to note Javaxišvili's unwilling admission of Vaxušt's error; J IV 93-7.Google Scholar

113 Cf. n.111.Google Scholar

114 Vaxušt, the discontented illegitimate scion of the eldest, Iberian line of the Bagratid dynasty (descended from Constantine II, q.v.) — which claimed the whole of once-united Georgia, but was deprived in the mid-eighteenth century of even its Iberian share by the Kings of Kakhetia (descended from Alexander I of Kakhetia, , q.v.) — must not have been loath to cast doubts on the Bagratid origin of the more successful, and rival, Kakhetian and Imeretian lines of the dynasty (cf. also note on Bagrat VI).Google Scholar

115 J IV 94-7, 139; Kakabaje, S., Sas. Sigel. 31; idem, Sak'art'velos Mokle Istoria: Axali Saukuneebis Epok'a (Tiflis 1920) 4; Gugushvili, , Chron.-Geneal. Table 129-30; but cf. Allen, , History 140.Google Scholar

116 His father's charters from 29 Sept. 5 Alex. I. 1417 (Z 223-4), where he follows George (VIII), to 1439 (résumé, IV Rapp. 16); Hist. Alex. I. 891; cf. Tables 1.625; cf. J IV 17-8.Google Scholar

117 It will be recalled that in the Churches of the Byzantine Rite, Catholic as well as non-Catholic, the episcopate is recruited from the regular clery.Google Scholar

118 Hist. Alex. I. 888-91 records the Katholikoi Theodore and Šio and, upon the latter's death, c. 1447, the Katholikate of David, son of Alexander I. Whatever David's official position, it is obvious that in view of his age, the situation as presented by this source is substantially exact. It is interesting to note the introduction into the Georgian Church of what amounted to an imitation of collegial sovereignty. The attempt to explain away the collegial Katholikate of David, Theodore and Šio by suggesting that, while David was of Iberia, the others were of Abasgia (J IV 18 n.1) is inadequate. Alexander I's charters of 1431, 1433 [III], 1434, 1440, and 1441, dealing with the last-named prelates, definitely indicate that they belonged to Eastern Georgia, i.e., Iberia. The katholikal list of Abasgia is far from complete, especially prior to the end of the fifteenth century; Tamarati, M., L'Eglise géorgienne des orgines jusqu'à nos jours (Rome 1910) 409. — Some historians have fancied three different persons in our one David, according to the three chronological groups of documents, i.e., David II or III (1426–1428), David III or IV (1435–1439), and David IV or V (1447–1457), Alexander I's son; Tamarati, , op. cit. 409; Natroev, A., Mcxet i Ego Sobor Sveti-Cxoveli (Tiflis 1900) 343-55; Janin, DThC s.v. ‘Géorgie’ 1271. (The difference in the numerals after the above katholikal names is due to the fact that some historians, like Tamarati 408, omit the Katholikos David of 859-861 while others count him; thus Natroev 454.) An acceptance of this division, on the ground that only the last group of documents could refer to the King's son (born c. 1417), is precluded by the very first document of the first group, the charter of 1426, wherein King Alexander mentions among his children ‘Our son, the Katholikos of Iberia, David’ Thus there was but one Katholikos, David III, in the years 1426–1457.Google Scholar

119 His father's charters from 21 Jan. 16 Alex. I. 1428 (Z 231) to 1442 (Z 253); cf. J IV 18. The last-mentioned charter is a seventeenth-century copy; before this, Zaal is mentioned in that of 1433 [I] (K 3.15-16; Z 239-40); cf. ibid. n.2.Google Scholar

120 Cf. below, chap. II.Google Scholar

121 Cf. note on Bagrat VI.Google Scholar

122 Hist. Alex. I. 890; cf. notes on Bagrat VI and George VIII.Google Scholar

123 Contin. II. 950.Google Scholar

124 Cf. J IV 80-4; Gugushvili, , Chron.-Geneal. Table 129. Others repeat the version of Vaxušt, more of which below.Google Scholar

125 Barbaro, , Viaggio alla Tana 14.98; Viaggio nella Persia 27.109, 28.110; Contarini, 2.115, 5.119, 6.120.Google Scholar

126 Contin. II. 895.Google Scholar

127 Sylvius, Aeneas, Epist. 850–1; George VIII's letter is dated Nov. 5. Nor is Bagrat mentioned in the similar letter of David, Emperor of Trebizond to the Duke of Burgundy, of Apr. 22, 1459; ibid. 849. — For the other Georgian princes about to join the crusade, cf. above, n.99.Google Scholar

128 Pancratius [rex] Iberorum, qui nunc Georgiani vocantur; (speech of the Ambassadors of the Eastern allies) Annales Minorum 13.153 (Pancratius, are the traditional Graeco-Latin renderings of Bag[a]rat ; cf. Justi, , Iran, Namenbuch 57). The rebellious Bagrat, as is clear from his charters, used the royal style of the Kings of Georgia — a style which expressed, in its central formula, the perennial dichotomy of the realm composed of Western and Eastern Georgia: ‘King of Abasgia and Iberia’=ap'xazt'a da k'art'velt'a mep'e (Abasgorum et Iberorum rex); cf. below, chap. II n.28. It is curious that Bagrat, who revolted in Abasgia (Imeretia) should in the West have been called rex Iberorum or Georgianiae (cf. following note), whereas George VIII, who was reduced by that revolt to Eastern Georgia alone, should have been referred to as rex Persarum, which, in Javaxisvili's opinion, is a mutilation of rex Abasgorum: J IV 67-9. Wadding describes George as rex Persarum, et maioris Armeniae, et minoris Iberiae, cui pater fuit Alexander; Annales Minorum 13.153. It is interesting that, whereas the letters mentioned only George and omitted Bagrat, the Ambassadors referred only to Bagrat and left out George. The claims of the two were mutually exclusive.Google Scholar

129 This is clear from the speech of the Ambassadors from the Eastern Princes before the Pontiff, in 1460; Annales Minorum 13.153. Indeed, in his rescript to Lewis of Bologna, of Oct. 4, 1458, Pius II mentions both Bagrat (rex Georgianiae) and George (rex Persarum); ibid. 60. This is the only time the two rivals are mentioned together.Google Scholar

130 Tamarati, , L'Eglise géorgienne 458–9.Google Scholar

131 Cf. J IV 77-9.Google Scholar

132 Zedg. Petition (cf. above, n.32) 22; cf. J IV 89. George VIII's charter of 1463 (AG 3.462) was granted at Kutais.Google Scholar

133 Zedg. Petition 22–3; cf. J IV 89.Google Scholar

134 Fragment of an undated fifteenth-century MS, ed. Z 301; cf. J IV 89-90.Google Scholar

135 Zedg. Petition 24; the charter of co-King Constantine II of 1466 (AG 2.13; for the date, cf. J IV 92-3) confirms the story of his uncle's capture; cf. J IV 91-3.Google Scholar

136 Zedg. Petition 24; cf. J IV 91-2, 93; above, note on David, son of Bagrat V In confirming, in 1472, a diploma of George VIII of 1460, Bagrat states that he ‘took possession’ (davipqarit‘) of the 'Royalty of the Two Thrones’ (on the royal style, cf. above n.128; below, chap. II n.28).Google Scholar

137 Hist. Alex. I. 891–2.Google Scholar

138 Contin. II. 892.Google Scholar

139 For the Western Dukes, cf. above, n.99.Google Scholar

140 History (HG I 2) 685–7, (II 1) 249-51.Google Scholar

141 The Annals of the Book of Hours No. 2; The Č'xeije Chronicle of the Hymnal No. 6 (ed. Z) 281; both record only the battle of Č'ixori; The Chonicle of the Ikort'a Horologion (ed. Kakabaje, , K'ronika Ikort'is No. 6 Žamn-Gulanisa, Tiflis 1911) 4; as well as Chron. géorg. 2-3/4. In the last-named source, the phrase relating to the capture: mep'e daiǍira quarqvarem tqu<e>t' is incorrectly rendered by Brosset as le roi fit prisonnier Qwarqwaré. Qvarqvare is the agent, as is indicated by the dative-pronominal case ending -m(an). t' is incorrectly rendered by Brosset as le roi fit prisonnier Qwarqwaré. Qvarqvare is the agent, as is indicated by the dative-pronominal case ending -m(an).' href=https://scholar.google.com/scholar?q=The+Annals+of+the+Book+of+Hours+No.+2;+The+Č'xeije+Chronicle+of+the+Hymnal+No.+6+(ed.+Z)+281;+both+record+only+the+battle+of+Č'ixori;+The+Chonicle+of+the+Ikort'a+Horologion+(ed.+Kakabaje,+,+K'ronika+Ikort'is+No.+6+Žamn-Gulanisa,+Tiflis+1911)+4;+as+well+as+Chron.+géorg.+2-3/4.+In+the+last-named+source,+the+phrase+relating+to+the+capture:+mep'e+daiǍira+quarqvarem+tqut'+is+incorrectly+rendered+by+Brosset+as+le+roi+fit+prisonnier+Qwarqwaré.+Qvarqvare+is+the+agent,+as+is+indicated+by+the+dative-pronominal+case+ending+-m(an).>Google Scholar

142 History, loc. cit. Google Scholar

143 Hist. Alex. I. 891; Contin. II. 892-3.Google Scholar

144 The charter of 1466 (Z 287-8) founds an agape for the repose of the soul patronisa mamisa Ǎ'uenisa giorgisa. Google Scholar

145 For the Bagratid claims of Davidic origin, of which the above nomina gentilia were an expression, cf. MGHL 154-6; below, chap. II n.28; The Early Bagratids 22-3.Google Scholar

146 Cf. J IV 83-4, where the question is left undecided.Google Scholar

147 For the dynastic legitimism of the Georgians, cf. Vaxušt, , Geogr. Description 10/11; Vaxtang, VI, Code §152; Karst, , Comment. I 201, II (Strasbourg 1937) 250.Google Scholar

148 Contarini, , 5.119. Cf. J IV 105.Google Scholar

149 Javaxišvili makes much of it; J IV 80-4; cf. below, chap. II. — ‘Prince’ is used here for rendering patron, an honorific epithet of a very restricted kind, applied to sovereigns, members of their families, and to the highest personages in the feudal-administrative and ecclesiastical hierarchies. The word corresponded to ‘Prince’ in its inchoate state — an epithet rather than a fixed title as it has been preserved, e.g., in the ceremonial nomenclature of Great Britain, where it is used (‘Puissant Prince’ etc.) of the sovereign, his family, as well as the three upper degrees of the peerage. It also corresponded to the medieval patronus, from which it was derived, in its feudal sense of suzerain; in this sense it had a wider application.Google Scholar

150 Cf. below, chap. II.Google Scholar

151 Cf. note on George and Bagrat and n.72 (charters from 1414 to 1424).Google Scholar

152 Hist. Alex. I. 891 : (lines 23-4) mep'is alek'sandres jma bagrat, (line 27) jma alek'sandre mep'isa, da je mep'is giorgisa bagrat; Contin. II 892: je giorgisa bagrat, jma mep'isa alek'sandresi. King Alexander is called son of ‘King George’ also in the Bridge Chr. 887 (cf. above, n.33).Google Scholar

153 Bagrat, je giorgisa alek'sandre mep'isa jmisa. Google Scholar

154 Cf. note on Alexander; above, nn.152, 33.Google Scholar

155 For these eighteenth-century compilations, cf. MGHL 159-61, 179, n.13; cf. above, nn.3, 4; below, n.202.Google Scholar

156 For the Seljukid house of Imeretia, cf. note on Alexander I.Google Scholar

157 Cf. note on Demetrius III and n.90.Google Scholar

158 The ‘King George’ of the Chron. géorg. (‘of Imeretia’ in the Bridge Chr.), who was killed in 1392, may disguise George VII; or this may constitute the only reference, outside Vaxušt, to the Imeretian duke of that name (cf. note on Alexander I: Stemma of the Seljukids). Č'alağni, unidentifiable in Georgian historical geography (Cf. Brosset, , Chron. géorg. 130), may very well have been the place north of the Araxes where Constantine I perished (q.v.).Google Scholar

159 Contin. II. 905: she is mentioned as the mother of Alexander I of Kakhetia (qv.); cf. note on George VIII.Google Scholar

160 On the Commission of King Vaxtang VI for the redaction of the Annals, cf. MGHL 163-6, 179–81.Google Scholar

161 Cf. Tables 4; Allen, , History 135.Google Scholar

162 J IV 78-105. But the author reaches no conclusion of his own.Google Scholar

163 Cf. below, chap. II.Google Scholar

164 In this document, the name of Constantine's parents, Demetrius III and Gulašar, are placed in such a way as to leave no doubt that they are not Bagrat's parents: ‘son of the King of Kings Demetrius Constantine, and Our mother of mothers, the Lady Gulašar [bestow] this firm, irrevocable, and indisputable [instrument] and We King of Kings Bagrat etc.’ Google Scholar

165 Even in his own day, Bagrat's act of courtesy towards his co-king caused some confusion for the literal-minded. The copyist of George VIII's charter of 1460, which Bagrat VI confirmed in 1472 (AG 1.2; Z 278-9), referred to the former as the latter's ‘uncle’ Google Scholar

166 History (HG 1 1) 685-7, (II 1) 249–51.Google Scholar

167 Bagrat was called ‘the Second’ as a successor to the Seljukid Imeretian duke, Bagrat I; cf. note on Alexander I. — Whether Bagrat VI had been Duke of Imeretia before his revolt, as is asserted by Vaxušt, , op. cit. 685, 249, or whether that assertion be merely a consequence of his imputed Seljukid origin, is beyond proof. — The genealogical arrangement proposed by Kakabaje, S., in Genealogia Bagrat IV Ap'x. da K'art' Mep'isa (Tiflis 1912) is nothing but a rearrangement of the Seljukid stemma as found in Vaxušt : Bagrat I the Little is made identical with King Alexander's brother of the same name; and Bagrat VI, son of the latter's son named George; cf. J IV 81-2. — A criticism of the sources on the basis of a scale of values, so as to give precedence to the primary sources over the secondary ones, would have spared Kakabaje his speculations; and realization of the fact that the two Continuations depend on one of the sources of the Chron. géorg., and contain unwarranted insertions, would have prevented Javaxišvili from the attempt to reconcile the divergencies of the chronicles regarding the battle of Č'ixori and the capture of George VIII; cf. J IV 85-6, 91-2.Google Scholar

168 The Georgian Monastery on Mount Calvary was ransomed from the Muslims at a high price by King Bagrat, son of a brother of Alexander; it was later restored by Leo I, King of Kakhetia, in 1535, according to Timothy Gabašvili who visited the Holy Land in the eighteenth century; Brosset, , Additions 201.Google Scholar

169 Contin. II. 905; but Chron. géorg. 4/5 records her death in 1507; for this error, cf. note on Alexander II of Imeretia. Brosset, citing the latter sources, gives Nov. 30, 1519 Tables 4.642.Google Scholar

170 Brosset, , XI Rapp. (3e livraison) 45 c and d ; Kondakov, N. and Bak'raje, D., Opis' Pamǎtnikov Drevnosti v Někotorǐx Xramax i Monastǐrǎx Gruzii (St. Petersburg 1890) 34. — For Amirindo I, cf. Korbelašvili, P, ‘Amilaxvart'a Sagvareulos Istoriuli L'Ancienne Géorgie 2 (1911–13) 2.109; he is mentioned also in a document of 1453 (Z 267).Google Scholar

171 Chron. géorg. 7/6 (Brosset incorrectly translated miic'vala mxec'ije sargis as mourut Mkhhétzi, fils de Sargis); Contin. II. 914. He also figures in a document of 1488 (Sas. Sigel. 25). The House of Mxec'ije, later Mxeije (P'xeije), of ancient Abkhazian dynastic origin, has survived to the end of the Georgian kingdoms and the Russian Empire; Vaxušt, , Geogr. Description 46/47; Spiski 63.Google Scholar

172 Brosset, , VI Rapp. (3e livraison) 117; Xaxanov, , Ěkspedicii na Kavkaz 46-7; Mik'elaje, A., ‘Sel. Kvemo-Čala, Goriysk. Uězda, Tiflissk. Gubernii,’ Sbornik Materialov 29 (1901) 39–41. This inscription must belong to Amirindo I and not to Amirindo II who was Prince-Master of the House and Constable of Upper Iberia in the years 1754–1774 (AG 3.219; cf. Mik'elaje 39). In the first place, in the seventeenth and eighteenth centuries, the name of the hereditary office-fief of Prince-Master of the Horse (Amilaxor, Amilaxvar) had become the surname of the heads of the House of Zedginije (as, e.g., in contemporary Spain, the Dukes of Medina de Rioseco were called by their hereditary office of Almirante de Castilla); so that they would be called ‘the Amilaxvar Amirindo’ or ‘Amirindo the Amilaxvari’, without the surname of Zedginije. The inclusion of the latter in the inscription marks it as considerably anterior to Amirindo II. Likewise, in the available sources of the time of Amirindo II one never finds the conjunction of the office of Amilaxvar with that of Palatine of Gori: the former office appears in them either alone or in conjunction with that of Constable (of the Right). Finally, the family tradition of the Amilaxvaris has it that a son of Joatham I (and Amirindo I was his son) married a princess royal (cf. below, n.174); the above Mxec'ije inscriptions confirm this.Google Scholar

173 Cf. above, note on Bagrat V and n.9. — This probably made Korbelašvili (Amilaxvart'a Sagv. Ist loc. cit.) recognize the existence of Gulk'an, Bagrat VI's sister; however, he represents her as the wife of another Zedginije; whereas he gives for the wife of Amirindo I the name of Helen, for no other reason apparently than that her daughter was so called. Moreover, he takes the inscription at Čala to refer to still another member of the house: Joatham I's uncle, whom he makes, gratuitously, Prince-Master of the Horse (before the enfeoffment of Joatham I!) and Palatine of Gori and to whom he ascribes a wife named Gulk'an; op. cit. 163 n.4, 109. On the whole, Korbelašvili is not very reliable when dealing with the matrimonial alliances of the Zedginijes.— Tables 4.642 mentions Eléné, fille d'une soeur du roi Bagrat, mariée à Sargis Mkhétzidzé and places her mother, N fille, as a sister of Bagrat VI's grandson, Bagrat III of Imeretia! Google Scholar

174 Cf. Dolgorukov, , Ross. Rod. Kniga III 469–70. Joatham I's son, T'aqa II, is said to have married the Princess, which is an error.Google Scholar

175 The earliest recorded member of this family is Doğorsnel Zedginije, in the mid-fourteenth century, in an inscription in the synodicon (MS No. 54) from Mt. Sinai; C'agareli, , Pamǎtniki Gruzinskoy Starinǐ 220; Korbelašvili, , op. cit. 121 n.43. One of his sons, Janibeg, , ‘a Prince greatly renowned in Iberia’ (Hist. Inv. Tim. 875; Šaraf ad-dīn II 242) submitted to Timur in 1400. He was a Palatine of Gori (Korbelašvili, 102 n.3) and Šaraf ad-dīn records in the next chap. the capture of the fortress of Zarīt(=Gori?); cf. Minorsky, , Encycl. Islam s.v. ‘Tiflis’ 757.Google Scholar

176 Cf. notes on George VIII and Bagrat VI. — Joatham I was Doğorsnel's great-grandson; Korbelašvili 102, 104, 107, 109. His relative, the author of the Zedginije Petition, describes how Joatham (his name is not given in that fragmentary MS) took George VIII's place in the royal bed, in order to save the incredulous King from the plot and to prove to him its existence; and how he was, as a result, severely wounded by the conspirators. Vaxušt, who had access to the Amilaxvari archives, but who must have used other sources than the Petition (J IV 88-91) relates practically the same story; History (HG I 2) 685-7. He gives us the name of Joatham and states that he was murdered and that it was his children who profited by the royal largess. Possibly Joatham died soon after of the wounds he had received. Cf. Almanach de Gotha 1930 ff. (‘Amilachvary’) : there George VIII is called VII; also Dolgorukov, Ross. Rod. Kniga, loc. cit. Google Scholar

177 Samilaxoro or Samilaxaro, comprising Gori, Samt'avisi, Kaspi, K'vemo-Čala etc., was situated in Inner Iberia, on the left bank of the Kur; Vaxušt, , Geogr. Description 240/241ff.; Saamilaxvros Davt'ari (ed. Kakabaje, , Tiflis 1925). George VIII must have elevated Joatham's eldest son, T'aqa II (or Joatham himself before he died) to the new honors immediately after the event, because soon afterwards the King was captured by Qvarqvare III and never again ruled in Georgia.Google Scholar

178 In the seventeenth and eighteenth centuries, this house ranked as third (after the Dukes of Aragvi and of K'sani) among the ‘undivided’ princes of Iberia (mt'avar-s cf. above, n.8). It was then that the name Amilaxvar became a surname of the heads of the house; the cadets being called Amilaxvarisšvili. After the Russian annexation of Georgia, the family was received among the princes of the Empire under the name of Amilaxvari. Cf. Vaxtang VI, code §35; Vaxušt, , Geogr. Description 42/43 46/47); Aktǐ 2.39; Karst, , Comment. I 228; Ivanenko, , Graždanskoe Upravlenie Zakavkaz'em 9; Dolgorukov, , Ross. Rod. Kniga III 496-7, II 62-3; Spiski 6; Almanach de Gotha 1930 ff. (‘Amilachvary’) (cf, above, n.61). — The enfeoffment of the Zedginijes with the immediate office-fiefs of the Crown was taken, in later tradition (cf. Vaxušt and the last three of the above works), to signify their elevation to the princely status. Nothing could be farther from the truth: in Georgia, before the Silver Age, as, let us say, in pre-Petrine Russia, the princely statu*** was, at least in theory, a matter of dynastic origin, not of creation by the Crown (cf. above, n.8). Joatham I's great-uncle Janibeg was called prince in the sources (cf. above, n. 175). The might of this house is sufficiently demonstrated by the fact that, far from suffering for its loyalty to the defeated George VIII, it had to be won over by his*** successful rival, Bagrat VI, by way of a matrimonial alliance.Google Scholar

179 In his joint charter with Bagrat VI, of 1475 (AG 2.31), Constantine makes reference to his father, the King of Kings Demetrius and his mother Gulašar (cf. above, n.164); in that of 1466 (AG 2.13; for the date date cf. J IV 92-3) he calls George VIII his uncle and speaks of ‘Our aid’ Gulašar; and in the charter of the nun Nino of the K'vat'axevi Convent, of 1477 (K 2.20-1), he is referred to as ‘son of Demetrius’.Google Scholar

180 Hist. Alex. I. 891; Contin. II. 895.Google Scholar

181 J IV 136; Kakabaje, , Sas. Sigel. 31 and Sak'art'velos Mokle Istoria 4; Gugushvili, , Chron.-Geneal. Table 126-7.Google Scholar

182 Vaxušt, , History (HG II 1) 11; Tables 1.626; cf. Allen, , History 138.Google Scholar

183 Cf. below, chap. II.Google Scholar

184 Charters of 1487 (Z 305-6), 1488 (Z 306), 1492 (Z 309-10; K 3.41-2); cf. J IV 134-5.Google Scholar

185 Chron. géorg. 3/4; Contin. II. 895; a paschal chronicle ed. Z 298.Google Scholar

186 Vaxušt, , History (H II 1) 13–5, 251-3; J IV 140-2; Kakabaje, , Sas. Sigel, 26 (his genealogy is incorrect).Google Scholar

187 Vaxušt, , op. cit. 149, 253, 211; J IV 158-69; Gugushvili, , Chron.-Geneal. Table 131.Google Scholar

188 Dates 384; Tables 1.626; —or in 1503: Chron. géorg. 3/5; Contin. II. 904.Google Scholar

189 Tables 1.626; cf. J IV 136; cf. above, n.8.Google Scholar

190 Tables 1. — Of the House of Georgia, the elder Royal line became extinct in 1658, and the younger line of the Princes of Muxrani still flourishes, as the eldest surviving line of the Bagratid race, having been received among the princes of the Russian Empire under the name of Bagration of Muxrani. And elder branch of the line of Muxrani, now extinct, succeeded in 1659 to the throne of Georgia, but was replaced on it by the House of Kakhetia in 1744 (cf. note on Alexander I of Kakhetia). Its descendants bore the Russian titles of Princes of Georgia and Princes Bagration. Cf. ibid.; Dolgorukov, , Ross. Rod Kniga II 5-14, III 9, 458-9, 17-22; Spiski 10, 12, 31-2.Google Scholar

191 Phrantzes, , Annales 3. 1–2; cf. Lebeau, , Histoire du Bas-Empire XXI (Paris 1836) 219-22; de Muralt, E., Essai de chronographie byzantine (St. Petersburg 1871) 873; J IV 44-6.Google Scholar

192 Javaxišvili, , loc. cit., is inclined to doubt the identity of the imperial fiancée with the wife of George Šaburije. He contends that the former, being at least thirteen or fourteen years of age at the time of her betrothal of Constantine XI, would have been judged too old for marriage by the standards of the time in 1465, at twenty-six or twenty-eight; and he is of the opinion that there must have been another daughter to George VIII, who married Šaburije. But George VIII married in 1445; his daughter, even if the elder of his children, could not have been more than about seven at the time of her betrothal (cf. Bagrat V marrying Anne Comnena, aged ten); so in 1465 she was about twenty. Duke Vameq's protestations that he was ‘greatly unworthy’ of this alliance might have been called for by the bride's quasi-imperial status. For the difficulties in connection with this marriage, cf. above, n.8.Google Scholar

193 Hist. Mong. Inv. 763; Tables 1.624. —Vaxtang III reigned from 1301 to 1307.Google Scholar

194 T'aqaišvili, , ‘Arxeolog. Ěkskursii,’ Sbornik Materialov 43 (1913) 57–9; Bak'raje, D., ‘Kavkaz v Drevnix Pamǎtnikax Xristianstva,’ Zapiski of the Society of the Friends of Caucasian Archaeology (Tiflis 1875) 41-2. From the House of Šaburije, the Duchy of Aragvi passed to those of T'urmanije and Čarmeuli, and, finally, in the sixteenth century, to the House of Sidamoni; T'aqaišvili, , loc. cit. ; Sadzagelov-Iverieli, G., ‘Ananurskiy Uspenskiy Sobor,’ Materialǐ po Arxeologii Kavkaza 7.70.Google Scholar

195 Cf. note on George VIII; J IV 43-4.Google Scholar

196 Cf. below, chap. II.Google Scholar

197 Cf. note on George VIII. — Contin. II. 894, 905, ascribes to Alexander an elder brother Vaxtang, who is said to have reigned between George VIII (I) and Alexander and to have died in 1510. This is quite spurious. George VIII's own diplomata mention only one, ‘first-born’ son, Alexander, and those of the latter prove him to have begun his reign in 1476, the year of George's death. This reference to Vaxtang is a trace of the less exact earlier stages of the K. V Redaction of the Annals (cf. MGHL 163-4) and is due to a confusion with the children of Bagrat VI (qq.v.).Google Scholar

198 Vaxušt, , History (HG II 1) 149; Dates 384; Tables 2.634; Chron. géorg. 4/6; Contin. II. 906. George also blinded his younger brother Demetrius. Easter fell on Apr. 20 in 1511; cf. de Mas Lattrie, L., Trésor de chronologie (Paris 1889) 467.Google Scholar

199 Tables 2 634; cf. IV Rapp. 32; note on George VIII and n.108.Google Scholar

200 The Č'olaqašvilis constituted one of the two ‘undivided’ princely houses of Kakhetia (cf. above, n.8); after the Russian annexation, they were confirmed among the princes of the Empire under the name of Čelokaev (Čolokaev); Vaxušt, , Geogr. Description 46/47; II Heraclius, Short Description (ed. Iveria 1884 No. 5) 33-4; Ioseliani, P, Rod Knǎzey Čelokaevǐx (Tiflis 1866); Dolgorukov, , Ross. Rod. Kniga III 477-8, II 62-3; Aktǐ 2.38-43; Spiski 96, 99 (cf. above, n.61).Google Scholar

201 Tables 2. — To this house belonged the last Kings of Georgia, Heraclius II (1762–1798) and George XIII (XII) (1798–1801), whose posterity bore, after the Russian annexation, the Russian title of Princes of Georgia. A collateral branch, descended from Alexander I's blinded, younger son Demetrius (cf. above, n.198) was received into the princely nobility of the Russian Empire under the name of (Davidov-)Bagration(-ov); cf. ibid. ; Dolgorukov, , Ross. Rod. Kniga III 17-22, 458, 471-4; Spiski 32, 34.Google Scholar

202 Tables 4.642. — The confusion of this Vaxtang with his brother Alexander's son of the same name, who revolted against his brother, Bagrat III of Imeretia (Chron. géorg. 4/5) and even issued diplomata as king (Introduction cciii), produced a composite figure, ascribed by the Contin. II as an elder brother to Alexander I of Kakhetia (cf above, n.197); cf. Brosset, HG II 1.322 n.3, 329-30 n.1. This is an added proof of the dependence of the two Contin. of the Annals on (a source of) the Chron. géorg.; cf. note on Bagrat VI.Google Scholar

203 Cf. note on Constantine II. Yet Alexander's charter of 1509 dates his regnal years from 1494, which is probably an error; J IV 141. — Alexander is called ‘the Second’, because a Seljukid duke of Imeretia, who revolted there in 1387, is counted as Alexander ‘the First’ For the Imeretian dukes cf. note on Alexander I of Georgia.Google Scholar

204 Contin. II. 905; Dates 384: both give 1510 (but without the month or the day); Chron. géorg. 4/5 has April 1, 1507. The year of the latter date is obviously incorrect: the entry G. E. 195 (=:A.D. 1507), recording the deaths of Alexander, his wife, and his mother, and the struggle for the succession between his sons, Bagrat III and Vaxtang, is found between the years 1509 and 1511. The error results from writing the k'oronikon 195, in Georgian (mxedruli) r.ž.e., instead of the correct 198=1510, in Georgian r.ž.ē. Cf. Brosset, HG II 1 322 n.3, 329-30 n.1.Google Scholar

205 Chron. géorg. 4/5; Dates 384; Contin. II. 905. The latter work has April 1 for her death, but no month or day for that of Alexander, which is obviously due to a misreading of the sources.Google Scholar

206 Cf. above, n.8.Google Scholar

207 Tables 4. — The House of Imeretia was subdivided in several branches bearing, after the Russian annexation, the titles of Princes Bagration of Imeretia, Princes Bagration, Princes of Imeretia, and Princes Bagration-Davidov; cf. ibid. ; Dolgorukov, , Ross. Rod. Kniga III 5-8; Spiski 11, 44. — Instead of the above-mentioned children of Bagrat VI, as revealed by the documentary sources, the Constin. II. 892 gives Bagrat two sons, named Alexander and David. In view of the unreliability of this eighteenth-century account and in the absence of confirmation by some other sources, the existence of David cannot be accepted.Google Scholar

1 Though semantically a contradictio in adiectu, this term is no more self-contradictory than the thing itself. ‘Oligarchy’ or ‘polyarchy’ are inadequate, for they have decidedly too non-monarchical a connotation; and, for that matter also, ‘oligarchic’ or ‘polyarchic monarchy’ It is to be emphasized that, whatever the number of its holders, the power under such a system remains unimpairedly monarchical.Google Scholar

2 Cf. J II 419-23; Avalichvili, Z., ‘La success***ion du curopalate David d'Ibérie, dynaste de Tao,’ Byzantion 8 (1933); Gugushvili, , Chron.-Geneal. Table 120-2. For the Abasgian conquest of Iberia, cf. above, chap. I n.30.Google Scholar

3 The quarrel, in 988, between Bagrat the Foolish and his cousin David of Tao on the one hand, and, on the other, Gurgen and his son Bagrat, may have been caused by Gurgen's attempt to relegate his father to a de jure position of co-king. Cf. Chron. Iber. QA 174-5= QM 240-1; Asołik, Stephen, Universal History 3.28 (tr. Macler, F, Paris 1917, 134).Google Scholar

4 Cf. notes on Demetrius III and George VIII in chap. I; and below.Google Scholar

5 Cf. MGHL 174 n.63.Google Scholar

6 Cf. Žordania, , K'ronikebi 1 (Tiflis 1892) 255–8, 259, 266-7, 268-9, 300ff.Google Scholar

7 Ibid. 294.Google Scholar

8 Hist. Mong. Inv. 582–3.Google Scholar

9 This is revealed by the charter of 1355 (résumé, IV Rapp. 8), issued by the ‘King son of the Great King David’ Brosset (ibid.) was inclined to doubt the authenticity of this document solely on the ground that Bagrat could not have been King in 1355, since his father, David VII, was still alive. Brosset was obviously ignorant of the system of co-optation of the heir, which is here presented with a special clarity in the correlation of the terms ‘King’ and ‘Great King’; cf. below.Google Scholar

10 This is attested by George VII's charter of 1393 (résumé, IV Rapp. 9), written in his 24th regnal year. Brosset (ibid.) again was at a loss before this renewed manifestation of co-kingship. And yet the Hist. Inv. Tim. refers to George as king in the lifetime of Bagrat V; 857, 866, 867 (in 1386 and 1393).Google Scholar

11 Cf. above, chap. I, note on Alexander I.Google Scholar

12 It will be recalled that the early Trapezuntine sovereigns used the Eastern Roman imperial style of and that it took a considerable diplomatic pressure on the part of the Palaeologi, then newly established at Constantinople, to induce one of them to change, after 1282, his title to In other words, the early Grand Comneni regarded themselves, theoretically at least, as the legitimate heirs of the Roman Empire (of the East), or rather as titular Roman Emperors, merely resident at Trebizond in expectation of Constantinople; for in the Byzantine ideology of the period, there could be but one Roman Empire. Cf. Vasiliev, , The Foundation of the Emp. of Trebizond 30ff.; Miller, , Trebizond 26ff.; Zakythinos, D., Le chrysobulle d'Alexis III Comnène en faveur des Vénitiens (Paris 1932) 92.Google Scholar

13 Bury, J. B., The Constitution of the Later Roman Empire (Cambridge 1910) 16ff. For Roman imperial collegiality cf. also Kornemann, E., Doppelprinzipat und Reichsteilung im Imperium romanum (Leipzig and Berlin 1930); Palanque, J. R., ‘Collégialité et partages dans l'Empire romain aux IVe et Ve siècles,’ Revue des études anciennes 46 (1944) 47-64, 280-298.Google Scholar

14 Cf. Paillard, A., Histoire de la transmission du pouvoir impérial à Rome et à Constantinople (Paris 1875) 7–188.Google Scholar

15 Cf. Bréhier, L., ‘L'origine des titres impériaux à Byzance,’ Byzantinische Zeitschrift 15 (1906) 161–78.Google Scholar

16 Thus, e.g. Seleucus I (305-280) co-opted his son Antiochus I in 293; the latter (281-262/1), in 266, his son Antiochus II (261-247) etc. Likewise, the following Lagids practiced association of the heir: Ptolemy I (305-283/2) and his son Ptolemy II Philadelphus (284; 283-245); the latter and his son Ptolemy III (Nov. 12/13, 247; 245-221) as well, perhaps, as another son, the problematic co-king Ptolemy (266-258); Ptolemy IV Philopator (221-203) and his son Ptolemy V Epiphanes (209; 203-181); Ptolemy VI (181-169) and his two sons Ptolemy Eupator (153/2-c.150) and Ptolemy Philopator (bef. 145;d. 145); and Cleopatra VI (51-30) and his son Ptolemy XIV Caesar (44-30).Google Scholar

17 The theogamic collegial kingship of the Lagids (for the most part between brother and sister) can be observed in the following cases: the Soteres Ptolemy I and Berenice I (posthumously); the Adelphi Ptolemy II and Arsinoe II; the Euergetae Ptolemy III and Berenice II; the Philopatores Ptolemy IV and Arsinoe III; Ptolemy V and Cleopatra I; the Philometores Ptolemy VI, Cleopatra II, and Ptolemy VII; the Euergetae Ptolemy VII, Cleopatra II, and Cleopatra III; the Philometores Soteres Cleopatra III and Ptolemy VIII; the Philometores Ptolemy IX and Berenice III; the Philometores Soteres Cleopatra III and Ptolemy IX; the Philadelphi Philometores Soteres Ptolemy VIII and Berenice III; Ptolemy X and Berenice III; the Philopatores Philadelphi Ptolemy XI and Cleopatra VI; the Philopatores (?) Cleopatra VI and Ptolemy XII; and the Philopatores Cleopatra VI and Ptolemy XIII.Google Scholar

18 Maurice, J., ‘Les Pharaons romains,’ Byzantion 12 (1937) 71–103.Google Scholar

19 Cf. Seston, W, Dioclétien et la Tétrarchie (Paris 1946) 249: ‘Résumé: La Tétrarchie a un fondement tout religieux.’ The possibilities of Egyptian influences, however, are not taken into account.Google Scholar

20 Bury, , Constitution of the Later Roman Empire, loc. cit. and History of the Later Roman Empire (London 1931) I 5–7. — Exceptions to this rule are found in the (usually two) parallel and territorially distinct imperial organizations which existed between the reforms of Diocletian and the end of the Western imperial line, and in the rise and the Byzantine recognition of the Holy Roman Emperors, which was tantamount, in theory at least, to a resuscitation of the Western half of the Roman Empire: Bury, , Constitution 16-7. Another exception was the joint rule, in the East, of the last Basilids, Zoe and Theodora, in the eleventh century.Google Scholar

21 Philotheus, , Kletorologion, apud Constantine Porphyrogenitus, De cerim. (PG 143) 1308.Google Scholar

22 Constantine, , De cerim. 989, 993, 1000, 1028, 1029, 1036, 1173.Google Scholar

23 Bury, , Constitution 21; DuCange, , Gloss. graec. I (Lyons 1688) 156-7.Google Scholar

24 Bury, , loc. cit. Google Scholar

25 Bury, , Hist. Later Rom. Emp. I 7 n.1.Google Scholar

26 It must be noted, however, that one modern Georgian historian, S. Kakabaje, has recognized the existence of co-kings in Georgia — though not, it seems, of the system of co-kingship.Google Scholar

27 The title of King of Kings was adopted by Gurgen, father of Bagrat III (cf. above) upon the death of his own father Bagrat II, in 994. Gurgen was then King of Iberia and his son, King of Abasgia; cf. J II 412; Gugushvili, , Chron.-Geneal. Table 122. In the case of Georgia, this title implied not so much what the genitive case suggests, i.e., ‘king over other kings’, as the superlative degree (of kingship) occasionally expressed by that case (Marr, N. and Brière, M., La Langue géorgienne [Paris 1931] 79 §96.2b), i.e., ‘great king, from among other kings’; in a word, a monarch wholly independent of others, an autocrat. Thus mep'et'-mep'e was a semantic equivalent of the Byzantine of which it also became a constitutional counterpart in the fifteenth century. The consort of a mep'et'-mep'e was dedep'alt'-dedop'al (‘queen of queens’). Georgian being devoid of the distinctions of gender, a queen-regnant (like the great Thamar) might be styled mep'e or mep'et'-mep'e. — For instances of the above-mentioned uses of royal titles, cf. below; for ‘Great King’ cf. above, n.9.Google Scholar

28 The royal style of the fifteenth-century Kings of Georgia was composed of geopolitical formulae and honorific epithets. The former were fixed, the latter variable. The formulae reflected the historical development of the Georgian State: the union of Western and Eastern Georgia, perennially separated by the Lixi mountains, and the acquisition of other kingdoms. They were as follows: (A) Of Abasgia (and) Iberia, (B) Albania, Kakhetia, and Armenia, King; (C) Šahanšah and Širvanšah [the title of the Kings of Armenia and that of the vassal Kings of Shirvan]; (D) Lord of the Two Thrones and Realms, Hither and Thither of the Lixi-s. The epithets included: (E) King of Kings [acquiring under the system of collegial sovereignty the specific constitutional meaning of king-regnant as distinct from his co-kings]; (F) God-crowned (or variants); (G) Sovereign and Autocrat of All the East (and variants). The relative position of these formulae and epithets varied considerably. Epithet E was conjoined with the name of the monarch as was also the dynastic surname of Jessian-Davidian-Solomonian-Bagratid. Cf. Brosset, , IV Rapp. 39–48; Kakabaje, S., ‘Gramota Caricǐ Tamarǐ Velikoy na Imǎ Gelat ot 1193 g.’, Bulletin de l'Institut caucasien d'histoire et d'archéologie 3 (1925) 111-20; idem, ‘Gramota Gruzinskogo Carǎ Georgiǎ III po Povodu Vosstaniǎ Knǎzey Orbel-i v 1177 g.’ ibidem 4 (1926) 123-5.Google Scholar

29 For examples, cf. below.Google Scholar

30 J IV 19, 36-7, 102-3, 135.Google Scholar

31 Cf. above, chap. I, note on Alexander I.Google Scholar

32 Above, chap. I, n.21.Google Scholar

33 Hist. Mong. Inv. 734, 745; Bek'a II's other daughter married, in 1297, the Emperor Alexius II of Trebizond, nephew of the queen of Demetrius II; ibid. 758; Panaretus 6.269; Pachymeres, , De Andronico Palaeologo 4.7 Google Scholar

34 The scantiness of documentary sources for the period forces one to rely exclusively on the near-contemporary Hist. Mong. Inv. of the anonymous Meschian Chronographer (MGHL 176). The following chronological table is based on it (the names of co-kings are indented) : Google Scholar

Demetrius II the Devoted, †1289 Google Scholar

Vaxtang II, 1289–1292 (Hist. Mong. Inv. 744-7) Google Scholar

David (VI) 1291 (ibid. 746) Google Scholar

David VI, 1292–1301 (ibid. 749-63) Google Scholar

George (V) in Iberia, 1299 (ibid. 758-9) Google Scholar

Vaxtang III, 1301–1307 (ibid. 758-9) Google Scholar

David (VI) Google Scholar

George (V) Google Scholar

George VI the Little, 1307–1318 (ibid. 781-5) Google Scholar

George (V) Regent (ibid. 782) Google Scholar

David (VI) †1310 (ibid. 783) Google Scholar

George V the Illustrious, begins as king-regnant, 1318 (ibid. 785) Google Scholar

Needless to say, the numbers of different kings are those of the traditional historiography stemming from Vaxust, and stand in need of revision. Thus, George the Illustrious should be ‘the Sixth’ and George the Little ‘the Fifth’ All this is due to the inability of older historians to accept the fact of collegial sovereignty in Georgia. Brosset, however, admitted the following: (1) David VI: roi du vivant de Vaxtang II, but déposé et remplacé par Vaxtang III. This is contrary to the Hist. Mong. Inv., where it is said (766) that, upon becoming king, Vaxtang III ‘did not go against his brother David’ and where the latter is mentioned as king at his death (783). However, Vaxtang, not David, was the king–regnant thereafter, as is evident from the context of pp. 762-81, as well as from the narrative of his war against the Sultan of Egypt (QM 772-7, notes;cf. MGHL 178).— (2) Vaxtang III: règne du vivant de ses frères David and George V.—(3) George VI the Little: roi sous la tutelle de George V.— (4) George V the Illustrious: roi du temps de David V, 1299–1301 puis 1318–1346 But the Hist. Mong. Inv. does not indicate that he was ever deprived of his royal dignity after his elevation to it in 1299. He was set up, it is true, by the Mongols with the intention ‘to replace’ David, as king-regnant (758-759). Since, however, David remained sovereign throughout Georgia, and George only held Tiflis, he was at best a co-king appanaged therein; or else an anti-king. Cf. Tables 1.624.Google Scholar

35 Cf. above, n.33.Google Scholar

36 The typically Byzantine character of the system adopted in Georgia will become apparent upon comparing the above list of the kings (n.34) with any table of the Eastern Emperors. Of course, although collegial sovereignity is distinguished from co-optation of the heir by its purely honorific character, it has been resorted to occasionally for purely practical considerations also.Google Scholar

37 Cf. above, chap. I, note on Bagrat V Google Scholar

38 Miller, , Trebizond 73, refers to pluralistic monarchy as old-established in Trebizond, yet Manuel III is the first Trapezuntine sovereign indicated as having practiced it.Google Scholar

39 Cf. above, n.10 and Chap. I, note on George VII.Google Scholar

40 Marr, and Brière, , La Langue géorgienne 266.Google Scholar

41 Above, chap. I, n.72.Google Scholar

42 The following Byzantine instances of collegial sovereignty will be recalled in this context: Heraclius (610-641) and his sons Constantine(III)-Heraclius, co-opted in 613, and Heracleonas, in 638; Constans II (641-668) and his sons Constantine (IV), Tiberius (IH) and Heraclius in 659; Basil I (867-886) and his sons Constantine in 869, Leo (VI) in 870, and Alexander in 871; Romanus I Lecapenus (919-944), his sons Christopher in 921, Stephen and Constantine in 923, and the former Emperor-regnant Constantine VII Porphyrogenitus; John V Palaeologus (1355–1390) and his sons Andronicus (IV) in 1379 and Manuel (II) in 1386; cf. DuCange, , Familiae augustae byzantinae (Paris 1680) 117–9, 120, 138-41, 142, 146-7, 239-43; Muralt, , Essai de Chronographie byzantine, 395-1057 (St. Petersburg 1855) 270-2, 290, 446, 450, 452, 498, 502; ibid. 1057–1453 (St. Petersburg 1871) 700, 708. — Constantine I's maternal grandmother, Theodora Cantacuzena, was a cousin of the Roman Emperor John VI Cantacuzenus (1347-1355), who in 1348 reduced the reigning Emperor John Palaeologus to co-emperorship, and elevated to it his own son Matthew; cf. below, n.47 Google Scholar

43 This charter led Kakabaje to assume that Vaxtang and perhaps Demetrius were co-kings from that date; Sas. Sigel. 12. As has been said above in the Introduction, charters (letters-patent) conferring or confirming privileges and donations were issued in feudal Georgia not only by reigning sovereigns, but also by members of their families and lords spiritual and temporal.Google Scholar

44 Javaxišvili hesitates to accept the fact of Vaxtang's co-optation and endeavors to explain it away by declaring as doubtful the charter of 1438 [I]; J IV 19. But the act of 1433 [I], for instance, wherein Vaxtang is also mentioned as king (and the authenticity of which Javaxišvili does not question), alone suffices to prove that fact. The hesitancy of the learned historian is due, no doubt, to his lack of insight in the system of collegial monarchy as practiced in fifteenth-century Georgia.Google Scholar

45 Javaxišvili, as is his wont, attempts to explain away the fact of Demetrius's co-kingship by dismissing his charter of 1445 as unreliable, because it is a later copy, and by urging the fact that the original act of George VIII of Dec. 25, 1449 (K 3.29-32; Z 261) mentions Demetrius without the royal dignity: this must signify, according to him, that Demetrius was never king; J IV 36-7 He confuses here two distinct problems: (1) Demetrius' co-kingship and (2) his de jure status as king-regnant. As regards the latter, the Georgian scholar's argument is invalidated by the diplomatic sources adduced below. As for the former, it is borne out by other evidence than the charter he adduces. In 1446 George reduced Demetrius to the de facto position of a co-king, hence the omission of the latter's royal title (not infrequent in the cases of co-kings) in 1449.Google Scholar

46 It is true that the Hist. Alex. I. 890, 891, states that George was established as King of Kakhetia in 1445; but this is completely outweighed by the charter of March 7, 1446. Moreover, Brosset mentions an inscription on the altar of the church of the Abbey of the Archangels at Gremi (Kakhetia), dated 1441, ‘under the truly faithful King, the Despot George’; IV Rapp. 22. Javaxišvili considers the information of the above chronicle erroneous on the ground that George became King of Georgia in 1446 (1447 is a lapsus calami); J IV 36. But the point in question is not his becoming King of Georgia, but his being made a co-king in Kakhetia. Kakabaje is right in considering 1433/4 as the date of the elevation of George to the co-kingship with Kakhetia as appanage; Sas. Sigel. 12.Google Scholar

47 For Byzantine examples, one will recall the reduction to co-emperorship of the lawful Emperor-regnant Constantine VII by Romanus I Lecapenus; that of Basil II by Nicephorus II Phocas and, later, by John I Tzimisces; that of Isaac II Angelus by his son Alexius IV; and that of John V Palaeologus by John VI Cantacuzenus: DuCange, , Familiae 149, 150, 153, 204, 239, 260; cf. following note.Google Scholar

48 Cf. above, chap. I, note on Alexander of Kakhetia. For a similar Byzantine instance, comparison can be made with the action of Romanus I, who reduced Constantine VII to co-regency and elevated to it his own three sons; DuCange, , Familiae 142, 146-7.Google Scholar

49 On the significance of this title, cf. above, chap. I. n.149.Google Scholar

50 Javaxišvili discards the charter of 1466 as unreliable, but, being unable to dismiss that of 1477, argues that perhaps Constantine succeeded Bagrat in that year, and not in 1478; J IV 135. This, of course, is contradicted by the regnal years of Constantine's acts (cf. note on Constantine II), and the Georgian historian finally abandons his argument. But, then, he ignores the charters of 1467 and 1475, which it would be difficult to explain away. On the other hand, he makes much of the joint charter of Bagrat and Constantine of 1468, in which the one is ‘King’ and the other patron; J IV 102-3. But that title was applicable to kings (cf. above, chap. I. n.149) and in this charter the royal style follows the two names (cf. below).Google Scholar

51 The Byzantine instance of Alexius I Comnenus, co-opting Michael VII's son Constantine Ducas upon dethroning the usurper Nicephorus III Botaniates, will be recalled: DuCange, , Familiae 164.Google Scholar

52 Cf. above, chap. I, notes on George VIII and Bagrat VI.Google Scholar

53 In this way the title patron, though belonging to all the members of the royal family (cf. above, chap. I, n.149), was often applied, in the singular, to the eldest of a sovereign's sons. What is more, one often finds in the documents the singular of the noun ‘son’ (je) applied to the eldest only of all a monarch's sons.Google Scholar

54 This is precisely the point combatted by Javaxišvili. In his opinion it is, e.g., decisive that Constan tine II is merely called patron (cf. above, nn.50, 53) in his joint charter with Bagrat VI of 1468 (AG 3.20) even though the royal style follows after the names of both. Else, he argues, one would have to consider as cases of a co-optation all the cases when in official documents the royal style followed the names of both the reigning sovereign and his sons, which, to his mind, would be an absurdity; J IV 102-3. Yet this is precisely the historical truth. We have seen that Constantine was a co-king of Bagrat VI and that the sons of Alexander I were co-opted in the very year when the royal style began to follow their names in that King's acts.Google Scholar

55 Cf. Ceret'eli, , ‘ArxeologiǍeskaǎ Progulka etc.’ Materialǐ po Arxeologii Kavkaza 7 84–85; Uvarov, Countess P S., ‘Koblianskoe UšǍel'e,’ Materialǐ po Arx. Kavk. 4 (1894) 72-4: the royal title of Bagrat of Muxrani evoked the author's astonishment (p. 74). — Bagrat was the founder of the branch of the Princes of Muxrani, the eldest extant line of the Bagratid dynasty; cf. above, chap. I, n.190.Google Scholar

56 For these fiefs, cf. above, chap. I, n.99.Google Scholar

57 Cf. above, chap. I, note on Constantine II.Google Scholar

58 Unless we accept as a precedent the doubtful case of George V in Iberia, in 1299; cf. above, n.34 (4).Google Scholar

59 It is quite possible that, upon the elevation of his elder sons (and of George) to the co-kingship in 1433, Alexander may have sought to protect the future of his third son, who was his first by the second marriage, against possible encroachments on the part of his elder half-brothers, by assuring him and his eventual posterity of an appanage — like that of the Dukes of Orléans in France. The co-optation was, to be sure, as usual extended only ad personam, whereas the appanage would have remained in the princely, no longer royal, posterity of George (who, in the normal course of events, had no prospect of succeeding to the throne). Precisely the same situation can be witnessed in the case of Constantine II's third son and co-king Bagrat, who was given the Principality of Muxrani in appanage, in 1512, and whose posterity constituted the house of the Princes of Muxrani, first princes of the blood of Georgia. (cf. above, n.55; chap. I, n.190). The choice of the easternmost province of Kakhetia may have been determined by the desire to remove George from any contacts with his mother's heritage, Imeretia, where the connection of local separatism with the ambition of the Seljukids had only too often proved detrimental to the unity of the realm. As we know, this precaution was of no avail. — It is interesting, too, that the title of the Princes of Muxrani was baton (muxran-baton), an equivalent of the Palaeologan and the early Kings of Kakhetia were likewise referred to by that title in some Georgian sources.Google Scholar

60 Cf. Bury, , History I 5–7, 16-8.Google Scholar

61 For Georgian feudalism, cf. Brosset, , Introduction lxv–lxvi; Allen, , History chaps. 19, 21, 22.Google Scholar

62 In the preface to Baltrušaitis, J., Etudes sur l'art médiéval en Géorgie et en Arménie (Paris 1930).Google Scholar

63 Cf. N. Adonc', Armeniǎ v Epoxu Yustiniana 383; Kherumian, R., ‘Esquisse d'une féodalité oubliée,’ Vostan 1 (1948–1949) 7–56; Christensen, A., L'Iran sous les Sassanides (2nd ed. Copenhagen 1944) 15-27, cf. 510-4.Google Scholar

64 Brosset, , Introduction lxv; Karst, , Code géorg.: Comment. I 220; cf. above, n.28: either the title of Šahanšah was counted as distinct from that of King of Armenia, or it was remembered that the Kingdom of Kakhetia, annexed by David II, was originally ‘of Kakhetia and Heret'i.’ Google Scholar

65 Cf. Allen, , History 138; Minorsky, , Encycl. of Islam s.v. ‘Tiflis’ 758. The enemies were the Aq-Qoyulu chieftains, Uzūn-Hasan and his son Sultān Ya'qūb.Google Scholar

66 Among the posterity of Asot the Great, Curopalates of Iberia (813-829), until the unification of Bagrat III (978-1014).Google Scholar

67 The relations of Alexander I of Kakhetia and Constantine II may give an indication that something very vaguely like dynastic condominium existed during the reigns of these two monarchs. In his charters of Jan. 23 and 24, 1479 (Z 300-1 and 301), 1503 (Z 319), and 1505 (Z 325; K 3.42) the former entitles himself merely King; this was no doubt due to an accord with Constantine, which found it legal expression in the Treaty of 1490; J IV 139. But the Imeretian rulers never — from the final instalment of Alexander II to the reign of the last king, Solomon II — ceased to regard themselves as Kings of Kings of Georgia. What is more, Constantine II appears to have been constrained to abandon Formula D of the royal style (cf. above, n.28) after his treaty with Alexander of Imeretia in 1491; J IV 138. Thus the old geopolitical dichotomy of Georgia appeared anew. Cf. Introduction lxxxiv.Google Scholar

68 J IV 94-9; Vaxušt refuted that notion, but it had arisen soon after the Partition, in 1589 King Alexander II of Kakhetia told the Russion Ambassador, Prince Zvenigo*** skiy that the unity of Georgia had been broken into three kingdoms by his forefat*** Bělokurov, S., Snošeniǎ Rossii s Kavkazom I (Moscow 1889) 169–70. The statement f*** the Hist. Alex. I. referred to above (chap. I, n.90) has been often interpreted in that se*** Google Scholar