Византия, болгары и народы Украины в 890 г.
This paper offers some observations about the context of Crimean history at the end of the ninth century. It continues the story of developments in the steppe world outlined elsewhere in this volume by Alexander Aibabin and Constantine Zuckermann.
Збережено в:
| Дата: | 2000 |
|---|---|
| Автор: | |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Кримське відділення Інституту сходознавства ім. А.Ю. Кримського НАН України
2000
|
| Назва видання: | Материалы по археологии, истории и этнографии Таврии |
| Теми: | |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/169990 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Византия, болгары и народы Украины в 890 г. / J. Howard-Johnston // Материалы по археологии, истории и этнографии Таврии: Сб. научн. тр. — 2000. — Вып. VII. — С. 342-356. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-169990 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1699902020-07-04T01:28:00Z Византия, болгары и народы Украины в 890 г. Howard-Johnston, J. История This paper offers some observations about the context of Crimean history at the end of the ninth century. It continues the story of developments in the steppe world outlined elsewhere in this volume by Alexander Aibabin and Constantine Zuckermann. 2000 Article Византия, болгары и народы Украины в 890 г. / J. Howard-Johnston // Материалы по археологии, истории и этнографии Таврии: Сб. научн. тр. — 2000. — Вып. VII. — С. 342-356. — англ. 2413-189X http://dspace.nbuv.gov.ua/handle/123456789/169990 en Материалы по археологии, истории и этнографии Таврии Кримське відділення Інституту сходознавства ім. А.Ю. Кримського НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| topic |
История История |
| spellingShingle |
История История Howard-Johnston, J. Византия, болгары и народы Украины в 890 г. Материалы по археологии, истории и этнографии Таврии |
| description |
This paper offers some observations about the context of Crimean history at the end of the ninth century. It continues the story of developments in the steppe world outlined elsewhere in this volume by Alexander Aibabin and Constantine Zuckermann. |
| format |
Article |
| author |
Howard-Johnston, J. |
| author_facet |
Howard-Johnston, J. |
| author_sort |
Howard-Johnston, J. |
| title |
Византия, болгары и народы Украины в 890 г. |
| title_short |
Византия, болгары и народы Украины в 890 г. |
| title_full |
Византия, болгары и народы Украины в 890 г. |
| title_fullStr |
Византия, болгары и народы Украины в 890 г. |
| title_full_unstemmed |
Византия, болгары и народы Украины в 890 г. |
| title_sort |
византия, болгары и народы украины в 890 г. |
| publisher |
Кримське відділення Інституту сходознавства ім. А.Ю. Кримського НАН України |
| publishDate |
2000 |
| topic_facet |
История |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/169990 |
| citation_txt |
Византия, болгары и народы Украины в 890 г. / J. Howard-Johnston // Материалы по археологии, истории и этнографии Таврии: Сб. научн. тр. — 2000. — Вып. VII. — С. 342-356. — англ. |
| series |
Материалы по археологии, истории и этнографии Таврии |
| work_keys_str_mv |
AT howardjohnstonj vizantiâbolgaryinarodyukrainyv890g |
| first_indexed |
2025-07-15T04:58:21Z |
| last_indexed |
2025-07-15T04:58:21Z |
| _version_ |
1837687637768404992 |
| fulltext |
Материалы по археопогии, истории и этнографии Таврии. Вып. VII
J. HOWARD-JOHNSTON
BYZANTIUM, BULGARIA AND THE PEOPLES OF UKRAINE IN THE 890s
This paper offers some observations about the context of Crimean history at the end of the
nintti century. It continues the story of developments in the steppe world outiined elsewhere in
this volume by Alexander Aibabin and Constantine Zuckermann. Attention is first directed at the
Balkans where war broke out between Byzantium and Bulgaria in the 890s. Associated diplo
matic activity by both belligerents reached deep into the steppes beyond the Danube. There it
triggered, or at any rate conbibuted to a dramatic set of interrelated events, which transformed
the face of Ukraine. The Pechenegs replaced the Hungarians as the dominant power between
the Don and the Danube and, consequently, as the people who thenceforth wielded authority
over the greater, steppe part of the Crimea. The ramifications of this geopolitical change ex
tended to central Europe, where the Hungarians, ejected from their previous homeland in Ukraine,
established a durable state in the Carpathian basin and secured their position there by a flurry of
wide-ranging, devastating raids further west and south.
But before we can observe these events, let alone interpret them, we must subject the
principal historical source, which reports them to close critical scrutiny. The beginning of
wisdom for the historian of the steppes as for the historian of Byzantium is willingness to look
at every piece of scarce evidence with a cold, clinical eye and, if necessary, to discard what
ever is demonstrably or probably unreliable - although such a procedure will inevitably de
plete yet further our already meager store of primary matehaP.
' В.Н.Златарски “Известията за българите в хрониката на Симеона Метафраста и Логотета”,
Сборник за народна умотворения и книжнина, XXIV.1 (1908), рр.82-147 laid the foundations for
all subsequent discussion, including his own История на българскита държава през средните
векове, 1.2 (Sofia, 1927), рр.282-323. The relatively abundant literature on the subject produced in
the middle and late twentieth century includes: S.Runciman, A History of the First Bulgarian Empire
(London, 1930), pp.143-149; G.Kolias, Leon Choerosphactes, magistre, proconsul et patrice:
biographie - correspondence (texts et traduction), Texte und Forschungen zur byzantinisch-
neugriechischen Philologie, XXXI (Athens, 1939), pp. 21-42; G.Cankova-Petkova, “Dererste Krieg
zwischen Bulgarien und Byzanz unler Simeon und die Wiederaufnahme der Handelsbeziehungen
zwischen Bulgarien und Konstantinopel”, Byzantinische Forschungen, III (1968), pp.80-113;
D.Obolensky, The Byzantine Commonwealth (London, 1971), pp. 104-106; R.Browning, Byzantium
and Bulgaria (London, 1975), pp. 57-60; A.Bo\ziov, “A propos des rapports bulgaro-byzantins sous le
tzar Sym6on (893-912)”, Byzantinobulgarica, VI (1980), pp. 73-81; J.VA.Fine, The Early Medieval
Balkans (Ann Arbor, 1983), pp. 137-140; and J.Shepard, “Symeon of Bulgaria - Peacemaker”,
Гэдишник на Софийская Университет “Св. Клиьяент Охридски”, Научен център за славяно
византийски процчивания “Иван Дуйчев”, LXXXIII.3 (1989), рр. 14-19. These historians are all
ready to place considerable faith in the principal source, the Logothete’s chronicle.
342
I. The chronicle o f Sym eon Logothete
The principal source for the history of the west Eurasian steppes in the 890s is the Byzantine
universal chronicle, which has been ascribed, rightly or wrongly, to Symeon Logothete, following
an attribution first made by a twelfth-century scribe. It survives more or less complete in eight
manuscripts. Another seven retain only the last part, covering the period 842-948, and attach it
to the end of the often-intemperate chronicle of George the Monk, which comes to a halt in 842.
Fragments of this continuation of George the Monk are preserved in another five manuscripts^.
A critical edition is at long last in preparation and should be published, at the latest, by the
beginning of the next millennium.
The Logothete’s Chronicle (as I shall call it henceforth, for convenience) supplies the only
connected account of the history of Byzantium and its relations with neighbouring peoples from
the death of Basil I in 867 to the fall of Romanos I Lekapenos in 944, in effect most of the crucial
first phase of a period of sustained cultural and political revival which extended into the middle
of the eleventh century. This last section of the chronicle (which ends with a cast-forward to the
death of Romanos in 948) was analysed with admirable care and good sense by the late Alex
ander Kazhdan^. Quite rightly he stressed that it was given shape by the various sources which
successively contributed material. Quite rightly he distinguished the original text from a revised
and stylistically upgraded edition preserved in four manuscripts, which extended the coverage
to 963 and introduced a pro-Phokas spin into late ninth- and tenth-century history*.
Kazhdan was wrong, though, to think that the last section was written by a contemporary. 1
find no trace of first-hand, eyewitness experience in the text Even a fleeting comparison of the
Logothete’s account of the Rus attack on Constantinople with that of Liudprand, whose step
father was present in the city at the time, is telling: Liudprand’s account is the more vivid and the
more narrowly focussed, while the Logothete presents a well-ordered overview of what hap
pened, probably based on official communiquiis issued at the time®. There is every sign that
written sources were being quarried for the last section of the text as they had been beforehand.
This suggests that the Logothete, whoever he was, was not at work soon after the last recorded
event, the death of Romanos Lekapenos in 948, unless he had an unusual capacity to cauterise
his memory. Compilation of the chronicle should probably be dated a generation or so later, and
placed in the early years of Basil ll’s reign, in the era of concerted and dangerous rebellion by
Anatolian aristocrats against the crown (976-89)®.
̂A.Sotiroudis, Die handschriflliche Oberiieferung des “Georgius Continuatus”(Redaktion A) (Thessaloniki,
1989). The attribution to Symeon Logothete is made first in Mosquensis Bibl. Syn. gr 251 (dated 1152),
next in the fourteenth-century Vindobonensis Historicus gr. 37 (Sotiroudis, pp. 52, 46).
’ А.П.Каждан “Хроника Симеона Логофета”, Византийский временник, XV (1959), рр. 125-143.
“ A.Markopoulos, “Sur les deux versions de la Chronographie de Symeon Logoth6te”, Byzantinische
Zeitschhft, LXXVI (1983), pp. 279-284 adds a fourth manuscript to the three discussed by Kazhdan.
® Liudprandi Antapodosis, V.15, ed. J.Becker, MGH, Scriptores in usum scholarum (Hanover, 1915),
pp. 137-139; Georgius Monachus Continuatus, ed. I.Bekker, Theophanes Continuatus, loannes
Cameniata, Symeon Magister, Georgius Monachus, CSHB (Bonn, 1838), pp. 914-916 (cited
henceforth as GMC).
® This contention entails redating the sixth book of Theophanes Continuatus, which uses and amplifies
the last part of the Logothete’s chronicle and therefore provides a terminus ante quern for its
composition. This concluding book of Theophanes Continuatus is usually dated to the reign of
Nikephoros Phokas (963-69), since his successor John Tzimiskes is described merely as Domestic
of the Scholai (a post which he held under Nikephoros). The reference to Tzimiskes may, however,
have been lifted and mechanically reproduced from a biography of his relative John Kourkouas,
written by a certain Manuel protospatharios and judge, which is summarised by Theophanes
Continuatus (ed. Bekker, pp. 426-429, with the reference to Tzimiskes at p. 428.15-20). If so, it provides
a terminus ante quern for the biography, not for Theophanes Continuatus.
______ Howard-Johnston J. Byzantium, Bulgaria and the Peoples o f Ukraine in the 890 s.
343
Familiarity with the Logothete’s Chronicle does not inspire unquestioning confidence. It is
highly polemical, in effect an anti-history of the Macedonian dynasty founded by Basil I (867-86).
It is so lurid in places that it prompted occasional exclamations of horror from the scribes who
copied it out in the later Middle Ages^. Psogoi, pieces of invective containing much invention
and targeted on named individuals, are recycled. Kazhdan highlighted a full-blown diatribe against
Basil I, which forms the core of the chronicle’s account of his reign. Other butts include the
Emperor Alexander (912-13), dismissed in two pages as a debauchee and madman, and the
patriarch Nicholas Mystikos, accused (in my opinion, quite unjustifiably) of crowning the Bulgar
ian ruler Symeon Byzantine emperor in autumn 913®.
There are gaping holes in its coverage. Nothing, for example, is said about relations with the
Rus around the years 907-911 when Byzantium negotiated two treaties with them®. There is
nothing about Symeon’s successful invasion of Serbia (probably to be dated to 922) and the
subsequent consolidation of Bulgar rule there over a seven year period'®. Nor is a word breathed
of the military operations, directed by John Kourkouas, which followed up the capitulation of
Melitene in 934 with a sustained offensive push into the Arab marches, although it was these
campaigns which first gave real impetus to Byzantine expansion in the eas t". These are simply
three of the most striking omissions, which may be detected. The chronicler seems to have
been at the mercy of the sources available to him, and, towards the end, to have relied out of
choice increasingly on a single source, a biography, almost hagiographic in character, of Romanos
Lekapenos.
This last part of the chronicle is also marred by chronological waywardness, as even the
staunchest defenders of its accuracy, Jenkins and Treadgold, are forced to concede. There are
demonstrable errors, and, no less disconcerting and confusing, abrupt swings backward and
fonvard through time. The narrative jerks the reader hither and thither, rather than taking him on
a steady progression through the years as might be expected in what is ostensibly an annalistic
chronicle. The general principal of arrangement is, indeed, chronological, but the Logothete has
an evident penchant for grouping related notices into thematic clusters, which cut across time.
There is so much flouting of chronological order that it is hard to believe in the thesis pro
pounded by Jenkins and Treadgold and defended by them with great ingenuity, that the Logothete
drew much of his material from a set of annals which supplied precisely and accurately dated
items of information'^. The very existence of such annals may also be doubted, since there is
' Expressions of shock are to be found in Mosquensis Bibl. Synod, gr. 251 (1152), Constantinopolitanus
(Topkapi Serai) 37 (14th century) and Vaticanus gr. 1807 (14th century): Sotiroudis, pp. 52, 54, 42.
' Kazhdan, p. 139 (Basil I); P.Karlin-Hayter, “The Emperor Alexander’s Bad Name”, Speculum, XLIV
(1969), pp. 585-596; Nicholas Mystikos is attacked twice during his period as co-regent from June
913 to February 914 (GMC, pp. 877.5-9, 878.3-6).
® S.H.Cross & O.P.Sherbowitz-Wetzor, tr.. The Russian Primary Chronicle. Laurentian Texf (Cambridge,
Mass., 1953), pp. 64-69; S.Franklin & J.Shepard, The Emergence of Rus 750-1200 (London-New
York, 1996), pp. 103-107.
Gonstantinus Porphyrogenitus, De Administrando Imperio, ed. Gy.Moravcsik, tr. R.J.H.Jenkins
(Washington, D.C., 1967), c. 32.99-132 (cited henceforth as DAI). The dating is mine, reckoning the
reign of Paul (installed as puppet ruler by Symeon) as lasting three years in total (917-20) and placing
the two Bulgar attacks on his successor Zacharias in 921 and 922.
” The historian’s appetite is whetted by a summary of John Kourkouas’ achievements as retailed by
his encomiastic biographer Manuel (Theophanes Continuatus, ed. Bekker, pp. 426-427); a rather
blinkered account of events in the years 934-940, based on Arab sources, is given by A.A.Vasiliev,
Byzance et les Arabes, rev. M.Canard, 11.1 (Brussels, 1968), pp. 273-290.
R.J.H.Jenkins “The Chronological Accuracy of the “Logothete” for the years A.D. 867-913”,
Dumbarton Oaks Papers, XIX (1965), pp. 91-112; W.T.Treadgold “The Chronological Accuracy of the
Chronicle of Symeon the Logothete for the Years 813-845”, OOP, XXXIII (1979), pp. 157-197.
Материалы по археологии, истории и этнографии Таврии. Вып. VII______
344
no positive attestation of their existence and since they would have served little purpose in a
centralised bureaucratic state which had both the means and the incentive to disseminate infor
mation and news in regular bulletins''®. It follows therefore that there should be no presupposi
tion of general chronological accuracy on the Logothete’s part.
II. W ar in the Balkans and Ukraine in the 890s: the Logothete’s account
The Logothete devotes a long, consolidated notice to the dramatic events, which altered the
balance of power south of the Danube and transformed the geopolitical configuration of the
steppes to its east in the 890s. This constitutes the only extant connected account of these
related episodes in Balkan and Ukrainian history. For ease of reference, citations will be to the
text of a late but accurate manuscript of the Continuation of George the Monk, published in the
Bonn Corpus and therefore easily accessible’^ . There are no significant points of difference
between this and the text of the only early manuscript of the full chronicle to have been pub
lished, an eleventh-century manuscript, which a fifteenth-century scholiast mistakenly attrib
uted to Theodosius of Melitene’®. A comparison with the early Slavonic translation (also dating
from the eleventh century) published by Sreznevskij likewise reveals no divergence on key
points of the narrative’®.
The substance of the Logothete’s account may be summarised as follows:
1. Outraged at Byzantine intransigence over a trade dispute, caused by corruption in high places,
the Bulgar ruler Symeon sent an army to invade Macedonia (by which is meant the Byzantine
province immediately to the west of Adrianople). The imperial forces sent to intercept it were
defeated. Captured Khazars serving in the imperial bodyguard had their noses slit on Symeon’s
orders (GMC, p. 853.1-20).
2. The sight of his mutilated Khazar guardsmen so enraged the Emperor Leo VI that he promptly
organised a double attack on Bulgaria. An alliance was negotiated with the leaders of the
Hungarians, who are named as Arpad and Kursan (Kusan [Theod. Mel.], Kosan [Slav tr.j).
Hostages were taken (GMC, pp. 853.20-854.2).
3. A Byzantine field army, commanded by Nikephoros Phokas the elder, advanced as far as
Bulgaria (phrasing which indicates that the army reached but did not cross the Haemus moun
tains), while the imperial fleet sailed to the Danube. A final offer of peace was made by an
embassy headed by the Quaestor Konstantiniakes (Konstantinakes [Theod. Mel.], Konstantin
[Slav tr.]). His arrest and imprisonment by Symeon was the signal for war. The Bulgar army
had been drawn south to oppose Nikephoros Phokas. The Hungarians were therefore able to
cross the Danube and to devastate the whole of Bulgaria (GMC, p. 854.2-12).
4. Symeon counterattacked and engaged them north of the Danube. He was defeated and
forced to take refuge in Distra (Dristra [Slav tr.], modern Silistra) (GMC, p. 854.12-14).
5. At the suggestion of the Hungarians, the Byzantines bought the Bulgars whom they had
taken prisoner off them (GMC, p. 854.14-17).
______ Howard-Johnston J. Byzantium, Bulgaria and the Peoples o f Ukraine in the 890 s.
C. Mango “The Tradition of Byzantine Chronography”
1989), pp. 360-363.
Harvard Ukrainian Studies, XII-XIII (1988/
Parisinus gr 1708, written in the sixteenth century by a conscientious scribe: Sotiroudis, pp. 25-27.
Edited by Bekker (full citation in n.5 above).
Monacensis gr. 218, ed. T.L.F.Tafel, Theodosii Meliteni qui fertur Chronographia, Monumenta
Saecularia, K.Bayer. Ak. Wiss., Ill, Classe I (Munich, 1859), pp. 186-188.0.Kresten, “Phantomgestalten
in der byzantinischen Literaturgeschichte”, Jahrbuch der Ostenreichischen Byzantinistik, XXV (1976),
pp. 207-212; Sotiroudis, pp. 17-20.
В.Срезневский Хроника Симеона Логофета (St. Petersburg, 1905), reprinted as V.l.Sreznevskij,
The Old Slavonic Translation of the Chronicle of Symeon Logothete (London, 1971), pp. 116-117.
345
6. Symeon now sued for peace via the admiral commanding the fleet on the Danube. A new
ambassador, Leo Choirosphaktes, was sent off, while army and fleet were withdrawn. Hov/-
ever, instead of negotiating, Symeon incarcerated Leo in Mudagra (Mundraga [Theod. Me!.,
Slav tr.]) and attacked the Hungarians (on his own - there is no mention about atony steppe
allies). Without aid from the Byzantines, the Hungarians were all slaughtered (GMC, pp. 854.18-
855.4),
7. Symeon now released Leo Choirosphaktes and offered to make peace if the Bulgar prisor,-
ers-of-war (now in Byzantine hands) were repatriated. The emperor agreed to these terms,
and the prisoners were duly released (GMC, p.855.3-7).
S. Claiming that there were other prisoners in Byzantine hands, Symeon broke the peace and
attacked the Byzantines. The Emperor responded by mobilising the full field army {‘all the
themata and the tagmata') under a new commander-in-chief in the Balkans. A battle was
fought at Bulgarophygon in which the Byzantine was routed. The emperor was especit i' »
grieved at the death of his Protovestiarios, Theodosios (GMC, p. 855.8-16).
9. The notice halts abruptly, without explaining how peace was restored or on what terms.
A number of preliminary observations may be made at this stage. A great deal has been
squeezed into a small space. The whole complicated sequence of episodes is placed between
the death of the Patriarch Stephen on 17th/18th May 893 and the fall of Korcn, a castle in
Cappadocia, tc the Arabs on 5th August 897^^. The pace of the narrative is fast, too fast indeed,
as if a film has been speeded up. By the time it reaches Symeon’s revenge attack on the
Hungarians, it has accelerated far beyond the historically possible. It also has a tabloid tone.
History is personalised and trivialised. The war is a confrontation between two individuals,
the two rulers. Specific incidents are to the fore together with the rush of emotions which they
generate in the main protagonists. While military operations and diplomatic negotiations
must feature, the viewpoint is that of the court at Constantinople, attention focussing on the
machinations which instigated the trade dispute at the origin of the crisis and the notable
personages subsequently involved in the field. Finally, there is a fundamental implausibility
in the overall picture presented of Byzantine statecraft in the reign of Leo VI. The empire has
become a foolish, fumbling thing, prey to extraordinary diplomatio nanve®. Symeon thus
has no difficulty in picking off his enemies one by one. He is allowed a free hand to deal with
his steppe adversary and can then instigate a new war of revenge against Byzantium at a
time of his choice and inflict a serious defeat.
Материалы по археопогии, истории и этнографии Таврии. Вып. VII______
Ш. An alternative narrative: the evidence o f other sources
Since there are good reasons for handling all parts of the Logothete’s chronicle with wary
caution, the implausibility and superficiality evident in his account of Balkan and steppe
history in the 890s should be taken as warning signs that something may be seriously amiss
with his version of events, that history may have been badly mangled in transmission. It is
vital, therefore, to seek out com.parative material with which to check the veracity of the
Logothete’s narrative. Recourse must be had to other extant sources of demonstrable worth.
The fragmentary but more reliable information, which they supply, must be pieced together
to form an alternative, more solidly based narrative by which the Logothete may be judged.
There are seven independent sources of information worth bringing into play. Pride of
place must go to the Tactics of the Emperor Leo VI, since their compilation was overseen by
a participant and initiator of events. A short notice provides valuable information about the
Hungarian invasion of Bulgaria'®. Next comes the diplomatic handbook, misleadingly intitled
" Jenkins, “Chronological Accuracy”, p. 104.
Leonis Imperatoris Tactica, XVIII, PG (Paris, 1863), CVII, col. 956 (cited henceforth as LT).
346
De Administrando Imperio, which reached its final form in the hands of Leo’s son, Constantine
VII Porphyrogenitus, in 951-52. Like the Tactics, which in many ways it complements, this was
a work put together by an imperial editor with full access to official sources of information. It
extends the range of coverage over the whole of Ukraine and can provide another view of
Pecheneg-Hungarian relations^®.
The third source, a collection of the correspondence of Leo Choirosphaktes, a senior diplo
mat of Leo Vi's, should perhaps be prized the most highly of all, since Leo negotiated the peace
agreement which brought the Byzantine-Bulgarwarto an end and the first fourteen letters in the
collection were written in the course of negotiations, eleven by Leo and three by Symeon, the
ruler of Bulgaria with whom he was negotiating. There is, however, a light, bantering tone to the
correspondence, which argues against taking it too seriously. Symeon is demonstrating his
mastery of Greek and his intellectual skills in a rather frivolous way, as he teases the Byzantine
ambassador with cleverly phrased, ambivalent letters. The responses (some of them rather
laboured) are clearly intended to demonstrate the mental and verbal agility of Leo Choirosphaktes.
Apart from a single dating indication, little light is cast on the circumstances or substance of the
negotiations, although a later letter, written around 910 from exile, is rather more informative:
out of favour and in exile, Leo Choirosphaktes was pleading for rehabilitation and cited the
various diplomatic services which he had rendered in the course of his career, beginning with
his first mission to Symeon®°.
The fourth and fifth sources, both Byzantine, refer in passing to a major Bulgar victory. St.
Luke the Stylite nad just begun his military' service as an eighteen-year-old. He took the first
crucial stop (that of being tonsured) which would lead eventually to a hyper-ascetic mode of life
as a stylite, after the Byzantine field army had been defeated and he had survived the danger
ous retreat which followed. This information is recorded by his biographer, a well-educated
Constantinopolitan who had coma to know him well over the last twenty-seven years of his life.
Since the date of his death can be calculated (see section IV below) and his age at the time is
put at over a hundred (a figure which should not be taken too literally), a rough date can be
obtained for the campaign^'. A second survivor of the battle was an Armenian soldier of fortune,
Melias, who was serving in the retinue of a noted Armenian commander of the time and later
went on to carve out a miniature principality under Byzantine protection in the Anti-Taurus^^.
Finally, two Frankish chronicles provide valuable notices about east European history in this
period. A connected but somewhat compressed account of the first two episodes of the Byzan-
tine-Bulgar war is presented in the last part of the Annals of Fulda. Its principal concern is with
the east Frankish court and the deeds of the reigning monarch, Arnulf (crowned king 887 and
emperor in 896), throughout the Carolingian world. But particular attention is paid to the prob
lems encountered by Arnulf on his eastern frontier in his dealings with Svatopluk ruler of Moravia
(who died in 894), and to the subsequent break-up of the Moravian kingdom. The final section
of the chronicle, which covers the years 882-901, was, it is generally agreed, put together in
eastern Bavaria^®. The second Frankish chronicle, that of Regino of Ргьт, was written at the
DAI, cc. 13.5-7, 37.5-14, 38.55-60, 40.7-27 (main notice), 40.32-34, 41.19-25, 51.109-125, with
the commentary of Г.Г.Литаврин и А.П.Новосельцев, Константин Багрянородный, Об управлении
империей (Moscow, 1989).
Kolias, Leon Choerosphactes, рр. 76-91 (letters 1-14, of which nos. 1, 3 and 5 were written by
Symeon) & 112-115 (letter 23), with discussion at pp. 34-42 (cited henceforth as LC).
Vie de saint Luc le stylite, cc. 10-12, ed. F.Vanderstuyf, PO, XI.2 (Paris, 1914), pp. 55-58, with
discussion of author and reliability of text at pp. 11-15 (cited henceforth as VL).
“ Constantinus Porphyrogenitus, De Thematibus, 1.12, ed. A.Pertusi, Studi e testi, CLX (Vatican,
1952), pp.75-76. DAI, c. 50.133-166 for his later career and its context.
” Annales Fuldenses, ed. F.Kurze, MGH, Scriptores in usum scholarum (Hanover, 1891), pp. 118-
135 (890-901), with main Bulgar notice at pp. 129-130 in the entry for 896 (cited henceforth as AF).
______Howard-Johnston J. Byzantium, Bulgaria and the Peoples o f Ukraine in the 890 s.
347
beginning of the tenth century in Lotharingia, far removed from the eastern margin of the
Carolingian lands. Nonetheless, it includes two notices dealing with eastern Europe. Both are
very loosely dated and include casts forward through time. The first, placed in the entry for 868,
summarises the history of Bulgaria from its conversion to the accession of Symeon. The sec
ond reports the expulsion of the Hungarians from their old territory by the Pechenegs under the
year 889, assembles a fair amount of ethnographic material about nomads, and takes the story
on to the Hungarian invasion of Italy in 899-900^^.
These are all sources of known provenance and demonstrable worth. Considerable confi
dence can be placed in the information which they supply. This confidence will be increased if a
coherent narrative can be constructed out of the discrete pieces of information yielded by them.
So let us see whether we can fit the individual pieces together to form a historically intelligible
picture and, if so, whether or not it tallies with the Logothete’s.
1. The Byzantines made peace with the Hungarians, a move which was taken very badly by the
Bulgars (AF, p. 129). This is an entirely different explanation for the outbreak of war from the
Logothete’s. But such was the speed and efficacy of the Byzantine diplomatic and military
response to the initial Bulgar attack in the Logothete’s version, that it may be inferred, quite
independently of the notice in the Annals of Fulda, that the Byzantines negotiated an offensive
alliance with the Hungarians and agreed a plan of campaign, before initiating the trade dispute
which provoked the war.
2. The Bulgars described by the annalist as ’fellow-citizens’ of the Byzantines, presumably
because they were now feilow-Christians and at peace with them - responded by invading
Byzantine territory up to Constantinople {AF, p. 129). There is general confirmation in the
Tactics of Leo VI (XV1II.42, col. 956), which has the Bulgars break the peace and raid the
districts of Thrace. Both these sources place the military operations in the Thracian approaches
to Constantinople, not, as in the Logothete’s version, in Macedonia which lay further west.
3. The Byzantine fleet then ferried the Hungarians across the Danube. Once the fixed defences
erected by the Bulgars on the southern bank of the river had been breached (in a heroic action
led by the captain of the admiral’s ship), the Hungarians invaded Bulgaria as far as Preslav,
shutting the Bulgarian ruler up in the fortress of Mundraga. Three battles were fought in the
course of the campaign, the first two resulting in clear Hungarian victories, the third a long
engagement causing heavy casualties on both sides and followed by the withdrawal of the
Hungarians (LT, XVIII.42, col. 956 [making out all three battles to be Hungarian victories]; AF,
pp. 129-130 [Characterising the third as a Bulgar victory won at heavy cost], and DAI, cc.
51.109-125,40.7-13). All three battles, it should be noted, were fought south of the Danube, in
contrast to the Logothete’s single Bulgar victory won to the north.
4. The Byzantines had recoursed to the Hungarians because their main forces were engaged in
the east against the Arabs (LT, XVIII.42, col. 956). This statement does not preclude action on
a modest scale by land in the Balkans, as reported by the Logothete, to activate Bulgar de
fences along the Haemus frontier and thus to draw their forces away from the Danube. A hint
to this effect is let drop by the Annals of Fulda (pp. 129-130), which refers to the mobilisation
of Bulgar forces elsewhere (posit! in expedicione) with the result that they could not put up
effective opposition to the initial Hungarian attack. There has, however, been some compres
sion in the notice, which gives the misleading impression that the Hungarian invasion oc
curred while the Bulgars were still engaged in their initial campaign into Thrace.
5. The Logothete’s account of troop transfers from Asia Minor and of operations which culmi
nated in a Bulgar victory over a large Byzantine field army at Bulgarophygon can be corrobo-
Reginonis Abbatis Pnimiensis Chronicon, ed. F.Kurze, MGH, Scriptores in usum scholarum (Hanover,
1890), pp. 95-96,131-133.
Материалы по археопогии, истории и этнографии Таврии. Вып. VII______
348
rated on four specific points by the Life of St. Luke the Stylite and the De Thematibus. 1) The
army in which Luke had recently been enrolled was, almost certainly, that of his home theme
the Anatolikon; it must have been transferred to Europe to take part in the campaign {VL, cc.
10-11, pp. 55-57). 2) Elite metropolitan forces also took part, since the Armenian officer whom
Melias served was in command of the tagma ofthe Exkoubitoi (De Them., 1.12.11-13). 3) Both
sources which report the presence of L.uke and Melias at the battle confirm that it resulted in a
general rout ofthe Byzantine forces {VL, c. 12, p. 57; De Them., 1.12.10-16). 4) The death of
the emperor’s Protovestiarios, in unknown circumstances on the battlefield, is confirmed by
the De Thematibus (1.12.14-15). This accumulation of independent testimony engenders gen
eral confidence in the Logothete’s account of the campaign, with one exception - the Byzan
tine mobilisation must have begun well before the start ofthe campaigning season, hence well
before a Bulgar army entered Byzantine territory. As for the victory won by the Bulgars, sur
prise probably played a large part. Symeon, a military commander of genius, did the unex
pected, marched south of the Haemus into Thrace, intercepted and engaged the full field
army of Byzantium in open country, something which none of his predecessors had ever
dared to do. There is every reason to think that the Byzantines were caught unprepared and
that the battle was lost almost as soon as it began.
6. The Emperor Leo VI now had to sue for peace. He sent Leo Choirosphaktes to negotiate with
Symeon. One of the key issues concerned prisoners-of-war, but they were Byzantines in
Bulgar hands, not the Bulgars previously taken prisoner by the Hungarians and later sold to
the Byzantines, as alleged by the Logothete. This is made absolutely plain both in the joking
letters exchanged by Symeon and Leo Choirosphaktes at the time and later in the latter’s
earnest plea for rehabilitation^®. Leo leaves us in the dark about the concessions made by
Byzantium in order to secure the release ofthe prisoners (who probably included many civil
ians, since he claimed that they numbered one hundred and twenty thousand). Later evi
dence, however, casts some light on Symeon’s terms for peace; besides the ransom paid
over for the prisoners and restoration ofthe traditional trading arrangements (on the assump
tion that the Logothete has not invented the trade dispute), Byzantium undertook to make
regular annual payments (in cash or kind) in future^®.
7. Having secured his southern frontier by this peace treaty and gained freedom of action,
Symeon set about organising a reprisal campaign against the Hungarians. DAI, c. 40.13-15 is
categorical about the order of events: Symeon renewed the peace with Byzantium before
turning his attention to the north. The Logothete is thus revealed once again to have made a
serious error in meshing his account of diplomacy in the north into his main political and
military narrative. The negotiations and preparations doubtless took time and were kept se
cret. DAI, c. 40.13-16 reports that Symeon made contact with the Pechenegs (a client-people
of the Khazars, assigned the task of defending the eastern approaches to the heartland’s of
the khaganate over recent decades). The misleading impression given is that the Khazars
themselves were not involved and did not sanction the plan which was agreed. The plan
involved a double attack on the Hungarians’ territory in Ukraine, by the Bulgars from the west
and by the Pechenegs from the east. The attack was launched at a time when many Hungar
ians were away on an unspecified expedition. It achieved great success, described in some
what exaggerated terms by DAI, c. 40.16-20 as the annihilation of the families of the Hungar
ians and the ruin of their land.
The same events are viewed from an eastern rather than a Balkan perspective earlier in the
DAPs dossier about the peoples of Ukraine. Here the Pechenegs attack the Hungarians,
______ Howard-Johnston J. Byzantium, Bulgaria and the Peoples o f Ukraine in the 890 s.
“ LC, letters 1-9, 13, 23.3-8, ed. Kolias, pp. 77-85, 89, 113.
“ Shepard, ‘Symeon of Bulgaria’, p.18.
349
under Khazar direction, and no mention is made of co-ordinated action by the Bulgars. This
version tallies with that of Regino of Ргьт who likewise concentrates on the Pechenegs^^.
8. The damage was such and the Pecheneg threat so great that the Hungarians migrated west
into the Carpathian basin, the Pechenegs occupying the territory in Ukraine which they va-
cated^®. These massive population movements inaugurated new phases in the histories of
Ukraine and Eastern Europe. A new formidable non;iad power was now interposed between
the Balkans and the Khazar khaganate, between the Black Sea littoral and the forested lands
to the north. To the west, the political configuration of Europe was soon changed out of all
recognition. The Hungarians brought about the rapid demise of the disintegrating Moravian
kingdom and launched damaging, deep-penetrating expeditions into central, southern and
western Europe^®.
IV. Chronology
The Logothete’s narrative of events has been shown to be at fault in several respects.
While he may be reliable on Byzantine military matters, reporting Balkan campaigns with
reasonable accuracy and in the correct order, he mauls history by failing properly to correlate
Ukrainian with Balkan events. Leaving aside a number of errors of detail (including a fair
amount of prosopographical confusion), we can isolate three major errors.
In general, he conceals the wider context of Balkan events. As is made very plain by
other sources, they were shaped by steppe diplomacy. His first grave error is to place
Byzantine negotiations with the Hungarians after rather than before the outbreak of fighting
in the Balkans. It has serious consequences. For it gives the impression that both Byzan
tine offensives, the first involving the Hungarians as the main strike force, and the second a
massive assault by the Byzantine field army from the south, were improvised, rather than
forming part of a carefully planned war of aggression. The evidence of other sources, how
ever, strongly suggests that Leo VI sought to break the power of Bulgaria, using all the
forces at his disposal (naval, military and nomad allies), and that the two campaigns were
planned together as a two-stage offensive, the first nomad attack from the north preparing
the way for a full-scale invasion by Byzantine forces from the south.
The second error is no less serious. It is both an error of arrangement and of substance.
Symeon is presented as playing for time after the Hungarian attack and opening (but not
completing) peace negotiations with the Byzantines. His crushing victory over the Hungar
ians (apparently unaided) is placed before, rather than after the Bulgarophygon campaign.
The Pechenegs have been entirely eliminated from the story. Byzantium is presented as
Symeon’s diplomatic plaything. The reality, as has been seen, was very different. Symeon's
initiative in the northern world followed the decisive victory which he achieved over Byzan
tium at Bulgarophygon. Considerable time, at least one, perhaps two or more years, must
be allowed for the negotiation of the anti-Hungarian alliance with the Pechenegs, the formu
lation of a joint plan of campaign and its execution. Time must also be allowed for the
consequential migrations of the Hungarians and Pechenegs.
' Finally, there is the convoluted tale of the prisoners-of-war in the Logothete’s deformed
version of history. Prisoners, many of them civilians, taken by the Bulgars, probably in the
course of their two campaigns on Byzantine territory, are transformed into Bulgars captured
by the Hungarians, sold to the Byzantines and then returned after Symeon has disposed of
Матернааы по археологии, истории и этнографии Таврии. Вып. VII______
DAI, cc. 37.8-14, 38.55-57; Regino, pp. 131-133.
DAI, c. 40.19-27, with cc. 37.8-14, 38.55-60.
DAI, cc. 13.5-7, 38.57-60, 40.32-34, 41.19-25 (destruction of Moravia), with C.R.Bowlus, Franks,
Moravians and Magyars. The Struggle for the Middle Danube, 788-907 (Philadelphia, 1995), pp. 239-
251; K.J. Leyser, Medieval Germany and Its Neighbours 900-1250 (London, 1982), pp. 43-49.
350
the Hungarians but before the battle of Bulgarophygon. History thus slides towards the non
sensical in the Logothete’s hands.
What, however, of the Logothete’s dating of the three successive Balkan campaigns
which he has recorded in the correct order? Three years of military action in the Balkans can
be fitted into the temporal gap left for them in the chronicle, which stretches from May 893 to
August 897. Historians have generally plumped for the period 894-96®°. But can confirma
tion be found in the other sources of proven worth?
It is hard to give a clear-cut, definitive answer to this question. For, while corroboration
can be found for the Logothete’s dating, there are worrying contrary indications in two of the
independent sources. On the surface, though, there seems to be nothing wrong. A first brief
notice about the Hungarian invasion of Bulgaria is placed in the year entry for 895 in the
Annals of Fulda (p. 126), only to be fleshed out in a fuller (and surely doublet) notice, that
analysed above, which is included in the entry for the next year (pp. 129-130). It should also
be noted that Symeon, who is reported, both by the Logothete and by DAI, cc. 40.7-19,
51.109-116, to have been Leo Vi’s Bulgar antagonist, cannot have succeeded his older
brother Vladimir before the second half of 892, since Arnulf dispatched an embassy to the
latter in September of that year {AF, pp. 121-122). Everything so far points to the preferred
period 894-96 which tallies with the Logothete’s chronology.
But a single chance remark in the collected correspondence of Leo Choirosphaktes should
give us pause. Symeon opens his first letter with a reference to a prediction made to him by
the Emperor Leo VI two years earlier (тгроттёриаг, 'the year before last’). The emperor had
foretold the precise time and length of a solar eclipse®^. Now there was oniy one eclipse
(annular) of the sun visible in Constantinople and the Balkans in the early 890s. It took place
on Sunday 8th August 891 and lasted for three hours. It is duly noted, in the correct place, in
the Logothete’s chronicle, and is the only solar eclipse commemorated in the Synaxarium of
Constantinople (which dates it carefully and notes that the stars became visible). It is hard to
escape the conclusion that the peace negotiations which followed the battle of Bulgarophygon
took place in 893, and hence that the war should be dated to 891-93®^.
Corroboration for this earlier dating can be found in the Life of St. Luke the Stylite. Luke's
death can be dated from the details given by his biographer. He died on 11th December in a
year when that date fell on a Thursday. The year can be calculated from an earlier passage
which describes the grim ordeal he underwent on his first column in his home theme, during
a spell of bitter winter weather which lasted one hundred and twenty days. That notorious
winter was recorded by the Logothete (who shows a consistent interest in unusual natural
phenomena) under the year 927-28, and played a part in triggering a prolonged agrarian
crisis in Byzantium (imperial legislators subsequently dated the start of the crisis to the first
indiction, 927-28)®®. The three years spent by Luke on his first column, which straddled the
______ Howard-Johnston J. Byzantium, Bulgaria and the Peoples o f Ukraine in the 890 s.
Златарски, Известията на българите, pp. 90-101 assembles the evidence, starting with the position
of the war in the Logothete’s chronicle.
®’ LC, letter 1.3-6, ed. Kolias, p. 77.
“ Jenkins, ‘Chronological Accuracy’, p. 104 and n. 64; Th. Ritter von Oppolzer, Canon der Finstemisse
(Vienna, 1887), tr. O.Gingerich, Canon of Eclipses {Uew York, 1962), pp. 200-201, n. 4995, chart 100;
GMC, p. 852.12-13; Synaxarium Ecclesiae Constantinopolitanae, Propylaeum ad AASS Novembris,
ed. H.Delehaye (Brussels, 1902), col. 878.
®® GMC, pp. 908-909; N.Svoronos, Les novellas des empereurs macedoniens concemantia terre et
les stratiotes (Athens, 1994), pp. 85-86. Contra R.Morris, 'The Powerful and the Poor in Tenth-Century
Byzantium; Law and Reality’, Past and Present, 73 (1976), p. 8 who accepts Vanderstuyfs redating of
the severe winter to 933-34 (introduction to VL, p. 22).
351
bad winter of 927-28, were immediately followed by forty-four years on his second column at
Eutropius, on the outskirts of Constantinople, which ended with his death. This must have
occurred on Thursday 11th December 973 (the nearest alternatives are 962 and 979) -
which would put his first three-year stint as a stylite in 926-29^. If Luke really was over a
hundred when he died, he must have been born in 873 and his service as an eighteen-year-
old on the Bulgarophygon campaign must be dated at the latest to 892. An error of one year,
either in his age at the time of the battle or at his death, is clearly easier to explain than one
of four years - which argues for an earlier (893) as against a later (896) date for the battle.
Given these two pieces of contrary evidence, the Annals of Fulda which provides power
ful corroboration for the Logothete's dating should be scrutinised yet more closely. Curiously
it is not the reigning Bulgar king who is the centre of attention (he is not mentioned or named),
but his predecessor, Michael, who initiated the conversion of Bulgaria and subsequently
abdicated and retired to a monastery. In their hour of crisis, the Bulgars appeal to him, and
he gives the advice which saves them. A penitential fast lasting three days prepares the way
for them to seek Cod’s help, which shifts the military balance and gives them a hard-won
victory in the third battle against the Hungarians. Taken by itself, there is nothing in the notice
to cast doubt on its placing by the annalist under 896, but there is an affinity between it and
an episode, described in the chronicle of Regino of Ргьт, which may give us pause.
Regino reports the conversion of Bulgaria under the year 868 and then includes a long
cast-forward: once again the central figure is ex-king Michael; he emerges from the monas
tery to which he has retired to save his people, this time the threat is domestic, namely a
revival of paganism under the active encouragement of his elder son and successor (Vladimir);
Michael has him arrested, blinded and imprisoned, and appoints his younger son (Symeon)
king with stern instructions to uphold Christianity®®. The suspicion arises that the two chroni
cles are reporting different facets of the same episode, that 'the injury they had done to the
Christians’ for which Michael insists the Bulgars do penance®® was the backsliding spon
sored by Vladimir, and that Vladimir was deposed and replaced by Symeon during 892. at a
late stage in the campaigning season. On this hypothesis, two acts were necessary to regain
God’s favour and success on the battlefield - an act of collective penance and the installa
tion of a true Christian ruler. The latter event cannot have happened much before September
892 when Arnulfs court believed that Vladimir was still on his throne and sent an embassy to
ask him to put an embargo on the sale of salt to Moravia®^.
A strong case can therefore be made in favour of an earlier chronology, at variance with
the Logothete’s dating of the Balkan war to 894-96. It has the additional advantage of provid
ing a partial justification of Leo Vi’s apparently unprovoked aggression against Bulgaria. A
revival of paganism, sponsored by the ruler, would have provided an acceptable pretext for
breaking the long-established peace®®. There are, however, a number of obstacles to be
surmounted before such a chronology can be accepted. First it must be supposed that the
Hungarians were in simultaneous diplomatic contact with the Byzantine and Carolingian
Материапы по археологии, истории и этнографии Таврии. Вып. VII______
VL, cc. 23-24, pp. 69-71 (first stint as stylite), cc. 78-82, pp. 132-139 (death after second forty-four
year stint as a stylite). Vanderstuyf, who indentifies the key dating elements in the Life (pp. 19-26),
plumps for 979 as the year of Luke’s death, on the assumption that Bulgarophygon was fought in 897.
Regino, pp. 95-96.
^AF, p. 130.
" AF. p. 121-122.
“ An early chronology (placing the war between 888 and 892) had been entertained before
Zlatarsky wrote his article (Златарски, Изветията за българските, р. 91).
352
courts, and undertook two major military actions far outside their territory in Ukraine in 892.
For, besides invading Bulgaria in alliance with the Byzantines, they took part in Arnulfs co
ordinated invasion of Moravia (securely dated to that year)^®. Arnulfs embassy to Bulgaria,
on this hypothesis, would have been taking advantage of Bulgaria’s temporarily weakened
state. Second, it must be conjectured that Constantine Porphyrogenitus (or a research as
sistant) has mistakenly retrojected Symeon’s accession to the beginning of 892, since he is
named as the Bulgar ruler who takes refuge in Mundraga when the Hungarians invade.
Such an error is far from inconceivable some sixty years on, in a text which is unsure of the
identity of the contemporary Hungarian ruler (Arpad at DAI, c. 38.55-60, his son Liuntika ate.
40.7-13). Symeon had, after all, led the Bulgar army to victory at Bulgarophygon (De Them.,
1.12.11-13) and then, after negotiating peace with Leo Choirosphaktes (his correspondence
provides a firm terminus ante quern for Symeon’s accession), prepared his counterstrike
against the Hungarians.
Finally, two contrary dating indications also have to be confronted and explained. The
placing of the first brief and somewhat misleading notice about the Hungarian attach on
Bulgaria in the entry for 895 in the Annals of Fulda, with the second fuller notice following in
the next year (pp. 126,129-130), three and four years respectively after the event according
to the earlier chronology, must be viewed as an error, arising perhaps from a delayed appre
ciation on the part of the annalist of the growing Hungarian threat to central Europe or per
haps from the late arrival of detailed information about the attack (perhaps brought by the
Byzantine embassy which, AF, p. 125 reports, was received by Arnulf at Regensburg late in
894). A notice under AH 283 (19th February 896 - 7th February 897) in the principal Arab
chronicle for the period, that of Tabari, must also be discounted”®. The Bulgar advance on
Constantinople and the arming of Arabs resident in the city (probably prisoners-of-war), a
desperate measure forced on the imperial government which is presented as vital in repel
ling the Bulgars, should perhaps be taken as a misplaced account of events in autumn 913
(Tabari indicating his doubts, by introducing the story with the phrase 'it is said’) rather than
as a very garbled version of the Bulgarophygon campaign”’ .
However, the more numerous and elaborate the argumentation needed to defend the
earlier chronology, the more confidence in it is eroded. It remains possible that the war was
fought from 891 to 893, three years earlier than is commonly supposed, but the Logothete’s
chronology may be judged equally probable. The chief difficulty - the eclipse to which Symeon
refers during his negotiations with Leo Choirosphaktes after the battle of Bulgarophygon -
may circumvented, but only by supposing that he uses the word тгротгёрил loosely to mean
'a year or two ago’ instead of ‘the year before last’”^ .
The dilemma cannot be resolved by reference to contemporary events in the Near East.
For neither in 892 nor in 895 is there any record of major military action on any sector of the
Byzantine-Arab frontier, which might have prevented the Byzantines from transferring troops
to the Balkans. The Emperor Leo’s remark, which implies that there was, seems to be disin
genuous, masking the true character of the war with Bulgaria which, as has been argued
above, was an an offensive carefully planned in two stages. The only reported Byzantine
operation on a large scale in the east in the early 890s was an attack on Theodosiopolis in
______ Howard-Johnston J. Byzantium, Bulgaria and the Peoples o f Ukraine in the 890 s.
“ 4F, p. 121.
”® Vasiliev, Byzance et les Arabes, II.2 (Brussels, 1950), pp. 11-12.
Contra Vasiliev, Byzance et les Arabes, 11.1, pp. 129-131.
”2 Jenkins, ‘Chronological Accuracy’, p. 104, n. 64, who rejects Kolias’ identification of the eclipse as
that of 7th June 894 since it was only total in Scandinavia and northern Russia (Kolias, Leon
Choerosphactes, pp. 33-34; Oppolzer, pp. 202-203, n. 5002, chart 101).
12 3a*. 2J9 353
western Armenia in 894, to which the Arabs seem to have replied with a successful raid in force
into Anatolia at the end of the yeah” , Then came a two-year interlude in the fighting (895-96),
during which a formal exchange of prisoners was arranged. Traditionally this has been seen as
clearing the way for the Bulgarophygon campaign, but it may be placed no less plausibly after
wards - the campaign into western Armenia being intended to restore Byzantium’s prestige
immediately after the defeat in Europe and a truce then being negotiated both to prepare the
way for the exchange and to secure a valuable breathing-space'” .
The only safe conclusion is to entertain both early and late chronologies as possibilities and
to suspend judgement on the question of whether or not the Logothete has placed his con
nected notices about the Balkan war in the right position in his text
V. The steppe dim ension
The Balkan war was part of a three-act drama. The first act, negotiation of an offensive
alliance between Byzantium and the Hungarians, and the third, a co-ordinated Pecheneg-Bulgar
attack on the Hungarians’ territory in Ukraine with all its consequences, can be regarded merely
as prelude and aftermath of the Byzantine-Bulgar conflict. The southern, more settled powers
can be viewed as the prime moving forces in wider steppe history. The middle of the three acts
can thus be made out to be the most important in the causal sequence.
We must, however, beware of being trapped within the limited field of vision of the meagre
surviving written source material. Emanating as it does from Byzantium and the Frankish lands,
it may mislead us into neglecting what was happening in the steppe world well beyond the
Danube and the Carpathian, and into overlooking the possibility that established nomad powers
took diplomatic initiatives of their own. A moment’s reflection on the relative geographical scale
of the two political arenas, Ukraine and the Balkans, may suggest that developments in the
former are more likely to have influenced those in the latterthan vice versa. Such, indeed, is the
direction of causal thrust which is implied by the chronological sequence of steppe and Balkan
episodes, once they have been disentangled from each other.
Whichever of the two chronologies we may accept for the Byzantine-Bulgar war, it is as likely
as not that the first diplomatic move which sparked it off was made by the Hungarians and that
the emperor Leo VI saw it as proffering an opportunity too good to miss. On the earlier chronol
ogy, two alliances were negotiated by the Hungarians well before 892, the year when they were
involved, either simultaneously or in rapid sequence, in Joint operations with the Byzantines
against Bulgaria, and with the Franks, Alamans and Bavarians against Moravia. It is surely
more plausible to explain the contemporary negotiation (probably beginning in 890) of two alli
ances with the two main power centres of southern and central Europe, the courts of Leo VI and
Arnulf, as the result of a single Hungarian diplomatic initiative than of two independent Byzan
tine and east Frankish initiatives. On the later chronology, the Hungarians had already demon
strated an interest in the frontier regions of Europe, by their intervention in force beyond the
Carpathians in 892, before they negotiated the alliance with Byzantium (in 893 on this chronol
ogy). It is surely more likely that they were the driving force behind the planned double attack on
Bulgaria (a weakened Bulgaria would be in no position to oppose future Hungarian activity in the
Carpathian basin) rather than Byzantium which had been at peace with Bulgaria ever since the
country’s conversion to Christianity a generation earlier.
What then were the reasons behind this westward shift in Hungarian interest? Was it
simply the latest case of steppe drift towards Europe? Were they being drawn west by the
Vasiliev, Byzance etles Arabes, 11.1, pp. 122-125 (commentary) and 11.2, p. 11 (Tabari on 894). The
Byzantine attack on Theodosiopolis (Armenian Karin) is reported by Thomas Artsruni, History of the
House of the Artsrunik’, tr. R.W. Thomson (Detroit, 1985), pp. 294-295.
Vasiliev, Byzance et les Arabes, 11.1, pp. 126-128.
Материалы по археологии, истории и этнографии Таврии. Вып. VII______
354
prospect of extending their pastoral base to include the Carpathian basin and then of using it as
a platform to raid Europe? Or were they taking advantage of the internal divisions which were
sapping Frankish power? Were they aware that Viking attacks in the west were hastening the
breakdown of the traditional political order in the Carolingian world, as well as seriously deplet
ing its resources?'*® Were they planning to exploit this tempting opportunity to raid deep into
Europe? Or were they reacting to a steady thrust southward of the eastern Vikings, the Rus,
who were, by the 890s, establishing a firm grip over Kiev?'*® Had their prime concern previously
been to manage the wooded steppe zone and were they now preparing to make a strategic
withdrawal? However plausible they may seem, all such explanations for a re-orientation of
Hungarian foreign policy in the early 890s must remain in the realm of conjecture, for lack of
evidence.
______ Howard-Johnston J Byzantium, Bulgaria and the Peoples o f Ukraine in the 890 s.
It is time, finally, to turn to the third act in the drama - the appearance of the Pechenegs in
Ukraine and the forced migration of the Hungarians into the Carpathian basin. Both the date
and the wider context can be established on the basis of evidence supplied by DAI, c. 37.2-14.
In a piece of careless editing (by no means unique), Constantine Porphyrogenitus dates the
westward movement of the Pechenegs first fifty, later fifty-five years before the time, 951-52, at
which he was pulling text together in its final form. The second is surely the intended figure (the
dropping of the final digit being more explicable than its addition) - in which case a secure date
of 896-97 is obtained for the Pecheneg attack on the Hungarians in Ukraine and the westward
migration of the latter. Confirmation is to hand in the Annals of Fulda (p. 131), which reports that
Arnulf spent the whole autumn of 897 on his eastern frontier. The decline of Moravian power,
which had begun after the death of Svatopluk in 894, was continuing apace, but that of itself
provided no reason for Arnulf (now emperor) to take personal charge of frontier affairs. His
absorption in the east is rendered intelligible if he was watching over the arrival of a new great
power in the Carpathian basin.
This seismic change in the geopolitics of eastern and central Europe followed either a year
or (on the earlier chronology) four years after the end of the Balkan war. There was a direct
connection with the war, since the Bulgars took part in the Pecheneg campaign against the
Hungarians, in reprisal for their invasion of Bulgaria. But the DAI (backed by Regino of Ргьт)
makes it plain that once again the prime moving force was located in the steppe world, although
in this case it was not the Pechenegs themselves but the Khazar khaganate, which took the
initiative. Both Pechenegs and Hungarians had been clients of the khaganate since the last
Khazar-sponsored re-organisation of their nomad neighbours around 830, the Pechenegs guard
ing the eastern approaches to the central lands of the khaganate, the Hungarians controlling
Ukraine to the w esF. Now, with the aid of the Oghuz Turks, the Khazars transferred the
Pechenegs from their eastern to their western periphery, thereby bringing them into conflict with
the Hungarians. The Khazars, it may be supposed, were also responsible for arranging the co-
'*® J.L. Nelson, Charles the Bald, (London-New York, 1992), pp. 254-264; P.H. Sawyer, Kings and
Vikings (London-New York, 1982), pp. 78-97; R. McKitterick, The Frankish Kingdoms under the
Carolingians 751-987 (London-New York, 1983), pp. 228-236.
'*® Franklin & Shepard, The Emergence of Rus, pp. 91-111.
DAI, c. 38.3-55. The view expressed in the text runs counter to current opinions, for example
those of G.Kristo, Hungarian History in the Ninth Century (Szeged, 1996) and C.Zuckermann, ‘Les
Hongrois au pays de Lebedia. Une nouvelle puissance aux confins de Byzance et de la Khazarie
ca. 836-889’, in N.Oikonomides, ed., Byzantium at l/Har (forthcoming). I hope to justify it in future,
with an extended commentary on the DAI's steppe dossier, in which due account will be taken of
certain confused and confusing editorial comments on geographical matters, made probably by
Constantine Porphyrogenitus himself.
355
ordinated action by the Bulgars which made the task of defending their Ukrainian territory hope
less for such Hungarians as were not away on campaign**®.
As for Khazar motivation, we can only resort to speculation, in the light of general knowledge
of contemporary developments in the surrounding world and of the subsequent behaviour of
Khazars, Pechenegs and Hungarians^®. The Hungarians, it may be conjectured, had been
showing too much independence. A growing preoccupation with Europe, a growing appetite for
the rich pickings offered by decaying polities in Moravia and in the Frankish lands®, was dis
tracting then from performing their primary task in the Khazar world order, namely that of polic
ing the Viking Rus in the wooded steppe and forest zones. A system of checks and bal
ances, in which client-peoples were used as countervailing powers against each other, was
vital to the maintenance of Khazar hegemony, and any serious disturbance to it, such as that
caused by the Hungarians by the 890s, must be put right in a decisive manner®'. Or else one
peripheral people - in this case the Rus - might become overmighty and begin to rival the
Khazars themselves.
In the event, the Pechenegs dealt effectively with the Hungarians and then probably
remained loyal clients of the Khazars®^. Their presence in Ukraine can be argued to have
inhibited the extension and consolidation of Rus power there and thus to have delayed the
time when the Rus would be able directly to challenge the Khazar khaganate. But we are in
danger of being lured into the wider history of the steppe and northern worlds in the tenth
century, and gazing at successive episodes, each as full of drama as that which we have
scrutinized. It is time to stop and, like the Emperor Leo VI around the year 900, to contem
plate with awe the new political configuration of Ukraine and central Europe, which Hungar
ians, Pechenegs and Khazars had brought about®®.
Материалы по археопогии, истории и этнографии Таврии. Вып. VII______
DAI, cc. 37.2-14, 38.55-60, 40.13-27; Regino, pp. 131-133.
Kristo, Hungarian History, p. 182 suggests that the Khazars were spurred into action by events in
the steppes beyond the south-east extremity of their sphere of influence. But the campaign to which
he refers, that of Isma’il ibn Ahmad, new Samanid ruler of Transoxiana, directed against a Qarluq
Turkish town (Taraz) in 893, is unlikely to have had such far-reaching effects (W.L.Treadwell, The
Political History of the Samanid State (Phil., Oxford, 1991), pp. 91-92).
“ Hungarian interest in central Europe was apparent well before the 890s; raids into east Frankish
territory are recorded under 862 and 881: Bowlus, Franks, Moravians and Magyars, pp. 236-239;
S.Nikolov, ‘Constantine and Methodius in the Steppes’, Byzantine and Modem Greek Studio, XXI
(1997 - forthcoming).
P.B.Golden, An Introduction to the History of the Turkic Peoples. Ethnogenesis and State-Formation
in Medieval and Early Modem Eurasia and the Middle East, Turcologica, IX (Wiesbaden, 1992), pp.
232-244 for an outline history of the Khazar khaganate. The khaganate will only be properly understood
as a political and administrative structure when archaeological data and the written sources are
exploited in tandem in a systematic way. Future interdisciplinary research of this sort will, I hope, back
the proposition put forward in the text.
“ There is nothing to indicate otherwise in the two sections of the DAI which deal with the north
(cc. 1-13, 37-42).
“ After the disastrous failure of his aggressive policy towards Bulgaria, Leo VI was forced to take a
fresh look at the northern world, dominated as it now was by a menacing Bulgar-Pecheneg axis, to
see what diplomatic possibilities it might offer. Some of the information gathered can be found in the
main northern dossier in DAI (cc. 37-42), the core of which was probably put together towards the end
of the reign of his reign, to be supplemented much later (between 948 and 952) by his son, Constantine
Porphyrogenitus in a second editorial phase (see my 'The De Administrando Imperio: a re-examination
of the text and a re-evaluation of its evidence about the Rus’, forthcoming in Travaux et Memoires).
The chief policy outcome was a decision to draw the Rus into alliance with Byzantium which resulted
in the trestles of 907 and 911 (Russian Ppmary Chronicle, pp. 64-69).
356
|