Posted by on Jan 11, 2017 in Library | Comments Off on “BAREFOOT AND NAKED”: WHAT DID THE BEDOUIN OF THE ARAB CONQUESTS LOOK LIKE?


The Syriac churchman Bar Penkaye, who wrote about familiar from Lawrence of Arabia and countless Holly 690, held the Arab invaders to have been “naked men wood films; but as far as the bedouin of pre-Islamic riding without armor or shield.”1 In the same vein Arabia are concerned, it would seem that we are wrong.
Michael the Syrian (d. 1199) reports that a certain Though “naked” may be a little hyperbolic, both liter Hiran sent by the last Sasanid emperor to spy on the ary and iconographie evidence suggests that it is not Arabs told his employer that the invaders were “a far from the truth.
barefoot people, naked and weak, but very brave.”2 To start with the literary evidence, Ammianus Mar A Muslim text dating from, perhaps, the later eighth cellinus, commander of the eastern armies about 350 century similarly insists that the invaders were “barefoot AD, tells us that the Arabs of the Syrian desert were
and naked, without equipment, strength, weapons, or provisions.”3 In all three texts the word “naked” seems
to be used in the sense of poorly equipped and lack
ing body armor rather than devoid of clothes, and all how literally one should take him: were they wear three depict the Arabs as poorly equipped in order ing Roman army issue, passed down from relatives to highlight the extraordinary, God-assisted nature of and friends who had served in the Roman army, or the Arab conquests. “I have a sharp arrowhead that alternatively stolen from unlucky soldiers? (“When
penetrates iron, but it is no use against the naked,”
as Rustam says in the Sh?hn?ma, in his premonition
of the fall of the Sasanids.4 But precisely what did him with the command, Ishlah yd walad, ‘Strip, boy!’
from the settled areas, it was their custom to accost the Arab invaders wear? It would be the first ques meaning that they intended to rob him of his cloth
tion to spring to Oleg Grabar’s mind. Under normal
circumstances it would be the last to spring to mine,
for as Oleg is fond of telling his colleagues, historians
tend to ignore the concrete physical manifestation Arab, this time one in Roman service at Adrianople, of things; in particular, they do not think of the way he says that he was long haired and naked except for
things looked and so miss an important dimension of a loincloth.7 In the same vein Malka, a fourth-century
the past. I have always pleaded guilty to that charge. Syrian who was captured by bedouin between Aleppo Having benefited from Oleg’s lively company and warm and Edessa and whose adventures were recorded by heart for over ten years, however, I shall now try to Jerome, describes how the Ishmaelites descended upon
make amends, if only with a trifling offering: how his party of about seventy travelers “with their long
should we tell a filmmaker who wanted to screen the
hair flying from under their headbands.” He did not think of them as wearing turbans or kaffiyehs, then, or as shielding their heads from the sun by any kind of head cover at all. Like Ammianus, he says that they the bedouin will not have been dressed in the same wore cloaks over their “half-naked bodies,” but he adds
way as the settled Arabs, and I should like to keep that they wore broad military boots {caligae)? Again
story of the Arab conquests to depict the conquerors? More precisely, how should we tell him to depict the
desert Arabs who participated in the conquests, for
things simple.
Most of us would probably reply that the hypothet
ical filmmaker should depict the bedouin warriors as men in kaffiyehs and flowing robes, along the lines
one wonders if they were wearing Roman army issue. They transported Malka into the desert and set him
to work as a shepherd, and there he “learned to go naked,” he says, presumably meaning that he learned
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms
“warriors of equal rank, half nude, clad in dyed cloaks as far as the loins.”5 The word he uses for their cloaks
is sagulum, a short, military tunic, and one wonders
bedouin raiders in the desert encountered someone
ing,” as Jabbur says of the Syrian bedouin many cen turies later.6) Ammianus does not tell us what, if any thing, the warriors wore on their heads, but of another


Fig.1.Ivorycarving,rightarm oftheChairofMaximia NY)
tocoverhimselfwithamereskin:thisseemstoh style of Lawrence of Arabia that bedouin were envisaged
beenallthatslavesworinesixtih-ncentupryrSyreia.-IslamicArabia.9O would infer that he had handed over his clothe
his captors.
We now turn to the iconographic evidence, look
ing at it region by region.
To start in Syria, there is a representation of semi naked bedouin in an ivory carving from a chair
made in the first half of the sixth century in Anti och or (under Syro-Palestinian influence) Alexandria
(fig. 1).10 It depicts Joseph’s brothers selling Joseph to two Saracens: the brothers are represented by the three figures on the left, Joseph stands in the middle,
and two Saracens appear with two camels behind them to the right. The Saracens, who are armed with a bow and a spear respectively, have long, apparently plaited
hair and wear nothing on their heads or their upper torsos, merely loose garments wrapped around their
waists, which reach as far as their ankles but expose one of their legs as they walk. The brothers are also scantily clad, but in more military-looking outfits, and
it is they rather than the Saracens who are wearing boots. The Saracens are shod in sandals. There is
of course no guarantee that the carving is based on observation rather than artistic convention, but one point is clear: it was not as heavily clad figures in the
Another ivory carving on the same chair shows the Saracens selling Joseph to Potiphar (fig. 2). Here Joseph is seen twice, first on a camel (on the left)
and next between Potiphar and one of the Saracens, to whom she is handing money. Potiphar is wear ing classical-looking robes. The Saracens’ robes also
appear more flowing than in the first panel, but here as there their lower body wraps are split in the mid dle, exposing their legs, and their arms are bare. In fact, their entire upper torsos could be bare, though
it is hard to tell. The short tunic that Joseph is wear
ing clearly includes a drape over one shoulder, and the adult Saracens could have a similar item on their
shoulders.” Maybe the artist dressed his characters in classical clothes in order to conjure up a bygone age. In any case, he depicted the Saracens with the same
long, apparently plaited hair as in the first panel, and he gave them sandals, too, but not any kind of head gear. One would take it to have been long hair of this kind that Malka saw flowing under headbands.
Yet another sixth-century carving, also a Syrian or Syro-Egyptian work, depicts two brothers sell
ing Joseph to a Saracen.12 Joseph and his brothers are wearing short tunics similar to those in which rural people are depicted on the mosaic floors of
sixth-century churches in Madaba.3 The Saracen is wearing a mantle that leaves the left part of his chest exposed, but what he is wearing underneath
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms


Fig. 2. Ivory carving, right arm of the Chair of Maximianus. Museo Arcivescovile, Ravenna. (Photo: Alinari/Art Resource, NY)
is not clear. All four are barefoot and bareheaded.
Finally, we have the depiction a man armed with a bow, sword, and whip, leading a camel (fig. 3); this appears on the mosaic floor of the church of the mon astery of Rayanos at cUyun Musa, at the eastern top of
the Dead Sea, dated by Piccirillo to the second half of
the sixth century.14 In Picirillo’s words, the man “is half
naked, wearing a long loincloth reaching beneath his knees with a cloak thrown over his left shoulder that
covers his forearm.” Picirillo suggests that he was an auxiliary soldier and deems the representation to fit
the “exaggeratedly dramatic” literary accounts of Arab
soldiers given by authors such as Ammianus Marcelli nus and Malka in Jerome.15 Whether the Arab was an auxiliary soldier or not, however, the representation actually seems to be quite different. The most dra
matic feature of the mosaic is the Arab’s bulging chest.
Neither Ammianus nor Jerome says anything about chests, but both highlight the long, flowing hair of the Arabs; though damage to the mosaic makes it impos
sible to say what, if anything, the soldier is wearing on his head, it is at least clear that he does not have
hair (or a kaffiyeh) coming down to his shoulders. The clothes involved are quite different, too. Ammianus’ Arabs were wearing short military tunics, Jerome’s were
dressed in cloaks and boots, but the soldier in the
mosaic is wearing a waist wrap and shawl along with sandals. This could well be based on observation, for
the waist wrap and shawl (iz?r and r?da3) are the two
chief items of male clothing in pre-Islamic poetry.16
The main feature that the three representations have
in common is the skimpiness of the outfits described.
Pitched against a horsemen encased in iron, Arabs such as these would indeed have come across as naked.
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms


Fig. 3. Mosaic from the church of Kaianos at cUyun Musa, Mount Nebo. (Photo courtesy of Michele Picirillo)
In sharp contrast to these representations, an image The text gives the name of the person commem on a piece of Coptic tapestry dating between the sixth orated, presumably identical with the person repre
and eighth centuries and said to show Joseph and an Ishmaelite merchant on a camel depicts both Joseph
and the Ishmaelite as thoroughly wrapped up.17 But the
alleged camel may well be a horse,18 and the alleged
Ishmaelite seems to be wearing trousers. So this can be left out of consideration.
sented, as Adhlal ibn Wahab’il but does not otherwise tells us anything about him.19 Macdonald wonders whether the incense burner is a funerary object rather
than a dedicatory one (as suggested in the catalogue of the exhibition in which it was most recently dis played20), for the inscription does not mention any deity, only a name and a patronym, and the vast major
ity of funerary stelae in both North and South Arabia only give the deceased’s name and patronym. If the object is funerary, the relief might in Macdonald’s opin
If the inhabitants of the Roman empire envisaged ion represent the naked soul of the deceased riding the Saracens as wearing nothing on their heads and his camel on the Day of Judgment.21 But as Macdon not much on their bodies, how were they seen by the ald himself stresses, this is highly conjectural. Besides,
Arabs themselves? We may start in the south. did the pagans of South Arabia believe in the resur Here the first image to capture one’s attention is rection? There is nothing to suggest that the deceased
a crude relief on an alabaster incense burner fromwas a Jew or a Christian. And the people depicted on
Shabwa in the Hadramawt, probably dating from other funerary reliefs are fully clothed. On the whole,
around the third century AD (fig. 4). It depicts a man it seems more likely that a bedouin of the Hadrami
riding on an unsaddled camel, positioned in front of plateau is being depicted here, for there are plenty the hump; he holds a short sword or a camel stick or of naked Arabs in the rock reliefs, as will be seen.
some such implement in his right hand and the reins Why such a man should figure on a Shabwan incense in his left, and a water skin or shield is attached by a burner is another question.
strap to the rear of the hump. He is stark naked, and, A fully clothed camel rider appears on a funer
apart from the reins, the camel is as naked as he is. ary relief, also of alabaster, dated to roughly the first
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms

sense of propriety of the settled people. Of decently dressed camel-riders, presumably soldiers in the local armies, we also have an example in a relief from Dura Europos that shows such a rider seated on a saddled camel, armed with a long lance, and wearing a tunic and mantle.24 But he is bareheaded, and maybe the South Arabian was too: Calvet and Robin interpret his apparent head cover as a hair style.25
In another funerary relief, a Sabaean alabaster of the second or third century AD, the lower panel shows
a horseman with the north Arabian name of cIjl ibn Sacdallat touching a camel with his spear, the act by
which a camel raider appropriates a camel. The upper panel shows the deceased sitting at a table with his
wife and child in attendance, or perhaps the deceased at a banquet, and both the stool and the table indi cate that we are in a settled environment, as also sug
gested by the fact that the nisba of the deceased was Qryn: he may have come from Qaryat al-Faw or from Wadi -Qura.26 He was not a bedouin raiding camels,
then, but rather a sedentary Arab engaged in what one would assume to be camel catching staged as a sport.27 All the figures are fully clothed, the deceased
in a long robe and the other two in shorter garments, and the deceased seems to be wearing some kind of head cover, though his putative wife and children are clearly bareheaded. The deceased’s headgear, if it is not simply hair, looks like some sort of stiff bon net, certainly not like a turban. South Arabian reliefs,
which usually show people bareheaded, do not in fact seem to depict any turbans at all.
Moving slightly north to Qaryat al-Faw, which flour ished from roughly the second century BC to roughly the fifth century AD, we find a bronze statue of a man
wearing nothing but a loincloth, but he is kneeling rev erently, presumably in prayer, and his outfit is more
likely to be a form of ihr?m than bedouin dress.28 Also at Qaryat al-Faw we find two drawings on plaster walls of horsemen hunting or raiding camels. One horse
man could be naked, but the other is wearing some thing like a tunic or at least a skirt. Whether they have
headgear is impossible to tell.29
incense burner, this camel rider was presumably a sol by the inhabitants of the desert themselves. The most dier in the local army, dressed in conformity with the striking image among these is a drawing of a horse
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms

Fig. 4. Relief from an alabaster incense burner. British Museum, ANE 125682. (Reproduced with the permission of the Trustees
of the British Museum)
to the third century AD, with an incription identify ing the deceased as Mushayqar Hamayat ibn Yashuf (fig. 5).22 He too is holding a short spear or camel stick in his right hand and the reins in his left, and
he is sitting on a fine camel saddle of a type also
attested on a bronze figurine of a camel thought to
be from Yemen.23 Unlike the wild bedouin on the That leaves us with the countless rock drawings left


Fig. 5. Alabaster funerary relief. Louvre, AO 1128. (Photo courtesy of Michael Macdonald)
either naked or wearing skimpy clothes “mainly meant to cover the private parts,” as Nayeem puts it.32 But these drawings are difficult to date, and though some
are Safaitic,33 many of them are likely to be much older than the period under consideration here.
There is an example of what the makers of rock art wore in a Thamudic drawing from the Tabuk region
of northern Arabia, which depicts a horseman and two men in a chariot?a driver and an archer (fig. 7).34 The horseman, who is riding in front of the chariot, appears to be every bit as naked as the camel on the
Sabaean stela, though one should perhaps envisage him as wearing a loincloth. He also seems to have long, flowing (rather than bushy) hair. The driver could be naked, at least as far as his upper torso is concerned
(the lower part of his body is hidden from view), but maybe the draftsman simply refrained from trying to depict his clothes. He could be bareheaded, but his head is pointed, perhaps to suggest the conical hel
met worn by Assyrian soldiers.35 The footsoldier who is pursuing the chariot and shooting arrows at it, how
ever (fig. 8), is dressed in a long waist wrap, with a slit at the side or the front to allow freedom of move
ment, along the lines of those depicted on the ivory panel of Saracens buying Joseph from his brothers (see fig. 1). He too seems to have long hair.
This drawing is likely to be very old. The chariot points to ancient Near Eastern times, perhaps the sev
man hunting an oryx with a short spear (fig. 6). He
is wearing a waist wrap similar to that of the Arab
soldier in the sixth-century mosaic; the thickened found on images of Arabs on Assyrian reliefs (although lines across his shoulders could be taken to suggest
that he is also wearing a r?da, and he has bushy or
kinky hair that, although quite long, sticks straight
out from his head, in a style that is quite common in
Safaitic drawings.30 Unless we take his hair actually to shot at by another.37
be some sort of hat, he is not wearing anything on his
head. Other drawings do depict headgear, sometimes
very elaborate, but apparently in the form of plumes,
which are hardly intended here.31 The author of the
Safaitic inscription on the same stone claims to have way for over a millennium before the rise of Islam.
made the drawing, which is thus roughly datable to In a drawing by W. Boutcher of a detail from the the period from the first century BC to the fourth Assyrian reliefs showing the campaign of Ashurbani century AD. By then, it would seem, the pre-Islamic pal (688-627 BC) against the Arabs, the Arabs, with
enth to fourth century BC,36 and the footsoldier has a long, pointed thong between his legs, a feature also
precisely what it is meant to represent is unknown). Indeed, one wonders if the occupants of the char iot should not actually be identified as Assyrians (or
perhaps Babylonians) pursuing one Arab while being
The age of the drawing notwithstanding, the cloth ing and hairstyle of the Arab archer are not drasti cally different from those examined above, suggest
ing that the desert Arabs dressed in much the same
plaited hair, are shown dismounted from their cam els and dressed in wraparounds, each with an open ing to allow freedom of movement (fig. 9). Their By the standards of the rock drawings, this horse wraparounds are not flowing like those of the Sara
“uniform” of tzar and r?da was in place, but without the turban or other headgear by which it is usually taken to have been complemented.
man is well dressed, for most drawings depict males as cens who purchase Joseph from his brothers (fig. 1),
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms


Fig. 6. Rock drawing depicting a Safaitic horseman. (Photo courtesy of G. M. H. King)

Fig. 7. Rock drawing. (Photo courtesy of Michael Macdonald)
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms


Fig. 8. Rock drawing, detail. (Photo courtesy of Michael Mac donald)
and their hair looks shorter and a good deal neater, too, but given that there are more than a thousand years between the images, the continuity is nonethe
less striking. To a somewhat lesser degree, the same
holds true when one compares the Assyrian represen
tations with the Safaitic rock drawings and the Mad aba mosaic.
In sum, what did the bedouin participants in the conquests wear? The answer seems to be generally not very much at all: either bits and pieces of what their settled neighbors?whether the latter were Byzantines,
Arabians, or (one assumes) Iranians?wore, or a wrap around and a nda covering part of their upper torso, and perhaps even sandals, but rarely, insofar as one can tell, anything on their heads. It is the absence of headgear that is the most surprising. Whatever the vari ations, all the desert dwellers seem to have looked a
good deal more like their ancestors of Assyrian times
than like Mush” s Rwala.38 As far as desert clothing is concerned, Arabia on the eve of Islam seems still to have been rooted in the ancient Near East.
When and why did the desert Arabs start cover ing themselves up? I cannot claim to know. My guess would be that they started doing so in the centuries
after the rise of Islam, and in consequence of the rise of Islam, for Islam drew the bedouin closer together to the settled people, giving them shared religious
and other norms. Wrapping up was what the people who mattered did, and so the bedouin came to do so
too (at least when they could afford it). According to Ibn al-Kalbi (d. 819 or later), the Tanukh who met the caliph al-Mahdi (d. 785) in Qinnasrin were wear ing turbans. They were trying to look their best on this occasion.39 A Byzantine miniature of ca. 976-1025
depicting Simeon Stylites venerated by Arabs shows Simeon in a hooded monk’s habit and the three
Arabs wearing turbans, now apparently as a matter of course.40 But I had better leave this question for another birthday.
Institute for Advanced Study Pr?nceton, NJ
Author’s note: I should like to thank Michael Macdonald for invalu
able help with images, inscriptions, and bibliographical references
alike. Insofar as this article has any merit, it is really due to him.
(The same most definitely does not apply to the shortcomings.)
I am also grateful to Mika Natif for teaching me to navigate the
Index of Christian Art, to Michael Cook for reading and com
menting on the paper, and to Julia Bailey for spotting visual clues that I had overlooked.
1. Bar Penkaye in A. Mingana (ed. and tr.), Sources syriaques, Leipzig, n.d. [1907?], 141; trans, in S. P. Brock, “North Mes opotamia in the Late Seventh Century,” Jerusalem Studies in Arabic and Islam 9 (1987): 58.
2. Michael the Syrian, Chronique, ed. and tr. J.-B. Chabot, 4 vols. (Paris 1899-1910), 4:417, 2:421.
3. D. Sourdel, “Un pamphlet musulman anonyme d’?poque cabb?side contre les chr?tiens,” Revue des ?tudes islamiques
34 (1966) 33 (text), 26 (trans.). For a reconstruction of the text from which the fragment comes see J.-M. Gaudeul, “The Correspondence between Leo and ?Umar,” Islamochristiana 10 (1984), 109-57, with the passage in question on 155. The
transmitter is Isma’il b. ‘Ayyash.
4. Firdaws?, Sh?hn?ma, ed. E. E. Berteis, 9 vols. (Moscow,
1960-71), vol. 9: 1. 119 (drawn to my attention by Masoud Jafari).
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms

i n i e
e w
the Deacon Thomr as, whole floori); 337 (Steaphanos spearcing
a lionn, wearing “a sleeveless orbiculaAted tunic.. .tied to ther
right shoulder”sthat seems to, be identical with thatTof the h brother on the left); 338-39 (donkey driver, soldier defend
ing himself against a bear), 343 (date); 345, 347 (Church of Saints Lot and Procopius, whole floor).

*, *
4 -6 – ;0
Fig. 9. W reproduc
Excavations 1967-1997 (Jerusalem, 1998), 333 (Church of 5.Ammia
6. Jibra’il Sulayman Jabbfr, The Bedouins and the Desert: Aspects
of Nomadic Life in the Arab East, trans. L. I. Conrad (Albany,
1995 [Arabic original 1988]), In, with vivid illustrations on 2-3.
7. Matthews, Roman Empire of Ammianus, 348, with reference to Ammianus, xxxi, 16, 6.
8. Jerome, “Vita Malchi Monachi Captivi,” paragraphs 4-5, in J.-P. Migne, Patrologiae Cursus Completus, Series Latina, 221 vols. (Paris, 1844-64), 23: cols. 57-58, trans. in Segal, “Arabs in
Syriac Literature,” 103; cf. I. Shahid, Byzantium and the Arabs in the Fourth Century (Washington, DC, 1984), 284ff.; Mat thews, Roman Empire of Ammianus, 348.
9. G. Jacob, Altarabisches Beduinenleben (Berlin, 1897), 44 (with reference to ‘Antara’s Mu’allaqa).
10. See 0. M. Dalton, East Christian Art: A Survey of the Monuments (Oxford, 1925), 172, 205ff.; idem, Byzantine Art and Archaeol
ogy (New York, 1961, orig. publ. 1911), 203ff.
11. Cf. Dalton, Byzantine Art and Archaeology, 206.
12. Berlin, Staatliche Museen, inv. no. 566; cf. Dalton, Byzantine
Art and Archaeology, 208; W. F. Volbach, Elfenbeinarbeiten der Spftantike und desfriihen Mittelalter (Mainz, 1952), 80-81; pl. 54, no. 172.
13. M. Piccirillo and E. Alliata, Mount Nebo: New Archaeological
14. M. Piccirillo, Madaba, le chiese e i mosaici, 207-8; Piccirillo and
Alliata, Mount Nebo, 356-58, with a better photo (fig. 224). 15. Piccirillo, in both Madaba, le chiese e i mosaici and Mount Nebo,
and with reference to Ammianus and Jerome in Madaba, le chiese e i mosaici, 225 n. 10.
16. Jacob, Altarabisches Beduinenleben, 44.
17. A. Kakovkine, “Le tissu copte des VIIe-VIlIe siecle du mus6e
metropolitain,” GNttinger Miszellen 129 (1992): 53-59. It was formerly classified as showing the flight into Egypt.
18. Presumably it was classified as a camel on the basis of its peculiar head (which mostly looks like that of a dog) and the similarity of its hooves and tail to those of the camel at Dura
Europos (cf. the reference given below, n. 24). But it has no
hump, and its legs and harness are those of a horse.
19. St. J. Simpson, ed., Queen of Sheba: Treasuresfrom Ancient Yemen
(London, 2002), 97-98 no. 110; also in W. Seipel (ed.),Jemen: Kunst und Archiologie im Land der K6nigin von Saba’ (Vienna,
1998), 86 and 88 no. 20, both without comments on the absence of clothes; Rpertoire d’ipigraphie simitique, 8 vols. (Paris,
1900-1968), 7: no. 4690.
20. Simpson, Queen of Sheba, 97-98 no. 110.
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms

21. Michael Macdonald? personal communication with reference to a discussion at the recent congress “Rencontres sab?ennes 10,” in St. Petersburg.
22. Y. Calvet and C. Robin? Arabie heureuse, Arabie d?serte: Les anti quit?s arabiques du Mus?e du Louvre (Paris, 1997), 109-10 no. 20, where both the image and the text are reproduced along
with a transliteration, translation, discussion, and bibliogra phy.
23. Reproduced in Simpson, Queen of Sheba, 99 no. 113.
24. A. Perkins, The Art of Dura-Europos (Oxford, 1973), fig. 40. 25. Cf. the reference given above, n. 22 (“Il porte une coiffure
arrond?e avec une sorte de pendant ? l’arri?re”).
26. Louvre, AO 1029: see Calvet and Robin, Arabie heureuse, 107-8
no. 18 (image, text, transliteration, translation, discussion, and bibliography); A. Caubert, Aux sources du monde arabe: L Arabie avant l’Islam, collections du Mus?e du Louvre (Paris, 1990) 28
and 39 no. 3 (where the upper panel is interpreted as a ban quet scene). For the meaning of the gesture with the spear see M. C. A. Macdonald, “Camel Hunting or Camel Raiding?” Arabian Archaeology and Epigraphy 1, 1 (1990): 24-28, with a reproduction of the stela on 26.
27. This seems at least as likely as that the deceased should be shown as engaged in camel-raiding, perhaps as a desire to claim links with a real or nomadic past, as suggested by Mac
donald, “Camel Hunting or Camel Raiding?” 25-26; idem, “Hunting, Fighting, and Raiding: The Horse in Pre-Islamic
Arabia,” in Furusiyya, 2 vols., ed. D. Alexander (Riyad, 1996), 1:76. Either is compatible with the conjecture that he was a
caravaneer (Calvet and Robin, Arabie heureuse, 108).
28. A. R. al-Ansary, Qaryat al-Fau: A Portrait of a Pre-Islamic Civili
sation in Saudi Arabia (London, 1982), 109 no. 3.
29. Ansary, Qaryat al-Fau, 130-33 (where the rider called Salim
b. Kacb seems to be hunting rather than raiding, given that the camel appears to have been speared or shot with an
30. G. M. H. King, The Basalt Desert Rescue Survey: Safaitic Inscrip
Hons (forthcoming; my thanks to Dr. King for allowing me to reproduce the image).
31. Cf. Macdonald, “Hunting, Fighting, and Raiding,” 76, 77 fig. 5b, where the upper tier of the headdress looks like giant feathers.
32. M. A. Nayeem, The Rock Art of Arabia (Hyderabad, 2000), 337. For some striking examples of naked people see Macdonald, “Hunting, Fighting, and Raiding,” 72 nos. 3, Id, lg, lh. Unfor tunately, these drawings are known only from hand copies, and there is no way of telling how accurately they represent
the originals.
33. M. C. A. Macdonald, “Reflections on the Linguistic Map
of Pre-Islamic Arabia,” Arabian Archaeology and Epigraphy 11 (2000): 45.
34. Cf. Macdonald, “Hunting, Fighting, and Raiding,” 74, 76ff? with the complete composition on 224-25.
35. Macdonald, “Hunting, Fighting, and Raiding,” 78.
36. Macdonald, “Hunting, Fighting, and Raiding,” 78; idem, “Wheels in a Land of Camels: Another Look at the Chariot
in Arabia,” forthcoming in Arabian Archaeology and Epigra phy.
37. The main objection to this proposition is that the man on horseback is identified in the inscription above him as hrb,
taken by Macdonald to mean enemy warrior on the basis of modern bedouin dialect. But this is clearly conjectural, and the word may not even have been correctly deciphered (cf.
Macdonald, “Wheels in a Land of Camels,” text no. 9).
38. Cf. A. Musil, The Manners and Customs of the Rwala Bedouins
(New York, 1928).
39. Shahid, Byzantium and the Arabs in the Fourth Century, 431. 40. Rome, Bibliotheca Vaticana, gr. 1613. The ninth-century min
iature ofjoseph’s brothers selling Joseph to a Saracen is unin formative, since no attempt seems to have been made to dis tinguish the Saracen from the other figures: all are wearing
the same long cloaks and all are bareheaded (cf. A. Grabar, Les miniatures du Gr?goire de Nazianze [Paris, 1943], pl. LXI).
This content downloaded from on Mon, 12 Sep 2016 20:39:36 UTC All use subject to http://about.jstor.org/terms