Historical Aspects of "The Lay of the Emir" Huxley, George Greek, Roman and Byzantine Studies; Fall 1974; 15, 3; ProQuest pg. 317

Antecedents and Context of 
Digenes Akrites 

George Huxley 

I 

Historical Aspects of The Lay of the Emir 

THE HISTORICAL CRITICISM of the epic Digenes Akrites begins with 
the recognition that the poem consists of two distinct parts. In 
the first part the centre of attention is the Saracen emir who, 

having fallen in love with his beautiful captive Eirene, an East Roman 
general's daughter, is willing to accept baptism in order to marry her; 
they settle in Cappadocia, where he becomes the father of Digenes. In 
the second part the hero is Digenes Akrites, the <Twyborn borderer', 
himself, and the action is centred further to the east in Commagene, 
beside the Euphrates in the neighbourhood of Samosata, where, at 
Trosis, Digenes builds his palace and after many exploits dies. This 
part, the Digeneid proper, cannot have come into existence before the 
imperial advance eastwards from Cappadocia during the campaigns 
of John Kourkouas, who had reached Melitene in A.D. 931. On the 
other hand, the historical context of the first part, the Lay of the Emir, 
is the Cappadocian frontier a generation or more earlier, before the 
imperial advance had begun. The Lay, indeed, was originally a dis­
tinct poem; it was extended by the author of the Digeneid, but it is 
now lost, as is the original Digeneid. 

The Lay and the Digeneid have recognisable historical contexts. It is 
with the background of the former that this paper is concerned; but 
epics with historical contexts are not necessarily epics about historical 
individuals, and we should resist the temptation to regard Digenes as 
a poetical representation of John Kourkouas, George Maniakes, or any 
other victorious agent of the central government in Constantinople. 
In so far as Digenes represents anything, he exemplifies heroically the 
military power, independence and riches of the akritic nobility in the 
century after the conquests of Kourkouas, before the incursions of the 
Seljuks began. The hero of the Digeneid gives every sign of being a 

317 



318 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

purely poetical creation of a learned Commagenian dependant of an 
akritic family, who lived sometime between about 940 and 1040;1 in 
Digenes the akritic life takes on a superhuman quality, but neverthe­
less H. Gregoire thought that he had identified the historical proto­
type of Digenes. 

In 788, according to Theophanes,2 a capable officer called Diogenes, 
Tourmarch of the Anatolikoi, fell in battle at Kopidnados against 
Arab invaders of the Anatolic thema. Here, claimed Gregoire,3 who en 
passant changed Kom8vcf8ov to II o8av86v in the text of Theophanes, is 
the original of Digenes. The claim was strongly criticised by Mavro­
gordato,4 but it has even been favourably mentioned in textbooks.s 
There is, however, nothing to commend it. The TOVPf.LcfPX7Jc Diogenes 
is not an a'krites; he belongs to the central organisation of the themata. 
He dies in war, not at home in peace. The name LltOYEV7]C has no 
etymological or semantic connexion with LltYEV1JC, 'Twyborn'. 
Diogenes the Tourmarch does not belong to the period in which the 
original Digeneid took shape, but more than a century and a half 
earlier. 

Yet Digenes has a prototype-his father, the Emir Mousour­
Ioannes. Both are borderers. Both carry off brides from households of 
the Doukas family, or at any rate, from households of a Doux. Both 
are great warriors who campaign far beyond their homes near the 
frontier. Both bring their mothers to live with them in their second 
homes-the Emir from Edessa to Cappadocia, Digenes from Cappa­
do cia to Commagene. Both die Orthodox Christians. Both are 

1 L. Politis, A History of Modern Greek Literature (Oxford 1973) 24, suggests that the poet of 
the lost original of the Digeneid lived at the end of the tenth or the beginning of the eleventh 
century. Since Digenes brings peace to the frontier, John Mavrogordato proposed to 
place the poet in the time of Constantine IX Monomachos (1042-1055), when there was 
peace on the frontier for the last time before the Seljuk onslaught (Digenes Akrites [Oxford 
1956] lxxxiv); but Digenes reflects not the existence of peace but his creator's longing for it. 
The mention of imperial tribute to Ikonion in the Grottaferrata version (G 1993-94 Trapp) 
need not be a reference to the Seljuk Sultanate of Rum; but even if it is, that is not a 
reason for dating the lost original of the Digeneid later than the East Roman defeat at 
Manzikert in 1071. I cite the earliest versions of the Digeneid from the synoptic edition of 
E. Trapp, Digenes Akrites (Wiener Byzantinische Studien 8,1971); though his reconstruction 
of the manuscript tradition takes too little account of the interaction of oral and written 
versions of the poem, he makes much vital evidence conveniently accessible. 

I p.463, 15-21 de Boor. 
s Byzantion 11 (1936) 608, and elsewhere. 
'op.cit. (supra n.!) lxxi-lxxii. 
GA. A. Vasiliev, History of the Byzantine Empire I (Madison 1952) 370; R. J. H. Jenkins, 

Byzantium, The Imperial Centuries (London 1966) 65. 



GEORGE HUXLEY 319 

<twyborn': the Emir's father was Chrysocherpes, whose historical pro­
totype was the Paulician heretic Chrysocheir, and his mother was the 
Muslim lady Spathia; Basil Digenes Akrites had the Emir for his father, 
and for his mother the East Roman Orthodox lady Eirene Doukas. 

The Emir, then, is the original, a fictitious original, of the equally 
non-historical Digenes. It follows that the complete Digeneid originated 
later than the Lay of the Emir. We have an approximate terminus ante 
quem for the composition of the Lay in the campaigns of John Kour­
kouas. A more precise terminus post quem is provided by the mention 
of Chrysocheir or Chrysocherpes, the father of the Emir; with the 
Emir must be considered Ambron, his maternal grandfather and 
Karoes his mother's brother. The Lay presents the Emir's ancestry 
consistently in the Grottaferrata version thus: 

AMBRON 

I 
I I 

PANTHIA = CHRYSOBERGES KAROES MOUSOUR OF TARSOS 

I 
EMIR6 

The flesh-and-blood original of Ambron is the historical Emir of 
Melitene, Omar, who for long distinguished himself in war against 
the Empire, but in the end he was surrounded by the imperial forces 
under Petronas in 863 near the Halys and killed.7 Omar's incursions 
had been made all the more damaging by his alliance with the 
Paulicians in the eastern frontier lands; they were led by Karbeas and 
Chrysocheir, renegade imperial officers, whose epic embodiments in 
the Lay are Karoes and Chrysoberges (or Chrysocherpes).8 The high 
rank achieved by Paulicians under the Iconoclast emperors shows 

• See Mavrogordato, op.cit. (supra n.1) 254. For variants in the genealogy see A. Pertusi, 
op.dt. (infra n.35) 538-40. 

7 Genesius p.96, 20, ed. Bonn; Theophanes Continuatus p.183, 6 ed. Bonn. 
S The Paulician vestiges in the Lay were first detected by C. Sathas and E. Legrand, who 

published the Trebizond version of the Digeneid in their book Les Exploits de Digenis Akritas 
(Paris 1875). Recently there has been much writing about the Paulician heresy. Relevant 
texts are re-edited by Ch. Astruc, w. Conus-Wolska, J. Gouillard, P. Lemerle, D. Papa­
chryssanthou and J. Paramelle in "Les Sources grecques pour l'histoire des Pauliciens 
d'Asie mineure," Travaux et Mbnoires 4 (1970) 1-227. The book of Nina G. Garsoian (The 
Paulidan Heresy [The Hague/Paris 1967]) must be read in conjunction with the compre­
hensive study ofP. Lemerle: "L'Histoire des Pauliciens d'Asie mineure d'apres les sources 
grecques," Travaux et Memoires 5 (1973) 1-144. Allusions to Paulicians in Akritic epic are 
considered by Lemerle on pp.llD-13 of his study. 



320 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

that Paulicianism cannot have been a purely rustic or proletarian 
phenomenon. Karbeas himself had been the adjutant (protomandator) 
of the thema Anatolikoi, and Theodotos Melissenos, the strategos who 
appointed him to that position, can hardly have deplored his religious 
sympathies.9 But under the rule of the widowed Empress-Regent 
Theodora, Paulicians were put to the sword and their property was 
taken by the state. Amongst the victims of Theodora's troops was the 
father of Karbeas, who was impaled.10 

So distressed was the adjutant at his father's fate that he fled, and 
five thousand with him, to the protection of Omar of Melitene. As 
general of the Paulicians, Karbeas decided to move the sectaries 
gathering at Amara and at Argaoun nearby, to Tephrike, a strong­
hold conveniently situated for raids deep into Romania and Armenia. 
In Tephrike a nascent Paulician state enjoyed the combined 
advantages of political independence and a military alliance with the 
Emir. 

From the point of view of Orthodoxy there was nothing incom­
prehensible in the alliance of Muslims with Paulidans. Muslims 
honoured Christ, and Islam was regarded as a kind of Arianism to be 
placed on a level with other Christian heresies.l1 Muslims and Pauli­
dans were in alliance against imperial Orthodoxy, and accordingly 
the author of the Lay and his more self-conSciously Orthodox fol­
lower, the creator of the Digeneid, felt no difficulty in linking Chryso­
cherpes or Chrysocheir in marriage with the Moslem lady Spathia, 
whose name, indeed, may recall the Paulician fort Spathe.12 

When Karbeas died in battle against the East Romans, he was 
succeeded as leader of the Paulicians by Chrysocheir,13 who soon after 
the usurpation of the Emperor Basil I led them across Asia Minor to 
Nikomedeia and to Ephesos, where he used the Church of St John the 
Divine as a stable for his horses.14 In an attempt to put a stop to such 

11 N. G. Garsoian, DOPapers 25 (1971) 99. 
10 Theoph.Contin. pp.165-66 ed. Bonn. 
11 Vasiliev, op.cit. (supra n.5) I 207. 
12 See N. Adontz, BZ 29 (1929/1930) 224, on Theoph.Contin. p.267, 16, ed. Bonn. In G 255 

Trapp her name is llavOla; Kyriakides, BZ 50 (1957) 143, suggests that Pantheia was the 
original name and, somewhat implausibly, that it is taken from Xenophon's Kyroupaideia: 
rijc llav8€l«c > rijc E7T«v8€lac > rijc E7TaOl«c. The variants may well have arisen during 
dictation. 

13 A. J. Toynbee, Constantine Porphyrogenitus and his World (London 1973) 655. 
u Genesius p.121, 19-22, ed. Bonn. 



GEORGE HUXLEY 321 

raids Basil attacked the Paulician redoubts beyond the frontier, but 
he failed to take Tephrike,15 and his army was forced to retreat. 

The distant raids of Chrysocheir are vividly recalled in the Lay of 
the Emir. They are recalled not because the author of the Lay found 
mentions of them in chronicles, but because the campaigns of the 
Paulidans had deeply impressed the folkmemory of Christians and 
Muslims in Asia Minor. Indeed, since the poet of the Lay lived before 
the campaigning of Kourkouas to and beyond the Euphrates, he com­
posed his poems within two generations of Omar's death, and may 
even have witnessed some of the Paulician wars. In the Escorial ver­
sion of the Digeneid, a late and perfunctory, dictated work16 which 
nevertheless preserves some details of the original poem, there is a 
specific mention of Chrysocheir's KOVpCOV to Nikomedeia. The Emir's 
mother asks him to remember his father's raiding: 

K' ,~, 8 ~, ,,~., 

OVOEV VJLacaL, TEKVOV JLOV. TL E1TOtKEV 0 1TaT'TJP cov; 

'fl' rT' " "" "A .L 0 novtQV EKOVPCEVCEV JLExpL Kat ELC TOV JLJLOV. 

'7\T ,~ "-1..8 " II' "Q EtC 1 vtKOJL'TJooav E-r aCEV K EtC paLVETOV E1TEt-''TJV. 

, " "" T • 8'\ , 'I' t , 17 K av OEV 'TJTOV 'TJ al\acca, 1TaVTa ELXEV v1TayaLVEL. 

The Emir's mother recalls that Chrysocherpes had refused an im­
perial offer of patrician rank because he would not forsake his 
religion by throwing down his sword (G 375-77), and such a peace­
offer may well have been made in fact to Chrysocheir in Tephrike. 
She also, when trying to persuade the Emir not to transfer to Cappa­
docia, reminds him of the slaughter and enslavement of Romans by 
his grandfather Ambron (E 245-47); her plea is further strengthened 
by a mention of the successful raids of her brother, Mourses or 
Mousour of Tarsos (G 383-85), "who led his army to Smyrna by the 
sea, ravaged Ankyra, the city of Abydos ... and from these victories 
returned to Syria."18 [She is also made to say that Mourses attacked 
"Aphrike (i.e. Tephrike), Terenta, and the Six Towns" (TI]v cE~a_ 

KWJLlav). Tephrike and Terenta (Taranta) were Paulician forts, and 
there was a KaTW cE~a1To'\Lc beside the Euphrates not far from Meli-

15 Georgius Cedrenus 2.206-07, ed. Bonn. 
18 C. A. Trypanis, Gnomon 45 (1973) 615. See also A. B. Lord, The Singer of Tales (Cambridge 

[Mass.] 1960) 215. 
17 E 248-51 Trapp: npalVI!TOV Pertusi, Aevum 36 (1962) 44: n'pVI!TOV E. 

18 Z 621 has instead of Abydos -riJv 'AKWUV (that is, Egin), but Arab raids reached as far as 
Abydos. 



322 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

tene.19 This looks like an inept addition from a catalogue of Comma­
genian names, made by a poet long after the Paulicians' alliance with 
the Muslims had been broken and their redoubts conquered by the 
imperial forces.] Ankyra, Abydos, Nikomedeia and Ikonion were all 
reached in Saracen raids,20 and the mention of them in the Lay adds 
greatly to the verisimilitude of the poem. The Armeniak thema was 
exposed to such frequent attacks that when the Emir's mother 
declares Mousour to have ravaged 'Armenia' (E 257), the poet almost 
certainly refers to the thema; this was not in Armenia, but further 
west, having been formed after troops stationed formerly in Armenia 
had retreated before the Arab advance in the mid-seventh century. 

In 872 the Paulicians met their match in Christopher, Domestic of 
the Schools and the emperor's son-in-law. At Bathyrrhyax north of 
the Halys river on the road to Tephrike the returning rebels were 
caught between the armies of the themata and the picked troops of the 
Domestic. Chrysocheir was killed while trying to make his escape, and 
the Paulician state was now broken: 1T€cOV-rOC OVV 'TOU XpVC6XHpoc 
CVVa1T€pa}J-avO'Y} mxca ~ avOouca -rfjc T€CPptKfjC €vavopta, remarks Ced­
renus.21 

The Emir in the Lay belongs to a context one generation later than 
Chrysocherpes and two later than Ambron. He does not fit into the 
context of the Paulician campaigns but into a later one, later than 872 
but before the victories of Kourkouas; his heroic figure is not, how­
ever, to be forced into an inflexible chronological straitjacket. His 
achievements surpassed even those of his father and uncle in Romania, 
for he claims to have wiped out Herakleia (in Lykaonia); to have 
seized Amorion 'as far as Ikonion' (that is to say, he took Ikonion on 
the way home from a KOUPCOV, since Amorion is northwest ofIkonion); 
to have subdued Syria; and to have captured Kufah, KoiJ4>€p in 
southern Mesopotamia (G 263-66). He even claims to have fought his 
way out of an east Roman encirclement at Mellokopia (Melegob),22 

11 E. Honigmann, Die Ostgrenze des byzantinischen Reiches von 363 bis 1071 (=A. A. Vasiliev, 
Byzance et les Arabes III [Brussels 1935J) Map II. Z 622 has n}v 'E1T7aKw/-dav. 

10 H. Ahrweiler, "L'Asie mineure et les invasions arabes," RHist 227 (1962) 9-10 (repr. 
in her Etudes sur les structures administratives et sociales de Byzance [London 1971] ch.9). 

II Georgius Cedrenus 2.212, 21-22, ed. Bonn. 
II Different forms of the name are found. Ibn Khurdadbih says that the name means 

'quarry of millstones', so he assumes MV)..oKO'TTla to be correct. See H. Gelzer, Die Genesis 
der byzantinischen Themenverfassung (repr. Amsterdam 1966) 85 n.l; H. Gregoire, Byzantion 
5 (1929) 338. For the Greek village see R. M. Dawkins, Modern Greek in Asia Minor 
(Cambridge 1916) 23-25. 



GEORGE HUXLEY 323 

south of the Halys (G 675), thus avoiding the doom of both Omar and 
Chrysocheir. In the poem the Emir symbolizes the Muslims' un­
diminished and confident aggression before the long-delayed East 
Roman counteroffensive brought imperial arms to the gates of 
Edessa and broke the spirit of the frontier Emirates; he has no iden­
tifiable prototype in late ninth or early tenth-century Edessa, and his 
change of religion was an event of a kind often witnessed by akritai on 
both sides of the frontier. 

Frontier emirs and their households could change allegiance with­
out penalty and even religion without pain because the East Roman 
administrators, recognising the Saracens to be superior to other bar­
barians, arranged for generous treatment of converted Muslims who 
settled in Romania.23 Each convert was given a grant of money by the 
military authorities for himself and his ploughing-oxen and a fixed 
amount of corn for sowing and rations. Intermarriage with converts 
was also encouraged: a family that had adopted as son-in-law a 
Muslim who had accepted baptism was entitled to three years' im­
munity from the produce-tax and the hearth-tax, whether the family 
was a military or civilian one. Thus it is with no little verisimilitude 
that in the Lay the Emir is willing to accept baptism in order to 
marry the daughter of Andronikos Doukas and to live with her in 
Cappadocia. 

Since the Lay came into being between the defeat of the Paulicians 
and the victories of Kourkouas, it is reasonable to claim as prototype 
for the Emir's father-in-law the Andronikos Doux or Doukas, an 
ambitious magnate prominent in the reign of the Emperor Leo VI. 
Almost certainly Andronikos had aimed to become emperor himself, 
but having been outwitted by the eunuch Samonas, a minister of Leo, 
he fled with his family and dependants to Kaballa near Ikonion.24 He 
seized the castle there, but later defected with his followers to the 
Saracens of Tarsos, and later still he went to Bagdad. In the Lay 
Andronikos (or, to give him his 'Syrian' name, Aaron) is said by his 
eldest son to have been exiled to the frontier because of military 
insurrection (E 135): Tdv KVptv fLac JgdJp£ccxv 8£!X fLOVp'T£K!X c/>ovccaTa.25 

23 Constantine Porphyrogenitus, De Cerimoniis 2.49, pp. 694-95 ed. Bonn. 
24 R. J. H. Jenkins, Studies on Byzantine History of the 9th and 10th Centuries (London 1970) 

X.230 For the career of Andronikos Doux (?-ca 908) see D. 1. Polemis, The Doukai (London 
1968) 16-21; Polemis, however, doubts that Andronikos aimed at the throne. 

16 8ux f'ovf1Tt!, Kd 8ux c/>ovccara E. COTTo H. Gregoire et M. Letocart: see Byzantion 14 (1939) 
211-26. f'oOpro > f'oOAro, Lat. tumultus. See also H.-G. Beck, Geschichte der byzantinischen 



324 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

Since it is clear that the defection of the historical Andronikos Doukas 
is recalled by these words, the composition of the Lay can be dated 
within narrower limits, between 906 and the victories of John Kour­
kouas. 

There is one more historical detail to be considered, though it does 
not prove what has been claimed from it. When the Emir's mother 
reproves him for having given up all for the love of a Roman girl, she 
asks: "Is there not amongst us the p.aJlSOuJI of Naaman, who was king 
among the Assyrians and deemed worthy of miracles because of the 
number of his merits?" (G 758-60). Gregoire,26 arguing that these 
words refer to the famous p.aJlS~AtoJl of Edessa upon which was 
preserved miraculously the impression of Christ's face,27 held that 
the passage was composed before 944, when the kerchief was brought 
to Constantinople after the army of John Kourkouas had threatened 
Edessa. The city bought immunity for herself, and for Samosata, 
Harran and Sariij also, by giving up the relic.28 

Gregoire's argument would have looked more convincing if the 
p.aJl&-}AtoJl with the impression of Christ's face were the only relic kept 
at Edessa, but the place was a famous repository of religious artifacts 
and ideas. There are three relics to be considered in connexion with 
the plea of the Emir's mother: (1) the Letter of Christ to King Abgar 
of Edessa; (2) the perfect Impression (Ekmageion) of Christ's face upon 
a linen-the p.aJlS~AtoJl; (3) the p.aJlstAtJl of Naaman, mentioned only 
in the Lay (the form of the noun p.aJlotAtJI is close to Arabic mandil and 
corresponds to Latin mantele, our 'mantle'). 

Since a Muslim lady would not wish to support her argument with 
a purely Christian relic if a Muslim, or jointly Muslim and Christian, 
one could be used, it is far from certain that the poet had the famous 
p.aJlO~AtOJl of Christ in mind. Nor should we emend p.aJlo tAtJI (,kerchief') 
to p.aJl'TCXTo ('letter') in G 758, because having replaced a cloth with a 
letter, we would still be left with the name of Naaman, and there 
would be no reason to assume that the poet was alluding to the Letter 

Volksliterarur (Munich 1971) 72. In the original Lay the emperor against whom Andronikos 
rebelled may not have been named. In the Digeneid the author added the emperor's name. 
He was either Basil (G 1007) or Romanos (Z 1319), but these names have no bearing on the 
date of the Lay . 

• 1 Bnantion 6 (1931) 486-90. 
17 For the history of this relic see S. Runciman, CambHist] 3 (1931) 238-52. 
sa Toynbee, op.cit. (supra n.l3) 319. See also J. B. Segal, Edessa 'The Blessed City' (Oxford 

1970) 215-16. 



GEORGE HUXLEY 325 

of Christ to King Abgar of Edessa. It is best to regard the <Kerchief of 
Naaman' as a relic known by the poet to have been venerated by the 
Muslims of Edessa. They could claim that it was older than the 
Letter and the Impression, and a towel fits well into the story of 
Naaman and Elisha, since the Assyrian would have used a p,cxvo{f\£V to 
dry himself after he had obeyed the prophet's command to bathe in 
the River Jordan.29 Elisha was venerated by the Christians, but he 
would have been also a figure of interest to the Muslims, and Naaman 
was a name respected by the Arabs of the Euphratesia even in pre­
Muslim times. Naaman, the tribal leader of the Skenite Arabs, fought 
as an ally of the Persians against the Empire late in the fifth century.30 

The mother of the Emir chooses her words carefully. She politely 
mentions a relic deserving of respect by her son as a Christian as well 
as by herself as a Muslim. A plain towel of Naaman would have no 
Ekmageion of the Assyrian upon it, and being aniconic would be more 
acceptable to Muslim sentiment than any Impression. We need not 
doubt that amongst the relics of Edessa there was a fLcxv8D\w said to 
have been used by Naaman when he had bathed in the River Jordan; 
but obviously the subject has no bearing upon the date of the Lay of 
the Emir, since there is no evidence that a <kerchief of Naaman' was 
ever removed from Edessa. 

The Kerchief in the Lay is significant because it shows that the poet 
knew local details from Muslim frontierlands. His interest in them is 
consistent with his lack of religious bias, which is also evident when he 
causes the brothers of Eirene to add to their plea for the return of 
their sister a polite expression of hope that the Emir may be found 
worthy to venerate the Prophet's tomb (G 74). In fact, as we learn 
from his mother, the Emir had visited the tomb with her and had 
seen in the night a miraculous radiance come from on high and fill the 
house with its light (G 748-51). Thus the akritic poet has knowledge of 
the light over the tomb of Mohammed that leads up to paradise. 
Informed and undogmatic interest in the Arabs and their religion is 
characteristic of the Lay, whose author reveals his fellow-feeling for 
the Saracen opponents of the East Romans by dignifying them too 
with the proud status of akritai: thus a Saracen retainer of the Emir is 
without hesitation called a Daylamite akrites (G 126). That the interest 
in Saracen life caused him to make use of Muslim epics from beyond 

29 Mavrogordato, op.cit. (supra n.1) xxxv. 
30 Theophanes p.141 de Boor. 



326 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

the akra is possible but cannot be proven. One such epic, which lies 
behind the Turkish romance Said Battal, celebrates the historical 
warrior Abd Allah Abu-'} Husain el Antaki el Batta1,31 who died in an 
Arab raid at Akroinos in 740.32 Certain it is however that the Cappa­
docian akra, like the passes of the Pyrenees, being disputed ground 
between Muslim and Christian, were a fertile seedbed of heroic 
poetry. 

In considering the poetical tradition to which the poet of the Lay 
belonged it is not helpful to refer in the first instance to the demotic 
TpayovSux in which Digenes is the hero, because we have no reason to 
believe that such songs about Digenes existed before the original 
Digeneid was composed and because the Lay antedates the Digeneid. 
That a vigorous tradition of poetry existed in Asia Minor indepen­
dently of the literary traditions of the capital is evident from a 
hostile remark of Arethas, Bishop of Caesarea in Cappadocia, who 
died in 932. He rails against "the cursed Paphlagonians who compose 
songs about the adventures of famous men and go from house to 

house singing them for a tip."33 Arethas is a prejudiced witness; 
singers had a respected role in society, and a story about a singer 
called Gyberion shows that their skills could be used even on behalf 
of the imperial government. 

After the defeat of Thomas in the civil war against Michael II some 
partisans of Thomas held out under their leader Gazarenos in Saniana, 
a fortress near the River Halys. The emperor's agent charged with 
the task of recovering Saniana composed a song for Gyberion, who 
sang it outside the gates of the town. The words were directed at the 
steward (oikonomos) of a church or monastery in Saniana and promised 
to him the bishopric of Neocaesarea in Ponrus if he would hand the 
town over: 

" _, I 
aKOVC€, KVPL OLKOVOJL€, 

'rQ1 , \1 
TOV VfJ€ptv, 7"L COV I\EYEL' 
" <:- A \ ~ , 

av JLOL OWC 'T'1}V ..::"avLavav, 
\ , , 

JL'Y]Tp01TOI\L'T'1}V C€ 1TOLCW J 

7\T " <:- , 
J. V €OKaLcap€LaV COL OWCW. 

When the message reached the ears of the steward, he took the hint, 

81 Mavrogordato, op.cit. (supra n.)) lxxii, following H. Gregoire. 
U For the date see E. W. Brooks,jHS 18 (1898) 202. 
33 S. Kougeas, Aaoypa<pla 4 (1912/13) 239. 



GEORGE HUXLEY 327 

and Gazarenos, having been shut out of the town, fled towards Syria; 
but he was captured and hanged.34 Thus poets might please 'country 
folk, or they might aid the imperial administrators. They might also 
be retainers of the akritic nobility. One such retainer, as his interests 
show, was the author of the Lay. His interests are those of his patrons, 
not of the peasantry, and we shall not be far astray if we suppose him 
to have lived and to have written his poem, and perhaps also to have 
sung it, in a castle of Cappadocia sometime in the first quarter of the 
tenth century. Because the author of the Digeneid, who lived perhaps 
half a century later, belonged also to the world of the akritic magnates, 
he adopted the Lay as his own and expanded it, so as to praise the 
akritic nobility, who had become the proud possessors of territories to 
the east of Cappadocia-in Lykandos, Commagene and the Euphra­
tesia.35 

3' Theoph,Contin. pp.71-72 ed. Bonn. See also J. B. Bury, A History of the Eastern Roman 
Empire (London 1912) 107-08, and for the text of the song, K. Krumbacher, Geschichte der 
byzantinischen Literatur (Munich 1897) 793. 

35 Being concerned in this study with the Lay, I have not discussed the Commagenian 
contexts in the Digeneid: the topography of the Digeneid proper was placed on a firm basis 
by H. Gregoire in his account of TrosisjTrusch near Samosata, Byzantion 6 (1931) 499-507. 
There is a substantial study of problems in the Digeneid by Agostino Pertusi, "La Poesia 
epica bizantina e la sua formazione; problemi suI fondo storieo e la struttura letteraria del 
<Digenis Akritas'," in La poesia epica e la sua formazione (AcadLincei quad. 139, Rome 1970) 
481-544. Amongst topics relevant to the Lay he examines the allusion to the mandilin of 
Naaman and points out (op.cit. 496) that it does not provide a secure terminus for dating 
the Digeneid. Other matters discussed by Pertusi are the relationship between the Lay and 
the story of Digenes (op.cit. 519ff.; see also H.-G. Beck, op.cit. [supra n.25] 70-74) and the 
hostile remarks of Arethas about Paphlagonian singers (Pertusi, op.cit. 502). 

Thanks are due to Mr John Haldon for several instructive conversations during my 
tenure in the summer of 1973 of a visiting fellowship at the British School at Athens, and 
to the Managing Committee of the School for having appointed me to the fellowship. 
Dr Judith Herrin kindly discussed the Paulician background with me. 



Digenes and the "Akritai" Huxley, George Greek, Roman and Byzantine Studies; Fall 1974; 15, 3; ProQuest pg. 328

II 
Digenes and the Akritai 

W HEN THE AUTHOR of the Digeneid took over the Lay of the 
Emir, he adapted his continuation of the story to a Com­
magenian context, because the imperial frontier had been 

advanced-during the second quarter of the tenth century-eastwards 
from Cappadocia into the formerly Saracen territories beyond the 
Euphrates. The Digeneid proper describes the precocious youth, heroic 
exploits and peaceful death of Digenes the Twyborn Borderer; as the 
Emir, the father of Digenes, grows old he is no longer the centre of the 
poet's attention. The purpose of the present study is to examine 
historical elements in the Digeneid in order to throw light upon poli­
tical and social circumstances in which it was created. We begin with 
some essential components of the original story, as the creator is 
likely to have told it. 

In the course of his early exploits as horseman and hunter, Digenes 
distinguishes himself in combat with apelatai (abactors or cattle 
rustlers) of the frontier lands, but not before he had expressed the 
wish to be like an apelates-Z'1}Tw K' EYW, va yl.vwlWt we de TWV 
a1TEA&TWv1-as if to emphasize the affinity between the akritic and 
apelatic ways of life on the borders of the Empire. The leader of the 
apelatai is the aged Philopappos, whose name recalls the ancient kings 
of Comma gene; the poet borrows the name from local folkmemory, 
or it is possible that he had read it in an inscription of the Commagen­
ian dynasty surviving to his day. In either case, it is an authentically 
local detail, and the Commagenian background of these apelatai is 
again emphasised when Philopappos and his friends recall their vic­
tory Hat the Euphrates, down at Samosata" Elc TOV 'Acpp&T'1}V 1ToTCtJL6v, 
K&TW Elc TO ECtJLOVC&TO (E 1311). 

Digenes serenades Eudokia, daughter of the Roman general Doukas, 
and unaided, wins her hand. This episode is an obvious adaptation of 
the capture by the Emir, his father, of the daughter of Andronikos 
Doukas. Digenes journeys as an akrites on the borders, the akra, with 
his bride, and rests here or there with her in a tent. He meets the 
emperor, who had been campaigning against the <Persians', by the 

1 Digeneid Z 1569 ed. Trapp (Digenes Akrites [Vienna 1971] 175.) 

32.8 



GEORGE HUXLEY 329 

Euphrates (G 1926); 'Persians' here would be Seljuk Turks, as so often 
in eleventh-century Byzantine historiography, but in Z 2307 the em­
peror is said to be campaigning against the MayovAtot : the name is a 
metathesis of FovAa/LtoL, Arabic ghulam, 'soldiery', and it takes us back 
to the final campaigns against the Saracens in the tenth century, much 
earlier than the Seljuk wars of the eleventh. In the Grottaferrata (G) 
and Russian versions of the story, the emperor's name is Basil, but in 
other texts, including the Andros and Trebizond manuscripts (Trapp's 
Z tradition), the emperor is ROlnanos. No chronological inferences 
about the date of the original Digeneid may be drawn from these 
names, since it is not clear that the creator of the epic gave a name to 

the emperor who honoured Digenes. The title BactA€vc TlJJv'Pw/La{wv 

would have sufficed.2 The poet draws attention to the political and 
military independence of Digenes by telling how the emperor con­
ferred upon him by means of a chrysobull authority to rule the borders 
(G 2002-03 and 2040). The Digeneid thus reflects the ambitions of the 
akriticnoblemen who were the poet's patrons; in the tenth century 
the imperial government was almost powerless to check the growth 
of authority and riches amongst the rural magnates of Anatolia. 

The Digenes of the epic, like the Digenes of the ballads, is a super­
human figure, in contrast to the idealized humanity of the Emir, his 
father, in the Lay. Amongst his deeds of prowess was the rescue of a 
girl of Mepherke (Martyropolis), one of the Arab frontier towns. She 
was the daughter of Melanthia, who has a Greek name, and the 
Emir Haplorrhabdes. Having been abandoned by her seducer, the 
captured son of the Roman general Antiochos, she waited forlorn be­
side a spring and a date-palm in the midst of the waterless plains of 
Arabia. Digenes restored her to her faithless lover and required him 
to marry her, though not before he himself had succumbed to her 
charms.3 The uncertain loyalties of frontier life are well illustrated by 
this story: for Melanthia, the wife of Haplorrhabdes, shows sympathy 
with the captured Roman, and when her daughter gives horses to him 
after setting him free and even makes him an eminent ruler in Syria, 
Melanthia approves, having no fear of her husband the Emir, who 
was away fighting in unceasing wars (G 2116-20). 

I R. M. BartOOan, Vizantiiskii Vremennik 25 (1964) 165-66, points out also that there is 
no allusion to Nikephoros Phokas in the Digeneid: in Z 4205 0 JI£KTJ<Popoc is not the name of 
the emperor but an epithet. 

3 The girl appears with Digenes or with her lover in the desert on plates of about 1200 
from Corinth; Alison Frantz, Bnantion 15 (1940/41) 90-91 and fig.s. 



330 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

When Digenes has defeated Philopappos and his band of apelatai, 
they persuade their kinswoman the Amazon Maximo to help them, 
but Digenes overcame her too. The chief of Maximo's troop of horse­
men, 7TPWTOV a.yovpov EgapxovTa TWV a.AAWV, is called Melimitzes or 
Melimentzes (G 2760-61, Z 3304, E 1390); he is the first to charge 
Digenes, but the hero quickly unhorses him with his staff. It is note­
worthy that the apelatai communicate by means of a system of bea­
cons, a typical detail of life at the frontier passes. 

Digenes in his wandering with Eudokia had come to Trosis in 
Commagene, a day's march west of Samosata. Hereabouts, not far 
from the Euphrates, Philopappos had first seen her (G 2739) and 
hoped to win her for the apelates Ioannikios or Ioannakes, 'Little John'. 
In this neighbourhood, too, after the defeat of the apelatai, the scene 
of the rest of the poem is set. Digenes puts his tomb on a single-arched 
(p,ovoKtpaTov) bridge (E 1648-49) and builds a palace which he sur­
rounds, as any self-respecting East Roman magnate would, with a 
beautiful garden. The body of his father, who had died in the family's 
old home in Cappadocia, is laid to rest in a stone tomb in the grounds 
of the new palace near the Euphrates, and Digenes' mother Eirene 
comes to live with him and Eudokia; evidently the hero's conquests 
had made it safe to remove her eastwards. Before long Eirene also 
dies. 

There is a prominent Commagenian monument on the hill of 
Sesonk near Trosis, and this may well have suggested to a local poet 
the idea of placing the palace of Digenes in the neighbourhood. There 
seems to be, however, no suitable monument near Trosis to represent 
the tomb of Digenes with its sarcophagus under four columns and a 
cupola on the arch of a bridge. The descriptions in G 3608 and E 
1648-544 are, however, sufficiently distinctive and consistent to show 
that the poet had a real structure in mind. 

For the tomb on the bridge we may, with Gregoire,S search no 
further than the splendid single-span structure which crosses the 
Bolan Sii (the ancient Chabinas), a tributary flowing southwards into 
the Euphrates in Commagene. Over the bridge passed the road from 
Melitene to Perre and Samosata. This, the most impressive structure 
of its kind in the heartland of the Digeneid, is surely the original of 
Digenes' arch and tomb. According to the accompanying inscriptions 

, The texts are analysed by A. Xyngopoulos in 'E>J..TJvtKtX 20 (1967) 24. 
6 By{antion 6 (1931) 504-05. 



GEORGE HUXLEY 331 

the bridge had been rebuilt in the time of Septimius Severus. At each 
end two columns were placed; they support nothing, but they may 
well have given to the poet the idea of a canopy supported by four 
columns with four arches over the sarcophagus. 

The death of Digenes came in peacetime, not in war. The hero had 
brought such security and quiet to the akra that his authority, the 
poet asserted, extended into the formerly Saracen domains beyond 
the Euphrates. When the hero had been entertaining some Orthodox 
kinsmen from Amida (Emet), he caught a disease-07TLc8oTovoc, 'the 
back-bending', perhaps a form of meningitis-while taking a bath in 
his garden pool. As he lay dying, he recalled his exploits to Eudokia, 
and after uttering sentiments of unexceptionable Orthodoxy, passed 
away, though not before perceiving that Eudokia, who expired with 
him, had been spared the sorrows of widowhood. 

To his wake came rulers of the East-Charzianians, Cappadocians, 
Boukellariots of the themata;6 worthy Podandites from the kleisoura 
on the Cilician frontier; men of Tarsos and of the Black Mountain of 
the Lebanon (by the latter displaced Mardaites may be intended); the 
elect of Bagdad; men of Bathyrrhyax (where troops of the Armeniak 
thema assembled); nobles from Babylon and many from Amida 
(G 3570-75). 

Not only does the poet show by means of his list of mourners that 
Digenes was honoured on both sides of the frontier, he almost wishes 
it away, for he thinks of the Twyborn Borderer as an independent 
ruler whose power was felt far beyond Commagene; his view of 
Digenes suits the historical context, since clearly defined boundaries 
in mountainous territory are for the most part inventions of modern 
nationalist states. The poem even calls Digenes {JacLA€VC, a title 
jealously guarded in Constantinople for the emperor himself; many, 
he says, did rejoice in the kingship of Digenes (G 3365). It is clear that 
the original poet was keen to idealize the aspirations to autonomy 
amongst his patrons in the frontier nobility, of whom it has been well 
said: "The great aristocratic families of the eastern marches were a 

6 The text of G 3572 has KOuKouAdJaptwrat, which Mavrogordato takes to be a corruption 
of BovKEA/..aptwrat, to match the other names of themata (Digenes Akrites [Oxford 1956] 244). 
D. Wood suggests that KOUKovA£Oap£wra£ means <men from the Monk's-Hood Rocks' 
(from KOUKovM£Ov, <cown-that is, from the district of Cappadocia (Byzantion 28 [1958] 

91-93). The poet, however, knows people of this district as TpwYAoMra£ (G 17), and there 
was a place called BOVKOU Al80e <*BOUKouA£8ap£, east of Caesarea, in the Lykandian district 
(Cedrenus [Skylitzes] 2.421 ed. Bonn). 



332 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

constant and a growing menace to the throne during the ninth, tenth 
and eleventh centuries. Under a weak sovereign their heads were 
practically independent despots, as were the Arab Emirs who con­
fronted them across the border."7 The poet's phantasy even allows 
his hero to take Ikonion from the Empire, so that Digenes may remit 
to the emperor tribute once paid to the city by the imperial govern­
ment (G 1993-95); in return for the Emperor's honouring of Digenes 
the people of Ikonion are to pay back «as much again from their un­
willing hands" as the Empire has paid to the city. There is no need to 

regard the tribute paid to Ikonion as an allusion to the Seljuk Sultan­
ate of Rum, since the dues paid to Digenes could well have entered 
the story long before the disaster at Manzikert in 1071, which led to 

the Seljuk conquest of central Anatolia. 
The elaborate descriptions of Digenes' palace are a prominent fea­

ture of the G and Z traditions of the poem. They reveal signs of 
literary reworking, and indeed so close are they to the allegorical 
poem To Sophrosyne of Meliteniotes that G, Z and Meliteniotes must 
share a common source-if in fact G and Z do not borrow directly 
from Meliteniotes. The author of To Sophrosyne may well be the 
Megas Sakellarios Theodoros Meliteniotes who died ca 1397,8 but his 
poem can tell us nothing about the description of the palace near 
Trosis in the original Digeneid. According to the E version, in which 
a connexion with Meliteniotes is not evident, there were battlements 
(7T'pop.axul)JJ€C E 1618) in the garden wall. The wall itself was so care­
fully made of polished stones in their op()op.app.apwctc that they 
gleamed from afar, and it was adorned with strange carvings, for it 
was g€JJOxapayoc.9 Within the well-watered demesne there were trees 
as well as flowers and vines. In the trees parrots sang, hailing him 
with his bride (E 1639-40), and in the 'TpayovSLa, but not in extant 
versions of the epic, birds foretell the hero's death. The garden and 
castle are closely linked in an akritic song from Pontos-' AKpt'T€C 
KaC'TpOJJ EX'TL'EJJ, • AKpt'TEC 7T'EPL{36'ALJJ .10 

The original Digeneid may have been content to describe briefly the 
demesne and castle; but that there were contemporary tenth-century 

7 R. J. H. Jenkins, Studies on Bnantine History of the 9th and 10th Centuries (London 1970) 
X.219. 

8 H..G Beck, Geschichte der byzantinischen Volksliteratur (Munich 1971) 125, following 
F. DOlger. 

• A. Xyngopoulos, Aauypaq,la 12 (1948) 560, explains the adjective. 
10 N. Polites, Aaoypaq,la 1 (1909) 245, no.35. 



GEORGE HUXLEY 333 

estates in Anatolia worthy to be described in elaborate detail and to 
be claimed in the poem for the akrites himself is evident in the 
Armenian history of Thomas Artzruni and his continuator. They des­
cribe the buildings at Vostan on the south shore of Lake Van and on 
the isle of Aght' amar in the lake, where the Church of the Holy 
Cross was built between A.D. 915 and 921 by King Gagik of Vaspura­
kan.n The palaces have been destroyed, but the church still stands. 
Gagik built walls about his demesnes; he led water to his gardens; 
his palaces had courtyards and high towers; his buildings and estates 
in fact provide an apt prototype for Digenes' palace, especially since 
they had come into existence not long before the creator of the 
Digeneid was at work in the tenth century.12 Their fame must have 
spread far beyond the shores of Lake Van to reach the Greeks and 
Armenians of Commagene and the Euphratesia two hundred and 
fifty miles to the west. 

The author of the Digeneid gives no sign of having been an Armenian 
himself; but he was interested in Armenians nonetheless. He or the 
author of the Lay gives the Armenian name Mouselom to the father 
of Andronikos Doukas, the father-in-law of the Emir; and Melimen­
tzes, the leader of Maximo's band of apelatai, also has an Armenian as­
pect. It is cognate with Armenian Mleh (or Melias), and Mleh-me(n)tz 
in Armenian would signify <great Mleh'.13 Amongst the several 
Armenian warriors called Mleh or Melias the obvious prototype for 
Melimentzes in the epic is the frontier general Mleh, who in the time 
of the Emperor Leo VI resettled the desolate town of Lykandos and 
ruled its neighbourhood as a kleisoura. Later during the imperial 
drive to recover the Euphrates limes, Lykandos became a thema and 
Melias its strategos.14 The country between Cappadocia and Melitene 
prospered under his rule, and so Tzamandos was also replenished 
with Armenian dependants of Mieh. The Lykandian thema was well 
suited to stockraising and horse breaking, as Constantine Porphyro­
genitus makes clear in a glowing description: Kovp6TPoc/>OC oVcCt ayCt8~ 

"'f3 ' , f3 ' , ~A.. ' , , '" 15 T€ KCtt t7T'7T'O OTOC KCtt 7T'CtVTOtWV OCK'Y]ILCtTWV EtC TpV<f1'Y]V €7T'tT'Y]U€toC. 

There is realism in the presentation of Mleh in the epic as an 
11 For the church and the account of Thomas Artzruni see Sirarpie de Nersessian, Aght-

'a mar, Church of the Holy Cross (Cambridge [Mass.] 1965). 
12 A. A. M. Bryer, Antiquity 34 (1960) 295-97. 
13 N. Adontz, BZ 29 (1929/30) 216. 
14 E. Honigmann, Die Ostgrenze des byzantinischen Reiches (Brussels 1935) 64-67. 
15 De thematibus 1.12 (p.72 Pertusi). 



334 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

apelates. In frontier societies cattle-droving and cattle-stealing can be 
overlapping activities, and as governor of Lykandos, far from the 
supervision of the central bureaucracy, Mleh would have possessed 
opportunities and authority to campaign on his own behalf, even 
against other akritic magnates. The appearance of Mleh of Lykandos 
in the Digeneid is thus a historical vestige in the poem datable before 
the mid-tenth century. The name Mleh the Great was long remem­
bered on the frontiers, and he is recalled to this day in the name of the 
nomadic tribe who live on the borders of Cappadocia and Cilicia and 
are called <sons of Melimentzes' (Melementzoglou or M€A€JL€l'S'6y. 

AOV).16 

Lykandos is also mentioned in the Digeneid, though not in connexion 
with Melimentzes. When Digenes has taken Eudokia from her 
parents' home, the general urges his men to hasten in pursuit of the 
<all-bastard' (E 912-13): 

'- " - A ' ,-, , , fJ" ••• cxyovpot CX1TO TOV VKavTOC, cxyovpOt CX1TO TrJl' 'y/\cxv, 

tfJ 0 ' t" ", - , ~, OT] T]CCXT€ €tC TOl' 1TCXYK01T€/\OV, €1TTJP€V TO 1TCXWtV JLOV. 

Lykandos was waste land before the resettlement by Mleh. The poet 
thinks of the country as watched by the retainers of Doukas the 
General, and as the mention of the fJtyAcx in E 912 shows, he treats 
Lykandos as though it embraced a kleisoura at the limit of the general's 
lands in Cappadocia. 

In the service of Doukas the General had been a certain Soudales, a 
Saracen. He was killed by Digenes as he tried to recover Eudokia for 
her father after the flight from the castle,17 A historical Soudales was 
one of the generals sent out by the Empress-Regent Theodora against 
the Paulicians. Skylitzes gives the persecutors' names as Leon son of 
Argyros, Andronikos son of Doukas and Soudalis,18 but in Theophanes 
Continuatus they are called <son of Argyros and the son of Doux and 
Soudales'.19 Mavrogordato suggested that <Soudales the Saracen' in 
the epic is taken from Skylitzes, because the conjunction of the names 
Soudalis and Andronikos Doukas is found in Skylitzes but not in 

16 P. Karolides, 'E7TLC7TJJ.LOVLK7J 'Em::TTJPtc 'E9vLKOV IIavE7TLc7TJJ.Liov 1905/06 (Athens 1906) 243; 
Mavrogordato, op.cit. (supra n.6) liii. 

17 Z 1987, E 919. 
18 Cedrenus 2.154 ed. Bonn. 
19 p.165, 17-18, ed. Bonn. Skylitzes supplies the name Andronikos himself, not from a 

source: see D. 1. Pole mis, Byzantinoslavica 26 (1965) 76. 



GEORGE HUXLEY 335 

Theophanes Continuatus ;20 and he argued that because Skylitzes is a 
contemporary of Psellos, the presence of the two names Andronikos 
Doukas and Soudales in the poem is a hint of an eleventh-century date 
for the composition of the Digeneid. The argument is not a strong one, 
because in the poem the father-in-law of Digenes is called Doukas the 
strategos, not Andronikos Doukas, and Soudales the Saracen is said 

to be a retainer of this Doukas, the strategos. Andronikos Doukas in 
the epic is based on the rebel against the Emperor Leo VI, not on the 
account in Skylitzes of the Paulicians' persecutors. It is quite likely 
that Doukas the strategos and Sou dales the Saracen are borrowed 
from the Doux and Soudales of Theophanes Continuatus, so that the 
name Soudales could have been taken into the epic as early as the third 
quarter of the tenth century, since the anonymous author of Books I 
to IV of Theophanes Continuatus worked under the supervision of 
Constantine Porphyrogenitus.21 

If the poet took the name Sou dales out of Theophanes Continuatus, he 
did not have to worry that it had been borne by a persecutor of the 
Paulicians. Nor is there any difficulty in a Saracen being a retainer of 
Doukas the General. Many a frontier family would have used renegade 
Arab akritai as lookouts because they knew the ground. The epic 
recognises that Soudales the Saracen is a {3ty>"&'TWp, because it states 
that he came &7T6 T~V IUpav Bly>..av (E 919, Z 1987); the name Sou dales 
does not have a Hellenic aspect, and may even be of Saracen origin.22 

Digenes, too, keeps watch in a {3ly>..a (G 2809, Z3350), as many an 
akritic nobleman must have done when campaigning against apela­
tai or mounting guard in the hope of lifting booty. Plunder, indeed, 
must have provided much of their income. The frontier magnates 
were as eager as the SwaTol in Anatolia generally to tax the peasantry 
on their estates, and thus they hastened the economic ruin of the 
soldiering folk from whom the armies of the themata had been re­
cruited; but on the borders there were also repeated opportunities to 
acquire property by military force, either by raiding with private 
armies beyond the East Roman territories or by attacking Muslim 
invaders as they returned homewards laden with booty. Prisoners 

20 Mavrogordato, op.cit. (supra n.6) lxxxiii. 
21 Gy. Moravcsik, Byzantinoturcica J2 (Berlin 1958) Quellen, 540. 
22 Karolides compares Arabic sad, soudoun (='tribal chief'): Epet. 3 (1926) 332. Soudalis 

could represent (1) Syed Ali or Sad Ali; (2) Said Ali (said from sa'ad meaning 'happiness' 
and 'fortune'); or (3) Saud Ali (saud, as in the name King Saud, again from sa'ad). I thank 
my colleague Dr S. Q. A. M. A. Hossain for advice about these names. 



336 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

were the most valued of such prizes because they could be freed on 
receipt of a ransom or sold into slavery. 

It is difficult to overestimate the value of the Digeneid to the social 
historian. In the poem we are given an authentic and indigenous con­
ception of the akritic spirit. Doukas the General has a large armed 
force of his own (G 1560). In his castle there is a numerous staff of 
housemaids (G 1564). His daughter Eudokia's dowry includes twenty 
centenaries of ancient coins, silver vessels, a wardrobe worth five 
hundred litrai, thirty-six estates with their income, four hundred 
prizewinning animals, eighty grooms, fourteen cooks, fourteen bakers, 
and one hundred and fifty other serfs (.f;vx&.pw.) (G 1657-65). The poet 
understands the significance of dowries in strengthening the ties be­
tween powerful families, though the hero claims that he desires 
Eudokia only for her beauty (G 1695); the bonds are made tighter 
still when the contract for the dowry was completed with the emir 
(G 1844) and the general and his family presented other magnificent 
gifts (G 1845-82). Amongst the general's contribution were mules, 
palfreys, grooms, twelve hundred hunting leopards,23 twelve hawks 
brought from Abasgia and twelve falconers with a falcon each. The 
importance of hawking in the akritic life is plain from the frequent 
mentions of hawks and other birds in the imagery of the poem: in the 
Lay of the Emir Eirene's youngest brother even dreams of hawks and 
of a falcon chasing a dove (G 447-54, Z 713-20). Akritic life on the 
northwest frontier of India is similarly devoted to falconry; the re­
mark that Dir (between Chitral and Swat) is a state where practically 
every hawk has a man's wrist under it24 is quite in the spirit of the 
Digeneid. Falcons were prized as much by the Saracens as by the 
Romans, as we see from the Arab historian Al Tabari, who reports 
that when Nikephoros I ransomed from slavery a girl of Herakleia 
whom he desired as a daughter-in-law, he sent to Al Rashid 50,000 
Islamic drachmai upon a bay horse, one hundred silk garments, two 
hundred embroidered garments, twelve falcons, three hunting dogs 
and three horses.25 

13 A trained feline, perhaps a hunting leopard, is shown walking beside a man on a 
plate from Corinth: Frantz, op.cit. (supra n.3) 89 and fig.6. A plate from the Athenian 
Agora shows a falcon perched on the scabbard of a warrior who may be Digenes (Frantz, 
art.cit. 88-89 and fig.3). 

14 Peter Mayne, quoted by Terence Creagh Coen, The Indian Political Service, A Study in 
Indirect Rule (London 1971) 178 note. 

26 E. W. Brooks, ERR 15 (1900) 746 (under A.H. 190 = 27 Nov. 805-16 Novo_806). 



GEORGE HUXLEY 337 

The riches of a dowry such as Eudokia's could be maintained only 
in a large household and vast landed estate, and when the poet 
provides Digenes with a palace, garden and lands fit to vie with the 
finest in the Empire, poetical licence hardly overpasses realism. The 
East Roman nobility were proud of their gardens. using them to 
express their power and riches, and the frontier aristocracy shared the 
pride in full measure. In his will written in 1059 the magnate Eusta­
thios Boilas records that when he moved from Cappadocia to found 
a new estate in Taiq, an Armeno-Georgian district of the thema of 
Iberia H ••• I built my house and the holy temple from the founda­
tions, and <laid out) meadows, parks, vineyards, gardens, aqueducts, 
small farms, watermills ... "26 The estate of Boilas was magnificent, 
yet he was oppressed by neighbours even more powerful than himself, 
and it is easy to understand why the central government was not 
able significantly to curb the inherited privileges and to deplete the 
riches of the frontier nobility, even in the reign of the vigorous 
Emperor Basil II. Digenes the {3aCLA.€VC with his private band of 
warriors, his conquests of the apelatai and his fellow-feeling for them, 
is a heroic exemplar of the challenges and opportunities afforded by 
frontier life. Digenes' palace is a fit dwelling-place for a fictitious 
Borderer of might and distinction in the period of confidence between 
the Byzantine advance beyond the Euphrates and the first raids of the 
Seljuks on Vaspurakan (which may have begun as early as 1029).27 

In Asia Minor concentrations of wealth in the estates of the (}vva'TOL 

caused a ruinous diversion of resources from the central treasury of 
the Empire, but confiscations were exceptional, and riches begat more 
riches at the expense of the peasantry and of the government. The 
tenth-century poet has no thought for the peasantry; he does not 
even trouble to introduce a Thersites, so devoted is he to themes 
designed to please his noble patrons; moreover, the imperial govern­
ment is a peripheral, and for the most part irrelevant, aspect of his 
world. 

The power of the akritic nobility was barely touched by the social 
legislation of the Macedonian emperors, and the exemplary punish­
ments of certain families by the Emperor Basil II had no lasting 

26 S. Vryonis, DOPapers 11 (1957) 263-77, at 266. 
17 For this date: A. J. Toynbee, Constantine Porphyrogenitus and his Age (London 1973) 

410 n.l. 



338 ANTECEDENTS AND CONTEXT OF DIGENES AKRITES 

effect upon their peers.28 The Svva/Tol were almost beyond challenge 
once the peasantry had decided that it was better to be the dependant 
of a local magnate than a taxpaying, but nominally free, subject of 
the government. The ruin of the theme-system was an inexorable 
consequence of this transfer of allegiance by the peasantry during the 
tenth century. The will of the Greek rural population to resist foreign 
invasion of Asia Minor was consequently sapped, so that after the 
battle of Manzikert in 1071 the Seljuk was seen as a deliverer by many 
humble Greeks of the countryside. Such was the grey reality of rustic 
life amongst the 7TtV'T'J'Tf:C. In contrast, the Digeneid gives us a poetically 
composite, yet authentic, picture of the akritic nobility; theirs was a 
rich and colourful world, worthy to be praised by an epic poet. The 
author of the Digeneid was not concerned with social and economic 
wrongs; like Homer, he aimed to please powerful patrons, and 
Hesiodic railings against 'gift-devouring princes' were not his business. 

THE QUEEN'S UNIVERSITY OF BELFAST 

March,1974 

18 Basil n's Novel no.29 of 996 has an addendum naming as possessors of excessive 
wealth amongst the aristocratic landowners the families of Phokas and Maleinos in Cappa­
docia, and an Arab itinerary reports that the estates of the Maleinoi were enormous 
(G. Ostogorsky, DOPapers 25 [1971] 7); in Ostrogorsky's words (loe.cit.), "Basil n was the 
last Emperor who attempted to check the upsurge of the landed aristocracy." In G 1925 
an Emperor Basil, who, during his 'Persian' campaign, desires to meet Digenes, is said to 
have "buried with himself the royal glory"--O Ka~ cw8&t/Jac P.€O' EaVTov -ri}v f3actAnov aogav. 
This can hardly be other than a reference to Basil II by the XIVth century redactor of the 
Grottaferrata text, but the words do not of course, prove that the original version of the 
Digeneid proper came into being later than the death of Basil II in 1025. For the social crisis 
caused by the growing power of the landed 8vva-rol in the Xth century see Toynbee, op.cit. 
(supra n.27) 145-76, and especially G. Ostrogorsky, History of the Byzantine State, tr. Joan 
Hussey (Oxford 1958) 269-83.