EVENTS IN ARABIA IN THE 6TH CENTURY A.D.

By SIDNEY SMITH
The following abbreviations are used for the texts and reference books cited :—
BH. = The Booh of the Himyarites, edited by Axel Moberg.
CIH. = Corpus Inscriptionum Semiticarum, Pars Quarta.
CMH. = The Cambridge Mediaeval History, vol. i.
LRE. = J. B. Bury, History of the Later Roman Empire. 2 vols.
SE8. = Repertoire d'epigraphie semitique, tomes V-VII rediges par G. Byckmans.
Arabische Frage = M. Hartmann, Der islamische Orient, vol. ii.
Budge, Ethiopia. = Sir E. A. W. Budge, A History of Ethiopia, Nubia and Abyssinia.
Chrestomathia = C. Conti Rossini, Chrestomathia arabica meridionalis epigraphica.
Cosmas = The Christian Topography of Cosmos, ed. by E. O. Winstedt.
Dittenberger = W. Dittenberger, Orientis Oraeci Inscriptions Selectae.
Fakhry = Ahmed Fakhry, An Archaeological Journey to Yemen. 3 vols. I I : Epigraphical
Texts, by G. Byckmans.
Glaser = Zwei Inschriften liber den Dammbruch von Marib. Mitteilungen der vorderasiatischen-
agyptischen Gesellschaft, vi, 370-488.
Institution = J. Ryckmans, VInstitution monarchique en Arable meridionale avant VIslam.
Malalas = Johannis Malalae Chronographia ex recensione L. Dindorfii.
Mas'udi = Masoudi, Les prairies d'or: texte et traduction par Barbier de Meynard et Pavet
de Courteille.
Mordtmann—Mittwoch = Rathjens—von Wissmannsche Sudarabien-Reise I. Sabdische In-
schriften: bearbeitet von J. H. Mordtmann und Eugen Mittwoch.
Muir—Weir = The Life of Mohammad, by Sir William Muir: a new and revised edition by T. H.
Weir.
Noldeke = Th. Noldeke, Oeschichte der Perser und Araber zur Zeit der Sasaniden, aus der arabischen
Chronik des Tabari ubersetzt.
Periplus = Hjalmar Frisk, Le Periple de la Mer trythrie.
Procopius = Opera, ed. J. Haury. 3 vols.
Theophanes = Theophanis Chronologia ex recensione Joannis Classeni.
von Wissmann—Hofner = H. von Wissmann und Dr. Maria Hofner, Beitrage zur historischen
Geographie des vorislamischen Sudarabien.
I. ECONOMIC CONDITIONS

T WO accounts of trade conditions in the 6th century show that there had
been a great change since the ' Periplus Maris Erythaei' was written, about
A.D. 50. The writer of that manual for merchant skippers was precise as to the
location of the incense-bearing lands.
After Kane, as the land continues on, there opens out another, very broad,
gulf, stretching a considerable distance in depth. It is called Sakhalites, and
the ' libanos-he&Tmg land '. It is mountainous and bad for landing. The air
is thick, dust-laden with the libanos blown down from the trees. These trees
that bear libanos are of no great diameter, and are not tall. They produce
the libanos in a solid form on the bark, just as some of our trees in Egypt
weep gum. The libanos is handled by the royal slaves and by those sent for
punishment. These places are dreadfully infectious and plague-ridden, even
for those just sailing along the coast, but for those working there death is in
the air, and they are downright destructive because of the insufficiency of
food.1
While Hadramawt remained an independent kingdom, as it was in A.D. 50, it
is reasonable to suppose that the kings maintained this industry by slave
1
Periplus, 29, p. 9.
VOL. XVI. PABT 3. 30
426 SIDNEY SMITH ;,

labour. Thefirstking' of Saba' and Dhu Raydan and Hadramawt and Yamanat * |
was Sammar Yuhar'is, whose date cannot be considered fixed.1 By the 6th i
century the 'fo'&awos-bearingland ' was the African coast of the Gulf of Aden.
In one of his calculations of distance by stages, backed by a reference to
Scripture, Cosmas Indicopleustes stated the facts incidentally.
From Aksum to the ends of Ethiopia Barbary, called the ' libanos-
bearing land ', which lies before, but is not beside, the Ocean, including the
extensive land Sasu, the last territory of the Ethiopians, is 40 halts, more
or less.2 . . . The K&«mos-bearing land is at the ends of Ethiopia, inland, the
Ocean lying on the further side. From there the inhabitants of Barbary,
since they are nearest, go up inland and by way of business transport from
there most of the incenses, libanos, cassia, calamus, and many other sorts.
They then transport it further by sea to Adulis, to Homerites, to inner
India, and to Persia.3
He adds that ' Homerites is not two days' journey from Barbary, the sea
intervening : the rest of Barbary is Ocean on the far side, called Zingion'.
The southern Arabian coast trade was with Taprobane, Ceylon, that island
' receiving the products of Sind, where there is musk, castoris, and androstachys,
and bartering with Persia and the Homerites and Adulis, and in turn receiving
from each locality named, and bartering with the Indians of the interior,
exporting its own produce to each market at the same time '.* The main staple
brought from Taprobane was metaxion, silk, most of it bought up by Persian
merchants.5
In the Byzantine Empire silk was bought by the public fisc for the state
factories, gynoilcia, and only sold to private traders if there was a surplus. The
great sums which had to be paid to Persia were largely spent on the army.
Justinian I tried to reduce these sums by fixing the price at 15 gold pieces
a pound. The Persians retaliated by limiting sales. The Emperor then further
reduced the market price for silk stuffs to 8 gold pieces. Private traders were
ruined ; manufacture became a strictly controlled state monopoly. At some
date not long after 552 the eggs of bombyx mmi were brought to Byzantium.6
By the 7th century sericulture was established in Asia Minor. The collapse of
the Empire after Justinian meant new borders, new taxes. Khusrau I in the
treaty of 562 tried to stop smuggling by restricting trading to fewer places on
the border. Later Persia lost control of many routes. But before 570 the silk
trade was important.
1
Two slightly diflFerent proposals, that in Institution, 311, the beginning of the 4th century,
and that of Dr. A. F. L. Beeston in BSOAS., xvi, 41, about A.D. 280-300, both depend on the
mention of Xagran in the 'Imru 31-Qais inscription from Namara as ' town of Sammar'. It is not
clear that Sammar Yuhar'is is meant. M. J. Ryckmans has rightly preferred to consider Sammar
in the inscription Ry 508/4 a known geographical name, see Museon lxvi, 334. The 'Imru
5
1-Qais inscription may indicate the relation of Nagran to an important region north of it in
the years before 328.
2
A typical example of the use of this expression by Cosmas, see p. 454.
3 4
Cosmas, II, 27 C-D : pp. 69-70. ibid., XI, 447 D-448 A : p. 322.
6 6
ibid., II, 27 A : p. 89. Procopius, Anekdota, 25, 13 : see LRE, ii, 330-2.
 GHASSfliN

O Zm PA* A net en, t to*** *&

N a. r R d.

INDIAN SEA

Farittf p. 42«] ARABIA IN THE SIXTH CENTUBY. M. W. Smith : 1064.

 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 427

Procopius, who acted as secretary of Belisarius till 540, became a critic of

Justinian. In his' History of the Wars ' there are innuendoes ; in the Anekdota,
or ' Secret History ', published after the Emperor's death, the attack is bitter,
often scandalous. To Christianity he adopted an indifferent pose, claiming
Attic culture. His account of the policy of Justinian in negotiations with
Ethiopia and Himyar is partial, and superficial, but correct as to intention.
At that time, when Hellestheaios * was reigning over the Ethiopians and
Esimphaios2 over the Homeritai, the Emperor Justinian sent an
ambassador, Julianus, asking 3 that both should combine with the Romans,
by reason of their common faith, in making war on the Persians, the object
being that the Ethiopians, by buying silk, metaxa, from the Indians and
re-selling it to the Romans, should possess themselves of great wealth,
while profiting the Romans only in that they would no longer be forced to
part with their own wealth to the enemy. (This metaxa is the stuff commonly
used for making dresses, which the Hellenes of old called Medic, but is now
termed serike, Chinese.) Further it was proposed that the Homeritai should
set up the fugitive Kaisos as phylarch over the Maddenoi, and should
invade Persian territory with a large army of the Homeritai themselves and
of the Maddenoi Saracens. (This Kaisos belonged by birth to a family of
phylarchs, and was eminently successful in war; but after killing some
relative of Esimphaios, hefledto a land completely destitute of men.) Each
king accepted the request3 and undertook to carry it out, and then dis-
missed the ambassador; but neither of them carried out the promises.
For it was impossible for the Ethiopians to buy the metaxa direct from the
Indians, since the Persian merchants commonly purchased all the cargoes,
being always at the very ports where the Indians' ships first put in, seeing
that they dwell in the adjoining land. And the Homeritai too considered
it a hard bargain, if they were to get, in exchange for marching against men
much better atfightingthan themselves, a desert, and at that one extending
over a distance of many days' journey. Even Abramos, later, when he was
most securely established as ruler, though he frequently promised the
Emperor Justinian to invade Persian territory, only started out on that
expedition on one occasion, and retired immediately. Such was the course
of events concerning the Ethiopians and Homeritai for the Romans.4
The tone of the passage is unmistakable. Justinian's policy was a failure,
doomed from the start, both by practical conditions and the unreliability o
barbarians. ' Medic ' is dragged in to suggest the decline from classical stan-
dards, the days of Persicos odi puer apparatus and earlier. The double appeal
of Justinian to religion and greed is hinted. There may be a sly reference, in
the words ' common faith', to the opposition between Monophysite and
Orthodox parties at the court, for Procopius elsewhere told of the struggle.
1
8 for b, as in 'AvaaapBov for Anazarbus, Malalas, p. 444, corrected by Theopbanes, i, p. 263.
Professor Guillaume has proved the alternation of/and t in some early roots, BSOAS., xvi, 1-12.
2
Haury did not admit the variant Bsimiphaios; it seems a secondary error. The form,
which occurs only in Procopius, shows how such names were treated.
s
a(iwv might be translated ' demanding ', but Justinian was not in a position to demand,
and aX-rqmv subsequently shows the correct nuance.
4
' Wars', I, xx, 9-13.
428 SIDNEY SMITH

Justinian's policy implies that Ethiopia and the Homeritai were already |
purveyors of luxury goods, partly dependent on consumption in the Empire.
Some effects in the Red Sea lands are to be found in a passage which is an
example of the real knowledge Procopius had of current affairs and of his
fundamental ignorance of matters long known to Roman merchants.
The boundaries of Palestine stretch eastward to the sea called Red. This
sea begins from the Indian parts and ends at that point of the Roman
Empire. And a city lies on its shore called Ailas, where, I am told, the sea
ends, becoming a very narrow strait. As you sail into the sea from there the
Egyptian mountains on the right turn south ; on the other side a country
deserted by men stretches northward for a great distance. As you sail on,
the land on either side is visible as far as the island Yotabe, not less than
a thousand stades distant from Ailas; Hebrews have long dwelt there in
independence, but under the rule of this Justinian they have become Roman
subjects. From there on comes a great open sea, and those sailing on here
no longer see the land on the right; they put into anchorage on the left
shore when night falls, for it is impossible to navigate in this sea in the dark,
since it is everywhere full of shoals. But there are harbours there, and great
numbers of them, not made by the hands of man, but by the natural
contours of the lands, and for this reason it is not difficult for sailors to
harbour at any point. This coast immediately beyond the boundaries of
Palestine is held by Saracens who have been settled in the Palm Grove since
long ago. The Palm Grove is in the interior, and stretches across a deal of
country, where absolutely nothing is produced except palms. Abochorabos
presented this Palm Grove to the Emperor Justinian; he was the ruler of
the Saracens there and the Emperor appointed him phylarch of the Saracens
in Palestine. He continually guarded the land from plundering, for Abo-
chorabos always seemed a man to be feared, and exceptionally energetic,
both to the barbarians over whom he ruled and to the enemy no less so.
Formally, therefore, the Emperor holds the Palm Grove, but it is impossible
for him to control it in the slightest, for a land completely destitute of
human beings, and extremely parched, lies between, extending for a distance
of ten days' journey. Moreover the Palm Grove is not worth consideration
at all. The gift Abochorabos presented was purely nominal, and the Emperor
accepted it with full knowledge. The position about the Palm Grove was,
then, something like that. Other Saracens, adjoining these, held the coast-
land ; they are called Maddenoi, subjects of the Homeritai. These Homeritai
inhabit the land further on, by the sea-shore. Beyond them there are many
other stocks, ethne, settled, they say, as far as the cannibal Saracens. After
them are the races, gene, of the Indians, but of these let each man talk as
he may wish.1
Procopius was writing as he wished, more inaccurately than some who wrote
about the Indians. The gulf of 'Aqaba is not east of Palestine. The Palm
Grove, Phoinikon, must be the Tabuk-Taima-Madain Salih region; Saracens
there may have been in the kingdom which held the coast, but they were not
the same people as those of the coast. The statement that the Maddenoi held
the coast from the latitude of Madain Salih to Himyar is contrary to all that
1
'Wars', I, xix, 2-16.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 429

is known, and quite untrustworthy. Procopius was obviously ignorant of

events in which Yotabe played a part shortly before his time. The nonsense
about cannibal Saracens is some vague recollection of the Ichthyophagoi
mentioned in histories of Alexander the Great's campaign along the Persian
coast.
Though this account of the Red Sea compares badly with that of the
Periplus, the statement that ships hugged the Arabian shore is. correct, for it
is confirmed indirectly by the account of the route from Alexandria to Aksum
in Cosmas. In the Periplus, the Arab shore from Leuke Rome to the Burnt
Island is said to be in the hands of Ichthyophagoi and of the Kanraitai,
dangerous pirates ; ships avoided putting in there.1 The complete change of
conditions must be due to the introduction of settled rule; the consequent
trade developments were presumably the reason for the invasion of an unknown
king of Aksum in the 2nd or 3rd century. Political borders and the taxes at
different ports had their effect on the choice of disembarkation points by
traders. Economic conditions changed too ; minted issues, at one time used
only for marginal values in barter, came to be the regular means of exchange.
By A.D. 500 Arabia was prosperous, by comparison with earlier and later
periods; the economy presented some of the features found in peripheral
regions supplying a populous centre. Such prosperity is always unstable.
II. CHRONOLOGY
Of the three chief Greek sources, Procopius is the most reliable as to the
order of the events he mentions. John Malalas of Antioch is not reliable as to
the order of events, but mentions incidents of interest to his readers in Syria.
The indication of date by regnal year is exceptional in either. Theophanes only
undertook his work at the request of his dying friend George the Synkellos ;
he arranged his material under dates, but in a haphazard way, partly no doubt
as a result of errors in converting dates in one system into another. In his
scheme, the 1st year of Diocletian was 5777 anno mundi, 277 anno CJiristi, the
15th year of Varachius king of Persia ; in our reckoning, this was A.D. 285,
and the 10th year of Bahrain II. Many dates in the 6th century depend on the
sequence of events.
The regnal years of the Persian kings were established by Noldeke, from the
figures given by Jacob of Edessa, more often correct than Agathias.2 These
are years according to the Seleucid calendar, but without intercalation after
A.D. 226. The Persian religious festivals were shifted to a month later in the
calendar every 120 years, but this did not affect regnal years, which in A.D. 500
began with our 19th July, in 600 with the 15th June. The first regnal year was
that of the accession, the second began with the next New Year day.
Hisam b. Muhammad al Kalbi gave the reigns of the kings of Hirah by
Persian regnal years ; his text is preserved by Tabari. The source must be a
1
Periplus, 20. The Kanraitai are otherwise unknown.
* Noldeke, 400-34.
430 SIDNEY SMITH

Sassanian register of years, used with remarkable fidelity. Byzantine sources

confirm Hisam. Mundir III was killed in battle near Qinnesrin, fighting Harit
b. Gabala the Ghassanid, in the 27th year of Justinian I,1 554. His son cAmr
was still alive in Indiction XII. 2 The Indictions were cycles of 15 years, named
so from a system of taxation introduced by Diocletian, by kind ; they began
either with A.D. 297 or A.D. 312, so that the indiction meant in this case began
in January 553, the 12th year in January 564. In October of that year Arethas
the phylarch, that is Harit, reported at Byzantium an attack of Abaros, that is
'Amr, on his settlements. Hisam assigned a reign of 16 years to 'Amr. These
dates are as well assured as any in Byzantine history.
The Syriac texts used the Seleucid era and the Syrian calendar of Antioch,
reformed to agree with the Julian year in the time of Augustus. Dates in them
TABLE A.—THE HIRAH DYNASTY

Sassanid
Yrs.3 contemporary Sel. A.D. Yrs.4 Sel. A.D. Lakhmid
4 Ardaair II 691-694 379-382 3 692-716 380-404 'Imru '1 Qais.
5 Sapur III 695-699 383-387 5
11 Bahrain IV 700-710 388-398 11
21 Yazdagird I 711-731 399-419 6
15 717-745 405-433 5 Nu'man I, b.
'Imru '1 Qais
19 Bahrain V 732-750 420^38 14
3
8J 742 -785 430-473 Mundjr I, b.
Nu'man
18 Yazdagird II 751-768 439-456 18
27 Peroz 769-795 457-483 17
10 786-805 474-493 'Aswad b. Mundir
4 Balas 796-799 484nt87 4
43 Kavad. 800-842 488-530 6
7 806-812 494-500 Mundir II, b.
Mundir
6
4
812-815 500-503 Nu'man II, b.
'Aswad
3" 815-817 503-505 'Abu Ya'fur b.
'Alqama
25 • 817-865 505-553 Mundjr UI, b.
Nu'man
47 Khusrau I 843-889 531-577 23
16 866-881 554-569 ' Amr b. Mund.ir
f 888'-892 577-580 Qabus b. Mundjr
12 Hormizd IV 890-901 578-589 3i
Suhrab
4 892-895 580-583 Mundir IV, b.
j.vj.uny^ir
7| 894S-916 592-604 Nu'man III, b.
Mundir
37 Khusrau II 902-938 590-626 14*
22 917-938 604-626 Persian governors
5 Rival claimants 939-942 627-631 -631 Mundir b. Nu'man
k. at Bahrain

1 2
Noldeke, 170, note 1, cites the references. Theophanes, i, p. 371.
3
Figures from Jacob of Edessa ; Agathias generally wrong if varying.
1
Synchronisms : Noldeke, Tabari, 78-9 : 85 : 132-3 : 169-72 : 345-8.
5
The figures presuppose coregencies.
6
Broken years counted with each reign.
7
7 years' interval not accounted for ; possibly Suhrab is in wrong order.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 431

can be converted with certainty into our reckoning. The only question involved
is the reliability of the writer.
Late Sabaean inscriptions are dated by an era which was introduced into
Himyar not later than the year 493.1 The year the nagasi invaded and defeated
Dhu Nuwas is known from Syriac texts, 525. An inscription at Hiisn al Ghurab,
dated 640, has been interpreted as referring to the death of Dhu Nuwas in
battle with the Abyssinians, so that era year 1 was approximately 115 B.C.2
A new view has been proposed by Dr. A. F. L. Beeston,3 who thinks that the
inscription refers to the attack of \Abraha on Sumu-yafa', Esimphaios, and that
era year 640 is approximately A.D. 530. The introduction of an era dating in
southern Arabia may have been connected with the use of the month names
found in the late inscriptions, which differ from those in the earlier periods.
There was intercalation, and the system must have been fairly accurate, to
judge from the correspondence of the era years with the Julian. There may
have been two alternative intercalary months, and the first month was probably,
in theory, that of the spring or autumn equinox. Though the month names
are connected with the seasonal year, it is not necessarily correct to infer that
they did in fact correspond; in correctly intercalated calendars such month
names are sometimes not the exact season described, as they were in use prior
to the introduction of intercalation.
III. THE CAREER OF 'ABRAHA
The accounts of this man in sources of disparate origin and character provide
a fair test of the reliability of the evidence. Procopius began his with the return
of Hellestheaios to Ethiopia after appointing Esimphaios, a native Christian
prince, king.
Of the Ethiopian army, many slaves and all who had a disposition to
lawless behaviour were unwilling to follow the king at all. Left behind, they
stayed there, out of a desire to acquire the land of the Homeritai, for it is
extremely rich. Not long after this mob, with some others, revolted against
Esimphaios and put him in prison 4 in one of the fortresses in that land,
appointing another king for the Homeritai, by name Abramos. This
Abramos was indeed a Christian, but the slave of a Roman citizen in an
Ethiopian city, Adulis, staying there for his commercial undertakings by
sea. On hearing of these events Hellestheaios, seriously desiring to requite
Abramos and the rebels for their treatment of Esimphaios, sent an army of
3,000 men against them, and one of his relatives as ruler. This army, men
no longer willing to perform their task and return home, but disposed to
stay there in a rich land, opened negotiations with Abramos without the
knowledge of the ruler, and came to terms with the opponents. When they
1
RES 3383, unpublished.
2
This was proposed almost as soon as the inscription was known, and the date ia given in
all the standard reference books, e.g. by J. H. Mordtmann, art. Himyar, in the Encyclopaedia of
Islam.
* B80AS, xvi, 37-40.
4
Ka8elp{av: Procopius wrote Attic Greek modelled on Thucydidea and Polybius, so the
natural meaning is not doubtful. There is no reference to a siege.
432 SIDNEY SMITH

were in action, they killed the ruler, joined the enemies' army, and stayed
there. Very angry, Hellestheaios sent another army too against them, which
actually fought an action with the followers of Abramos, but after suffering
a severe defeat in the battle returned home immediately. Thereafter the
Ethiopian king, out of fear, sent no other expedition against Abramos.
When Hellestheaios was dead, Abramos agreed to pay tribute to the
successor to the kingship over the Ethiopians after him. In this way he
secured legitimate rule.1 But this happened later.2
There is African evidence as to the date of the death of the nagasi Hellestheaios,
'Ella 'Asbeha. Silko, king of the Noubadai, was converted to Christianity, and
two missions, one Monophysite, from Theodora, one Orthodox, from Justinian,
were sent to him not long after 540. An inscription of Silko on a temple wall at
Kalabshah, in debased Greek, dates from his pagan years; his title was
'fiaoiXioKosof the Noubadai and all the Ethiopians \ 3 Even if this was a
boast, it shows that the power of Aksum had declined; the death of 'Ella
'Asbeha must be earlier than 540 by about 5 years. If Procopius is trustworthy
the recognition of 'Abraha must be earner than 540, but later than 535.
The earliest Syriac sources do not mention 'Abraha; the ' Book of the
Himyarites' states that the nagasi Kaleb, ('Ella 'Asbeha), conducted the
expedition himself, stayed 7 months, appointed a native Christian king, and
left Kushite administrators, camo yidVo, ' to guard the king against enemies'.
The two primary Arabic accounts are in agreement in not mentioning a native
king. According to Hisam, the naga&i sent two generals against Dhu Nuwas ;
after the victory one of them, 'Abraha, stayed at San'a, refused tribute and rid
himself of the other, 'Ariat, in a duel by a trick. The only important difference
in the account of Ibn Ishaq is that 'Abraha was in the army under 'Ariat. The
duel occurred after 'Ariat had been in command for either two or' some ' years.
The sources are reconcilable ; if there were three armies, the divergencies are
explicable.
A Sabaean inscription at Constantinople, badly broken at both ends, but
complete as to the number of lines, dates from the reign of Sumu-yafa' 'Aswa'.4
It began with an invocation of the Trinity, followed by the king's name and
titulature. In the next line are the names Martad-'ilan 'Ahsan and S§umu-yafa'
'Aswac, sons of Sarah-b-'il, that is Sarah-b-'il Yakmul, the king's brother.5 The
third line recorded the founding of some building, probably in obedience to
' their lords the nagasat (plural) of Aksum '. The remainder of the inscription
1
TTJV apxyv iKpa.Tvva.To: the translation in the Loeb edition, ' he strengthened his rule', does
not adequately render the nuance, see Thucydides iii, 82, 12.
8
'Wars', I, xx, 2-8.
3
The sources are John of Asia, ' Ecclesiastical History', iii, iv, 6—7, Dittenberger I, no. 201,
and Procopius, ' Wars', I, xix, 37, the account of the destruction of the temples at Philae by
command of Justinian.
4
Published partially in BES 3904, fully in Musdon, lix, 165-72, by Professor G. Ryckmans.
The restoration of the royal titulature after the names of the sons by M. J. Ryckmans in Inaitu-
tion, 242, quoted in Museon, Ixvi, 338, and B8OAS., xvi, 38, is not indicated in the preserved
text or in any source. Restoration is not evidence.
5
See p. 458.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 433

must have taken the common form of a series of ' when ' clauses, recounting
previous events. There is a reference to ' their armies, the royal one and that
of the qayl, as to which Dhu . . . gave orders, tqh', and then probably to an
expedition undertaken for ' that king with a mounted force, fyylm, and with
their army with the king'. A broken clause in which there is a mason's error
seems to mention the assumption of kingship by 'Ella 'Asbeha,1 followed by the
statement that he appointed ' kings for Himyar and deputies, 'qbtm, for the
nagasat of Aksum ', so that the Himyarites ' might serve the Aksumite kings '.
The wording distinguishes these Aksumite kings from the nagasat.2 The next
clause dealt with the time when they, presumably ^umu-yafac and his sup-
porters, followed, kSthw, someone, presumably 'Ella 'Asbeha, when he attacked
an enemy and ' overcame them at the surge of the sea '.3 After the prayer,
' Now may (The God) render Himyar prosperous', there is a string of names,
• of interest for re-constituting the genealogy of the Yaz'an family, and a final
invocation of the Trinity.
If Sumu-yafa* was the native, Christian king Esimphaios, the deputies for
the nagasat correspond to 'Ariat and 'Abraha in Arab sources. If, as the
inscription indicates, these deputies were called ' Aksumite kings ', Procopius
called a sub-king ' ruler', archon. The account in Procopius postpones till
after the revolt of Abramos the mention of Abyssinian rulers, who existed,
according to the ' Book of the Himyarites ', from the time of the departure of
the nagaii. But the ' Book ' did not carry the narrative on to the time when
the deputies were called kings, and applies a description implying subordina-
tion. The Arab writers do not mention the native king ; their sources, on this
point, were of northern origin, the territory in contact with the parts ruled
directly by the deputies. The plurality of kings in the nominally united kingdom
was no novelty ; previous rulers associated one or more sons with themselves
as co-regents. The inscription uses the plural nagasat where other sources
mention only one nagasi. The Ethiopic chronicles, and some king-lists, omit the
name of Beta 'Ezre'el, though a coinage of his is known; it has always been a
possible explanation of an apparent lack of information in the chronicles about
a known period that Beta 'Ezre'el was co-regent with 'Ella 'Asbeha.
The divergencies between the sources prove no more than that none gives
a full, correct account, but all are partially true, as might be expected in con-
temporary accounts. The consistent -undervaluation of the Arab sources is
partly due to their inclusion of stories, but also partly, I think, to the Arab
historians' standpoint. Apart from Ibn Khaldun, they had no general theories,
and admitted only contingent causes. After an event, there were, for them, only
1
(k)m imllen H'bfyh mile U>U: mlk is mulk, smlkn, which Professor Ryckmans emends to the
plural 'mlkn, is clear in a well-cut inscription. As the mason omitted the sad in the name, he
more probably omitted the infix t in stmlkn than mistook alif for sin.
2
The assumption that they are identical has obscured the relation of the inscription to the
Arab sources ; the deputies for the nagaidt are known to those sources, not the grant of kingship
to them by the nagaii. The only co-regents with Sumu-yafa* were Abyssinians.
' thlhmw If bbrn: Arabic E\S ' toflow', root ty<.
434 SIDNEY SMITH

statements, a final decision as to exact truth was beyond human power. For
modern Europeans, always willing to make positive decisions and state causes,
however variously, the citation of different accounts through a catena has no
appeal. But the jig-saw puzzle of the material about the 6th century requires
the method of the law-courts. The reliability of each source on each detail
must be tested. For this purpose the method of a Tabari has advantages over
that of Ibn Khaldun—or Procopius.
The duration of Abyssinian rule from the seizure of power by 'Abraha to
the death of his son by the wife of Abu Murra Faiyad Dhu Yazan, Masruq, is
given by the two sources in Tabari as 72 years, clearly an over-estimate. As
often, another account fits. The vahriz went to the Hadramawt twice, once
when he installed Macadi-karib b. Saif (Abu Murra) Dhu Yazan, once when he
went to avenge him. The return of the vahriz was approved by Khusrau I, not
later than 578. Ma'adi-karib ruled for 4 years before he was assassinated by
Abyssinians, 575-8. The battle of the vahriz and Ma'adi-karib is correctly dated
in the 45th year of Anusirwan. Masruq ruled 3 years, his half-brother and
predecessor Yaksum, 2 ; the combined reigns fix the date of the death of
'Abraha in 569/70.1 By then he controlled Taif and Mecca ; only the tribe
Khat'am, south-east of Taif, attempted to resist him.
The expression of doubt about this date continually recurs. The Year of
the Elephant is regarded as part of an extensive fiction. On this basis the
battle of Mons could be treated as unhistorical because of the angels. There
is contrary evidence, but it is instructive. Hisam stated that the Prophet was
born in the reign of cAmr b. Mundir (III) of Hirah, after he had ruled 8 years
8 months, that is in 561/2.2 The date must be wrong, possibly owing to corrup-
tion, the association of the birth with the reign of cAmr is not likely to be
fiction, for there is no motive. cAmr died in 569.3
Some limits will have to be imposed on modern criticism and reconstruction.
Conti Rossini reduced the story of 'Abraha's advance against the Ka'ba, and
the elephant, to some recollection of an Ethiopian king Aphilas of the pagan
period before 'Ezana,.4 There is no evidence to prove that the Meccans had any
recollection of such a time, or that Aphilas ruled so far north as the Meccan
border. As Noldeke saw, there are various Arab traditions, not a single
' Mohammedan ' account. A tribe like Hudail can never have had any contact
with Aphilas, and would not have preserved a tradition about him.
On one point the narrative of Ibn Ishaq can be connected with other sources.
He states that when 'Abraha was imposing Christianity on the northern Arabs,
he appointed Muhammad b. Khuza'i emir of Mudar, and adds incidentally that
Qais, who went with his brother to the territory of the Kinana, fled to 'Abraha
1
Mas'udi, iii, 172, 167, 162.
2
Noldeke, 172.
3
Table A shows that over 7 Seleucid years in the Persian reckoning were not accounted for
in HiSam's synchronisms. The record may have been deficient at this point.
4
He first proposed this theory in Journal Asiatique, 11 Serie, Tome xviii, 29-32; it was
repeated in Storia d'Etiopia, 189, 197.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 435

when Muhammad was murdered.1 According to Procopius this Qais, a prince

of Ma'add, had to flee from Sumu-yafa'. Justinian requested that king to
restore Qais at the same time that he asked for an invasion of Persian territory ;
he would renew that request to 'Abraha. The father of Nonnosos twice went to
Qais before he became a fugitive in Himyar, that is before 530. Nonnossos
himself went to Qais when he ruled some northern region; he must then have
been restored. Later Qais sent his son Mu'awiyah to Justinian, then gave his
emirate to his brother and followed his son ; Justinian made him phylarch of
Palestine.2 The Arab sources show that throughout Qais was in friendly
relations with 'Abraha.
Professor G. Ryckmans has published an inscription 3 which shows why
Qaisfledto Himyar, and how he had an opportunity to return.
By the might of the Merciful and His Messiah: the king, 'Abraha Za
Yabman, king of Saba' and Dhu Raydan and Hadramawt and Yamanat 4
and of ' their' 5 Arabs on the plateau and in Tihamat. They6 have written
this document, stating that 7 they raided Ma'add in the raid in springtime,
in the month of Dhu Tabtan. Further, that they subjected8 all the bany
'Amir. Now the king appointed 'Abi-gabar to the battle force with Kiddat
Wa'il, and Basir b. Husn with Sa'd. And they fought and battled at the
head of the army,9 Kiddat against Bany 'Amir, Murad and &a'd against
Wad . . . (and) Murran, in a wadi on the route to Turaban. And everyone
who applied himself and fought for the king in Haliban slew and took
prisoners.10 And as Ma'add continued (in retreat) they drew closer, dnw kzl
mcdm, and took hostages. And thereafter cAmr b. Muddir gave much and
pledged for them (the hostages) from him (the king), and accepted his rule
over Ma'add.11 And they returned from Ma'add. By the might of the
Merciful. Its month of... ,12 in year 662.
1 2
Noldeke, 203-4, with note. Miiller, Fragmenta Historicorum Graecorum, iv, 179.
3
Museon, lxvi, 275-84, Ry no. 506.
4
Location uncertain ; I take it to be the region north of early Saba'.
6
The plural suffixed pronoun is regarded by some as a plural of majesty, though mlkn takes
a singular verb. The singular suffix in ' Its month of' so-and-so is deictic. Arabs who were in
personal subjection to the king could hardly be put on a level with the kingdoms. The title is
a claim to rule over Arabs dwelling on the plateau and in Tihama as opposed to other Arabs.
• Indefinite for passive. Again sometimes taken as royal plural. The formula is common,
and occurs in other than royal inscriptions.
7
lc : clauses containing dependent statements not infrequently begin thus, instead of with
d, in the late Sabaean inscriptions.
8
qMw: the translation ' rebelled ' is not justified by GIH 541/9-10. For the social classes
q&d-'idsee Rhodokanakis, Altsabaische Texte, i, 49-50. On 3M as ' soldier ', Mordtmann-Mittwoch,
232-4, to my mind convincing proof that both classes were men of military age liable to the
levy. The distinction is between the ruling people and the subject provinces. The verb is both
active and static, ' to subject' and ' to be subject'.
• This preliminary to the main battle was a custom of the period ; the wording distinguishes
the tribes from the main army.
10
whrgw vPerw wmnmw dcsm wmhd mlkn: Professor Ryckmans takes the last w as introducing
the principal clause after a subordinate beginning with wmnmw. If csm has its normal Arabic
sense, mnmw is the subject oihrgw w'srw, the w before it emphatic or an error.
11
ws^hmw cmrm bn mdrn wrhnhmw bnhw wsthlfhw cly mldm: cAmr is the subject throughout,
the 3rd person suffix refers to 'Abraha. The -to- infix forms in the South Arabian inscriptions
commonly mean acceptance of the action expressed without the infix ; compare ntsr, to accept
protection, Urdw, to accept as pleasing, Hmm, to accept guidance.
12
The month name is omitted by error.
436 SIDNEY SMITH

The inscription was found by the Belgian expedition led by Professor RyckmanB
on the face of a granite rock near the well Murayghan. M. J. Byckmans hags
pointed out that the well is not on the route from Hima to 'Aflag and Yamama. -1
About 276 miles north of the well is a small town, Haliban, probably the place
mentioned in the text. Turaban will then be Turaba, about 80 miles south-east
of Taif. Ma'add was defending the approach to some route from 'Aflag to
western centres. The inscription celebrates the deeds of auxiliaries of the main
army of 'Abraha, presumably near their habitat.
Of these auxiliaries, Kiddat are the Kinda of the northern border of
Hadramawt. Murad, mentioned in the 7th century, for instance in connexion
with the chief conspirator in the assassination of cAli, may have occupied the
area round Murayghan. Sa'd may well be the sept of Qurais,1 though fighting
against the Murra, also Qurais.2 The other certain constituent element of
Ma'add, bany 'Amir, were bany Hawazin. It is tempting to restore the broken
name Wada', but the preserved traces do not favour that, and the section of
Kiddat called Kiddat Wada' belong to western Hadramawt. The enemy
confederation stretched northwards from eastern Hadramawt, well away from
western urban centres, as far as the latitude of 'Aflag.
Though described as a ghazzia, the intervention of cAmr b. Mundir proves
that this was a major war. Ma'add had previously recognized the suzerainty of
Hirah, so it was an invasion of Persian territory. M. J. Ryckmans has seen that
it corresponds to the expedition mentioned by Procopius. Accepting the view
that era year 640 was A.D. 525, he dates the inscription 547, and believes that
inexact equivalence of Julian and era years would still permit a connexion with
the war between Khusrau and Justinian from 540 on.3 The active operations
of that war in Syria and Mesopotamia ended with the Persian retreat from
Edessa late in 544, the truce was signed early in 545.4 The continuation of
hostilities in Lazica, the Crimea, would not justify a breach of the peace in a
year beginning not more than four months earlier than January 547. Justinian
sought peace in 544; the attack of 'Abraha cannot be dated later with any
probability.5
Mundir III did not die till 554-; cAmr must have been acting for him in the
south owing to the absence of his father. The part played by Mundir in these
1
Of the a4 dawahir: Mas<udi, iv, 122; for present location, H. StJ. B. Philby, Arabian
Highlands, 25.
2
Mas'udi, iii, 119, iv, 121. The arrangement of Qusaiy by which such tribes at Mecca were
al bitah may correspond to the territories originally held ; if so, there was a long-standing division
of the whole confederation.
3
Museon, lxvi, 342.
4
The dates are fixed by Procopius, ' Wars', I I : from v, 540; from xiv, 8, 541 ; from xx,
542 ; from xxiv, 543 ; from xxvi, 544 ; xxviii, 11, 545.
5
Justinian did not know that the war of the Lakhmid against the Ghassanid would continue.
In 545-6 Mun&ir killed a son of Harit, and later suffered a severe reverse. ' Wars', II, xxviii,
12-14. The ' sacrifice to Aphrodite' is an ' embroidery' like those found in Arab stories, and is
on a level with another by Procopius, the cannibal Saracens. It is surprising that it has been
accepted as historical evidence that Mund.ir was a savage pagan by some who reject historical
statements in the Arab sources because of the very human stories.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 437

Persian wars was active and constant; his forces could face Koman armies in
the field, and were such a threat to Syria that the garrisons could not be taken
on campaign. During 544 he was always in the Euphrates region; 'Abraha
probably timed his attack on that basis. The Abyssinian achieved his own
purpose, but not Justinian's; the cession of the suzerainty over Ma'add by
c
Amr was preferred to the withdrawal of Mundir's army from the north. Once
Ma'add was subject to 'Abraha, it is probable that he would install the former
emir, Qais b. Khuza'i, on whom he could rely ; it is equally probable that he
appointed Muhammad b. Khuza'i to impose Christianity in the Mecca region,
since Qais could support him from his adjoining territory. The evidence inter-
locks, though from quite unconnected sources.
Procopius stated that the expedition was undertaken after Abramos was
firmly established. The dam inscription of 'Abraha at Marib,1 constantly
referred to but rarely used in considering the sequence of events, records how
'Abraha secured rule over a united kingdom in a year when he was already
recognized as the legitimate king by great powers.

The Dam Inscription of 'Abraha

(Face A, 1-8: CIH 1-8.) By the might and aid and mercy of the
Merciful and of His Messiah and of the Holy Spirit. They 2 have written
this inscription: Behold 'Abraha who has been exalted, the king, the
descendant of men of Ge'ez, the ramaihis,3 Za Bayman,4 king of Saba' and
Dhu Raydan and Hadramawt and Yamanat and of ' their ' Arabs on the
plateau and in Tihamat.
(Face A, 9-24: CIH, 9-24.) Now they have written this inscription,
stating that:—He subjected and recognized as ruler by decree Yazid b.
Kabsat in their province, whom they accepted over Kiddat Wada'.5 He
was established as deputy for him and was subject; and with him were
the grayZ-princes of Saba', the men of experience, 'shrn, Murrat and Tamimat
and Hanis and Martad and Hanif Dhu Khalil, and the men of Za'an,6
qayl-princes, Ma'adi-karib b. S\imu-yafa' and Ha'an and their ' brethren',
bany 'Mam. 7 And further, that:—They sent in peace 8 Garrah Dhu ZBNR,9
1
Glaser, transliterated into Hebrew: CIH 541, re-transliterated: Chrestomathia no. 64:
Fakhry I, pp. 79-83, no. 4 bis. M. Fakhry's copy has been extensively used for this translation.
1
See p. 435, note 4. This instance might be thought decisive.
* Unexplained. Abramos, slave of a Roman, may have claimed Roman citizenship when he
revolted from Aksum. The hard breathing might represent the diaeresis in Greek 'Poi/talos.
£ is the Greek a in hrsls.
* Za Yabman at Murayghan ; the official inscription at Marib is likely to be correct. For
Ge'ez names with Za compare the king list printed in Budge, Ethiopia, i, 206-7.
6
Generally translated ' Kiddat and Da>». Compare Kiddat Wa'il. The Yazid family of
Za'an held the Wadi Rukhailah, north of <Azzan, By 340 in Musion, lii, 312-6.
* }z3nn, plural, posits this form; the region yz'n, in Arabic Yazan, has a not uncommon prefix,
see p. 443, n. 3.
7
This may be a tribal name in the plural; if so, Sulaim.
8
%u>: SSS 4193/8 requires this meaning: RES 4084/7-8, ' because the god replied in an
oracle that he would send them away in peace'.
* Ky 520/1, dbnnr might be connected, Museon, Ixvii, pi. 1.
438 SIDNEY SMITH

that he might be supreme* in the east. And the garrisons of Kidar killed
him and plundered.2 And Yazid collected those put under his command,
dhfhw, from Kiddat, and made war on Hadramawt, and seized Mazin, the
runaway,3 the descendant of men of the Dammar sept, and returned to
'Abran.
(Face A, 24-Face B, 18 : CIH 24-46.) Then the report reached them,
and the Habasat and Himyar called up 4 and collected their armies by
thousands 5 in the month Dhu Qayzan of 657. And the generals 6 of Saba3
set out and penetrated from Sirwah to Nabat in 'Abran. And when they
reached Nabat, Kidar had appointed their battle force (to posts). And
Himyar delayed and collected information,7 and their commanders were
Watih and 'Awdah, both Dhu Gadan.8 Then Yazid reached them,9 in
Nabat, and held back his hand from them (the enemy) before appointing
the battle force. And then, wk, a report reached them concerning Saba',
that the ramp and the dyke and the catchment basin and the frontal work10
were broken in the month Dhu Madra'an of (year) 7.
(Face B, 18-27 : CIH 46-55.) "Then after this mandate reached them,
the runners presented a covenantJ1 which the Arab (collective), sept of
Wada', who were under orders12 with Yazid, judged good, (to the effect)
that all of them should hold their hands back, and give them 13 pledges by
the runner, and as to the battle force Kidar had appointed, the gow^-princes
should join (it) to those who were subject.
1
y'fqn : Arabic 'afaqa : the glossary entry, Chrestomathia, incorrect.
2
shtw, as in Arabic, Akkadian sahatu, see Revue d'Assyriologie, xxi, 68-9, common in the
Ma'er archive, often wrongly transliterated with t.
3
hgn : root htvg. A northerner : an ancestor of Ghassan was so named.
4
sf*: the preliminary to a campaign, RES 4193/8-9, causative in CIH 308/19, of calling up
the levy for labour, CIH 541/94.
8
Armies formerly were levied byfifties,yhms : not verbiage.
6
mqly : plural construct of maqawil, (Kamus): not a preposition.
7
Hw wlmd : hendiadys. In CIH 540/78 read klmdw.
8
A section of Za'an, as noted by M. J. Ryckmans in Museon, lxvi, 337. On the location
see von Wissmann—Homer, 92-3.
9
This seems decisive ; Yazid had not rebelled at this time.
10
'rmn wcwdn whbsm wnufrft <Pfn: construction of carim. CIH 540/6-22, of cwd, ibid. 22-8.
Combined they correspond to Akkadian arammu, see F. Thureau-Dangin, Textes mathematiques
babyloniens, 21, no. 45. hbsm corresponds to md'bn in CIH 540/10-11, the reservoir. All had a
facing of quarried stone, grb, and brick, Ibt, here called mdrft, root dfr, used of lining wells, RES
2817/1, and of facing a mswd, RES 4198/2. In spite of the lack of mimation in mdrft, d'fn is the
periphrasis for the genitive : >/, in the form >n/, ' facade,' in the Mineaean RES 3535/1 and
3029/1.
11
brt: the meaning is settled by RES 3689/12 and Glaser 1399 in N. Rhodokanakis, Die
Inschriften an der Mauer von Koblan-Timnac, brtm wm'-brm, gyr brtm, covenanted or agreed without
covenant. Both are texts from Qataban. The word is Hebrew, not cognate with the brt which
represents Arabic buruz(un).
12
gb'w cm yzd: causative in Fakhry I, p. 117,fig.66, II, no. 121, 'to command by law', and
in the broken context CIH 398/8 ; either the causative or the simple form in the Qatabanian
text RES 3858/1 ; the simple form in RES 3908/5,' because (the god) protected him in what he
ordered him ' . For the passive compare CIH 621/8. In Ey 3/4, Museon, xl, 165, the causative
has the specific sense ' to have goods ordered in advance '. Not related in sense to the Arabic
tjaba'a, ' to turn back in a cowardly manner '. Compare Akkadian g/qabii, ' to say ', frequently
' to order '. Aramaic gb>, ' to impose a tax ' is cognate.
13
The other party.
 EVENTS IN AEABIA IN THE 6TH CENTURY A.D. 439

(Face B, 27-Face C, 4 : CIH 55-63.) And the king appointed a time *

incumbent on the tribes (for the return) to ditches and plastering and cut
terraces and bar'a-land 2 and pitch-pit, dflm, and workmen's shops, nhmt,
and mud-bank, sayhurim, for to repair the ramp and the dyke and the
breaches that were in Marib, and imposed a promise on them, in the month
Dhu Sarban of (year) 7.
(Face C, 4-16 : CIH 63-75.) Now after they had appointed the time and
the Arab judged (it) good, they made for the city Marib, and they prayed 3
at the church, b't, of Marib, for therein (in the city) were a priest, father of
a mission (?) 4 and his ' son '. They lifted the ramp away and dug till they
reached the rent,6 and at a point above the rent, to base the dyke.6 And
when they had let (the water) flow away, in order to base the dyke, there
was anxiety and care 7 for the tribes and the city. And all who saw how
this anxiety had come nigh upon the tribes made praises heard for them,
'dnw Ihmw,8 for the 'ahabis of (among) them and the Himyarites of (among)
them.
(Face C, 17-21 : CIH 76-80.) And after that they had made praises
heard for the tribes, the qayl-princes who had fortified themselves in Kidar
went down and when they reached the king 9 with the battle force which
they had appointed to join them,10 then the king and his son held back their
hands (from war with them).
(Face C, 21-28 : CIH 80-87.) The king decreed, gb', in the city Marib :
' The builders of the ramp and the qayl-jirinces who were there are illus-
trissimi \ n Then he had brought before him 12 their ' sons': Aksum Dhu
Ma'afir, the king's son, and Mar-gazzaf Dhu DRNH, and cAdug1S Dhu
Fayis,14 and SYWLMN,15 and Dhu Sa'ban and Dhu Eu'ain 16 and Dhu
Hamdan and Dhu Kula'an and Dhu Mahadd and others (entitled) Dhu, and
I c
ztoi: possibly related to the adverbial <J>^- V. ' never '.
* The source of the material mbr>, CIH 540/, 11, 24-5, 63,fictileclay.
3
qdsw : specifically Christian, contrast dUfor sll, line 92.
4
Glaser, dbmstlh : Fakhry >6 mstlh : meaning guessed.
5 c
rn, defective for <wrn, Hioarun, ' defect'.
6
The necessity of clearing a length of the dyke to repair damage there caused by the breach
in the ramp would increase risks in Marib.
7
Glaser, dllm w(wsm, a common phrase which does not mean pernicies et pestilentia, as CIH
renders : CIH 540/68-9, dllm wmwtm, ' anxiety, even death,' Fakhry, clearly bllm : if this is
correct, compare Arabic balla, ' to moisten ', then ' to exercise benevolence ', the metaphor
explained in Freytag I, 147 a.
8
^jl, ' admiratione affectus fui't', Freytag, i, 23 a. CIH: dimiserunt eos, but the sense
' to permit' would require an addition with 6.
9
Glaser, wkwsfyw .mllen : Fakhry, wkwsjiw hmlkn : CIH, Imlkn, conjecture. The preposition
h for I occurs only in inscriptions from Hadramawt, with the infinitive. At present, emendation
seems out of place.
10
Note the use of the military force on labour for the state.
II
Glaser, .Imtm : CIH slmtm : Fakhry, 'gltm : g and I are indistinguishable, t and y sometimes
confused in the copy. I guess 'glym.
12
mt: i.e. matta, to ask for access to a person : ' when ' makes no sense.
13 (
dl, an appellative, would normally have -m in a personal name.
14
Fayis was not a normal tribe, but a body of courtiers. cAdug was possibly an Abyssinian.
16
So Fakhry. Glaser, Dhu SWLMN.
M
On southern Bu'ain see von Wissmann-Hofner, index. Originally perhaps the stock from
which the dynasty of JJimyar sprang, Mascfldi, iii, 203.
440 SIDNEY SMITH

a strong one, clsm, Dhu Yaz'an and Dhu Dubyan, and thefaj&ir-officersof
Hadramawt and of qrnt.1
(Face C, 28-33 : CIH 87-92.) And it was then, wk, that there reached
them the embassy 2 of the nagaSi, and there reached them an embassy of
the Roman king, and a delegation, tnblt, of the king of Fars, and an envoy,
rsl, of Muddiran (al Mundir), and an envoy of Harit b. Gabalat, and an
envoy of 'Abi-karib b. Gabalat.
(Face C, 33-Face D, 3 : CIH 92-101.) And then after those who wished
had prayed 3 in praise of the Merciful, the tribes went according to the
previous (fixed) time that he (the king) should call them up, according to
their promise 4 for later. And when the tribes arrived at the term of the
later Dhu Da'awn,5 and when they sent them, the tribes, to their duties,
k'sywhmw 's'bn brhmw, they repaired what was broken from the dyke,
about which Ya'fur 6 had given orders, tqh, obligatory on Saba5 and the
gew/Z-princes who were cm tnkn.7 And he gave them instructions.8
(Face D, 4-15: CIH 102-114.) [Describes the repair of a measured
section of the dyke, the ditching to reconstruct the ramp and its plaster
facing, and the improvement of the area irrigated from the catchment basin
so that the outflow could run straight on, lyr 'qdmn, with two channels from
the point of divergence.9
(Face D, 15-27 : CIH 114-137.) And so they completed10 the work
within the period for which they were under obligation at their sections,
bn ywmn dbhw ycfw Igzwhmw. And Yadac-'il and a butcher u gave a blessing
1
Either a region, not included in Hadramawt, or a description, 'foreigners (literally,
opponents),' settled in the land.
2
nitekt: in inscriptions of this period iskt is used of a wife. The occurrences have been dis-
cussed by Dr. Beeston and Professor Ryckmans in Musion, lxv, 279, lxvi, 109-11, lxvii, 103. The
combination btek wr<P wthrg implies thatteledoes not mean' help ' or' instigation ' but something
similar, perhaps ' instruction '. In the damaged text RES 4194/2, >nimtekmfcs>bbythmw must
mean an employee of some sort, not one in command; fa3 is construct before a prepositional
phrase, and accusative, ' the man looking after the household quarters '. In Ry 520/5 it is per-
missible to doubt whether a worshipper of rhmnn would proclaim his polygamy. In the 6th
century women at Nagran engaged in business, tektm may describe the wife who looks after a
man's affairs, and need not be applicable only to a wife, wld in the same texts is not neces-
sarily confined to the literal meaning. On the whole there is no reason to doubt that mh$kt,
not mentioned in these discussions, is cognate and means something like ' those conducting
negotiations for ' a king. Some significance must be attached to the distinction from tnblt.
3
Glaser, dUn bhmd : Fakhry, 411 bhmd, with no space for n, which is not required. 4tt here
for sll, which occurs in the simple form in CIH 540/79. Form: like ^^i for L^i.
4
In the pact; the previous work was voluntary.
5
Either an intercalary month or two months had the same name.
6
Sarah-b-'il Ya'fur; his decree, CIH 540/64-74, included Himyar and Hadramawt, and
was issued as a result of a breach immediately after his own reconstruction.
7
Unexplained. Possibly there was a rota of duty for princes to take charge of annual
plastering. If so, tnkn is a suffixed form from tin.
8
Not' to alter '. The Arabic nakira means ' to be ignorant of' something, but the adjective
nakir(un) seems to retain the sense required here,' intelligent'. The form is presumably intensive.
9
nmry mfggm : nmry dissimilation, dual construct of mamarr(un), Freytag, iv, 165 a. mfggm
root fagga, to diverge.
10
Glaser and Fakhry, rz'w, which would mean' assigned': CIH 540/82, rz'w hrsm... wnskm,
' they distributed gold and electrum '. This does not fit the present context. I read rt'w.
11
tbhm, though the beasts are called dbybm. Note the intentional restriction of the blessing.
The rest was not a sacrifice, but a feast.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 441

at the church and the dyke and the ramp over 5,806 of flour and 26,000 of
dates of the bqnt kind. There were 3,000 slaughtered beasts, both cows and
small cattle, and 300 camels, fast, darkish and with cut ears (she-camels)*
and 11,000 turtle doves, 'ghlb, . . . 2 And they perfected 3 their building
work 4 on the 58th day (that they were responsible for their sections B) in
the 11th hour, in the month Dhu Ma'an of 658.
This narrative establishes a sequence of events. 'Abraha originally held only
the west, Himyar and Saba'. Before era year 657 he reduced Kiddat Wada',
western Hadramawt; by then Sumu-yafa' had been removed, since his son was
with Yazid, the deputy 'Abraha appointed. The next step was to control
eastern Hadramawt and tribes north of it. Resistance was not expected,
but arose in a region that can be located if cAbran is al cAbr.6 Kidar, the enemy,
is otherwise unknown 7 ; it was obviously a considerable confederation from
the measures taken. Some of the Arabs must have been closely related to
Kiddat Wada', Yazid's force. As a preliminary measure Yazid led his own
men to deal with a rebel, of a family associated with the old kingdom of
Ma'in. When 'Abraha's main army arrived, the enemy position caused delay.
Then news of the disaster at Marib produced one of those rare occasions when
Arabs forget minor issues and unite. The initiative came from Yazid's own
men. A covenant fixed a date for all to work at Marib, but the united forces
voluntarily undertook the preliminary steps. Thus 'Abraha was, in the words
of Procopius,firmlyestablished, but by consent, not through general acceptance
of the Abyssinian or of Christianity.
According to one view, era years 657-8 should be 547-8, according to
another 542-3. If era year 662 is not later than 544, 'Abraha received the
embassies not later than 539. There should be a check on this in the political
situation further north.
IV. HIRAH, PHOINIKON, KINDA
In the Lakhmid genealogy a reign of incredible length is assigned to an
historical figure, 'Imru '1-Qais, called al Bad' as founder of a dynasty. This
' king of the Arabs ' ruled the desert border up to Syria, and as far south as
1
Reading "blm sfym wgrbbrn wgsym. The first and the last adjectives are common, but I do
not know the significance of the dark colour.
2
sqym dtmrm : if this applies to the doves, I can make nothing of it.
3
Adopting the restoration in CIH, k(ml)w.
4
There are two nouns mqb, see Mordtmann-Mittwoch, 178-9. ' Booty ' is surely from Iqh,
as Conti Rossini gave it. The meaning ' building work ' I would associate with the Arabic
participle W» yn Freytag, iv, 941a ' expertus '.
5
Restoring btmnyt whmsy ymim wqlw Igzwhmw.
6
von Wissmann—Hofner, 123-4. The eastern end of the incense route, Periplus, 24 : ' The
exports from there (Mouza, Mukha) are either local goods, myrrh, both the choice and the virgin,
from Abeira and Minaea, (or) limestone, and all the cargoes already mentioned from Adulis '.
7
Not to be confused with the kdr mentioned in the inscription of Karib-'il Watar, which von
Wissmann—Hofner, 38, locate near Timnac, unless that was a tribe and had moved. The
narrative of 'Abraha's dam inscription proves that cAbran was a considerable distance east,
perhaps a little north, of Sirwah.
VOL. XVI. PART 3. 31
442 SIDNEY SMITH i

Nagran, more or less the old Nabataean kingdom of the first centuries B.c :
and A.D.1 The extension southwards points to the early importance of Nagran
as the western base of a route to the Persian Gulf. The founder died in A.D. 328,
19 years after the accession of Shapur II who, at the beginning of his active
career, perhaps after 325, dealt severely with the Arabs in southern 'Iraq, and
conquered Bahrain and Yamama almost up to Yatrib.2 The problem of con-
trolling the Arabs in this area was neatly solved; a second 'Imru '1-Qais, also
a founder and called al Bad', was the first king at Hirah of a dynasty which
accepted loyalty to Persia as a condition of existence. The arrangement
endured for two centuries. Though Shapur's campaign in Arabia was attributed
by Hisam to the desire to punish poverty-stricken raiders into 'Iraq, it is clear
that a major Persian interest could be served. By the 4th century the western
shore of the Gulf provided disembarkation points from which goods could be
transported to Mesopotamia and Syria without incurring Persian taxes. Shapur
stopped that. The change between 328 and 380 was such that a Lakhmid was
willing to move from west to east.
Ibn Ishaq stated that Khusrau I made Mundir III of Hirah ruler of Oman,
Bahrain, and Yamama as far as Taif in or shortly after 531.3 This was ap-
parently a re-assertion of the rule of the Persian client in that region. There is
sound reason to accept the statement. A coherent set of grave-stones has been
found within an area reaching from Warka in the north-east to the mainland
opposite Bahrain.4 Professor Ryckmans has suggested that the language is
Safaitic ; it is in any case North Arabic. The script is not that of Safa, but, with
a slight deviation, Sabaean. Unified terms for burials and similar customs are,
in these regions, before the rise of Islam, a symptom of political unity. When the
Lakhmid dynasty was finally driven out of Hirah by the Persians in 604/5, it
maintained a hold on Bahrain; Mundir b. Nu'man (III) b. Mundir (III) was
king there in the year of the embassies to the Prophet, 631.
The north-western border of Hirah, to use the name of the city for the
kingdom, is known from the dispute about the region called Strata, which
Harit the Ghassanid claimed on philological grounds, while Mundir adduced
the customary taxes.5 It lay about the middle of the route from Palmyra to
Damascus.6 From there southwards the western boundary can be deduced,
1
Two bronzes in a classical style, of the 1st century A.D., from 'Ukhdud, with a South
Arabian inscription, show clearly Nabataean influence : British Mueeum Quarterly, xi, 153-6;
G. Ryckmans in Muse'on, lii, 62-4 ; M. Rostovtzeff, Social and Economic History of the Hellenistic
World, ii, 855 and III, 1537 (not quite accurate); Philby, Arabian Highlands, 85. The inscrip-
tion : ' X has dedicated at the canal these recipients to the god Dhu iSamawi': is significant in
the future Christian centre ; consideration of the character of this ' lord of heavenly ones ' must
take local developments into account.
2
Noldeke, 56. » ibid., 238.
4
W. K. Loftus, Travels and Researches, 233 : Capt. Shakespeare in JUGS, 1932, 59, plate at
p. 325; Col. A. P. Dickson in Iraq, x, 1-8: G. Ryckmans in Museon, 1, 239-40 (correct Thay to
Thag): F. V. Winnett in BASOR no. 102 (April, 1946), 4-5.
5
Procopius, ' Wars', II, i, 798.
• Rene Dussaud, Topographic Historique de la Syrie, 255, 286.
 EVENTS IN ARABIA IN THE 6lH CENTURY A.D. 443

partly from the accounts of other kingdoms, partly from the uninhabitable
character of the volcanic stone desert.
'Abi-karib, who sent an envoy to 'Abraha in era year 657 as a ruler equal in
rank with Mundir of Hirah and Harit the Ghassanid, must be the Abochorabos
who made a nominal gift of Phoinikon, the Palm Grove, to Justinian, obviously
early in the Emperor's reign. The northern territory of this Roman phylarch
must have been along the border of the province called Palestine III by the
Roman administration as he was appointed phylarch of the Saracens in
Palestine. The date of this appointment should be about the same as that
known for the similar appointment of Harit or Arethas ; Justinian decided to
follow the Persian example * of creating client kingdoms among the Arabs in or
shortly after 527. It is possible that 'Abi-karib and Harit were brothers,2 and
that Justinian's policy was to use a family opposed to the Lakhmids. The
kingdom of Abochorabos included Yotabe,3 besides the Palm Grove, that is the
Tabuk region ; the father of 'Abi-karib, Gabalat, had also ruled there. In the
time of Anastasius, though Persia and Byzantium were at peace, Skenite Arabs
belonging to the phyle of Naaman, that is Nu'man (II) b. 'Aswad of Hirah,
500/1-503/4, made an incursion into Euphratesia; they had to be met by
a Roman army and were severely defeated by the strategos Eugenius. Theo-
phanes 4 dated this event anno mundi 5990, that is 498, certainly wrongly, as
the mention of Naaman shows. The earliest possible date is 500, the affair was
probably a preliminary step instigated by Kavadh, who began a war in Armenia
in August, 502. In the same year as the victory of Eugenius, Romanus, archon
of the forces in Palestine, undertook a war against Agar the son of Arethas, of
the Thalaban tribe, and took a large number of prisoners ; again, the date is
not earlier than 500. Before the war with Agar, Romanus had defeated and put
to flight another Skenite, Gamalos, who had invaded Palestine before the
appointment of Romanus. Theophanes added: ' Then the Romans freed
Yotabe, the island lying in the gulf of the Red Sea (cAqaba) and subject to no
light taxes to the Roman Emperor, but held in the interval by the Skenite
Arabs. In great battles Romanus gave Roman merchants once again the
opportunity to inhabit the island and to fetch cargoes from the Indians, and to
bring in the tribute appointed by the Roman Emperor.' The Roman aim is
clearly defined, direct access to the Indian trade by state-supported merchants.
Direct rule of Yotabe, restored before 500, lasted till about 527, when 'Abi-karib
was appointed, doubtless under special conditions. The incursion of Agar after

1
Perhaps following Leo's introduction of it, p. 444.
8
If there has been discussion of this point since that of Glaser, it is unknown to me.
8
Not infrequently spelt Jotabe in English reference works. Yo- is connected with the prefix
ya- in regional names already noted (see p. 437, n. 6). For examples in ancient and modern
southern Arabia see von Wissmann—Hofner, 113, Anm. 3, and 140. It is ancient, for Amorite
districts of the 18th century B.C. have such names, given in the Ma'er archive. One instance,
Yamutbal, also called Emutbal, proves that ya- is a separable element, since it is also called
Mutiabal.
« i, 217-8.
444 SIDNEY SMITH

the defeat of Gamalos, that is Gabalat, shows that his kingdom lay immediately
to the east.
The interval in direct Roman rule mentioned by Theophanes, before the re-
conquest, began with the intrusion of Amor-kesos, who is described as a Persian
adventurer, that is an Arab prince from Persian territory, Hirah. He chose to
move into Roman lands, and lived on raiding other Saracens, without at first
attacking Romans. Finally he was strong enough to take possession of Yotabe,
turn out the Greek customs officer, and thus gain considerable wealth. He then
opened negotiations with Byzantium, with a view to being appointed phylarch
of Arabia Petraea. As he sent an ecclesiastic, Amor-kesos was certainly sym-
pathetic to Christianity; at that period he may have been a Christian. The
Emperor Leo accepted the proposal, and in his last year, 473, entertained
Amor-kesos at his capital in a manner that gave offence to some Byzantines.1
In the peace treaty between Theodosius II and Bahram V in 422 this region
had been reserved to the Romans, with the right to exclude Persian Arabs, but
Leo now recognized Amor-kesos as ruler of Yotabe and several Arab tribes, with
the title phylarch.
If Amor-kesos represents ' 5Imru-'l Qais', as is generally assumed,2 the
' adventurer ' was probably a descendant of the first king of that name ; the
kingdom, which for lack of a better name must be called Phoinikon,3 following
Procopius, is practically the same. In that case he was related to the Lakhmids.
If Gamalos-Gabala was the father of Harit the Ghassanid, the move from Hirah
looks like the beginning of the Lakhmid—Ghassanid feud. Like the Nabataeans,
the kings of Phoinikon depended on control of transit from Yotabe, the principal
port, where the merchants, though doubtless Roman citizens, were mainly
Jews according to Procopius, to Palestine. Both Amor-kesos and Abochorabos
were pro-Roman in policy and disposition. From the accession of Zeno
onwards there were intermittent revolts by the Samaritans, sometimes in
combination with orthodox Jews, sometimes not; those in 476 and 529 took
a serious form. They are sometimes attributed to religious persecution, but the
sources are not explicit on this point; destruction of the synagogues may be
a symptom accompanying, rather than the cause of, the revolts. The Samaritans
must have expected assistance ; the mass flight of 529 shows they looked to
Persia, only likely to act through Hirah. In the period between Amor-kesos
and Abochorabos, Gamalos tried an anti-Roman policy; his attack was not
that of a desert shaikh, an explanation no more than facile, but one that
required a considerable administrative change. The situation is obscure, but
instructive.
The succession is sufficient to cover the years from before 470 to about 540.
The end of the reign of Abochorabos must be dated as early as possible, because
1
CMB 472. The source ia Muller, Fragmenta Historicorum Graecorum, iv, fr. 1 of Malchus,
De legibus gentium.
a
GMH, ibid. (E. W. Brooks).
3
Many places were so called ; one in Egypt, see D. Meredith in BSOAS, xvi, 1954, p. 237.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 445

Justinian appointed Qais phylarch of Palestine, obviously in succession to

Abochorabos.1 One date proposed for the dam inscription, 547, is too late for
'Abi-karib.
The Agar driven from the Palestinian border by Romanus is called the ' son '
of Arethas, the Hugr b. 'Amr b. Mu'awiya b. al Harit al 'asgar of the gene-
alogies.2 This early Harit b. 'Amr was, according to the ' saga', sent by Hassan
Tubba' into Ma'add to attack Hirah, and Hisam quoted a story that he killed
Nu'man b. 'Imru '1-Qais 3 ; Nu'man I died in 433. Hassan Yuha'min was, with
his brother iSarah-b-'il Ya'fur, co-regent with his father 'Abi-karib 'As'ad,
whom he accompanied on a hunt, when they rested in the Wadi Masil; his
province therefore bordered on Macadd. §arah-b-'il Ya'fur was sole king in
Saba', Hadramawt and Himyar by era year 564, probably A.D. 448.* The
chronology seems to confirm the ' saga'. Hassan Tubba' is said to have con-
ducted an expedition into Yamama, appointed his step-brother Hugr to rule
there, and kept 'Amr b. Hugr at his court; he was subsequently murdered by
his ' brother' 'Amr and his sister was given by cAmr to Hugr in marriage.
'Amr does not occur as a royal name in Himyar, the assassin was probably the
hostage. The inscriptions suggest6 that as co-regents Sarah-b-'il and Hassan
ruled distinct parts of the kingdom, the latter in the north-east. The ' saga ' is
probably not fiction, but in essentials correct about the origin of the Kinda
dynasty and its relationship with the Himyar dynasty.
Hugr 'akil al murdr,6 about 500, had a son 'Amr, barely mentioned outside
genealogies ; he seems to have been succeeded by Harit b. 'Amr. The family
was closely linked with Hirah in their time. 'Aswad b. Mundir, 474-493, married
'Umm al Malik bint 'Amr b. Hugr; Mundir (III) b. Nu'man married Hind bint
Harit b. 'Amr, and her son was 'Amr b. Mundir, 554-569. 'Amr acted for Mundir
in era year 662. Mundir married Hind as part of some alliance with Harit, an
independent ruler; the probability would seem to be that Harit succeeded
Hugr at some date later than 505. One date in his career can be fixed, if an
1
See pp. 428, 435.
2
References in F. Krenkow, Kinda, in Encyclopaedia of Islam.
3
Noldeke, 148. Extracts relating to two distinct men are put side by side. The confusion
caused by like names is constant.
4
The texts are : Fakhry, i, p. 105, fig. 48, ii, no. 60; Ry 509 in Museon, lxvi, 304, where
«&' is used of a peaceful expedition as in Ry 3 in Museon, xl, 165, line 5 ; CIH 540/52-4. This
'Abi-karib 'Ascad cannot be identified unhesitatingly with the co-regent of Maliki-karib Yuha'min
who was an older brother of Dara'-'Amar 'Ayman in era year 493; the interval is too long.
Otherwise Institution, 318.
6
So, if RES 4105 is to be restored as suggested in Institution, 222, and 'Abi-karib 'As'ad and
Sarah-b-'il were co-regent without Hassan being mentioned. Sarah-b-'il, named first in Fakhry
no. 60, must be the older brother, and subsequently ruled alone. Ry 509 by its nature does not
prove that Sarah-b-'il was co-regent, disappeared and restored.
' The bitter herbs meant must be the Akkadian irru, ' the bitter plant', $a(m)mu marru, see
Campbell Thompson, Assyrian Botany, 223-9, used to alleviate internal pains, identified by him
with the opium poppy, by B. Landsberger with cucumbers. Ibn Ishaq stated that bitter herbs
were introduced among the Arabs during the smallpox plague of 569-70 ; the allusion must be
to a drug, not to vegetables; Noldeke 219, Anm. 2, saw this. Perhaps the drug still used in
Yaman is meant.
446 SIDNEY SMITH

explanation of a confused passage in John Malalas,1 who wrote his Chronologia

about 565, is accepted.
At the same time (immediately after the victory of 'Ella 'Asbeha) it
happened that there was enmity between the dux of Palestine, Diomedos
the silentiarius, and the phylarch Arethas. Arethas in fear entered the inner
border towards the Indian parts, and on hearing this Alamoundaros, the
Saracen of the Persians, falling on the very phylarch of the Romans, slew
him, for he had 3,000 with him. And on hearing these things, the Emperor
Justinian writes to the duces of Phoenicia and Arabia and Mesopotamia, and
to the phylarchs of the eparchies to march against him and pursue him and
his crowd. And immediately Arethas the phylarch and Naaman and
Gnouphas2 and Dionysius the dux of Euphratesia and Sebastian the
chiliarch were marching off with the support army; and Alamoundaros
the Saracen on learning of it fled to the Indian parts with what Saracen
troops he had.3 And the duces of the Romans and the phylarchs with the
support army went into (his land) and as they could not catch him anywhere,
they attacked the Persian parts and seized his tents and took prisoner
a multitude of men and women and children and as many dromedary camels
as they found, and various chattels. They also burned four Persian camps,4
after seizing them and the Persians and Saracens in them, and returned to
the Roman parts after victory.
L. Dindorf, editing the text in 1831, naturally asked about Arethas, ' an a
mortuis revixit ?'. The Ghassanid Harit was certainly not skin by Mundir;
his son was, but in 545-6. The Ghassanid family would not be at enmity with
a dux of Palestine, no member of it wouldfleesouth to ' Indian parts ' and then
receive orders from Justinian, von Gutschmid proposed5 to see in the enemy of
Diomedos Harit b eAmr al Kindi. This is convincing. Though the statement that
Mundir slew a Roman phylarch at this time must be an error, the immediate
reaction of Hirah to the attack on Harit can be combined with other indications
of the true position about 525.
Qais was a refugee at the court of ^umu-yafa15Aswac by 530. He had been
prince of Ma'add. The reason why he became a refugee is apparent from the
later cession of Ma'add to 'Abraha by 'Amr b. Mundir, just as the reason why
he fled to Himyar is to be found in the earlier war of Ma'adi-karib Ya'fur with
Ta'labat and Mundir.6 The Roman Emperor desired the restoration of Qais
because that would stop the activities of Harit b. 'Amr, an adherent of Persia.
Arab sources mention a period when Mundir was driven out of Hirah by Harit
b. 'Amr, and a story narrated by Hisam implies that Kavadh approved.7 But
according to the Byzantine authors, Mundir played a considerable part in the
1
Malalas, pp. 434-^5 ; the end of this work was mauled by an epitomist.
1
Obviously Arabs, Nu c man and Gafna; presumably sons of Harit, whose wife counted
a Gafnah in her pedigree. Nu'man is strange, since it belonged to the Tunukh family, displaced
by that of Salih.
3
He avoided entering acknowledged Persian territory because Persia was at peace with the
Roman Empire.
* The kings of Hirah had Persian troops : Noldeke, 83.
5
Noldeke, 171, Anm. • See p. 461. ' Noldeke, 148-50.
 EVENTS IN ARABIA IN THE 6TH CENTUBY A.D. 447

policy and wars of the Persian king; throughout the years 527-32 he was
fighting in Mesopotamia or Syria. The restoration of Mundir to Hirah in the
Arab accounts was due to Khusrau I ; that was in 531-2, when the truce was
being negotiated. The Arab accounts can be explained. In era year 662 'Amr
acted for Mundir. In the absence of Mundir from Hirah before the winter of
531-2 Harit, his father-in-law, acted for him. Before Kavadh started the war
with Rome in 527, Harit was attacking the Palestinian border, while Mundir
attacked Syria. The gift of the territory of Harit to Mundir by Khusrau*
indicates the time when Harit and his family ceased to play any part in events.
The description of Agar as ' of the tribe Thalaban' has been regarded as
a reference to Taghlib.2 By 600 that tribe, like Bakr, held territory far to the
north, on the border of 'Iraq and Mesopotamia ; there is no indication that in
the first quarter of the century these tribes were in central Arabia. The bany
Ta'laba who acknowledged the Prophet in A.H. 8 occupied territory east of
the Qurais. That corresponds to the region where Ma'adi-karib Ya'fur fought
Ta'labat. From that area men might move into Himyar ; Harit and Martad
of Ta'laban are mentioned among the supporters of Sumu-yafa* 'Aswa'.3 Daus
Dhu Ta'laban fled from Nagran in 523 when the martyrdoms began. It is the
proximity of the territory of Hugr and Harit to the region of Mecca which
explains their prominence in the Arab sources ; actually they seem always to
have been subordinate to, and dependent on, the king of Hirah.4 Their western
border marched with Palestine, southwards past Phoinikon and the independent
cities to about Wadi Masil. North and east lay the provinces directly ruled from
Hirah. To the south was the fluctuating border of Himyar. Dumat aj Jandal,
never mentioned and obviously unimportant, may have been a base in the
north. The main centre lay round Hail and Riyadh.5
In era year 657 both the king of Pars and Mundir sent representatives to
'Abraha. If that year was 547,* there had been peace between Persia and
Byzantium for two years ; no political intention can be deduced. If the year
was 542, the Persian delegates were sent in the middle of a strenuous campaign,
but a political purpose can be divined ; the meeting with their enemies cannot
have been pleasant. If the year is 539, Kavadh and Mundir were planning war.
Justinian, 'Abi-karib, and Harit b. Gabala had a reason for seeking an alliance ;
five years later, era year 662, 'Abraha attacked Mundir's territory, at one time
ruled by Hugr and Harit b. 'Amr, in response to several requests of Justinian,
while Mundir was engaged in the north. All these envoys condoled while at
Marib; it is doubtful whether they were sent simply for that purpose, and
1
ibid., 168. » CMH, 481.
s
Muaion lix, 167, line 13.
* Neither Hugr nor Harit is named in Ry 510, as a recognized feing would normally be.
6
There is no phonetic objection to the identification of tlVbt with the Thalaban of Theophanes.
In Arabic the tribal name is spelt as often with fatha as with dhumma in the first syllable.
• This view would incidentally involve the denial of any connexion between the campaign of
era year 662 and the attack of Abramos on Persian territory described by Procopius. ' Wars 7 ,
I-II, xxviii, 11, contains no reference to any event later than 545 ; see LEE, ii, 420, note 2 . ~
448 SIDNEY SMITH

whether, when they were sent, the kings knew anything about the breach of
the dam.
V. THE EMBASSY OF JULIANUS
In 530 Justinian sent Rufinus to Kavadh to negotiate a peace, but the
Persian king, bent on further gains, gave no definite answer. In 531 Hermogenes
went on the same errand, but only arrived in Mesopotamia in time to take part
in the battle of Kallmikon.1 Procopius attributed the desire of Belisarius to
avoid the action to the advice of the Emperor and Hermogenes. Immediately
after the Persian victory Hermogenes went on to Kavadh. The hope of peace
may have been due to knowledge of the attitude of Khusrau, who accepted the
terms offered shortly after his accession ; a treaty was signed in the spring of
532. But in 531 Kavadh was still alive, and Hermogenes had no success.
Procopius introduced his excursus on the policy of Justinian in negotiations with
Ethiopia and Himyar, that is the attempt to cause a diversion of Persian and
Skenite forces to the south, between the account of Kallinikon and the im-
mediate sequel, the refusal of Kavadh to make peace.2 At that time the chief
concern of Justinian, aware of military inferiority, was to end the war at almost
any price ; an administrator of his calibre would not, at that crisis, have
dispatched an embassy to produce results that might interfere with his imme-
diate aim. Procopius wrote a history ; it is the more necessary to recognize
his bias. The excursus distracts attention from the results of the defeat of
Belisarius. Justinian was left in a weak position in securing terms.
The excursus is increased in length by the description of the southern lands
already quoted, and by an irrelevant discussion of the dealings of Diocletian
and Justinian with the Blemmyes and Nobatai, which calls attention to wasteful
expenditure on untrustworthy barbarians. Ths historical narrative falls into
three parts. The first does not begin with a precise date. ' Then, TOTC, the idea
occurred to the Emperor Justinian to use the Ethiopians and the Homeritai as
allies to the detriment of the Persians.' After the additions, the second section
begins with an account of the political conditions existing at the time of the
dispatch of the embassy ; that required an account of the victory of Helles-
theaios, introduced by the phrase VTTO TOVS \povovs TOV TroXefiov rovSe.
Since the victory is dated on other grounds in 525, the words in this connexion
can only mean ' near 3 the time of this war ' of 527-532, and are misleading,
not absolutely erroneous. Then later events, after the removal of Esimphaios
by Abramos, are narrated to demonstrate the failure of the policy. In the last
section Procopius returned to the step taken by Justinian when he conceived
the idea, ' Tore Se, at that time he sent an ambassador, Julianus'.
Procopius mentions only one embassy to Esimphaios, not several.4 Nonnosos
went on such an embassy, but no precise information emerges from the preserved
1 2
Procopius, ' Wars', I, xviii, 16. ibid., I, xxi, 1.
3
imo generally means ' just after ', but also ' just before ', i-no VVKTO..
4
The statement in BSOAS., xvi, 38, may be based on ' Wars ', i, xx, 13, which deals with
Abramos.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 449

fragments, and there is no mention of him in the chronographers. Nonnosos,

his father Abram (a name which implies Semitic descent) and his grandfather
were used as envoys to Arabs, obviously because they spoke the language;
they did not belong to the aristocratic class of officials. Julianus did, and
finally rose to important posts ; in 540 he was sent as being a secretis with the
son of Eufinus on an embassy to Khusrau. His brother Summus was commander
of the troops in Palestine about 539, when an attempt was made to arbitrate
between Ghassanid and Lakhmid claims to Strata. There is nothing to show
that Procopius knew more than that, from the point of view of Belisarius,
Julianus was sent on a fruitless mission. Malalas copied from an ambassador's
account.1 If Nonnosos accompanied Julianus as interpreter it may be his,
otherwise Julianus wrote it. It is not altogether credible.
The Eoman Emperor, on hearing from Eufinus (in 530) of the evasions
of Koad, sent the sacra (rescript) to the king of the Auxumitai. That king
of the Indians, after making an attack on the king of the Ameritai Indians
and winning a great victory, took his kingdom and his entire land. And he
appointed instead of him as king one of his own race, Anganes, since the
kingdom of the Ameritai Indians was also now subject to him. The Eoman
ambassador, after sailing to Alexandria, reached the Indian parts by the
Nile and the Eed Sea.2 And when he came into the king's presence, the king
of the Indians 3 was carried away with joy, because he had long thought it
right that he should secure an alliance with the Eoman Emperor. According
to the account given by the ambassador himself, when the king of the
Indians received him, he set down in writing the appearance of the royal
audience. The king was naked, round his body at the loins a cloth of linen
and gold, and he was wearing, on belly and shoulders, straps with pearls,
and more than five armlets, and gold rings on his hands. A gold and linen
fasciola * was bound on his head, with four ribbons hanging on either side,5
and there was a golden collar on his neck. He stood above four elephants,
which bore a platform with four wheels, and above as it were a high car
bound round with golden leaves, as are the cars of the archons of the
eparchies,6 and he stood thereon holding a small shield, gilded, two little
lances, also gilded, in his hands. And all his notables were there in arms and
flutes making music for chanting.
And the Eoman ambassador, on being presented, made obeisance by
bending the knee. ' And the king of the Indians ordered that I should rise
and be led to him.' 7 And on accepting the sacra of the Eoman Emperor he
kissed the seal. On further accepting the presents sent by the Emperor, he
1
Malalas, pp. 457-9.
2
Then by ship to Adulis along the Arab shore, the regular route.
3
Either Ameritai or Auxumitai; the later reference to Elesboas is suspect.
4
</>a.Ki6Aw. presumably metal with openwork ribs like rods.
5
Part of the Sassanian insignia, with an earlier prototype in Assyrian reliefs. Such ribbons
adorn a baitulion in the relief published by Professor W. B. Henning in Asia Major, New Series, ii,
plate x. The inscription, p. 174, no. 3, states that the relief was cut by ' townsmen of (the settle-
ment named after) 'Am-wafa3, son of Bod-'Aqi, of the Bany Kuzai'. In the Aramaic of southern
'Iraq the name is equivalent to southern Arabian cAm-wafay or cAm-wafa£.
6
That is, the Arab phylarchs of the Empire.
7
The first person indicates direct copying of the original text.
450 SIDNEY SMITH

admired them. After breaking the seal and listening to the letter in trans-
lation, he found that it included proposals that he should arm against Koad,
the Persian king, and ravage the land bordering on his, and, in future,
should have no further commerce with him, but conduct business with
Egypt in Alexandria, through the land of the Ameritai Indians he ruled, by
way of the Nile. And immediately the king of the Indians, Elesboas, under
the inspection of the Roman ambassador, set war in motion against the
Persians. After sending in advance even the Saracens, Indians subject to
him, he marched into Persian territory on behalf of the Romans, sending
word to the Persian king that he should expect the attack of the king of
the Indians, at war with him, and that he would pillage all the land over
which the Persians ruled. And while these events were in progress, the
king of the Indians, after ordering the presence of the Roman ambassador
and giving him the Mss of peace, dismissed him with every consideration.
For he also sent sacra by an Indian ambassador, and presents for the
Emperor.
A king in Aksum could not take the immediate steps against Persian territory
described ; his deputy in Himyar could. An ambassador in Aksum could not
inspect preparations in Himyar. An Aksumite king, above all a Christian king
there, would not appear naked ; an Abyssinian or an Arab at a Red Sea port
might. The insignia are not those of a nagasi as known from later pictures;
they are Asiatic in type. There are then legitimate reasons for doubting the
account in the text as it stands. On the other hand the attempt to prove that
Malalas concocted the story from Procopius and Nonnosos is unconvincing.
The original suggestion that Procopius himself erred in giving the name Julianus,
whereas the embassy was that of Nonnosos, had to be withdrawn.1 The use of
the word translated ' fasciola ' is found in Nonnosos ; but the word describes a
thing, and may have been commonly used for that rarity. The form Ameritai
for the otherwise usual Homeritai is also found in Nonnosos ; that proves that
Malalas was using a Syrian form, not using Procopius. The account cannot be
dismissed as fiction; the inconsistencies would be removed if the name of
Elesboas were eliminated as a gloss. Malalas, who wrote to entertain, seems to
have reduced a longer narrative, and confused an interview with ' Anganes' by
introducing ' Elesboas '. There is no adequate reason to doubt that Julianus
delivered his rescript, was satisfied that steps were being taken 2 and saw a
ruler in some Himyarite city. The date cannot be later than early 531.
At the time of the embassy, 'Ella 'Asbeha was still the acknowledged suzerain
of Himyar, Sumu-yafac still the native king. One inconvenient result of
equating era year 640 with 530 A.D. is that the proposed interpretation of the
Husn al Ghurab inscription would prove that 'Abraha was akeady in that year
the acknowledged leader of the Abyssinian troops, and therefore in revolt
against 'Ella 'Asbeha. The account in Procopius is then not historical.
In the dam inscription of 'Abraha, a ' year 7 ' is twice mentioned. The
1
J. H. Mordtmann in ZDMG., xxxv, 694.
8
The statement of Procopius that the promise was not carried out will also be correct. When
negotiations for peace began, the Himyarite campaign was abandoned.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 451

narrative deals with era years 657 and 658, and the wording requires the
equation of year 7 with era year 657. M. J. Byckmans has suggested that
' year 7 ' is not an unusual abbreviation but a regnal year.1 There is no other
occurrence of a reckoning by regnal years in Sabaean inscriptions. If the
proposal is correct, era year 651, that of 'Abraha's assumption of kingship, is
more likely to be 533 than 536. According to Procopius 'Ella 'Asbeha lived
some time after the revolt of 'Abraha ; he died before 540 by a few years.2 It
would seem that 533 is a fair approximation for the revolt.
An Arab source, Waqidi, gave the cause of 'Abraha's rebellion 3 : Ariat,
who corresponds in description though not in name to the Anganes of Malalas,
' rewarded the ga^Z-princes and humbled the poor'. That probably means that
he allowed the Abyssinian troops no licence, for the ' Book of the Himyarites '
gives an account of the terror they inflicted before 'Ella 'Asbeha left.4 The
opposition 'Abraha met before he appointed Yazid in Hadramawt was doubtless
inspired by the yoyi-princes said to be ' with Yazid ' when he was deputy.

VI. THE INVASION OF 'ELLA 'ASBEHA

Martyrologies vary considerably in character. However strong the prejudice
they rouse, they are not, as a class, unhistorical, though particular cases may
be. The Martyrium Arethae, the account of the martyrdoms at Nagran,
contains two distinct elements, one ostensibly historical, the other religious
edification along ordinary lines. The text is obviously much later than the
recorded events, but it contains factual information of a kind that can only be
from a source nearly contemporary with the events, a detailed list of the ships
collected by the Abyssinians at Adulis ; as Noldeke saw,5 this must be a Roman
merchants' list, drawn up when Justin I gave his consent to the impounding of
vessels, and sent Daus Dhu Ta'laban, the fugitive from Nagran, to 'Elk 'Asbeha.
Arab sources state the difficulty the nagasi found in securing ships, and Pro-
copius, in his excursus, describes the boats customarily used in the Red Sea in
a way that proves them unsuitable for heavy transport. This matter has been
scantly treated; it is generally assumed that 'Ella 'Asbeha only decided to
invade Himyar after hearing of the martyrdoms. The reason why Daus was
sent from Byzantium to Aksum may be that it was known that the naga&i
intended to invade, and was only prevented by lack of shipping. The Martyrium
gives two dates, October 523 for the march of Dounaas,6 that is Dhu Nuwas,
against Nagran when the martyrdoms took place, and May, 525, for the arrival
of the naga&i. Extreme critics, who distinguish a ' profane' and a ' church'
tradition and regard the latter as a serious exaggeration of the actual importance
of the events at Nagran, have, paradoxically, accepted the dates.

1 2
Institution, 323. p. 432.
• Noldeke, 215. * BH, Syriac text, 49, a, b.
s
Noldeke, 188, Anm.: a remarkable example of critical acumen.
• The attribution of this form (in the Greek text in the accusative) to Procopius in BSOAS.,
xvi, 38, must be an accident.
452 SIDNEY SMITH

No better example of the fallibility of some critical methods and of the need
for constant revision of opinion could be cited than the change introduced by
Professor Axel Moberg's discovery and study of the Syriac ' Book of the
Himyarites'. Moberg proved that this text was written by someone present
at the court of Mundir III in Ramlah, a member of the Monophysite mission.
The writer baptized converted pagans who fled to Hirah because they were
relatives of martyrs. He also showed that there is no essential conflict of fact
between the ' Book ' and the Martyrium, but a difference of statement which
proves independence. The ' Book ' begins the narrative at an earlier date than
the Martyrium. The writer of the Martyrium had access to information unknown
to the author of the ' Book '.
The third, much disputed, Syriac source is the work of John of Ephesus or
Asia, who was born at Amid and driven thence, probably, by the repression of
Monophysites in the time of Ephraem, patriarch of Antioch. He was in
Palestine in 534 and reached Byzantium in 535, where Justinian gave him
important tasks. When severe measures were taken against the Monophysites
by Justin II, from 572 on, he was one of the principal victims, and it was then
he wrote the third and last part of the ' Ecclesiastical History'. In it he
reproduced the text of a letter from Simeon of Beth 'Arsam to Simeon the
archimandrite of Gabtila, dated 20th January, 524. It became a famous docu-
ment, copied in many later books, with the slight differences transmission
causes. In every case, including the text attributed to Dionysius of Tall Mahre,
the source was John of Ephesus.1 He had known the bishop of Beth 'Arsam
personally. The assumption that the document is a forgery has always been
an aspersion on a remarkable character.
The letter, which is certainly not a literary exercise to display powers of
composition, begins with facts. On his way to Ramlah Simeon was mocked by
pagan Taiyaye and Ma'addaye who knew of the repression of Monophysites in
Syria, Persia, and Himyar. Then envoys arrived at Mundir's court from the
king of Himyar with a letter. Of this Simeon gave what he believed to be the
substance in direct form. No writer of the 6th century would have done other-
wise, or have expected a reader to believe that he knew the ipsissima verba ; in
Semitic languages, as in Greek literary tradition, this was the common style.2
The ' letter ' gives what Yusuf was at the time thought to have said ; it begins
with facts that were known. The Kushite king was dead. The king of Himyar
had seized the opportunity to make himself king before a Christian could be
installed. The main part consists of an account of the extermination of Chris-
1
See the Abbe Nau in JA., 9 Serie, viii, 346 ff., ix, 529. Noldeke's review in WZKM., 1896,
160-1.
2
Mordtmann in ZDMG., xxxv, 700, criticizing Noldeke, invoked the name of Bichard
Bentley. Bentley proved, by cumulative evidence, that Phalaris was a late school exercise, and
not a particularly good one ; he was contradicting those who stated that it was one of the most
ancient and admirable works of Greek literature. He did not claim to have proved that every
letter preserved in an ancient book is a ' forgery'. He would have distinguished between the
reproduction of Simeon's dated letter, and the ' letter ' in it.
 EVENTS m ARABIA IN THE 6TH CENTURY A.D. 453

tians and the story of the martyrdom of a woman, Rome, and her ' daughters '.
It closes with the suggestion that Mundir should follow this example. The
object of Simeon's letter is clear. He aimed at reporting the story in such a way
as to rouse indignation, so that no Christians could claim that the Monophysites
were indifferent to the sufferings of other Christians.
The opposition between a ' profane ' and a ' church ' tradition is only
justified by the absence of any mention of Justinian and political motives in the
three Syriac or Syrian sources. The ' profane' writers dealing with political
events do not fail to mention the massacres; Procopius did, though with an
aside to show how little his contemporaries knew of Greek culture when they
called barbarian paganism Hellenic. The similar treatment of the Arab sources
as a single ' Mohammedan ' tradition implies a unity in them which does not
exist. One agrees with the Martyrium that 'Abraha or Abram was with the
first Abyssinian army; another agrees as to his mildness of temper. The
common point in the Arab writers as against others is their silence about the
appointment of a native Christian king, and that is to be explained as due to the
limited extent of that king's control. The slight, not essential, differences,
increase the value as evidence. The 'Ariat episodes in the narratives repeated
by Tabari are not unhistorical; he corresponds to the relative of Hellestheaios
mentioned by Procopius and the Anganes of Malalas, possibly also to the
Aidog of Pseudo-Dionysius and the Anzug of Michael the Syrian.1 The reign
of &umu-yafac, the rule of 'Ariat as deputy for the nagasi, posit an interval
between the victory of 'Ella 'Asbeha and the revolt of 'Abraha which must
cover the years 525 to about 533.
Only the chapter headings of the ' Book of the Himyarites ' are preserved
for sections i-vii. The original text told, in chapter ii, of ' whence they (the
Himyarites) first received Judaism'; in iii-iv an account of Christianity in
Himyar ended with a mission of Bishop Thomas to the Kushites to inform them
of persecution. This was followed, in v, by the first coming of hywn' and the
Kushites, then in vii by the departure of the Kushites from Himyar. The
preserved text begins at that point and continues with the account of the ' first
persecution by Masruq '. The only natural interpretation is that there was an
Abyssinian invasion immediately before Masruq became king, and two distinct
persecutions by him. The hywn' who first invaded Himyar was an Abyssinian
leader earlier than 'Ella 'Asbeha ; Bishop Thomas reported a persecution which
was not the first persecution by Masruq. Professor Moberg considered the
possibility that hywn' is simply haywdn, ' the wild beast', and rejected it.2 In
discussing the date of Bishop Thomas, he pointed out that the Christian sources
are agreed that Bishop Paul died two years before the persecution of Masruq,
when his bones were disinterred and dishonoured; Moberg therefore left two
alternatives, either Thomas was not bishop of Nagran, or he was earlier than
the reign of Dhu Nuwas.3
1
J.-B. Chabot, Chronique de Michel le Syrien, tome ii, 184.
1 3
BH, xo. ibid., Ii.
454 SIDNEY SMITH

Ever since the time Ludolf's work was published it has been known that
there was an Ethiopian legend about the way 'Ella 'Asbeha rid his country of
a king who ruled for 9 years, called 'Arwe,' the beast,' 1 probably a laqab which
gave rise to the legend. Cosmas Indicopleustes, who was trading at Adulis
before the accession of Justin I in 518, mentions in the introduction to his
' Christian Topography' the instruction he received in his early years from
Patricius, later the Nestorian patriarch in Persia. Patricius was accompanied
everywhere by the teacher of divinity Thomas of Edessa.2 The appellation
episkopos is not necessarily decisive against the correlation of the sources.
At the beginning of the ' letter' of the Jewish king of Himyar to Mundir of
Hirah, in which Simeon of Beth 'Arsam reported the substance the original
document was thought to contain, there is a statement that the Kushite king
was dead and that this opportunity to seize the throne was taken. It might
seem natural to infer from this that the accession of the Jewish king was not
long before 524. But that is an inference, not necessarily correct. If the Jewish
king opened negotiations then, he would naturally, in this literary style, give
some account of his reign; the ' letters' between the Emperors and the Persian
kings in Procopius can be compared. It is admissible to connect the Kushite
king who died immediately before the accession of Dhu Nuwas with the hyvm?
who invaded Himyar.
In the second book of the ' Christian Topography ', written before 547 by
a few years,3 Cosmas told how he and another Egyptian Greek made copies of
inscriptions on two fallen stone monuments ' at the beginning of the reign of
Justin (I) the Roman Emperor, while I was present in those parts, 25 years
more or less before these (present) years, Ellatzbaas, the king then, intending
to march to war against the Homeritai of the opposite coast '.* Cosmas was,
apart from his conviction that deductions from Scripture required rejection of
the reasoning of pagan philosophers, a pragmatical person, especially about
figures. When he said ' the beginning of the reign of Justin' he meant 518, not
525 or even 523. His doubt, which led to the ' more or less', concerned the
interval between the writing and the reading of the book, ' these years.' Had
he known of the martyrdoms as the cause of the projected invasion, he would
have mentioned them. 'Ella 'Asbeha was nagasi by 518.
Cosmas thought that the inscriptions were continuous, but the two are
distinct. One was the broken triumphal inscription of a king of Aksum now
called the ' Monumentum Adulitanum '.5 The reason why Ellatzbaas wanted
1
Budge, Ethiopia, i, 261.
2
Cosmas, Logos B, 73 : p. 52.
3
The first five books were published separately, books VI-XII later in reply to adverse
criticism. The eclipses dated by modern calculation to 547 are described at the beginning of
book VI in an argument meant to refute attacks ; book II must be earlier.
4
Cosmas, II, 101 C : p. 72.
5
Dittenberger, II, no. 198, pp. 292—4. The date of this inscription is still uncertain. Conti
Rossini connected it with Aphilas. The Kinaidokolpitai, ' men of Kogues' Gulf', correspond to
the Kanraitai.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 455

this text can be discerned. The unknown earlier king said: ' I sent a naval
and land force across the Eed Sea against the Arrhabitai and Kinaidokolpitai
dwelling there, and imposed taxes on their kings. I ordered them to be tributary
for their land and to go by road or to sail peacefully. I made war from Leuke
Kome to the Sabaeans' country.' The region conquered was the pirate coast of
the Periplus. Ellatzbaas wanted a document proving ancient right to lost
territory.
The long interval between 518 and 525 is explicable as due to the shipping
difficulty. The opposite coast could be defended against open boats ; Roman
merchant vessels were equipped for defence. The Abyssinians did not dare, or
were unable, to impound such ships till Justin gave permission in 525.
The Husn al Ghurab inscription x begins with a list of names, headed by
that of £umu-yafac, not called king, many of them tribal. Miss Hofner's valuable
study 2 shows that they belong to Hadramawt. Former translations and
interpretations require revision.
S\unu-yafac 'Aswac and his sons Sarah-b-'il Yakmul and Ma'adi-karib
Ya'fur,3 bany Lahay'at Yarhum,4 sept Kula'an, and Dhu Yaz'an and Gada-
num (followed by 19 names) 5 and bany Milh and their tribes (3 names) and
Dhu Yaqtan (5 names) and the kabir-ofUceis and executive officials, mhrg,
of Sayiban Dhu Nasf.
They wrote this inscription on the fortress-rock, V, Mawiyat, stating
that:—He restored it, its surrounding wall and the gate and its watch-towers
and its paved stone ways.
And that:—He accepted the order to fortify, itsn', therein when they
were commanded, Jcgb'w, from the land of Habasat. And the 'ahabiS * sent
the fast detachment in the land of Himyar7 when the king of Himyar and
his gw/J-princes had caused men of Himyar and the men of Rahab to be
slain.8 Its month Dhu Higgatin of year 640.
Were the interpretation of this inscription as the record of an unknown siege
1
CIH 621 : Chrestomathia no. 65: K. Maker in WZKM., xxxiv, 54-75 : RES no. 2633.
2
von Wissmann—Hofner, 92-3.
3
Thefirstson named after an uncle ; the second after the predecessor of Yusnf, and associated
with Yazid, the deputy of 'Abraha.
1
bany is not used literally here ; this Lahay'at Yarhum should be the traditional founder of
the family, not the brother of the qayl Sarafc-'il Yaqbul, and not the son of Sarah-b-'il Yakmul.
* Including hsrn. Mlaker suggested that this tribe inhabited the Wadi Kasr, described in von
Wissmann-Hofher, 125 S. The writer of the Periplus heard that Eudaimon, Aden, was taken by
Kaisar after the time of Chariba-el, i.e. Karibi-'il Watar Yuhan'im. Frisk, Periplus, 110-1,
rightly joined others in rejecting emendation of Kaisar into Ilasar or anything else. Could a Dhu
Kaisaran be meant ?
• Abyssinians resident in Arabia, as opposed to Habasat.
' w'&yw 'frbin zrftn b}r4 tmyrm: the Syriac zriphutha is an abstract, and could not be the
object of a verb meaning ' to send '. zrft is the Arabic zarafatun, a small troop, here messengers.
8
khrgw mlk bmyrm w'qwlhw >hmrm w'ribn : usually translated ' when they (the Abyssinians)
slew the king of Himyar and his princes, men of Himyar and Rahab'. If 'frmrm were in apposition
to 'qwlhw, nunation would be required; the tautology after mlk Jymyrm. is improbable. The
distinctions ' (some) Himyarites ', ' the men of Rahab ', is due to the peculiar, but undefinable,
importance of Rahab, cf. CIH 540/79-80, where the reading seems to be slwhmw w'r'shmw
iD»bty rhbm. von Wissmann-Hofher, 23, connect the name with Khuraibat Rahaba near Sirwah.
456 SIDNEY SMITH

of £umu-yafac in Kane in A.D. 530 correct, 'Abraha had already revolted from
'Ella 'Asbeha in that year. The position when Julianus was sent from Byzan-
tium was not that described by Procopius. &umu-yafa who certainly took the
title king abandoned it when 'Abraha revolted, while still able to resist. Over
17 years elapsed between the revolt of 'Abraha and his first attempt to impose
rule on eastern Hadramawt. Procopius, whose work was published at latest in
551,1 referred to an expedition of 'Abraha in 552.
The equation of era year 640 with A.D. 525 arose because it was generally
assumed that there is a reference to the death of Dhu Nuwas in battle with the
Abyssmians at the end of the inscription. The inscription thus became the
record of an act of treason by £umu-yafac and his party, and of the defeat and
slaughter of his kinsmen. The narrative of Hisam 2 provides a different and
better explanation. After the massacre of Christians, when Dhu Nuwas heard
of the Abyssinian expedition and summoned the 'aqwdl to defend the land, they
refused. According to Ibn Ishaq, the Himyarites and subject tribes of Yaman
assembled but were disunited. The Husn al Ghurab inscription confirms and
supplements the sources ; in it &umu-yafac gave the reason for his acceptance
of orders from the Abyssinians and dissociated himself from the ' king of
Himyar', denying the full title otherwise always found in this period. The
year 640 should be prior to, not later than, the death of Dhu Nuwas. The last,
most violent phase of the massacres culminated at Nagran in 523. Sumu-yafa'
must have dissociated himself at an earlier stage, openly, from Yusuf. The
situation points to the equation of era year 640 with 522.

VII. DHU NUWAS

The ' Book of the Himyarites ' did not introduce the name of the ' crucifier '
till after the account of the first Abyssinian withdrawal. It calls him Masriiq;
the accursed name is written upside down.3 In the Martyrium he is given his
original title, Dhu Nuwas, not his name. The reliability of the Arab sources,
for all the divergence in detail, has now been partially demonstrated. A king
Lahay'at Yanuf called Dhu Sanatir,4 murdered members of the royal family
till one, the boy Zurcah, murdered him. A qayl of the time of i3umu-yafac was
named Zurcat Dhu Marhabum 5 ; it is then a possible name for a member of the
royal family. The boy took the name Yusuf; there was a king Yusuf 'As'ar at
1
K. Krumbacher, Geschichte der byzantinischen Litteratur, 42.
2
Ndldeke, 191, Ibn Ishaq : 194-5, Hisam.
3
The word is naturally to be connected with Syriac seriqo, ' worthless'. The name of
'Abraha's second son, also Masruq, is probably a pure Arabic laqab not the royal name.
4
This seems to be a play on the regional name Satiran, Fakhry II, no. 63, or the tribal Bany
Satiran ; in the pagan period a Icabir of the Fayisan was named Lahay'at Satiran, RES 3913/1
and Fakhry II, no. 61. It cannot be connected with Sanatruces, the Parthian prince of Hatra
who met Trajan : that name was discussed by Noldeke, 500, and has been found at Hatrah by
Sayid Fuad Safar, Sumer, VIII, 191-2, nos. 36 and 37, and IX, 19. Otherwise Arabische Frage,
490. Sanatir is not a place name, and none of the traditional explanations seems likely ; could
it be a laqab based on a combination of sana1 and satir,' infamous scoundrel' ?
5
Museon, lix, 167, line 12. Marhabum is presumably collective, Rahab.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 457

this time. Lahay'at Yanuf was, with his brothers Nawf and Ma'adi-karib
Yan'um, co-regent with his father Sarah-b-'il Yakuf in era year 582 x ; if he
lived till era year 595 or 600, Yusuf was a youth by his death. Only one element
of fact in the stories is demonstrably incorrect; Zur'ah cannot have been
proclaimed king immediately on the death of Lahay'at Yanuf, his predecessor
was Ma'adi-karib Ya'fur, called in the' Book of the Himyarites ' Ma'adi-karim.2
Two slightly different accounts of the same campaign, one of them originally
found by Mr. H. St.J. B. Philby, both of them now admirably published,
come from sites on the western side of the great desert. One is on a rock near
the well of Hima, the other on the rock Kawkab, some 50 miles to the north-
east, on the route from Nagran to al 'Aqiq, west of Sulayil. By the side of the
main inscription at Kawkab men with the principal qayl added their own
names. One Tamim, muqtawi, that is an officer of the household, of Lahay'at
Yarhum Dhu Gadanum, ended with a prayer : ' Now may the Merciful have
mercy on the Bany Malik,' a sept of Qurais 3 from which he obviously sprang.
Two others, Mu'awiyat b. Wali'at, and Na'amat b. Malik, similar officers, mqtt,
of the principal qayl, ended : ' By the Merciful, Lord of Jewry ' ; in spite of all
critical doubts in the past, the sources are confirmed as to the king's faith. The
longest adscript reads : ' The qayl Sarah-b-'il Yakmul has passed through the
wide areas of conflict, hagay 'ayhar haigin,*1 when he joined with his brother and
lord, the qayl Sarah-'il Yaqbul Dhu Yaz'an'.
To the confusion caused by the repetition of nomina and cognomina in
different generations there is added the doubt inspired by the use of bany.
When Sumu-yafa' 'Aswac and his two sons are called 'bany Lahay'at Yarhum,
sept Kula'an', Lahay'at Yarhum, in normal Arab usage, is an ancestor. If he
were the father of Sumu-yafa', he would not be the Lahay'at Yarhum who was
Dhu Gadanum, a sub-section, on campaign in era year 633. An undated
inscription 5 at the junction of Wadi Durra and Wadi 'Abadan records repair
of an irrigation system by :—
Sarah-b-'il Yakmul and Sarah-'il Yaqbul and Martad-'ilan 'Ahsan and
Sumu-yafac 'Aswac, and the sons of Sarah-b-'il, Lahay'at Yarhum and. $umu-
yafac 'Aswa', sept Yaz'an and Gadanum and . . . and Yasbur and . . . of
their tribe Dayafatan and Ratihum and the mahirat 8 andfcafoV-officersof
the tribe Sayiban.
This document established proprietary rights in water. Four heads of all
1
Institution, 341-6.
2
BH, Syriao Text, 43 b. In Akkadian as in Sabaean the root is Icarabu, in Arabic karama.
Lidzbarski's assertion, Ephemeris, I, 263, that m is ' original', paid little attention to the
phenomena. Alternation from 6 toTOnot only characterizes different languages of the Semitic
family, but can occur within one and the same language.
3
One of the a4 dawahir: Mas'iidi, iv, 122.
4
Boot hyg.
6
Ry 63, RES 4069, discussed with different results by M. J. Eyckmans in Institution, 245-50
and Mustkm lxvi, 337.
• The word mahir was used in ancient Egyptian of an official scribe, and in texts from Ras
as Samra of a household official of the goddess, <Anat, ii, 15, 21.
VOL. XVI. PAST 3 . 32
458 SIDNEY SMITH

Yaz'an agreed to the work. Sarah-b-'il is named first as the land concerned was
his. The unusual order,' the sons of Sarah-b-'il' before the next proper names,
is a matter of emphasis 1 ; these juniors were mentioned because they were
heirs to the land, superior to the distiict officers, not because their consent was
necessary. One of the sons was named after his uncle ; the recurrence of the
two names together in the Constantinople inscription proves that it was the
older man who became king.2 When he did, there was another son of Sarah-b-'il
named after an uncle, Martad-'ilan 'Ahsan. The undated text should be earlier
than era year 633, when Sarah-'il was the acknowledged ' lord '. Lost names
at the beginning of the Hima inscription are followed by:—' their sons,
Sarah-b-'il Yakmul and Ha'an, men of Nasa'an, and Lahay'at Yarhum and
Martad-'il Yaml(?)ad.' At the end is the statement that the qayl Sarah-'il Dhu
Yaz'an had the inscription cut. The proof that Yaz'an was united in support
of Yusuf in era year 633 is to be found in the text at Kawkab.3
The qayl Sarah-'il Yaqbul b. Sarah-b-'il Yakmul, banu Yaz'an and
Gadanum and Nasa'an and Habum and Gaba'. In this inscription are
written down, tstrw,* the things they accomplished in a campaign by wadi
beds and a river valley 5 for their lord the king Yusuf 'As'ar against the
'ahabis in Zafar. Now they overthrew the church, qlsn. Then the king
marched to 'As'aran and appointed him over an army. And he laid waste
Muhwan (Mukha) and killed all its inhabitants and overthrew the church.
And he laid waste all the forts of Sammar and its plain. And the king made
a surprise attack on 'As'aran. And all that the troops of the king slew and
took as booty was collected, tgm', 13,000 slain and 9,500 prisoners6 and
280,000 camels and oxen and goats. And thereupon 7 the king appointed
him to join against Nagran amongst the heads of families of bani 'Az'an,
qrm bn 'z'n, and among the tribes of Hamdan and their city, and their
Arabs, and the Arabs of Kiddat and Murad and Madhig. And the king
ordered that sslt mdbn (or rrddn: chains of servitude ?) should be fastened,
hrzy, onfilesof the Habasat, bmqrnt hbst, and to the ' man ' of §anca, wlsn'n.
In his armies and with him were Lahay'at Yarhum and Sumu-yafac •'Aswa'
and Sarah-b-'il 'As'ad, sept Yaz'an, with their tribe the men of Za'an. Its
month, Dhu Qayzan of 633. Now may The God to whom the heaven and
the earth belong protect the king Yusuf against his enemies, and may this
inscription be under the ban of the Merciful against any who would erase
or deface it. 0 Merciful, show Thy mercy for ever. Thou art Lord.
1
Emendations proposed neglect the nature of the document.
2 3
See pp. 432-3. Museon, lxvi, 295-303: Ry no. 508.
4
This passive construction is parallel to the indefinite
5 }
udk khm rcfm : the h must represent an adverbial ending in a ; Jchm is an unusual ortho-
graphy for km, representing the long vowel; rgcm is the Arabic rajcun, a place through which
water runs.
6
Thesefiguresmay mitigate criticism of the credibility of the Syriac sources ; they obviously
derived their figures from accounts current in the Himyarite kingdom. There is no reliable
criterion for any estimate of the population.
7
wbn dkyhw mlhn : this temporal adverb corresponds to the Phoenician locative adverb bn,
Eshmunazar 5, ' therein '. This use of deictic n is comparable to that in the compound preposi-
tion In, ' from ', where it anticipates a noun in apposition, and in the conjunction bcdn, where it
introduces a subordinate clause. Dr. S. Y. Bakr has proved to me that this corresponds to the
-ma mukqffah of the Arab grammarians.
 EVENTS m ARABIA IN THE 6TH CENTURY A.D. 459

If era year 633 was A.D. 523, !§umu-yafac 'Aswa', who was appointed king in
525 as a Christian, was among those responsible for the martyrdoms at Nagran
in what the ' Book of the Himyarites ' seems to have described as the second
persecution. It seems an unlikely qualification for the man whose name in the
' Book ', 'Aswar (?), is an easy corruption of 'Aswa'.1 If era year 640 was 525,
the Abyssinians still held Zafar, Muza, and Nagran in 518, though the statement
is Cosmas proves that the nagasi was planning an invasion then. Were there
reason to distrust the sources, these points could be evaded ; it has yet to be
proved that they merit distrust. If era year 633 was 515, the Abyssinian with-
drawal, after the ' first coming ' of which the ' Book ' gave an account, took
place three years before the nagasi's preparations for invasion.
The preserved text of the ' Book ' tells of the protest made to Yus"uf by
riSone d'ammeh, that is yayZ-princes, when he ordered that even the bodies of
the rich woman Ruhm, of her daughter and of her grand-daughter should be
dishonoured.2 Euhm had done much good to high and low, princes and common
folk. On one occasion Macadi-karim, the king, had been in need of money, and
she had lent him 12,000 dinars on interest. On hearing later that he was unable
to pay, she remitted the debt. The princes demanded that she should be
decently buried, and Masruq had to give way. If some were disgusted during
the second persecution, others, particularly Christians, may have become so
earlier. The logical sequence should be that the &uinu-yafac who fought for
Yusuf in era year 633 to drive out foreigners and overthrow their churches, was
in era year 640 opposed to the king of Himyar because native Christians had
been massacred ; he therefore accepted orders from the nagasi. Later, in A.D.
525, after the success of the Abyssinians, he was appointed as the Christian
king, but was not completely independent. The Constantinople inscription
gives his version of the events.3
The leaders of the 'Az'an in era year 633 were Sarah-'il Yaqbul Dhu Yaz'an,
Lahay'at Yarhum, Sumu-yafac 'Aswac, Sarah-b-'il 'As'ad, §arah-b-'il Yakmul,
all Yaz'an, Ha'an of Nasa'an and Martad-'il Yaml(?)ad, with others whose
names are lost. In the Constantinople inscription, Martad-'ilan 'Ahsan and
&umu-yafac 'Aswac, sons of Sarah-b-'il (Yakmul), are associated with the king
SSumu-yafa' 'Aswac; they must have joined the Christian party before era
year 640. In the dam inscription of 'Abraha, Hacan is named as a qayl with
Yazld ; he too must have joined the $umu-yafa' who became king, or he would
have lost his princedom. On the other hand the ' Book of the Himyarites '
often mentions ' the crucifier Dhu Yazan ' as the principal oificer of Masruq in
the second persecution; he must be Sarah-'il Yaqbul. There was a division
within the Yaz'an family between era years 633 and 640.
The expedition of era year 633 was directed by a king whose main force
1
Moberg's discussion of the reading in BH, clxvii-clxix. The corruption has been recognized
by M. J. Ryckmans.
2
BH, Syriac text, 43b.
' See pp. 432-3.
VOL. XVI. PAST 3 . 32*
460 SIDNEY SMITH

cons sted of Arabs and Hadramawt princes against Abyssinians. The first
attack was on the Abyssinian capital, Zafar, the church is called qlhn, ecclesia,
not b% as in the dam inscription of 'Abraha ; it was apparently treated as the
symbol of foreign domination. The massacres of Abyssinians were savage, but
not such as to shock national feeling. At Nagran the inscription mentions only
enslavement, a fate shared by some of the people of San'a. In the ' Book of the
Himyarites ', the account of the first persecution begins with the burning of
Abyssinians left in Zafar who had taken refuge in the church ; it then states
that Masruq sent orders to the provinces that all Christians who would not
deny their faith should be killed.1 There were two phases. Men who consented
to the massacre of Abyssinians and the overthrow of their churches might not
approve the extension to all Christians in the kingdom.
Some obviously did approve, Dhu Yaz'an for example. There must have
been some reason. Perhaps the story of Euhm will explain this transference of
hate from foreigners to fellow-countrymen. If she lent money to the king, she
did so to others, though she gave to the poor. If the household of the leading
Christian, the martyr Arethas, engaged in money-lending, other Christians
would. The same cause has led to the same result so often that the inference
hardly needs support.
The inscription at Kawkab is dated in Dhu Qayzan, earlier in the year than
that at Hima, dated in Dhu Madra'an. Since both record the expedition it was
over, and no inference can legitimately be drawn as to its course. But it is
significant that after the campaign these princes stayed in this remote region.
The fact is to be coupled with the description of the force which advanced
against Nagran. All 'Az'an, that is roughly western Hadramawt, and Hamdan,
the high plateau near San'a, some of whose inhabitants were treated like
Abyssinians, were united with Arabs of the north-east of the kingdom. The
princes retired to the north-east presumably because their strength lay there.
Two years earlier there had been a campaign in the reverse direction,
recorded in the Wadi Masil, halfway from Mecca to Riyadh.2
Ma'adi-karib Ya'fur, king of Saba' and Dhu Raydan and Hadramaut,
hdrmt, and Yamanat and of ' their ' Arabs on the plateau and in Tihamat.
This inscription is displayed, hwrw, and made lasting3 in Ma'^il Gumhan
concerning a campaign 4 on the heights5 of Kat'a, because the Arab
(collective), a subject, qsdm, excluded them,8 and Muddir made war with
them. Now they went on campaign with their tribes, Saba' and Himyar and
1
BH, Syriac text, 7 b.
2
Ry 446, now 510, Mtiseon, lxvi, 307-10.
3
wlf: land charters were called wtf as being grants for ever.
4 l
ly mhn sbHm : mhn is surely an indefinite pronoun, ma and >hn.
5 c
rg : 'cliff'. If the tribal name !§ubaic is correctly read, more probably the ranges of Philby,
Arabian Highlands, 109—10, than the hillocks mentioned ibid., 95, though they are called cArq
al Subai*.
6
Ihm dndynhmw crbn: Ihm surely consists of the preposition with a pronoun, combined
demonstratives, followed by the relative. The -n form of the verb cannot here be an infinitive,
the usual explanation.
 EVENTS IN ARABIA IN THE 6TH CENTURY A.D. 461

the Rahabat and Hadramaut and Yamanat, and began to rage.1 And with
' their' Arabs were Kiddat and Madhig. And with bany Ta'labat and
Muddir were &ubaic (?). In the month Dhu Qayzan of 631.
The Abyssinians were in this year in control of Zafar. The Emperor Anastasius,
a friend of the Monophysites, sent a bishop to Himyar,2 and the Abyssinians
were Monophysites. Ma'adi-karib Ya'fur attacked an alliance of Ta'labat and
Mundir of Hirah ; Ta'labat must be the southern end of the kingdom of Hugr
and Harit b. 'Amr. Abram, father of Nonnosos, twice went on missions to Qais,
the prince of Kinda and Ma'add, so Byzantium had some connexion with
southern Arabia in the time of Anastasius. There is a partial similarity between
the political position in era year 631 and that of 662, when 'Abraha al 'Asram,
after repeated requests from Justinian, attacked Ma'add and 'Ami b. Mundir
intervened. The reason why neither Hugr nor Harit is mentioned does not
appear ; there is no information about these kings to fill the gap between 500
and 525.3 So far from there being any difficulty about dating the total extrusion
of the Abyssinians from Himyar over a decade before the second coming of the
Kushites under Kaleb, scattered notices confirm the probability that the reign
of Ma'adi-karib coincided with an Abyssinian domination.
Since the invasion of 'Ella 'Asbeha was delayed for lack of shipping, the
possibility of an earner invasion might be questioned. Only the inscriptions at
Hima and Kawkab mention 'As'aran. M. J. Eyckmans has cited Hamdani to
prove that 'As'aran is the Tihama from south of Zabid to Mukha.4 Much
earlier, 'Ezana, in both his pagan and his Christian years, was entitled ' king
of Aksum and Himyar and Raydan and Habasat 5 and Saba' and Salhen' and
other African lands. As his inscriptions are triumphal, conquests in Arabia in
his time would be mentioned. Professor Enno Littmann must be correct in
concluding that the title was inherited.6 Perhaps by his time actual control in
Arabia had been lost; but there is no proof that the Abyssinians lost all the
ports. The route of ships from Adulis by the Arab coast to an Egyptian port at
the beginning of the 6th century indicates that they did not.
The king of Himyar in era year 614 was Martad-'ilan Yanuf. An inscription
of that year,7 about the building of a house at Marib on an irrigated terrace,
under the protection of the king, and with the aid and by the authority of
Lahay'at Yanuf Dhu Hasbah 8 and Rahab-'il 'As"ead Dhu Ma'afir, ends with the
1
urybn: I take this form to be for yavfoirm, root wbn, singular for pi. An unknown tribe
could not be joined to the nations preceding.
1
Theodoros Anagnostes (Lector), ii, 58, quoted LRE, ii, 322.
3
See p. 446. * Museon, lxvi, 333.
5
Invariably either Abyssinia, or Abyssinians in their own land. The order is therefore
significant. Part of Arabia was included with the African territory, as it is named between Raydan
and Saba'.
• Miscellanea Academiae Berolinensis, 97—127, especially 107-8. The counter-argument,
Institution, 309—11, takes no account of the character of the inscriptions.
' Fakhry i, p. 109,fig.55, ii, pp. 46-9, no. 74.
8
Hasbah was associated by M. Hartmann with the name of the father of 'Abraha b. as
Sabbah, the grandson of 'Abraha al 'Asram. On the 'Asbah and modern Subaihi see Dr. R. B.
Serjeant in Museon, lxvi, 126-9.
462 SIDNEY SMITH

broken words : ' May the Merciful be favourable. (Thou art L)ord '. This I
corresponds to the formula of Sarah-'il Yaqbul, and attests Jewish faith; it |
could not be used by an officer of a Christian king or his princes. A strong case
can be made out for believing that from 'Abi-karib 'As'ad to Martad-'ilan the 1
dynasty was Jewish in faith, and connected with Yatrib.1 The reaction against
Macadi-karib Ya'fur was led by a man who took a Jewish name, but avoided
a royal nomen or cognomen of the earlier Jewish kings.
At the beginning of the 6th century the struggles of inter-related royal
families were combined with pressure from foreign powers, and the need to
maintain trade. Adherents of the Christian and Jewish faiths became associated
with political parties.
VIII. THE KELIGIOUS STRUGGLE
At a time when an unfounded scepticism as to the existence of any sort of
Christianity, and as to the Jewish faith of Dhu Nuwas, was being expressed in
England, Dr. Tor Andrae published an essay which included a careful examina-
tion of the evidence.2 He rightly held in balance the references to, or implica-
tions about, both Nestorian and Monophysite influences in the introduction of
Christianity into southern Arabia. Though Nestorians were active missionaries,
for instance in India and China, the available sources give much less than
a reasonable account of their activities in the south owing to their nature.
There was a close connexion between Hirah and Nagran and Christian com-
munities at Hirah would, as elsewhere in Persian territory, be mainly Nestorian.
Syria, Palestine, Egypt, and Abyssinia were Monophysite and it is difficult to
decide whether some of the ' monotheistic ' Sabsean texts were professions of
the Jewish or the Christian faith,3 since the Monophysite faith could be ex-
pressed in a simple formula. The notable point is the introduction of a Trini-
tarian formula by Sumu-yafa' 'Aswa', and the continued use of it by 'Abraha.4
Both had close political connexions with Byzantium.
Side by side with the Christian missions there was active proselytizing by
the Jews, of which the inscriptions provide evidence beginning from the late
4th century. There is no reason to distrust Arab traditions that 'As'ad 'Abu-
karib, the 'Abi-karib 'As'ad of the inscriptions, w\b lived in the first half of
the 5th century, accepted the Jewish faith at Yatrib ; kings of the second half
of that century in Himyar professed it. The Talmudic and other Jewish
evidence has been collected and carefully discussed.5 The rise of the Jewish
community to power in Yatrib cannot be precisely dated; it was obviously
1
Clearly put in H. StJ. B. Philby, The Background of Islam, 116-9.
a
D. Tor Andrae, Der Ursprung des Islams und das Christentum, 7-24. Contrast D. S. Mar-
goliouth, Relations between Arabs and Israelites (Schweich Lectures), 59-83.
3
Texts published before 1945 were catalogued and discussed by G. Ryckmans in Miscellanea
Historica Alberti De Meyer, 194-205.
4
See pp. 432, 437.
6
I owe this information to the kindness of Professor B. Lewis who referred me to Hayyim
Ze'eb Hirschberg, Yisra'el ba-cArab : the Jews in Arabia from the fall of the Second Temple until
the Crusaders. (Modern Hebrew.) Unfortunately I have not had access to this book.
 EVENTS IN ARABIA IN THE 6 T H CENTURY A.D. 463

parallel to the rise of the Christian community at Nagran. From before 500 to
about 530 the important Jewish trading community was at Yotabe". Abochora-
bos the phylarch ruled there till 540; after that nothing more is said about
Yotabe. Ma'adi-karib Ya'fur, whose reign corresponded to the ' first coming '
of the Kushites, was supported by Christians in Nagran, Yusuf, of Jewish faith,
decimated that community, 'Abraha, who ultimately gained control of Taif
and Mecca, not previously included in any of the Arab kingdoms, is favourably
mentioned by Christian writers. The relations of the Jews at Yatrib with
Samaria and Tiberias is unquestionable, an inclination to look to Persia among
some Palestinians is shown by events. Late in 523, or at the beginning of
January, 524, Yusuf had to face the opposition of Christian princes in his
kingdom and the prospect of an Abyssinian invasion. He wrote to Mundir of
Hirah. He may have advocated extirpation of Christianity; what he must
have wanted to know was whether he could get help. In 525 he was killed,
shortly before Justinian appointed Abochorabos ; during the war with Persia,
the Samaritans rebelled. There is some underlying interconnexion.
There is no reason to believe that the faiths professed were not sincerely
held; a reduction of motives in this period to a single interest, religious,
political, or economic, is as naive as in any other century. It would be easy to
misinterpret the economic motive of the individual states. Abyssinia was not
interested in the Indian trade, as is clear from Procopius ; control of ports on
the Arab shore of the Red Sea was a vital interest. From the time kings of
Himyar added ' of " their " Arabs in Tihamat' to their title, that is from the
sole reign of 'Abi-karib 'As'ad, about 440, the political aims of a Christian in
Aksum and of a Jewish king in Himyar conflicted. According to Arab historians
both Sammar Yuhar'is and 'Abi-karib 'Ai'ad fought the Persians in India; the
details are incredible, the possibility of a conflict over Indian trade undeniable.
But it was Justinian's policy, partly economic, partly military, which brought
to an end the isolation of Saba' and Himyar, and their independence.

IX. THE ARAB SOURCES

The purpose here has been to interrelate as simply as possible isolated
references to events in Arabia during the 6th century. If, as I think, a logical
order is imposed by the events,1 the inscriptions, which supply indubitable
evidence, but only as to isolated facts, can now be used to assess the reliability
of the sources. In the case of the Greek authors, there will be no need for a
revision of opinion. The chronographers' weaknesses have long been known ;
not many under- or over-estimate Procopius. Not enough attention has some-
times been paid to the Syriac sources owing to their character; rationalist
dislike has induced neglect of incidental evidence of value in them. But the
first need is a critical appreciation of the Arab sources and their relative value.
The impatient method which swept them into a single class,' the Muhammadan
1
See Table B. The effect of reckoning the era from 115 B.C. or 110 B.C. can be judged by
shifting years marked E.
464 SIDNEY SMITH

TABLE B.—DATE LIST

(— before : + after : * approximate)
E 464 Sarah-b-'il Yakuf king of Himyar : Lahay'at Yanuf, Nawf and Ma'adi-
karib Yan'um co-regent.
473 Amor-kesos appointed phylarch by Leo I I : Yotabe abandoned.
474 'Aswad succeeded Mundir I at Hirah : married d. of eAmr b. Hugr.
476 The Samaritans revolted, set up Justasa as king, and attacked Csesarea.
The revolt was suppressed.
488 Accession of Kavadh.
? Himyarite saga : Hassan Tubbac sent his sister's son Harit b. 'Amr
against Ma'add, with Hirah as objective.
491 Accession of Anastasius I.
494 Mundir b. Mundir succeeded 'Aswad at Hirah.
E 496 Martad-'ilan Yanuf king of Himyar.
— 500 Romanus attacked Gamalos (Gabalat): direct rule at Yotabe.
500 Nu'man II b. 'Aswad succeeded at Hirah.
Skenite Arabs of Nu'man defeated by Eugenios in Euphratesia.
Agar son of Arethas (Hugr b. cAmr b. Mu'awiya b. Harit) driven from
border of Palestine by Romanus.
502 Kavadh began war with Anastasius.
503 Nu'man II of Hirah died of a wound fighting for the Persians; 'Abu
Ya'fur b. 'Alqama ruled 2 years at Hirah.
A Jewish plot to surrender Constantia to Kavadh failed.
505 Seven year truce concluded between Persia and Byzantium.
Mundir (III) b. Nu'man, also called b. 'Imru '1-Qais, king of Hirah;
married Hind, d. of Harit b. 'Amr b. Hugr.
— 513 First coming of Kushites to Himyar.
? Anastasius sent a bishop to Himyar.
E 513 Ma'adi-karib Ya'fur fought for transit through Kat'a against alliance of
bany Taiabat and Mundir III.
E 515 Campaign of Yusuf 'Ai'ar, king of Himyar, against Abyssinians in
Zafar, Mukha and Nagran.
518 Justin I Emperor.
'Ella 'Asbeha, already nagaU, planned invasion of Himyar.
E 522 Sumu-yafa' 'Aswa' fortified Kane on instructions from Abyssinia, when
the king and princes killed Himyarites.
523 Yusuf advanced against Nagran: second persecution. Martyrdoms.
Simeon of Beth Arsam informed by Taiyaye and Ma'addaye.
Envoy of Yusuf arrived at court of Mundir.
524 Simeon wrote to Syria. Daus Dhu Taiaban appealed to Justin. Daus sent
to Aksum. Roman ships impounded.
— 525 Qais (b. KhuzaH) of Ma'add, the refugee, fled from court of S\unu-yafac.
525 Yusuf, unsupported, killed by the Abyssinians in battle by the sea.
526 'Ella 'Asbeha appointed Sumu-yafa' king and Abyssinian deputies.
Flight of Harit b. 'Amr from Diomedos, dux of Palestine.
Mundir attacked Ghassan ; driven into desert by a Roman army.
527 Kavadh began war with Justin.
Accession of Justinian.
Harit b. Gabalat and 'Abi-karib b. Gabalat appointed Roman phylarchs.
 EVENTS IN ABABIA IN THE 6TH CENTURY A.D. 465

529 Julianus, a brigand, king of the Samaritans : Christians murdered.

Revolt suppressed. Mass flight of Samaritans.
— 529 Harit b. cAmr at Hirah (Arab sources).
531 Embassy of Julianus to 'Ella 'Asbeha and Sumu-yafa'.
Justinian opened negotiations for peace : refused by Kavadh.
Accession of Khusrau I.
532 Truce between Justinian and Khusrau.
Mundir returned to Hirah : responsible for Bahrain and Yamama.
533* Abraha seized sole rule in Himyar : refused tribute to nagasi.
534* 'Ella 'Asbeha sent expeditions against 'Abraha. (Procopius).
536* Death of 'Ella 'Asbeha : recognition of 'Abraha by his successor.
— 540 Silko ' basiliskos ' of the Noubadai and of all the Ethiopians.
539 Roman attempt to arbitrate between Harit b. Gabala and Mundir.
E Yazid deputy for 'Abraha in Hadramawt; opposition by Kidar.
E Breach of the dam at Marib : the covenant in Himyar.
E Embassies from Aksum, Rum and Fars : envoys of Roman phylarchs and
Mundir.
540 Khusrau began war with Justinian.
— 544 One or more further embassies sent to 'Abraha. (Procopius.)
E 544 'Abraha's expedition against Ma'add.
544-6 Cosmas Indicopleustes published books I-V of the ' Topography '.
545 Truce between Justinian and Khusrau.
546 Mundir defeated, after killing son of Harit b. Gabala.
551 Procopius published the ' Wars '.
Samaritans and Jews at Csesarea revolted and killed the pro-consul
Stephanus. Revolt suppressed.
554 Mundir killed in battle with Harit b. Gabala. cAmr succeeded.
? 'Abraha appointed Muhammad b. Khuza'i to enforce Christianity.
1 'Abu Murrah appealed to Khusrau against 'Abraha.
565 Justin II Emperor ; repression of Monophysites.
568 John Malalas of Antioch published his Chronography.
569/70 Year of the Elephant. Death of'Abraha. Plague of small-pox. Death of
'Abdallah b. cAbd al Muttalib. Birth of the Prophet.
570 Interregnum at Hirah.
Yaksum b. 'Abraha king in Himyar.
572 Masruq b. 'Abraha king in Himyar.
Flight of Ma'adi-karib b. Saif 'Abu Murrah from his half-brother Masruq
to Khusrau.
War began between Khusrau and Justin II.
575 The vahriz sent by sea to the Hadramawt; Masruq killed in battle.
Ma'adi-karib installed king. The vahriz departed.
577 Qabiis appointed king of Hirah.
Ma'adi-karib murdered by 'ahabis ; return of the vahriz.
578 Hormizd IV succeeded Khusrau.
Justin II abdicated; Tiberius began attempt at restoration.
580 Mundir IV b. Mundir succeeded at Hirah.
582 Maurice Emperor.
Nu'man III b. Mundir at Hirah.
590 Disorder in Persia.
604 Death of Nu'man I I I ; Persian governors appointed.
466 SIDNEY SMITH

tradition', supposed to derive entirely from Byzantine literature,1 is clearly

inadequate. They are disparate and difficult to classify. The attempts at
continuous historical accounts, compiled in much later periods, are generally
negligible. Scattered in them there is material of the first order. The chronology
of Hisam, much of the narrative in Hisam and Ibn Ishaq as quoted in Tabari,
some entries in Mas'udi, are demonstrably reliable. Fragments of early verse
quoted, so far from being the product of an age later than the Prophet, must be
nearly contemporary with events to which they refer. Passages about tribal
history 2 provide the means of understanding the events recorded in inscrip-
tions. Some, for instance, the roll-call of honour after Nihawand, or the ribald
rhymes of the girl of the sept 'Amir b. Sa'sa'ah,3 illustrate the tribal movement
from the south, and define a historical factor often treated too broadly. The
most difficult evidence of all is to be found in the ' saga ' ; as von Kremer said,
quoting Franz Eiickert, the seed must be sifted from the chaff,4 and there is
much more seed than some have thought. Hamdani is now generally accepted
as reliable. Some of the ' saga ' must derive from earlier sources, unknown to, or
not mentioned by, him.
Correct appraisal of events in the 6th century will add to or alter a paragraph
or so in dealing with the 7th. In 630-1 the Prophet delivered instructions to the
envoy of Himyarite princes, worded like a diplomatic ultimatum,5 showing
full knowledge of the Christian and Jewish parties, and of the survival of
paganism. Dr. R. B. Serjeant has collected Sabaean graffiti near the reputed
tomb of the prophet Hud which are not likely to be later than the 6th century 6 ;
with whom did the Prophet count these adherents of another prophet, whose
existence can no longer be regarded as a convenient fiction ? 7 Events of the
time of Qais, the Kinda and Mundir were known ; what was the reason for the
appointment of a Persian governor, which provoked the resistance of Qais b.
Maksuh and cAmr b. Ma'adi-karib %
Details may require consideration. Social conditions in Arabia demand a
1
The last re-statement of this view known to me is in BH, xlv-xlviii. Detailed arguments,
e.g. that of Moberg about the reference in the Koran to the Nagran martyrdoms, frequently
break down. 'Ukhdud explains itself.
2
e.g. Mas'udi, iii, 122 : ' there are many accounts about Qurais, Gurhum, Khuza'a and the
others of Macadd ' : iii, 391-2, the wars of Ghassan and Ma'add.
3
Mascudi, iv, 236-241 : vi, 137-55. Contrast the many cases cited in von Wissmann-Homer
which prove that tribes still occupy territory they held in pre-Islamic centuries.
4
von Kremer, Ueber die siidarabische Saga, vi. von Kremer himself frequently anticipated
points subsequently proved. So far as I can judge the references in the traditionalists are, by
comparison, worthless.
5
Ibn Hisam, 963, quoted in Muir-Weir, 456, as ' curious'. The pagan sticks arefiguredin
A. Grohmann, OottersymboU und Symboltiere.
• Published by Professor G. Ryckmans in Museon, lxvii, 181-5. The monogram in nos. 2
and 10, read hb, should I think be regarded as that of the tribe Habum, and compared with the
monogram yz for Yaz'an. The interpretation as a statement, ' he loved ', cannot be considered
natural in this case ; the occurrence of #6 in graffiti not in holy places is another matter. Habum
was a section of Yaz'an. There were adherents of Hud among the Jews and Christians there.
7
This seems to be the view taken in Muir-Weir, 51-2, and in D. S. Margoliouth, Mohammed,
131.
 EVENTS IN ARABIA IN THE 6 T H CENTURY A.D. 467

new treatment. The land was not, before the appearance of the Prophet, a
closed box, in which there were a few Jews and Christians, isolated from the
great states. Lop-sided views have been induced from quaint stories of the
jahiliyah, and the abiding Arab predilection for nomad ways. Only the acci-
dental literary emphasis seems to justify conclusions that would not apply to
Syria or 'Iraq. Paganism, suppressed and illiterate, survived in Syria till the
time of the Abbasids. Some inkling of the extent of pagan survivals in 'Iraq can
be formed from Lady Drower's work on the Mandseans. There were thriving
cities in Arabia, old foundations, as civilized as any in Syria or 'Iraq, and
perhaps as large, apart from Antioch and Madain. The Christian and Jewish
communities were large, and not mainly foreigners. Arabs had faced the
formidable Abyssinians. Military leaders had fought men trained in Persian
armies on equal terras. Princes had dealt with international affairs.
Mecca remained pagan, in a small region where each city remained indepen-
dent. But the rise of the Qurais to wealth and importance was sudden. 'Abraha's
attack on the Ka'ba was the first known recognition of the fact. There must be
many unknown causes to account for the phenomenon. A later event shows
the importance attached to Jiddah. An expedition, sent by the Prophet against
that town and the 'ahabis, pursued the enemy to an unnamed island 1 ; it should
be Yotabe. The decline in the commercial importance of that place may have
been due to the diversion of trading vessels to Jiddah, attracted by lower
charges and through transit to the Damascus region offered by Qurais. The
attraction of the annual fair at a pagan centre is sufficient explanation of the
stubborn maintenance of the old religion.
By the third quarter of the 6th century the economic system began to
break down. Governments collapsed. In Persia this took the form of a disputed
succession, as often before and since. The last of the Arab kings died at Hirah
in 604 ; the direct government by Persians produced the conditions of 634-5.
At the end of the reign of Justinian the industrial and financial conditions were
calamitous 2 ; the Empire never recovered fully. The peripheral regions were
affected. The kingdom of Aksum passed away. In spite of the a priori doubts
that have been expressed, the weakness in Abyssinia shown during the negotia-
tions with the Prophet accords with the lack of information from Ethiopic
sources. In southern Arabia, Persian rule, probably never extending far,
petered out to an unknown end ; the old kingdom split up. Colonel van der
Meulen and Mr. Ingrams have shown that Hadramawt cannot be maintained
by a closed economy ; in modern times it has exported men, in the 7th century
the men went north. Saba' disappeared, absorbed by Yamanat, not a part of

1
Ibn Sa'd, 117-8, quoted in Muir-Weir, 436. The attack implies that Jiddah belonged to
Qurais. The Abyssinians were not subjects of the nagasi, but settlers there, who frequently
served as professional soldiers and the like ; see Father H. Lammens, L'Arabie occidentals avant
VBegire, 244-57.
8
G. Ostrogorsky, Geschichte des byzantinischen Staates, 48 : ' wirtschaftlich und finanziell
vollig zerriittet'.
468 EVENTS IN ABABIA IN THE 6TH CENTURY A.D.

the earlier kingdom. The final breach of the dam at Marib was not the cause of i
more than a local, temporary, breakdown, but it was the symptom of decay in
a state comparatively prosperous for over a thousand years. When Arab kings
and princes in Himyar, Oman, Bahrain accepted Islam, their action was a tacit
admission that Mecca and Yatrib had become the leading cities in their world.
The Arab sources are not explicit about the main trends. When it is recog-
nized that much in their accounts is not incompatible with the known facts,
modern judgments on what is probable and improbable will have to be modified.
30th June, 1954.