M.C.A. Macdonald
Literacy and Identity in Pre-Islamic Arabia
CONTENTS
Preface
Acknowledgements
vii–viii
ix
PArt 1: LIterACy, LAnguAge And sCrIPts
I
Literacy in an oral environment
49–118
Writing and Ancient Near Eastern Society: Papers in Honour
of Alan R. Millard, eds P. Bienkowski, C. Mee and E. Slater.
New York/London: T & T Clark International, 2005
II
Nomads and the Ḥawrān in the late Hellenistic and
roman periods: a reassessment of the epigraphic
evidence
303–413
Syria 70. Paris, 1993
III
reflections on the linguistic map of pre-Islamic Arabia
28–79
Arabian Archaeology and Epigraphy 11. Copenhagen, 2000
PArt 2: AnCIent ethnICIty
IV
some reflections on epigraphy and ethnicity in the
roman near east
177–190
Mediterranean Archaeology 11. Sydney, 1998
V
Arabians, Arabias, and the greeks: contact and
perceptions
1–33
Published in Italian as ‘Arabi, Arabie e Greci. Forme di
contatto e percezione’, in I Greci: Storia Cultura Arte
Società. 3. I Greci oltre la Grecia, ed. S. Settis. Turin:
Giulio Einaudi editore, 2001, pp. 231–266
VI
‘Les Arabes en syrie’ or ‘La pénétration des Arabes en
syrie’: a question of perceptions?
La Syrie hellénistique (Topoi, Suppl. 4). Paris: De Boccard,
2003
303–318
vi
COntents
PArt 3: AsPeCts Of the hIstOry Of AnCIent ArAbIA
VII
Was the nabataean kingdom a ‘bedouin state’?
102–119
Zeitschrift des Deutschen Palästina Vereins 107. Tübingen,
1991
VIII On Saracens, the Rawwāfah inscription and the Roman army
Published in French as ‘Quelques rélexions sur les Saracènes,
l’inscription de Rawwāfa et l’armée Romaine’, in Présence arabe
dans le Croissant fertile avant l’Hégire, ed. H. Lozachmeur.
Paris: Éditions Recherche sur les Civilisations, 1995, pp. 93–101
IX
trade routes and trade goods at the northern end of the
‘incense road’ in the first millennium b.C.
1–26
333–349
Profumi d’Arabia. Atti del Convegno, ed. A. Avanzini. Rome:
‘L’Erma’ di Bretschneider, 1997
Addenda and Corrigenda
1–23
Index
1–36
this volume contains x + 420 pages
Arabians, Arabias, and the greeks:
Contact and Perceptions*
I The Idea of Arabia
We owe the concept of “Arabia” to the greeks. before them, the Assyrians had
sometimes referred to “the land of the Arabians”1 (māt aribi), a vague and always
“far-off” area whose location migrated with its nomadic inhabitants, but usually
they simply wrote of aribi, etc., “Arabians”. similarly, in the Old testament, the
terms ʿarāb, ʿarābî are used of peoples rather than of a place.2 In both cases, the
“Arabians” they referred to were, for the most part, nomads who were described in
stereotypical terms.3
but the initial greek experience of “Arabians” was rather different. the earliest
contacts were made through trade rather than through warfare, and were with settled
communities rather than with nomads. since, according to the principle enunciated
by herodotus, “egypt is all that country which is inhabited by egyptians, just as
Cilicia and Assyria are the countries inhabited by Cilicians and Assyrians”,4 it is not
surprising that he gave the name “Arabia” to the areas inhabited by Arabians. As
time went on, the greeks discovered Arabians in a number of widely scattered and
unconnected places, but this did not prevent them from referring to each of these
regions as “Arabia”, though inevitably this led to a certain amount of confusion.
by the common circular pattern of thought, once a place had been called “Arabia”, all its inhabitants were assumed to be “Arabians”, regardless of how they
thought of themselves. thus, when Alexander’s companion, nearchos, on his voyage from India to Susa, sighted Raʾs Musandam, at the mouth of the Persian Gulf,
and “those who had knowledge of this district said that the promontory belonged
*
this is a revised and slightly expanded version of a chapter first published in Italian as “Arabi,
Arabie e greci. forme di contatto e percezione”. Pages 231–266 in s. settis (ed.), I Greci. Storia
Cultura Arte Società. 3. I Greci oltre la Grecia. torino: einaudi, 2001.
1
I use the term “Arabian” here as a translation of Akkadian aribi and greek Ἀράβιος and Ἂπαψ
because it does not carry all the modern implications of the term “Arab”. We know relatively little
about the populations which were called “Arabian” by others in antiquity and it therefore seems safer
to avoid any suggestion that they are necessarily identifiable in all respects with those we call “Arab”
today.
2
for references to, and discussion of, the Akkadian and hebrew forms of these terms see
Ephʿal 1982: 6–7.
3
see below.
4
herodotus, Historia, 2.17. In this, he was arguing against the Ionians who claimed that “nothing but the delta is egypt” (2.15).
2
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
to Arabia,”5 this was a critical moment in the history of the region. for Alexander’s
naval expeditions soon showed that this promontory was part of a huge landmass,
“almost the size of India”,6 and the implication was obvious: if “the promontory belonged to Arabia”, then the whole landmass must be “Arabia” and, by the circular
argument, all its inhabitants must be “Arabians”. thus it was that the concept of the
“Arabian Peninsula” was born and the ancient populations of its south-west corner
– the Sabaeans, Minaeans, Qatabanians, Hadramis, etc. – came to be called “Arabians” by the greeks, even though they themselves would not have recognized such a
description. Indeed, we do not know whether any of the inhabitants of the Peninsula
in the early hellenistic period would have called themselves “Arabians”. It would
be many centuries before the majority of the inhabitants of the Peninsula would
regard themselves as Arabs and would themselves call their homeland Jazīrat alʿarab, “the Peninsula of the Arabs”.
before looking at greek contacts with, and perceptions of, the Arabians, it is
necessary to clarify a few terms. It is not at all clear how the word “Arabian” was
defined by those who used it in antiquity. It is generally assumed to have originated
as a self-designation,7 but we do not know what the criteria were at any particular
period for calling oneself, or being called “an Arabian”, and, as just noted, it was
often used of populations which would not have recognized the label as applying to
themselves. the term was used of so many different groups, with widely differing
life-styles, that the only common factor one could imagine is linguistic. unfortunately, however, we have little evidence of the language spoken by these groups
until late antiquity.8 thus, in this brief study I shall simply accept that the label was
given to these diverse populations by greek writers, without struggling to answer
the unanswerable question “why?”.
We need also to remember that “Arabian” is an ethnicon and is not the equivalent of “nomad”, “camel-breeder”, “raider”, or “brigand”, as is all too often assumed. One should not necessarily therefore expect to find nomads or camels in an
area called “Arabia”, nor violence and pillage among its inhabitants.9
finally, some greek geographical terminology for parts of the ancient near east
can be confusing. In the Classical, hellenistic and even the roman periods the
term “red sea” (Ἐρυθρὴ θαλάσσα) generally referred to the whole Ocean south
of Asia, i.e. the Indian Ocean. hence the Περίπλους τῆς Ἐρυθρᾶς θαλάσσας was
5
Arrian, Indica, 32.6–7. It is, of course, possible that Arrian, who was writing 400 years later,
simply assumed that information available in his own day was known at the time of nearchos’ voyage.
6
this report by hieron is quoted in Arrian, Anabasis, 7.20.8. It is possible that Persian seamen
had already circumnavigated the Peninsula in the time of Darius I (520–486 BC), see the excellent
discussion in Salles 1988: 79–86.
7
Ephʿal 1982: 7 and Macdonald 2003 (= paper VI in this collection): 303–308 and forthcoming.
8
As I have pointed out elsewhere, the language behind personal names is a very unsafe guide
to the ethnicity of, or to the language spoken by, those who bear the names (Macdonald 1998 [= paper
IV in this collection]: 187–189).
9
Such assumptions are, alas, all too common, see for example Donner 1986: 6, and Högemann
1985: 33ff. and passim, and the criticism in Macdonald 2003 (= paper VI in this collection): 308.
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
3
the “Periplus of the Indian Ocean”, not of the “red sea” in the modern sense. the
latter, which was known to the greeks from an early period,10 was either included
in theἘρυθρὴ θαλάσσα, as part of the Indian Ocean,11 or was called the “Arabian
gulf” (Ἀράβιος κόλπος). On the other hand, what we call the Arabian/Persian gulf
was sometimes called “the Persian gulf” (ὁ κόλπος ὁ Περσικάς),12 but more often
was simply regarded as an inlet of the Ἐρυθρὴ θαλάσσα, and was not given a special name.13 What follows is a brief survey of the growth of greek awareness of
Arabians and Arabias, followed by a look at some of the images which the greeks
created of Arabians and the places in which they lived.
II The Development of Greek Awareness of Arabians and Arabias
the earliest contacts between greeks and Arabians appear to have come about
through the trade in aromatics. The Greek words μύρρα (“myrrh”), λίβανος, and
λιβανωτός (“frankincense”) are clearly of semitic origin14 and the first two, at least,
are already found in the poetry of sappho (c. 600 bC).15 both the substance and the
word would have been brought north to the Mediterranean from ancient Yemen16
by its merchants and those from north Arabia,17 and, by the fourth century bC, the
first delphic hymn to Apollo refers to Ἄραψ ἀτμὸς “Arabian vapour”.18
greek geographical knowledge about the Arabians began to accumulate from the
end of the sixth century bC. In about 520, darius I, the Achaemenid king of Persia,
sent scylax of Caryanda on a voyage of exploration which started on the Indus and
ended in egypt.19 unfortunately, we know of scylax’s discoveries only at third hand.
In about 500 bC, they were used by his younger contemporary, the Ionian hecataeus,
10
see the reconstruction of the world map made by the Ionian, hecataeus in c. 500 bC incorporated into herodotus’ world view (figure 1).
11
It is surely in this sense that it is called Ἐρυθρὴ θαλάσσα in the inscriptions of the Ptolemaic
period, Dittenberger 1903: nos 69/6, 186/4, and in the Adulis throne inscription (ibid., no. 199/48).
12
see, for instance, Arrian, Anabasis, 7.20.9.
13
see, for instance, diodorus siculus, 2.11.2.
14
On μύρρα see most recently Sima 2000: 281–284. Oddly enough, a word *lbn has not yet
been found in Ancient South Arabian [ASA] texts. However, since words meaning “frankincense”
from the root lbn are found in hebrew, Classical Arabic, and the Arabic dialects of dhofar (the region
which produces frankincense) and other parts of south Arabia, as well as a number of east African
languages, it is highly likely that such a word existed in AsA and was the origin of the words in greek.
See Müller WW 1997: 193–195; 1978: coll. 703–709, 748.
15
for μύρρα and λίβανος see Lobel & Page 1963: fr. 44.30; for [λι]βανώτῳι see fr. 2.4.
16
I am using this term, instead of the normal “south Arabia”, to make clear the very sharp
cultural, linguistic and social distinctions which existed between the populations in the south-west of
the Peninsula (“ancient Yemen”) and those in other parts of the Middle East with whom the Greeks
often lumped them together as “Arabians”. The term “Yemen” is, of course, an anachronism in this
context since it does not seem to have come into use until much later, but it is employed here purely
for convenience to make an important distinction.
17
See Macdonald 1997 (= paper IX in this collection).
18
Powell 1925: 141, line 11.
19
herodotus, 4.44.
4
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
who was a geographer not an explorer and who made a map of the world (see figure
1). his work is also lost but was used, with many critical comments, by herodotus,
half a century later.
After reaching the mouth of the Indus, scylax followed the Indian and Iranian
coasts westward. It is not clear whether he entered the Persian gulf. since his mission was at the behest of the Persian state, it is inconceivable that he would not have
known of its existence, but no trace of such knowledge is preserved in hecataeus’
description of the southern coast of “Asia”, in which he merges the west coast of
India and the southern coasts of Iran and Arabia into a continuous line broken only
at the entrance to the red sea (figure 1).20 Presumably, scylax would have had with
him Persians who would have recognized the straits of hormuz, and so, rather than
turning into the Persian gulf, he may well have crossed them. he would then have
run down the eastern coast of the Oman Peninsula before following the southern
coast of the Arabian Peninsula, and eventually turning north and rowing the length
of the red sea, which he said took 40 days.21 If this is correct, scylax’s voyage was
not a “circumnavigation of Arabia”, since he travelled only along its southern and
western coasts.
We do not know what scylax considered the Arabian Peninsula to be, though it
is interesting that he says that Arabians inhabited the Kamarān islands off the Red
Sea coast of Yemen, north of al-Ḥudaydah.22 As usual, it is impossible to know exactly what he meant by this and whether he was contrasting the population of these
islands with the (non-Arabian) inhabitants of ancient Yemen.
As far as we can reconstruct the geography of hecataeus from the quotations
from his work which have survived, he saw the world as a more or less circular
landmass surrounded by “Ocean” (figure 1). this landmass was divided into three
parts, the northern half of the circle was occupied by europe, while the southern
half was divided between Asia and Libya.23 As we have seen, he regarded the west
coast of India, and the southern coasts of Iran and Arabia as a continuous line forming the southern limits of “Asia”, broken only by the “Arabian gulf (our red sea),
after which Asia continued into eastern egypt as far as the nile, which formed the
boundary with the third “continent”, “Libya”.24 In the hecataean system, “egypt”
20
It has been suggested that the fact that, two centuries later, Alexander the great needed to
send expeditions from the Persian gulf and the red sea to attempt a circumnavigation, shows that
knowledge of the Persian gulf – and thus that the land mass on the east of the red sea was part of a
peninsula – had long since disappeared from Greek geographical lore (Salles 1988: 85).
21
herodotus, 2.11. herodotus specifies “from its inner end to the wide sea”. however, scylax
would have been travelling from south to north, and was presumably forced to row because the prevailing wind in the red sea is from the north.
22
Jacoby 1923: 36 (Frag. 271), Καμαρηνοὶ νῆσοι Ἀραβίων.
23
herodotus, 2.16 “the Ionians ... and the rest of the greeks ... divide the whole earth into three
parts, europe, Asia and Libya.”
24
Pearson 1939: 85–87, argues from Herodotus 2.15–16, that Hecataeus “reckoned Libya as
part of Asia” and so did not regard the nile as the border between Asia and Libya. however, if I have
understood herodotus correctly, this does not seem to accord with the final sentence of 2.16: “If their
[the Ionians’] opinion is correct, then it is clear that they and the rest of the Greeks cannot add up when
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
5
consisted only of the nile delta.25 We know from other passages that herodotus
regarded the area between the nile and the red sea as “Arabia” and was very
vague about the land to the east of the red sea. It therefore seems likely that on
hecataeus’ map, Asia consisted, from east to west, of the lands of the Indians (in
the far east), then the Persians followed by the Assyrians (each occupying the entire
north-south breadth of the Asian landmass), followed (north to south) along the
Mediterranean coast by the Phoenicians, the Syrians of Palestine 26 and the egyptians (confined to the nile delta), with “Arabia” running from gaza or the eastern
delta southwards, between the nile and the red sea, as far as the Ocean. there is
no hint that hecataeus was aware of an “Arabian Peninsula”.27
herodotus also had a great deal of first-hand experience to report on the subject of
the Arabians. He had visited Egypt, sailed up the eastern Mediterranean coast from
Pelusium to tyre, and travelled thence to babylonia. for herodotus, “Arabia” is eastern egypt between the nile and the red sea.28 he describes this as a land of clay and
stones29 in which there is a long mountain chain running parallel to the nile, from heliopolis at the entrance to the delta, south to the “red sea” (i.e. the southern Ocean).30
herodotus thought that the eastern slopes of these mountains were the source of
frankincense.31 this shows clearly both that he had not visited the area east of the
mountains and that he was unaware of the true source of the precious commodity,
knowing only that it was somewhere east of egypt. his ignorance is hardly surprising, since he did not know of the existence of the Arabian Peninsula. his report
they divide the whole earth into three parts, Europe, Asia, and Libya; they ought to add to these a fourth
part, the delta of egypt, which would belong neither to Asia nor to Libya. for by their own description,
the Nile is not [i.e. cannot] be the river that separates Asia and Libya, for the Nile divides at the extreme
angle of this delta, so it is this land [i.e. the Delta] that must lie between Asia and Libya.” This surely
implies that herodotus thought that “the Ionians” (his usual way of referring to hecataeus) did believe
that the nile formed the border between Asia and Libya, but that, since the nile divides to form the
delta and the Ionians claimed that egypt is restricted to this area, they should have believed that the
two continents were separated by “Egypt” (i.e. the delta). thus herodotus is accusing hecataeus of
inconsistency in merely saying that the nile is the border between the two continents, when hecataeus’
own belief makes “egypt” (the delta) an area between the two, which in the hecataean system would
belong to neither. this sort of nit-picking argument is consistent with herodotus’ other “rather childish
quibbling argument” (Pearson 1939: 85) against the Hecataean view of Egypt (2.15).
25
herodotus, 2.15–16 (an opinion he ascribes to the Ionians and with which he strongly
disagrees).
26
that is the coastal strip, or “Philistaea”.
27
Herodotus, 4.38–39. His description is made more confusing by his use of the word ἀκτή
(“peninsula”) in these sections. herodotus, and presumably hecataeus, are taking an imaginary stance
in central Iran from which project two peninsulas, one to the west (i.e. Asia Minor) and the other to
the east and south (i.e. the landmass of “Asia”, formed by amalgamating India, Iran and the Arabian
Peninsula). Herodotus then describes the bar of land between Iran and the Mediterranean which links
these two peninsulas, and which was inhabited from east to west by Persians, Assyrians and then (from
north to south along the Mediterranean coast) by Phoenicians, Syrians of Palestine and Egyptians.
28
Herodotus, 2.8, 2.19.
29
Ibid., 2.12.
30
Ibid., 2.8.
31
Ibid., 2.8.
6
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
that it took two months to cross this north-south chain of mountains at its widest
point, and that frankincense grew on its eastern slopes, suggests that he thought the
Ἀράβιος κόλπος (our red sea) was much further to the east than it really is.32 Could
it be that his informants in this case were north Arabian merchants bringing merchandise from ancient Yemen across the Red Sea to Egypt, and that the two month
journey was in fact measured from the Sabaean capital, Mārib, or the Minaean
capital, Qrnw (see figure 2)?33
Certainly, herodotus’ statement that “Arabia is the most distant to the south of all
inhabited countries”34 gives the impression that the information came from someone
who knew the south-west of the Peninsula, beyond which stretched the seemingly
infinite Ocean. It also serves to introduce bizarre descriptions of how aromatics and
spices were gathered which, with their emphasis on the noxious creatures which protect the precious substances, sound very much like travellers’ tales designed to discourage the curious. the fact that he calls the source of aromatics “Arabia” is surely
only explicable if his informants were from the north of the Peninsula, or even from
eastern Egypt. A Sabaean or Minaean merchant would surely not have called himself an “Arabian”. We know that many Minaean merchants made the journey north,
since at least one was buried in egypt and others left inscriptions in several parts of
the near east, and even as far away as delos. A considerable number brought back
foreign wives from the north, six of whom were from egypt and one apparently
from Ionia, though she has a semitic name.35 however, the northern end of much
of the trade must usually have been handled by north Arabians and, as so often,
the customers associated the product with its (Arabian) retailer rather than with its
source (ancient Yemen), about which the retailer no doubt did his best to keep them
in ignorance. thus, frankincense became known to the greeks, not only by the technical terms λίβανος/λιβανωτός but as Ἄραψ ἀτμός, “Arabian vapour”. 36
herodotus does not make clear the northern limits of his Arabia, but there is
some evidence that he thought it stretched up to the Mediterranean. He writes that
its seaboard “is inhabited by syrians” (τῆς γὰρ Ἀραβίης τὰ παρὰ θάλασσαν Σύροι
νέμονται).37 His use here of the word θάλασσα, without qualification, implies that
32
On the other hand, thanks to the voyage of scylax, he was better informed on the size of the
red sea, reporting that it took 40 days rowing from one end to the other, but only half a day to cross
from one side to the other at its widest point (2.11).
33
graffiti written in scripts from ancient north Arabia have been found in a number of places in
eastern Egypt, e.g. at al-Muwayḥ (Robin 1995: 112–115) between Coptos (Qifṭ) and Myos Hormos
(al-Quṣayr al Qad�m), see Fig. 2.
34
Herodotus, 3.107.
35
the name is S¹lmt. Graffiti written in the script of ancient Yemen have been found in Wādī
Ḥammāmāt (Colin 1988: 33–36). For the tomb inscription of the Minaean in Egypt (Répertoire
d’Épigraphie Sémitique, Paris 1900–1968, no. 3427), see most recently Robin 1994: 291, 293–296.
for the inscription at delos, see Répertoire d’Épigraphie Sémitique no. 3570. For the women from
Egypt and Ionia married by Minaeans (in the so-called “Hierodulenlisten”) see Bron 1998: 119–120,
and the references there.
36
Powell 1925: 141, line 11.
37
herodotus, 2.12.
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
figure 1. A reconstruction of the map of Asia as envisaged by herodotus (c. 450 bC).
7
8
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
he is referring to the Mediterranean and, as noted above, he anyway seems to have
thought that the red sea was much further east than it really is. It is unclear whether
he was aware of the extent of the sinai Peninsula and, if so, whether he considered
it to be part of Arabia, but certainly there had been Arabians in northern sinai since
at least the days of Sargon (721–705 BC),38 and it was probably they who assisted
both Esarhaddon (680–669 BC)39 and Cambyses II (530–522)40 in their invasions of
egypt. thus, it is likely that, for herodotus, Arabia consisted of eastern egypt up to
just east of the delta, plus the whole of sinai and even the negev (if he was aware
of them) right up to the Mediterranean littoral along the coast of Sinai between the
Serbonian Marshes and Gaza (see Fig. 1). On this coast there was a short stretch
between Cadytis (Gaza) and Ienysus (probably el-Arīsh or Khān Yūnis) where “the
trading-stations on the sea belong to the Arabians” (τὰ ἐμπόρια τὰ ἐπὶ θαλάσσης
... ἐστὶ τοῦ Ἀραβίου), but gaza itself and the ports west of Ienysus belonged to the
“syrians of Palestine” (Σύρων τῶν Παλαιστίνων) in herodotus’ day.41 this would
then explain herodotus’ comment elsewhere that “egypt is like neither to the neighbouring land of Arabia, nor to Libya, no, nor to Syria [i.e. anywhere where Syrians
live] (for the seaboard of Arabia [i.e. the coastal area of the region inhabited by Arabians] is inhabited by Syrians).”42 the implication is surely that while syrians lived
along the coast (apart from the gaza-Ienysus stretch), the hinterland of northern
sinai, of the eastern delta, and of eastern egypt was inhabited by the Arabians.
this tiny stretch of coast belonging to the Arabians is less than 30 km long, if
Ienysus is el-Arīsh, and less than 10 if it is Khān Yūnis. Yet it must have served as
the outlet for much of the merchandise brought up the Peninsula from ancient Yemen and across the Negev to the Mediterranean. This window on the Mediterranean
freed the Arabians from the Phoenicians’ virtual monopoly of maritime trade in the
region and would have meant that the Arabians could deal directly with greek and
egyptian merchants and sea-captains. We do not know whether this situation was
created by the Persians or whether they merely confirmed an existing arrangement.
geographically, this “window” was situated within the fifth satrapy, or governorate, of the Persian empire which ran from the border of Cilicia and syria down the
eastern Mediterranean coast to Egypt, and included Cyprus.43 Yet it appears to have
retained its independence, for as herodotus makes clear, the Arabians, alone among
all the peoples of Asia, “did not yield the obedience of slaves to the Persians, but
were united to them by friendship (ἀλλὰ ξεῖνοι ἐγένοντο), as having given Cambyses passage into egypt, which the Persians could not enter without the consent
of the Arabians.”44 One suspects that political advantage played a greater role than
gratitude in the Persian decision to treat the Arabians as client allies rather than as
38
39
40
41
42
43
44
Ephʿal 1982: 93–94, 101–111.
Ibid., 137–142.
Herodotus, 3.5–9.
Ibid., 3.5.
Ibid., 2.12, my italics.
Ibid., 3.91.
Ibid., 3.88.
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
9
subjects. After all, it was unnecessary to conquer a region whose inhabitants were
friendly and were willing to pay handsomely to retain their independence.
for herodotus tells us that, unlike most subjects of the Persian empire, the Arabians were outside the normal tax administration of the central government.45 Instead they gave an annual “gift” (δῶρον), i.e. tribute, to the Persian treasury of
1000 talents (approximately 30 tonnes) of frankincense.46 this would have been
worth an enormous sum, far more than the 350 babylonian talents of silver paid by
the whole of the fifth satrapy which surrounded the Arabian enclave.47 the theme
of the Arabians’ love of independence and their refusal to bow the knee to conquerors is one which struck a chord with greek writers and recurs many times in later
works on Arabia. We shall return to it at the end of this chapter.
In the period between herodotus and Alexander, we get only glimpses of the
relations between these Arabians and the greeks. In 410 bC, Pharnabazus, the Persian satrap of hellespontine Phrygia, joined sparta in the war against Athens, contributing the Phoenician fleet to the struggle. however, at a critical juncture, when
the Athenian fleet was pursuing the spartans in the dardanelles, he refused to allow
his 300 Phoenician ships to go to the rescue “on receiving information that the king
of the Arabians and the king of the egyptians had designs upon Phoenicia”.48 In
diodorus’ source, ephorus, this is presented as a trick by Alcibiades who, though
on the spartan side, did not want Athens destroyed. Whether or not this was so, it
suggests that the Persians felt that they could no longer take for granted the loyalty
of the Arabians and that the latter had the means and motivation to seize an opportunity to attack their trading rivals, the Phoenicians.49
In the mid-fourth century bC a fragmentary text wrongly attributed to scylax
of Caryanda, gives a tantalizing glimpse of this Arabia and its relationship with the
egyptian authorities. by this time the length of the Arabian coastline had apparently increased to 1200 stadia from “the borders of syria” (at gaza?) to the mouth
of the easternmost branch of the nile “at Pelusium”.50 the passage ends tantalizingly with a damaged section dealing with the relations between the egyptians and
the Arabs.
45
Apart from the Persians, who paid no taxes of any sort, the only subject peoples outside the
normal tax administration were the “ethiopians nearest to egypt” (i.e. the nubians) in the far south
of the empire and the Colchians in the far north (3.97). These, like the Arabians, paid regular “gifts”
(δῶρα). the ethiopians and the Colchians were the most remote peoples of the empire, and it was
clearly far more efficient to treat them as semi-autonomous and to exact tribute from their leaders
than to try to subject them to the taxation system which applied in the rest of the empire. the case of
the Arabians, however, was very different since they were an enclave within a province in which the
normal tax system was operating.
46
Herodotus, 3.97.
47
Ibid., 3.91. It is clear from 3.89 and 3.95 that all the tribute in bullion was paid in silver, except
that from India which was 360 Euboïc talents of gold dust (3.94).
48
Diodorus, 13.37.3–6; 13.46.6.
49
For the commercial and political background see Retsö 1997.
50
Müller C 1855: 79–80. However, the figure of 1200 stadia (approx. 35 kms) seems too short
to account for the distance from gaza to Pelusium.
10
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
When Alexander laid siege to gaza, the Achaemenid governor is said to have
put up a stout resistance strongly supported by Arabian mercenaries.51 Indeed, there
is a story – almost certainly apocryphal – that one of these Arabians hid his sword
behind his shield and, pretending to be a deserter, flung himself at Alexander’s
knees. When Alexander bade him rise and join the Macedonians, the Arabian drew
his sword and wounded him in the neck.52 Once he had taken the city, Alexander
swept on through the coastal part of this Arabia and into egypt to receive the surrender of its satrap.53 he then marched south, “through the desert”, up the east
bank of the easternmost branch of the nile as far as heliopolis, near where the
delta begins, receiving the surrender of all the districts through which he passed.54
Most of these desert districts would have been in “Arabia”. Thus having marched
unopposed through the “seaboard of Arabia” and its northern districts, it is not surprising that he was considered to have “conquered the greater part of Arabia”.55 he
was apparently unaware, or unconcerned, that in fact it stretched far away up the
east of the nile and probably covered most of sinai, the negev and edom, while
the Peninsula was still terra incognita to him. Interestingly, however, he was aware
of the independent nature of the Arabians and when he appointed Cleomenes of
Naucratis as governor of “Arabia around Heroöpolis” (i.e. east of the Delta), “he
instructed him to permit the district governors (νομάρχαι) to govern their own districts as had been their way all along, but that he was to exact from them the tributes
(φόροι)”,56 an arrangement which, on the face of it looks remarkably similar to the
Achaemenid one.
but these were not the only “Arabians” whom Alexander encountered. before
he could advance on gaza he had had to undertake a seven-month siege of the
Phoenician city of tyre. needing wood to build a siege-tower, he sent expeditions
to Mount Lebanon to cut cedars. But peasants from among the Arabians (arabum
agrestes) attacked the Macedonians, killing 30 of them, and taking others prisoner.57 Leaving the siege operations in the hands of two of his generals, Alexander
immediately took a force of cavalry, light armed troops and archers and “marched
towards Arabia to the mountain called Antilibanus. Part of this country he captured,
part he received in surrender, and in ten days returned to sidon.”58 this “Arabia”
is a very different one from that between Palestine and egypt. It is situated in the
Anti-Lebanon in roughly the same area in which people whom the Assyrians called
“Arabians” had been grazing their flocks in the eighth century bC.59 but that was
four hundred years earlier and the “Arabians” that Alexander encountered were
51
52
53
54
55
56
57
58
59
Arrian, Anabasis, 2.25.4; 2.27.1.
Quintus Curtius Rufus, Historia, 4.6.15.
Arrian, Anabasis, 3.1. 1–2.
Ibid., 3.1.3.
Ibid., 3.1.2.
Ibid., 3.5.4.
Quintus Curtius Rufus, 4.2.24.
Ibid., 2.20.4–5
Ephʿal 1982: 97–100.
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
11
agrestes, “peasants, rustics,” rather than pastoralists. It is not clear whether they
were supporting the Phoenicians against Alexander, or were simply objecting to the
felling of the cedars, and no doubt acquiring some useful weapons in the process.
eratosthenes (c. 285–194 BC), writing in the following century,60 described
both the Lebanon and the Anti-Lebanon, the mountainous regions on either side
of the Massyas Plain (the modern Beqāʿ Valley), as being “held by Ituraeans and
Arabians, all of whom are malefactors (κακοῦργοι), but the people in the plains are
farmers”.61 he also describes the foothills of the Anti-Lebanon overlooking the region of damascus, as “the Arabian mountains”.62 While this confirms the presence
of Arabians in this area, it unfortunately tells us little else.
by contrast, an anonymous contemporary of eratosthenes,63 described the rich
agricultural land of western and northern Ammon and southern galaaditis (gilead),
probably up to the borders of Pella, as the “χώρα [territory or countryside] of the
Arabians”, and called Amman (later to be Philadelphia of the decapolis) “rabbatamana of Arabia”. here we seem to have a well-established agricultural and urban
population, which played off the Ptolemies and the seleucids in their struggles over
southern syria.
On his return from India to susa and babylon, Alexander set about discovering as much as possible about the landmass to the south-west, between Mesopotamia and egypt. the Achaemenid Persians, whose empire had stretched from India
to egypt and beyond, must have been familiar with the position and size of both
the Persian gulf and what the greeks called the Ἀπάβιος κόλπος (“the Arabian
gulf”, i.e. the red sea), and they may even have circumnavigated the Peninsula.
the greeks, on the other hand, seem still to have been working with herodotus’
model (figure 1) of an unbroken coastline between western India and what we call
the red sea. When Alexander took over the Persian empire he must quickly have
60
eratosthenes’ geographical work is lost but large sections of it were quoted by strabo, and
have thus been preserved.
61
strabo, Geography, 16.2.18. Strabo added that in his day (first century BC/AD), these
Arabians and Ituraeans had preyed on the merchants from Arabia felix until the romans established
security in the area (16.2.20).
62
Ibid., 16.2.16.
63
Preserved in Polybius, Historia, 5.71. 1–4. The events described took place in 218 BC during Antiochus III’s attempt to wrest southern syria (including modern Jordan, Palestine and Israel)
from Ptolemy IV Philopator. It is difficult to pinpoint the exact position of this Arabia but according
to Polybius (5.70.12, and see figure 2 here), Antiochus had been marching from Sidon, past the Sea
of Galilee and Mount Tabor and had reached Pella in the eastern Jordan Valley. It was somewhere
between here and Abila that he received the promises of support and provisions from “the inhabitants
of neighbouring Arabia” (οἱ τῆν παρακειμένην Ἀραβίαν κατοικοῦντες, 5.71.1). On the other hand,
rabbatamana was considered to be in “Arabia” and the Ptolemaic troops massed there were devastating the territory or countryside (χώρα) of the Arabians who had offered help to the seleucid side
(5.71.4). Thus, it would seem that this “Arabia” consisted of ancient Ammon but stretched far enough
north to be considered “neighbouring” to Pella. this suggests that it included at least the southern half
of galaaditis (gilead), south of the river Jabok (Ἰάβακχος, modern Wādī al-Zerqā), but probably also
gerasa (modern Jerash) whose territory apparently adjoined that of Pella, at least in the roman period
(Seigne 1997). For a different view see Gatier 1988: 160–161; 1995: 111–112 .
12
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
discovered the glaring discrepancy between the two geographies. he had already
sent one of his companions, nearchos, with a fleet to explore the coast between the
Indus and “Persia”.64 It seems clear from nearchos’ report, as preserved in Arrian’s
Indica, that Alexander was aware of the possible existence of the Persian gulf and
that one of the instructions given to nearchos was to look for it and, if possible,
to explore it. thus when the expedition reached the straits of hormuz, nearchos
changed course from west to “between west and north”.65 in order to enter what he
recognized as the Persian gulf.66 Interestingly, there was a dispute between nearchos and his lieutenant, Onesicritus, the steersman. the latter, who does not seem
to have been privy to Alexander’s orders, apparently assumed that the “bay” would
be of little significance and gave orders to sail across its mouth (the straits of hormuz) and down the east and south-eastern coasts of the Oman Peninsula. nearchos,
however, insisted that they must not “spoil Alexander’s undertaking” and that they
should enter and explore the “bay”.67 thus it was that, coasting up the Persian side
of the gulf, the expedition eventually joined Alexander near susa.68 One highly significant moment in nearchos’ voyage was the first sighting of the Oman Peninsula,
when “those who had knowledge of the district said that this promontory belonged
to Arabia”.69 for Alexander, whose knowledge of Arabia was based on hecataeus, modified and expanded by herodotus and his own experience, this could have
meant one of two things. either this was simply another place where the seemingly
ubiquitous Arabians lived, like the Sinai coast or the Anti-Lebanon; or it was “the
other side” of the Arabia in eastern egypt, from the farthest (and as yet unknown)
parts of which came the “Arabian vapour” so prized in the Mediterranean world
and with which Alexander himself was so liberal. After taking gaza, he is said to
have sent 500 talents of frankincense to his old tutor Leonidas.70 but gaza was only
the entrepôt, not the source. those who had told nearchos that the Oman Peninsula
was “part of Arabia” also said that it was from here that “the Assyrians [i.e. the
inhabitants of Mesopotamia] imported cinnamon and other spices”.71 When at last
nearchos and his companions reached the mouth of the euphrates they found that
the village of diridotis was where the merchants “collected together the frankincense from the neighbouring country and all the other sweet-smelling spices which
Arabia produces.”72
Arrian, Indica, 20.1, 32.10–11.
Ibid., 32.3.
66
Ibid., 32.8. “the bay (κόλπος) runs back into the interior”. Confusingly, Arrian calls it here
Ἐρυθρὴ θάλασσα “the red sea”, though in Anabasis 7.16.2. he calls it “the Persian Gulf, called by
some the red sea”.
67
Arrian, Indica, 32.10–13; Anabasis, 7.20.9–10.
68
Arrian, Indica, 33.1 to 42.10.
69
Ibid., 32.6–7, and see note 5 above.
70
Plutarch tells the story that Leonidas had criticized the young Alexander for throwing frankincense on the altar-fires with both hands and had told him to wait until he had conquered the lands
which produced the precious commodity before being so extravagant with it (Alexander, 25.4–5).
71
Arrian, Indica, 32.7.
72
Ibid., 41.6–7. Diridotis would appear to be Teredon, a “city” which, according to a tradi64
65
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
13
having established the existence of the Persian gulf73 and being aware of the
existence and dimensions, if not the true position, of the “Arabian gulf” (i.e. the
red sea), it looked highly probable that, as the Persians must have informed him,
the land between these two inlets was a Peninsula. Moreover, given the reports
that nearchos had collected, it must have seemed almost certain that this “Arabia” held the mysterious source of cinnamon, frankincense and other spices and
aromatics.74 With this in mind, Alexander dispatched three small expeditions to
explore the western (Arabian) coast of the Persian gulf and another three to attempt
a circumnavigation of the Peninsula, one from east to west and two from west to
east. none of these was entirely successful, though they all brought back valuable information. the first, led by Archias, got no further than the island of tylos
(ancient dilmun, modern bahrain).75 the second expedition was under the command of Androsthenes who “sailed round part of the peninsula of the Arabians (ἡ
χερρόνησος τῶν Ἀράβων)”, from east to west.76 It is not clear how far he got since
the records of his voyage which have survived are confined to the Arabian shore of
the Persian gulf, ending just short of the Oman Peninsula.77 the third voyage was
entrusted to hieron of soli who was specifically ordered to circumnavigate the Peninsula and reach Heroöpolis on the Suez isthmus. He did not reach it “but turned
back and reported to Alexander that the Peninsula was vast, not far short of the size
of India.”78 We do not know how far he went but if Arrian is correct in saying that
no one had succeeded in rounding Maketa (Raʾs Musandam on the tip of the Oman
Peninsula) then he did not even leave the Persian gulf.79 nevertheless, a sizeable
amount of information was gathered about the Arabian coast of the gulf and the
islands which lie off it.
Alexander also seems to have sent at least two expeditions to attempt the circumnavigation in the other direction.80 One seems to have started from the Ailanitic inlet (the gulf of Aqaba)81 and another from that near Heroöpolis (the Gulf
tion preserved in Eusebius (writing 800 years later), Nebuchadnezzar had established in southern
Babylonia “against the incursions of the Arabs” (Potts 1990, i: 349).
73
Arrian, Anabasis, 7.16.2. “...he discovered the Persian [Gulf] ...to be only a gulf of the
great sea.”
74
Cinnamon, like no doubt many of the other spices and aromatics, was not, of course, native to
Arabia, but was imported and transmitted to the Mediterranean by the ancient Yemenis and the North
Arabians.
75
Arrian, Anabasis, 7.20.7. He also visited the island of Failaka, off Kuwait, which Alexander
commanded to be renamed Icarus (ibid., 7.20.5).
76
Ibid., 7.20.7.
77
strabo, 16.2–4, 16.3.1 which are based on eratosthenes’ quotations from Androsthenes’ report.
78
Arrian, Indica, 43.8.
79
Ibid., 43.9: “But the cape which Nearchos says his party sighted running out into the sea
opposite Karmania, no one has ever been able to round and thus turn inwards [i.e. turn at 180°] towards the far side [viz. the Indian Ocean side of the Oman Peninsula].”
80
Ibid., 43.7.
81
Αἰλανίτος μυχός, strabo, 16.4.4.
14
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
of suez).82 these gathered much valuable information on the topography and populations of both sides of the red sea, including identifying at last, albeit rather
vaguely, ancient Yemen as the source of frankincense and myrrh. At least one of the
expeditions passed through Bāb al-Mandab (see figure 2) and ran a considerable
distance along the southern coast of Arabia, but apparently without being able to
complete the circumnavigation by entering the Persian gulf.83
by the time of Alexander, the greeks had also long known of Arabians in northern Syria and Mesopotamia. Xenophon reports that the “king of Assyria” (actually
nebuchadnezzar II, king of babylon) had subjugated the syrians and had received
the submission of the king of the Arabians,”84 but the exact location of these Arabians is not clear. Later, Xenophon lists the kings who formed an alliance with another
“king of Assyria” (actually nabonidus, the last king of babylon) against Cyrus the
great. these were, from west to east: Lydia, greater Phrygia, Cappadocia and Arabia. the last provided 10,000 horsemen and 100 chariots and a large number of men
armed with slings, the sort of army one would expect from a wealthy settled state.85
Indeed, of all the allies, only the king of Assyria and babylonia himself brought
more horsemen and chariots.86 thus, like all the other Arabians encountered by
the greeks up to the time of Alexander, these were sedentaries not nomads. Cyrus
defeated this vast army and in the rout which followed, the kings of Cappadocia
and Arabia, who resisted despite not having had time to put on their armour, were
massacred by Cyrus’ troops. Xenophon remarks that the majority of the dead were
Assyrians and Arabians “for as these were in their own land, they were very slack
about getting away”.87 this, again, suggests that these Arabians were sedentaries
not nomads, for the latter, having no land to defend, have always been notorious for
their ability to melt away into the desert when a battle turns against them. We do not
know where this battle took place. however, in 540 bC, after his conquest of Lydia,
Cyrus the Great seems to have moved eastwards across Asia Minor, subjugating the
Phrygians, Cappadocians and the Arabians on the way.88 the order in which these
ἀπὸ Ἡρώων πόλεως, theophrastus, De causis plantarum, 9.4.4.
Eratosthenes, preserved in Strabo, 16.4.4; Arrian, Indica, 43.7.
84
Xenophon, Cyropaedia, 1.5.2.
85
Ibid., 2.1.5. Compare Croesus, king of Lydia, who provided 10,000 horsemen and more than
40,000 peltasts (lightly armed troops), or the ruler of Greater Phrygia who provided 8,000 horsemen,
and not less than 40,000 spearmen and peltasts, or the king of Cappadocia who sent 6,000 horsemen
and no less than 30,000 archers and peltasts.
86
Ibid., 2.1.5. “the Assyrian himself, who holds babylonia and all Assyria besides ... brought
at least 20,000 horsemen and ... not less than 200 chariots and ... a great number of foot soldiers.”
the kings of Arabia and Assyria were the only members of the alliance to contribute chariots. Again,
à propos of Cyrus equipping his army with chariots, Xenophon mentions that the peoples of Media,
Syria and Arabia ... employ the chariot in the same way as the Cyrenians today.” (6.1.27). The chariot
is not a weapon of nomads and is only practicable in country where the land surface is relatively hard
and free of large stones, unlike most of north Arabia.
87
Ibid., 4.2.31.
88
Ibid., 7.4.16. In the past, this statement has caused problems to scholars who have assumed
that the Arabians must have been in what we think of as “Arabia” (i.e. the Peninsula), see, for instance,
82
83
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
15
three nations are mentioned suggests a route from sardis south-eastwards towards
the euphrates, which would only make sense if these “Arabians” were in northern
syria, somewhere between Cappadocia and the euphrates (see figure 2).
Over a century later, in 401 BC, Xenophon himself encountered Arabians in
central Mesopotamia during his famous march with the younger Cyrus from Sardis
to babylonia which ended at the battle of Cunaxa followed by the long march back
with the 10,000. Again, there is no hint that these were nomads.89
Alexander also encountered Arabians in southern Mesopotamia. They had been
settling in babylonia for centuries90 and by Alexander’s day there may well have
been a more or less continuous band of Arabian settlement in southern Mesopotamia and the north-east of the Peninsula. he sailed down the Pallacopas canal which
ran parallel with and to the west of the euphrates “to the lands of the Arabians”,
where he founded yet another Alexandria and populated it with volunteers and
disabled ex-soldiers.91 this may well be the point at which, according to nearchos
“there is an inhabited village which received the merchandise from Arabia; for the
seaboard of the Arabians borders on the mouth of the euphrates and the Pasitigris”
(see figure 2).92 If this is correct, it means that the Arabians were in the area east
of the mouth of the euphrates (as well, presumably, as the regions west and southwest of it) and so, in effect, controlled the whole head of the Persian gulf. Later,
this area came to be known as Mesene which Strabo, at the turn of the era, describes
as having on one side [i.e. west and south-west] “the desert of the Arabians; and on
another side [i.e. north-west, north and north-east] ... the marshes ... formed by diversions of water from the Euphrates; and on another side [i.e. south] ... the Persian
sea” (figure 2).93 I shall return to the Arabians in Mesopotamia, below.
thus, by the death of Alexander, or shortly afterwards, the greeks knew not only
of the existence of the Persian gulf but a great deal about the inhabitants of both
its shores. they also had a more or less clear idea of the shape of “the peninsula of
the Arabians” and had located the source of frankincense, myrrh and other aromatics in the south-west of the Peninsula. they knew that the north of this peninsula
joined Mesopotamia in the east and the Levant in the west, and that Sinai and eastern egypt were also the home of Arabians.94 they were also aware of Arabians in
the Anti-Lebanon and in northern Syria and in central and southern Mesopotamia.
thus the term “Arabia” as used by the greeks at the beginning of the hellenistic
period, referred to at least six different parts of the Middle East: (1) eastern Egypt,
Dandemaev 1989: 59–60. To conquer these Arabians would, of course, have caused Cyrus an enormous detour.
89
See Macdonald 2003 (= paper VI in this collection): 308.
90
Ephʿal 1982: 112–117.
91
Arrian Anabasis, 7.21.7.
92
Quoted by Strabo, 15.3.5.
93
Ibid., 16.4.1.
94
see Arrian’s summary of knowledge of the Peninsula Indica, 43.4–6, and the descriptions by
eratosthenes and Artemidorus (preserved in strabo, 16.4.2–4) much of which must have been based
on the explorations ordered by Alexander.
figure 2. greek knowledge of Arabians before Pompey’s conquest of syria (63 bC).
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
17
Sinai and parts of southern Palestine; (2) the Anti-Lebanon; (3) northern Syria; (4)
central Mesopotamia; (5) southern Mesopotamia and the head of the Persian Gulf;
(6) the Arabian Peninsula. Within the next century, they were to add the regions of
(7) Ammon and at least part of Gilead, in northern Transjordan and (8) southern
transjordan. some of these may in fact have been contiguous and greek geographers may have considered them as parts of the same “Arabia”.
All this marked a stunning advance on previous greek knowledge of the Arabians and their homelands, but it was almost entirely theoretical and was still mixed
with exotic legends. for at this stage, the contact between greeks and Arabians had
been minimal. however, Alexander’s conquests created a cultural conduit between
the hellenic world and the near east through which cultural, social and political
ideas flowed easily in both directions.
from this point onwards, greek knowledge was refined by subsequent explorations such as those of Ariston in the red sea, which formed the basis of Agatharchides’ work on the subject.95 Ariston identified various peoples along the west
coast of the Peninsula. though well aware of the existence of the Ailanitic Inlet
(gulf of Aqaba), he considers the west coast of sinai to be part of Arabia, which
suggests that by his day (at least) it was populated by “Arabians”.96 near the southern tip of sinai he locates a point to which Agatharchides says “the gerrhaeans97
and Minaeans bring, as is the report, frankincense and other aromatic products from
what is called upper Arabia”. the last two words show that by this time the location
of the source of aromatics was correctly located in southern Arabia, for on maps in
antiquity the south was usually at the top.
Along the coast, Agatharchides describes a number of different peoples of varying degrees of savagery, some hunter-gatherers (e.g. the banizomenes),98 some
pastoralists (e.g. some of the debai whose lives revolved around the camel),99 and
a few farmers (e.g. others of the debai).100 finally, he comes to the sabaeans “the
most numerous of the Arabian peoples.”101 note the unconscious circular argument:
for greek geographers “Arabia” is anywhere inhabited by Arabians. thus, because
the sabaeans live in a region which greek geographers have called “Arabia” they
must be “Arabians”.
the picture of the west coast of Arabia which emerges by the turn of the era is
one in which there are civilized, if exotic, societies at the northern and southern end
95
Ariston was sent on a voyage of exploration down the western coast of the red sea by Ptolemy II Philadelphus (284–247 BC). His report was used by Agatharchides of Cnidus in his book on the
red sea, lengthy quotations from which are preserved in diodorus siculus, strabo (via an abridgement made by Artemidorus of ephesus at the end of the second century bC) and the byzantine writer
Photius. See Burstein 1989.
96
Agatharchides, 87a (via Diodorus 3.42.5).
97
gerrha was on the east of the Peninsula, but its exact position is still disputed.
98
Agatharchides 92a, and b.
99
Ibid., 97a, b, c.
100
Ibid., 97a, and c.
101
Πολυανθρωπότατοι τῶν Ἀπαβικῶν ἐθνῶν ὂντες, ibid., 99b.
18
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
of the red sea – the nabataeans and sabaeans – separated by large areas inhabited
by peoples in various stages of “underdevelopment”. this is the picture presented
by diodorus and strabo, both of whom were, for the most part, making amalgams
of earlier reports.
by this time, the shape of the Arabian Peninsula was more or less clear, as was
its relationship to Syria and Mesopotamia. Thus Strabo states succinctly that the
whole of the land of the Arabians (except those in Mesopotamia) lies south of
Judaea, Coelë (i.e. southern) syria and babylonia. the part near the euphrates,
between Babylonia and Coelë Syria (as well as in Mesopotamia) is occupied by
the tent-dwelling Arabians (σκηνῖται Ἄπαβες) who are divided up into small sovereignties (δυναστείας μικρὰς) and mainly breed livestock (particularly camels) in
barren and waterless steppe. south of these lies an extensive desert, and south of
that are “the people who inhabit Arabia eudaimon”. thus “the northern side of Arabia eudaimon is formed by the above-mentioned desert, the eastern by the Persian
Gulf, the western by the Arabian Gulf [our Red Sea], and the southern by the great
sea that lies outside both gulfs.”102
eratosthenes (c. 285–194 BC), whom Strabo quotes extensively, already had a
good idea of the peoples who inhabited north Arabia as well. Working from west
to east, starting at the gulf of suez, he lists these as first the nabataeans, whom
we know to have been in the negev, the eastern delta of egypt, (probably) sinai,
southern transjordan and north-west Arabia. next he mentions the Chaulotaeans
(Χαυλοταίοι). These are probably the people who appear as Ḥwlt in the so-called
“safaitic inscriptions”, graffiti carved almost entirely by nomads on the desert rocks
of what is now southern syria, north-eastern Jordan and northern saudi Arabia, in
the roman period. the Ḥwlt usually appear as the enemy in these texts and are generally portrayed as making incursions into the pasturing grounds of these nomads.
there is a small amount of evidence that they came from northern Arabia,103 and
they are probably to be identified with Pliny’s Avalitae whose towns were domata
(modern al-Jawf) and Haegra (modern Madāʾin Ṣāliḥ), both in northern Arabia.104
the third people that eratosthenes places in the northern part of Arabia are the
Agraeans (Ἀγραίοι). these are almost certainly the people in northern and northeastern Arabia known as Hgr, whose king struck coins in approximately the third
century bC.105
Strabo, 16.3.1; cf. 16.4.2ff. However, he could also be confused, as apparently in the first
sentence of 16.4.21.
103
Macdonald 2000 (= paper III in this collection): 41–42. Since that paper appeared, I have
identified two more inscriptions by members of the Ḥwlt, this time in the hismaic script, and I shall be
publishing these shortly. It is possible that the Ḥwlt should also be identified with the land of Ḥawîlâ
in Genesis 25:18, see Müller WW 1992.
104
Pliny, Naturalis Historia, 6.32.157. In fact, by Pliny’s time, Haegra belonged to the Nabataeans, but it is anyway by no means certain how accurate his information was, or what exactly was
implied by his attribution of oppida to tribes.
105
Robin 1974: 92–111; Potts 1990, ii: 60–62; Callot 1990: 227–234, and references there.
102
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
19
eratosthenes continues, saying that the first people south of the syrians and
Judaeans are farmers, which would describe very well the nabataeans, who were
expert at conserving the small seasonal rainfall in their land and making the maximum use of it for agriculture.106 south of this, “the soil is sandy and barren, producing a few palm-trees, a thorny tree and the tamarisk, and affording water by digging
... and it is occupied by tent-dwelling Arabians and camel-herds”. this is a good
description of Arabia deserta, while the statement “the extreme parts towards the
south, lying opposite ethiopia, are watered by summer rains and are sowed twice,
like India...” presumably describes the monsoon rains in Yemen. 107 strabo also had
access to descriptions of both coasts of the Persian gulf and of the red sea and apparently a certain amount of information on the Indian Ocean coast of the Peninsula
as well.108
Under the Seleucids and then the Romans, Syria and Mesopotamia gradually became better known to the greeks. this is clearly seen in strabo’s treatment of syria.
he identifies a number of different groups as σκηνῖται, i.e. “dwellers in tents”. the
first are tent-dwelling Arabians in the northern Jazīrah (figure 2), where the area
between the euphrates and the tigris is at its widest. these he describes as “a tribe
of brigands and shepherds who readily move from one place to another when pasture or booty fail them.” the people “who live alongside the mountains” (i.e. in the
foothills of the kurdish mountains) are attacked from above by the Armenians and
from below by the Arabians.109 Interestingly, this is one of the very few descriptions
of nomadic Arabians in greek texts before the Christian era.110
A different group of Arabian σκηνῖται or tent-dwellers lived in Parapotamia, on
the right (i.e. “west”) bank of the euphrates.111 they were apparently independent
106
see, for instance, strabo 16.4.26 “most of the country is well supplied with fruits”.
see strabo, 16.4.2, for this whole passage.
108
Ibid., 16.3.2–7, 16.4.5–20.
109
Ibid., 16.1.26.
110
In Pseudo-scylax (c. fourth century bC) there is a tantalizing passage that may refer to the
Arabians between gaza and Pelusium. It occurs immediately after a statement of the distance between Coelë syria and Askalon and is followed, after a considerable gap, by the end of the section on
Arabia, but we cannot be sure that it refers to Arabians since all but an initial A (of Ἀραβία?) is lost at
the beginning of the fragment. Interestingly, the people it describes are said to be mounted nomads,
with the implication that they rode horses (νομάδες ἱππεύοντες) who raised small cattle of all sorts,
particularly sheep, go[ats] and camels (Müller C. 1855: 79). If this does refer to Arabians, it would be
by far the earliest description of them as horsemen.
111
Strabo, 16.1.28; 16.2.1, 16.2.10, 16.2.11. Parapotamia could, of course, refer to either bank
of the river, but strabo makes it clear that he means the right bank in this case (16.2.10, 16.2.11). Confusingly, strabo also describes a quite different group of people whom he also calls Σκηνῖται, though
he explains that they are “now called Malians”. Far from being nomads, these lived in and around “a
noteworthy city” called Σκηνάι, which was “situated on the borders of a canal towards babylonia.”
these people were “peaceful and moderate towards travellers in the exaction of tribute”. he does not
make clear whether or not they were Arabians (16.1.27). It is tempting to compare the name of this
city, Σκηνάι i.e. “tents, encampment” with that of the later city of al-Ḥīrah, in southern Iraq, whose
name is generally derived from syriac ḥirtā “encampment”.
107
20
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
of the romans and “gave ear” to rome or to Persia as they thought fit.112 As an
example, strabo mentions the chief of one such tribe, “Alchaedamnus, king of the
Rhambaeans, who were nomads on this side of the Euphrates ... [who] was a friend
of the romans, but upon the belief that he was being treated unjustly by roman
governors, retired to Mesopotamia and then became a mercenary in the service of
Bassus [who had rebelled against the Romans].”113
finally, central syria, east of Apamaea on the Orontes114 as far as the euphrates, was dominated by “the Arabian chieftains [φύλαρχοι]” of Parapotamia.115 the
area south of Apamaea as far as the Ituraean kingdom116 “belongs for the most part
to tent-dwellers [σκηνῖται] ... similar to the nomads in Mesopotamia”. However,
strabo seems to distinguish these tent-dwellers from Arabians since, though he
regards both groups as “less civilized” than those in close “proximity to the syrians”, he says that the Arabians have governments that are better organized, citing
Arethusa (Emesa, modern Ḥomṣ) as an example. Thus, it would appear that Strabo
considered there were Arabian cities in central syria as well as Arabian nomads.117
When, in Ad 106, the romans annexed the nabataean kingdom and it became
Provincia Arabia, the terms “Arabian” and “Arabia” received yet another meaning. slowly the bewildering number of different “Arabias” disappeared from greek
writing and the term came to be confined (approximately) to the inhabitants of the
Province or of the Peninsula, regardless of whether or not this was how they thought
of themselves. elsewhere, the Arabian quality of some areas seems to have been
forgotten. In egypt, for instance, the term Ἀραβάρχης which must originally have
signified “a ruler of Arabians”118 came to mean “a tax or customs collector” of any
ethnic background and in any part of egypt.119 the way of life of those described by
greek writers became a more important identifier than their ethnicity (however that
was defined) and so the terms “tent-dweller” and “nomad” came to be widely used
without the qualification “Arabian”, etc. by the fourth century, these terms had
largely been replaced by Σαρακηνοί (saracens), a term which probably originated
in the Arabic120 word šarqiye for “those who migrate each year to the inner desert”,
Ibid., 16.1.28; 16.2.1.
Ibid., 16.210.
114
Ibid., 16.2.10.
115
Ibid., 16.2.11.
116
This was situated in the Anti-Lebanon and the Beqāʿ valley and had its capital at Chalcis
– hence strabo calls the region ἡ Χαλκιδικὴ, 16.2.11 – and its religious centre at heliopolis (baalbek).
strabo writes of ἡ Χαλκιδικὴ that it “extends down [i.e. northwards] from Massyas [the Beqāʿ valley,
see 16.2.18], and all the country south of the Apameians, which belongs for the most part to σκηνῖται”
(16.2.11).
117
strabo, 16.2.11.
118
Indeed, this is apparently how it is used by Cicero in his sarcastic reference to Pompey (Epistulae ad Atticum, 2.17.3).
119
Lesquier 1917 100–102.
120
It probably also existed in Ancient north Arabian, where we find the verb ʾs²rq “to migrate
to the inner desert [regardless of direction]”. For the distinction between Arabic and Ancient North
Arabian see Macdonald 2000 (= paper III in this collection): 28–32.
112
113
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
21
i.e. nomads121. this then came to be used for all those who had previously been
called “Arabians”,122 and after the rise of Islam, for all Muslims. Only the Roman
Province and the Peninsula retained the name “Arabia” and the latter is today one
of the enduring legacies of ancient greek encounters with Arabians.
III Greek Images of Arabias and Arabians
the ways in which Arabians were portrayed in greek writings developed almost independently of the progress of greek contact with and knowledge of them. Almost
from the earliest surviving descriptions, pictures of different groups of Arabians
were constructed from a few (often garbled) facts, some stereotypes, topoi, and folk
etymologies. As we shall see, some of these have been remarkably enduring and, in
an only slightly modified form, survive in the West today.
the first of these images is that of the far-away and exotic land inextricably
bound up with frankincense and other luxury goods. the enormous sums which
Arabian merchants charged for their wares in the markets of gaza, egypt, Phoenicia and greece created the impression that the country from which they came
must be fabulously wealthy.123 such a view, of course, ignored the immense cost of
transporting the goods and the heavy tolls paid at every stage of the journey. then
as now, producer countries seldom benefited from the high retail prices of their
commodities in far-off markets.
the fairy-tale nature of Arabia eudaimon was fixed long before its whereabouts
were finally discovered by Alexander’s explorers. herodotus, who thought frankincense grew on the eastern slopes of the mountains in eastern egypt, but who also
reported that “Arabia is the farthest to the south of all inhabited countries,”124 tells
bizarre stories about the gathering of aromatics and ends “there is a marvellous
sweet smell from all this land of Arabia”.125
Thus, when Agatharchides gave his description of ancient Yemen which purported to be based on the first-hand reports of Alexander’s explorers, he remarked
that “a natural sweet odour pervades the entire country because practically all the
things which excel in fragrance grow there unceasingly ... As for persons sailing
along the coast, although they are far from land, that does not prevent them sharing this kind of pleasure, for in summer ... there is an offshore breeze.”126 even at
this distance, the strength and sweetness of this “mixture of the noblest perfumes”
See Macdonald 1995a: 95, and the English version which appears as paper VIII in this collection. the term does not mean “easterners” as has sometimes been supposed, but those who migrate to
the inner desert, whatever the direction.
122
See, for instance, the Roman historian Ammianus Marcellinus (c. AD 330–395) who said of
the saracens that their “original abode extends from the Assyrians to the cataracts of the nile and the
frontiers of the blemmyae” (14.4.3).
123
Agatharchides, 104b (preserved in Diodorus, 3.47.5).
124
Herodotus, 3.107.
125
Ibid., 3.113.
126
Agatharchides, 99b (preserved in Diodorus, 3.46.4).
121
22
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
in their fresh state makes those “who partake of its special qualities think that they
have enjoyed the mythical ambrosia.”127 the fertility of the land and the wealth of
its inhabitants are described in lavish terms. they surpass “in wealth and all the
various forms of extravagance not only the nearby Arabians but also the rest of
mankind”128 needless to say, Agatharchides here is not providing information gathered empirically, but is elaborating on a literary topos to provide what his readers
would expect of the fabulous land of “Arabia”.
so strongly rooted and so wide-spread was this idea, that, in describing Alexander’s march south from Arbela (figure 2)129 through Mesopotamia, Quintus Curtius
rufus said that the army passed Arabia “famed for its aromatics and fertility”.130
Curtius was writing in the first century Ad, after the ill-fated expedition led by
Aelius gallus down the west coast of the Peninsula had confirmed the true position of ancient Yemen. Yet so strong was the association of the name “Arabia” with
aromatics and fertility that Curtius inadvertently applies the description to a quite
different area inhabited by “Arabians”, at the opposite end of the Middle East.
thus the picture of Arabia eudaimon is like something out of One thousand
and one nights. It is a country at the ends of the earth. A land of unimaginable
wealth and fertility, where perfumes, which in the real world would cost a fortune,
simply blow on the breezes, and cinnamon and cassia are used as firewood.131 the
inhabitants live in extraordinary opulence and have strange sexual habits,132 another
favourite topos among greek and roman geographers. they have a king who lives
“in effeminate luxury” and although he is the “authority in lawsuits” he is a prisoner in his palace. the land is filled with weird and dangerous animals. herodotus has
a whole list of them, 133 and even Agatharchides describes a peculiarly unpleasant
species of snake in the “frankincense forests” which leaps into the air as it strikes
and is probably descended from herodotus’ flying snakes.134 It is not an ideal, nor
is it a sort of “Paradise”, it is simply fascinatingly exotic.
ibid.
Agatharchides, 104b (preserved in Diodorus, 3.47.5).
129
Arbela is north-east of the Tigris, between the Great and Little Zab rivers, in what is now
north-eastern Iraq.
130
Quintus Curtius Rufus, Anabasis, 5.1.11. the passage says that as Alexander was marching
south from Arbela (towards Babylon, 5.1.17), he passed Arabia on his left (euntibus a parte laeva Arabia). Curtius does not say at what point in his journey Alexander crossed the tigris, but his statement in
the very next paragraph (5.1.12) that the route was “through plains in the land lying between the tigris
and the euphrates” suggests that he did so as soon as possible after leaving Arbela, before turning south.
If “left” is not simply an error for “right”, as has often been assumed, this would mean that Curtius
thought of this “Arabia” as lying on one or other side of the tigris. It is worth noting that Adiabene (between the Great and Little Zab rivers) on the eastern side of the Tigris, part of which Alexander had just
left, was an area in which Arabians were known to live at other times (see Macdonald forthcoming).
131
See Zambrini 1997: 485–487.
132
strabo, 16.4.25.
133
Herodotus, 3.107–113.
134
Agatharchides, 99; Herodotus, 3.107.
127
128
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
23
Parallel to this image of Arabia felix was an equally fabulous one of Arabia
deserta. As noted above, the earliest “Arabians” whom the greeks encountered
were merchants, and it seems that it was some time before they came across Arabian nomadic pastoralists. Moreover, when they finally did so, the nomads were
not in the areas which the greeks had traditionally called “Arabia” (i.e. southern
Palestine, northern Sinai, eastern Egypt, or the Peninsula) but in parts of Mesopotamia and central Syria. Yet soon this image was to become the dominant one of the
“Arabian” not only in the greek but in the Western mind in general. the nomad,
who lived in tents in the desert, far away from civilization, who bred camels, sheep
and goats, and who raided law-abiding merchants and farmers:135 this became the
image of the “Arabian” par excellence and we have already seen it being built up in
the descriptions by greek writers of the hellenistic and roman periods.
the greek “fabulous” picture of the Arabian nomad reaches its apogee in diodorus’ descriptions. here, the nomad is depicted almost as the “noble savage” and
placed in an idealized and exotic environment. the desert is a refuge into which he
can retreat from those who would try to subdue him. Only these Arabians know the
“places of hidden water” in the desert, which they can open up to obtain “drinking
water in abundance”136 thus, the desert – so cruel to their enemies – is gracious to
the nomads. “It is spacious as air”137 and “is ranged over by a multitude of Arabians
who are nomads and have chosen a life in tents. these raise great herds of animals
and make their camps in plains of immeasurable extent.”138 Indeed, “there is ... such
a multitude of herds that many tribes which have chosen a nomad life are able to
live very well, experiencing no want of grain but being abundantly provided for by
their herds.”139 this desert is full of strange and wonderful animals, which, though
he says they are fierce, Diodorus does not seem to regard as harmful. Many are
hybrids, such as the struthocamel140 and cameleopard,141 which, he explains, result
See, for instance, Diodorus, 19.94.2–10.
Ibid., 2.48.3.
137
Ibid., 2.54.2.
138
Ibid., 2.54.1.
139
Ibid., 2.50.2. It is not absolutely certain that diodorus is referring here to Arabia Deserta
rather then Arabia Felix which he described in chapter 49, but both the content and the sentences
which immediately follow this quotation suggest that he was.
140
Ibid., 2.50.3–6. this was probably the ostrich.
141
Ibid., 2.51.1. the “cameleopard” (καμηλοπάρδαλις) is often said to be the giraffe, despite
the fact that the latter does not, of course, live in Arabia. but diodorus describes the cameleopard as
being smaller than a camel and having a shorter neck. It has sometimes been suggested that this passage should be emended to make it similar to Agatharchides’ description of a cameleopard, which,
he says, is the same size as a “camel ... and its neck is so long that it obtains its food from the tops of
trees” (73a, and cf. 73b). It is surely obvious that these authors are describing two different animals.
the cameleopard described by Agatharchides was in Africa and is clearly a giraffe observed from
nature. On Greek and Roman knowledge of the giraffe see Gatier 1996. Diodorus’ cameleopard, on
the other hand, is in Arabia and is in a list of imaginary hybrid animals. It is almost certainly shown
in the first century BC frescoes in the necropolis at Marissa, Palestine, see Peters &. Thiersch 1905:
25, Pl. VIII, and Jacobson 2005: 34, no. 4, pls A5 and 14. However, the authors of both works appear
to be unaware of the distinction and simply regard it as a badly drawn giraffe. As late as Ad 531, it
135
136
24
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
from the land breathing “in a great deal of the sun’s strength, which is the greatest
source of life, and ... for that reason it generates breeds of beautiful animals in great
number and of varied colour ... [and] also outcroppings of every kind of precious
stone which are unusual in colour and resplendent in brilliancy”.142 In this desert,
there is no hint of thirst, or of labour, or of privation. Its inhabitants have all chosen
their way of life and enjoy the sort of abundant blessings showered on Job before,
and particularly after, his afflictions.
but there was also a negative aspect to these descriptions, even when, like those
of diodorus, they were favourable in all other respects.143 for even though he painted an idealized picture of nomadic life, a cultured, urban greek intellectual could
hardly regard it as ideal. thus, it was difficult for the outsider to see the difference
between raiding and brigandage. Indeed, to the victims, the two must have been
indistinguishable, even though to the perpetrators the motives and conventions may
have been quite distinct. thus diodorus, says of the “Arabians, who are called
nabataeans,” “they lead a life of brigandage, and overrunning a large part of the
neighbouring territory they pillage it.”144 Indeed, they extended their “brigandage”
from the land to the sea, and, in the gulf of Aqaba, they harassed Ptolemaic and
roman shipping which must have been threatening their monopoly of the northern
end of the frankincense trade.145 To the Egyptians and the Romans it was piracy; to
the nabataeans it was, no doubt, self-defence.
this negative strain goes back to a much older tradition, found in the Assyrian
annals of the eighth and seventh centuries bC, in which Arabian nomads, along
with many other peoples, are made to represent the opposite of the self-perception
of the society which views them. they are “the other”, “the enemy,”146 they live in
deserts not in towns, in tents rather than houses, they have strange sexual customs,
they break faith, and above all they refuse to accept imperial authority. In the roman and especially the byzantine periods, this use of the image of the Arabian
nomad, along with that of other “enemies at the gates”, increasingly replaced the
earlier positive, “libertarian” view. the nomad’s refusal to bow to authority, which
had been idealized by greeks of the fifth century bC and the hellenistic period,
came to be seen, in an imperial age, as wilful disobedience to the empire’s “mandate of heaven” and rejection of its mission civilisatrice, just as the Assyrians had
characterized it a millennium before.
the association of “Arabians” with camels became almost obligatory. Agatharchides in the second century bC describes those debai who were nomadic147 as
“camel-raisers who rely on this beast for all the most important necessities of life.
appears in a mosaic at Mount Nebo in Jordan (Piccirillo 1991: 64, fig. 31).
142
diodorus, 2.51.3 and 2.52.1. he also mentions brilliantly coloured birds (2.52.5).
143
Ibid., 2.48.2.
144
Idem.
145
Ibid., 3.43.5.
146
See Zaccagnini 1982; Malbran-Labat 1980.
147
the debai lived on the red sea coast of the Arabian Peninsula at more or less the latitude of
Mecca (see figure 2). While some were nomads, others were farmers.
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
25
they fight against their enemies from these animals, and they easily accomplish all
their business by transporting their wares loaded on their backs. they get their food
by drinking their milk, and they roam their whole country on racing camels”.148 this
idea of the camel which supplied all the nomads’ needs became a favourite topos
and is also found in diodorus who describes the animal’s many uses in more fanciful detail.149 At the end of the second century Ad, Clement of Alexandria describes
the relationship between the Arabian nomad and the camel in similar terms, though
praising the camel as having a “better nature than the barbarian [i.e. Arabian]”.150
this indissoluble link between the Arabian and the camel produced another literary topos and possibly some bizarre events in military history. herodotus tells
us that when Cyrus was faced by Croesus’ superior cavalry at the battle of sardis
(546 bC), he stripped the baggage camels of their loads and mounted men on them
“with gear appropriate to cavalrymen” and sent them ahead of his army to throw
Croesus’ cavalry into confusion, “for the horse fears the camel and cannot abide the
sight or the smell of it.” 151 the trick worked and had two, contradictory long-lasting
effects. One was the doctrine that horses have a particular antipathy to camels, for
which there is no biological evidence,152 though it is understandable that a charging
mass of large ungainly animals, of whatever species, would be enough to frighten
the best-trained horses. nevertheless, this idea became a topos which was repeated
by writers in greek and Latin for centuries afterwards.153 the second effect was
Agatharchides, 97b (preserved in Strabo, 3.45.4).
Diodorus, 2.54.6–7. He implies (2.54.6) that these camels are in “that Arabia which lies along
the Ocean ... situated above [i.e. south of] Arabia Felix”, which could refer to Ḥaḍramawt (2.54.4).
however, he may either have forgotten to revert to Arabia deserta, or the relevant words have dropped
out of the text. he mentions both the dromedary, which has one hump and is native to Arabia, and the
δίτυλος or bactrian (two-humped) camel, which is not. however, occasional examples of the latter
species, probably left behind by caravans from Central Asia, must have found their way to southern
syria and northern Arabia since they are occasionally depicted on rock drawings there, see the discussion in Macdonald, Muʾazzin & Nehmé 1996: 471–472. Diodorus describes the use of the camel in
war as follows: “and also in their wars the same animals carry into battle two bowmen who ride back
to back to each other, one of them keeping off enemies who come on them from in front, the other
those who pursue in the rear” (2.54.7). This is, of course, entirely fanciful, but it is probably based on
a misunderstanding of a genuine practice among nomads, in which a warrior would often ride pillion
(but facing to the front not backwards, of course) on a friend’s camel as far as the battlefield and then
get off the camel to fight either on foot or, if he had one, on his horse (which had been led unridden,
while he was riding the camel on the journey, to keep it fresh for battle). similarly, in flight, foot soldiers would often scramble up behind a camel-rider and, if desperate, try to ward off pursuit with their
bows. this is what is happening in the famous reliefs showing Arabians on camels fleeing from the
Assyrians, vainly trying to ward off the pursuit with their bows (British Museum WAA 124926, see
Barnett 1976: pl. XXXIII). This relief has often been misinterpreted in the past. For a brief discussion
of “Arabian” warfare, see Macdonald 1995b: 1363.
150
Clement of Alexandria 1905, paedagogus, 3.3.25.1.
151
Herodotus, 1.80.
152
Personal communication from Juliet Clutton-brock, to whom I am most grateful.
153
to take just one example, as late as the second-third century Ad, Aelian was repeating the
story that “the Persians, since the battle which Cyrus fought in Lydia [some 700 years before Aelian
was writing], keep camels together with their horses, and attempt by so doing to rid horses of the fear
148
149
26
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
that, from this time onwards, military commanders in the Middle East apparently
thought that camels could be used as mounts in regular armies.154 In fact, of course,
as any nomad could have told them, while camels are extremely useful in taking
warriors and baggage to and from the battlefield, their only use in a pitched battle
is (as Cyrus recognized) to drive a wedge through the enemy’s front and scatter his
cavalry. As far as we can tell, nomads have never fought from camel-back, for the
camel is too difficult to manoeuvre and too high to be useful in a mêlée, where it
is all too easy for an infantry man to hamstring it or stab it in the belly. Instead, the
camel is used by nomads to get them to and from the battle. nevertheless, so strong
was the association of Arabians with camels in the greek mind that writers after
herodotus assumed that an Arabian contingent in a regular army must have been
mounted on camels.155 but greek intellectuals (and possibly army commanders)
being unaware of the camel’s strengths and weaknesses treated the animal tactically
as if it were simply a large horse, with the proviso that it had to be kept separate
from the rest of the cavalry. this had disastrous effects, as at the battle of nisibis in
AD 217 when the camels deployed by the Persian king, Artabanus, were brought
down by the caltrops scattered by the romans, fell over each other and formed huge
mounds of bodies which gravely impeded the fighting.156 All this was because the
image of an “Arabian” was judged to be incomplete without a camel. Yet, in fact, as
Xenophon reported, Arabian contingents often consisted of large numbers of horsemen and even chariots,157 and in the hellenistic and roman periods Josephus never
mentions camels in the (nabataean) Arabian army, though he, and others, report
large contingents of cavalry.158
which camels inspire in them” (On the nature of animals, 11.36).
154
It is not entirely certain whether this idea was fixed in the minds of the military commanders
or whether it was a topos of the historians who reported the battles, often many years later.
155
Herodotus describes the Arabians in the army which Xerxes raised for the invasion of Greece
in 480 BC, as armed with long bows, some on foot (7.69) and some on camels (7. 86), but there is
no suggestion that the camels were to be used in battle. however, his younger contemporary, the
unreliable Ctesias of Cnidus (late fifth century BC, preserved in Diodorus Siculus [first century BC],
2.17.2) has “men mounted on camels with swords 4 cubits [2 metres] long” in the army of Semiramis.
this topos is repeated by Livy (37.40.12) who refers to camels ridden by Arab archers carrying “slender swords four cubits long “ in the army of Antiochus III at the battle of Magnesia in 189 BC. Herodian’s account of the battle of Nisibis (AD 217) goes a step further, referring to the “καταφράκτοι
[the Persian heavily armoured cavalry, the “tanks of antiquity”] on horses and camels” wielding long
spears. A man in heavy armour on top of a camel and armed with a long spear would need all his
strength and attention simply to maintain his seat, especially in a charge, and would be far more of a
danger to himself than to the enemy. Moreover, the horses of the kataphractoi were also armoured, but
sufficient armour to protect the vulnerable areas of a camel would more or less immobilize it.
156
herodian, 4.15.5.
157
Xenophon, Cyropaedia, 2.1.5.
158
For instance, in 87 BC Aretas II, king of the (Nabataean) Arabians attacked the Seleucid king
Antiochus XII with 10,000 horsemen near Qanawat in the Ḥawrān (Josephus Antiquitates Judaicae,
13.15.1); in 65 BC, Aretas III led an army of 50,000 cavalry and infantry against the Jewish ruler Aristobulus (ibid., 14.2.1); in 47 BC, Malichus I provided Julius Caesar with cavalry for the Alexandrian
war (Caesar, De bello alexandrino, 1.1), etc. even though the figures are probably fanciful, they are
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
27
but perhaps the qualities which left the most profound and long-lasting impression on the greeks, were the Arabians’ love of freedom, their independence and
their refusal to bow the knee to any conqueror. these qualities were seen as characteristic both of the settled merchant communities and of the nomads. sometimes,
fate helped them maintain their liberties by interposing huge distances between the
Arabians and the greed of would-be conquerors, as Agatharchides notes in the case
of the sabaeans.159 On the other hand, diodorus stresses that the nomads of Arabia
deserta chose their isolation and made considerable sacrifices to maintain it.160
Other Arabians, such as the merchant communities on the coasts of the Mediterranean and the Persian gulf had to resort to other means to protect their independence. thus, herodotus states that within the Persian empire, after Cambyses’
conquest of egypt (525 bC), “all Asia” was subject to the Persian crown, “except
the Arabians [in the Gaza-Ienysus region161]; these did not yield the obedience of
slaves to the Persians, but were united to them by friendship.”162 It was suggested
earlier that this “special relationship” may have resulted in the Arabians, at least
those of the gaza-Ienysus seaboard (and presumably the hinterland in the negev),
being granted the, apparently unique, status of an autonomous enclave within the
fifth satrapy.163 however, as I also pointed out above, it seems unlikely that the
Persians granted this autonomy simply out of gratitude, and it is clear that the Arabians paid dearly for the privilege, their annual “gift” to the Persian treasury of
1000 talents of frankincense being worth considerably more than the annual tax of
350 talents of silver paid by the whole of the fifth satrapy.
Unlike the Arabian nomads, the Arabian merchants on the Mediterranean coast
possessed many of “the things which are valued among other peoples”164 and were
not able to preserve their liberty by melting into the desert beyond the reach of the
covetous. the long and dangerous routes by which their goods had to travel from
ancient Yemen to the Mediterranean coast also made them peculiarly vulnerable to
blackmail by those who possessed the means to disrupt the trade. It seems likely
therefore that these Arabian merchants persuaded the Persians that the interests of
the empire would best be served by permitting the Arabians to continue running
their trade in their own way, in return for a huge annual payment, rather than by
trying to administer and tax a sullen and hostile people with endless opportunities
clearly meant to convey considerable forces of cavalry.
159
Agatharchides, 104b (preserved in Diodorus, 3.47.8). He ends his description of the country
and its riches with the comment “they have maintained this prosperity undisturbed for a great period
of time, however, because they live completely apart from those people who because of their own
greed, consider the wealth of other people their own godsend.” Another version adds “since slackness
is unable to preserve freedom for a long time” (104a, preserved in Photius, Cod.250.102, 459b).
160
Diodorus, 19.97.3–4.
161
Presumably, this also included the Arabians in eastern egypt, since no mention is made of
them in herodotus’ description of the sixth satrapy, which consisted of egypt and the regions around
it (3.91).
162
Herodotus, 3.88.
163
Ibid., 3.91, 97.
164
Diodorus, 19.97.4.
28
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
for circumventing the authorities along the desert route from ancient Yemen. In
effect, therefore these Arabians negotiated and bought their liberty. this could be
regarded as a testament to the determination and endless ingenuity of the Arabians
in preserving the one thing they prized most highly.
According to Polybius, when, at the end of the third century bC, Antiochus III
attacked the Arabian merchants of gerrha on the Persian gulf coast of the Peninsula, they struck a similar deal with him. by this, Antiochus agreed to allow the
gerrhaeans to retain their “perpetual peace and freedom” and in return the gerrhaeans “passed a decree honouring Antiochus with the gift of five hundred talents of
silver, a thousand talents of frankincense, and two hundred talents of the so-called
‘stacte’ [oil of myrrh].”165
even the nomadic nabataeans were prepared to buy off aggressors when necessary, at least according to diodorus siculus. In his description of how, in 312 bC,
Antigonus I sent his son, demetrius Poliorcetes, to attack and plunder the nabataeans, diodorus puts into their mouths the speech on liberty quoted below which
includes the words, “We therefore beg both you and your father to do us no injury
but, after receiving gifts from us, to withdraw your army and henceforth regard the
nabataeans as your friends.”166 demetrius agreed and received from them “as gifts,
the most precious of their products”.167
this independence expressed itself in other ways too. When Alexander was preparing to invade the Arabian Peninsula he is said to have used as the casus belli
the claim that “the Arabians [presumably those in the Persian Gulf region] were
the only people on earth who did not send ambassadors to him”.168 Whether or not
it was true, for the claim to be plausible this must have been the sort of act which
people would regard as characteristic of Arabians.
A similar story is told of another group of Arabians, 500 years later, when the
Roman emperor Septimius Severus was campaigning in northern Mesopotamia.
the Arabians in the region, who had joined their neighbours, the Osrhoëni and Adiabeni, in a bid for independence in the reign of the previous emperor, Pescennius
Niger, sent envoys to Severus in AD 196 suing for peace. However, the Osrhoëni
and Adiabeni were shown no mercy, on the grounds that the terms they offered
fell far short of complete submission. While their neighbours were being punished
for their effrontery, the Arabians sent a second mission to severus offering “more
reasonable terms”, but “they did not obtain what they wanted, since they had not
come in person.”169 It is clear that severus was determined to teach the whole area
a lesson and nothing short of complete submission could have saved the Arabians
from sharing their neighbours’ fate. but it is interesting that the ostensible grounds
165
166
167
168
169
Polybius, Historia, 13.94–5.
Diodorus, 19.97.4.
Ibid., 19.97.6.
strabo, 16.1.11.
Cassius dio, Roman History, 75.2.1.
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
29
for refusing their “more reasonable offers” should have been a story which presumably reflected behaviour expected of “Arabians”.
After describing how, “in the waterless region,” the nomadic nabataeans “dug
wells at convenient intervals and have kept the knowledge of them hidden from the
peoples of all other nations”,170 diodorus neatly summarizes his picture of these
“Arabians” in the following words, “Consequently, the Arabians who inhabit this
country,171 being difficult to overcome in war, remain always unenslaved; furthermore, they never at any time accept a man of another country as their over-lord and
continue to maintain their liberty unimpaired, Consequently, neither the Assyrians
of old, nor the kings of the Medes and Persians, nor yet those of the Macedonians
have been able to enslave them.”172
While diodorus is here referring to the nomadic nabataeans in the negev and
southern transjordan, he uses almost identical words of the “Arabians” who, he
says, were allies of the mythical king ninus, the founder of the Assyrian empire
and husband of semiramis.173 Within the terms of the ninus story, diodorus must
surely have envisaged these “Arabians” as being those in northern Mesopotamia.
however, while the statement about the independence of the Arabians and the difficulties their land presented to would-be conquerors fits well within his description of the nabataeans, it is completely irrelevant to the story of ninus, in which
the Arabians are simply presented as his allies in the conquest of the whole of the
Middle East. This suggests that the term “Arabian” automatically evoked – at least
to diodorus but presumably also to his readers – the image of an unconquerable,
independent-minded people, regardless of which group of Arabians was meant or
of the context in which they were mentioned.
Where did the greeks get this view of the Arabians? It is obviously impossible
to know for sure but it is very interesting that it coincides very closely with the
qualities by which the modern bedouin distinguish themselves, that is autonomy
and jural equality.174
Observers from the outside are likely to have been struck most powerfully by
the emphasis on autonomy, on every man making his own decisions and on no
one being able to command another. Only on deeper acquaintance would the jural
equality become apparent. Many westerners are unaware that a Bedouin sheikh
leads, and has always led, only by agreement and has no power to command, except
in specific circumstances (such as on a raid) when individuals voluntarily agreed
to accept orders. to societies accustomed to hierarchical social and military structures, these concepts have often been difficult to grasp. Yet there are hints of a
The excavation and use of these wells is described in more detail in Diodorus, 19.94.6–10.
In 2.48.4–5, this refers to the Arabians “between Syria and Egypt”, i.e. on the Mediterranean
coast south-west of gaza (and probably the negev and southern transjordan, since he includes the
nabataeans) and in eastern egypt. however, in the almost identical passage at 2.1.5, he is apparently
referring to the Arabians in northern Mesopotamia.
172
Diodorus, 2.48.4–5.
173
Ibid., 2.1.5–6.
174
Lancaster W. & F. 1988.
170
171
30
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
recognition, though not perhaps an understanding, of it in strabo’s description of
the nabataean king “who often renders an account of his kingship in the popular
assembly; and sometimes his mode of life is examined,”175 and much later in the
description by Ammianus Marcellinus of the Saracens among whom “all alike are
warriors of equal rank”.176
this view is seen at its most romantic in diodorus, when he puts the following eloquent speech into the mouths of the nabataean Arabian nomads whom he
describes as being “exceptionally fond of freedom.”177 We “live in the desert and
in a land that has neither water nor grain nor wine nor any other thing whatsoever
which among you would be regarded as the necessities of life. for, since we are
in no way willing to be slaves, we have all taken refuge in a land that lacks all the
things which are valued among other peoples and have chosen to live a life in the
desert and one altogether like that of wild beasts, harming you not at all.... You cannot force us to live a different life and [if you try to] you will only have a few captives, disheartened slaves who would not consent to live among strange ways.”178
diodorus (or his source, hieronymus of Cardia) does not pretend that these were
the actual words of the nomads, but he presents the speech as the sort of thing he
would have expected them to say.179 thus, it provides a particularly valuable insight
into a hellenistic greek view of the Arabian character and is near the beginning of
a long tradition, which has continued sporadically to the present day, in which the
perceived virtues of the Arabian nomads are used either as examples of the most
treasured ideals of the society which views them, or as a salutary contrast to its
moral failings.180
References
unless otherwise stated, all quotations and translations of Classical sources are taken from the
Loeb Classical library editions.
Agatharchides see Burstein 1989.
barnett, r.d.
—. 1976. Sculptures from the North Palace of Ashurbanipal at Nineveh (668–627 B.C.). London:
British Museum.
bron, f.
—. 1998. Inventaire des Inscriptions Sudarabiques. 3. Maʿīn. Paris: boccard / rome: herder.
175
strabo, 16.4.26.
Ammianus Marcellinus, 14.4.3.
177
Diodorus, 19.94.6.
178
Ibid., 19.97.3–5 [translation slightly altered from that in the Loeb edition, for the sake of clarity], and see 19.94.2–9.
179
Ibid., 19.97.6, ῥηθέντων δὲ τοιούτων λόγων.
180
for an interesting study of the image of “the bedouin” in the works of the early european
travellers to the Middle East, see Tidrick 1981.
176
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
31
Burstein, S.M. (translator and editor)
—. 1989. Agatharchides of Cnidus on the Erythraean Sea. Translated and edited by S.M. Burstein.
London: hakluyt society.
Callot, O.
—. 1990. Les monnaies dites «arabes» dans le nord du Golfe arabo-persique à la fin du IIIe siècle
avant notre ère. Pages 221–240 in Y. Calvet & J. Gachet (eds), Failaka Fouilles Françaises,
1986–1988. Lyon: GDR / Maison de l’Orient.
Clement of Alexandria
—. 1905. Clemens Alexandrinus. I. (die christlichen schriftsteller der ersten drei Jahrhunderte).
Leipzig: hinrichs.
Colin, g.
—.1988. A propos des graffites sud-arabiques du Ouādi Hammāmāt. Bulletin de l’Institut Français
d’Archéologie Orientale 88: 33–36.
Dandemaev, M.A.
—. 1989. A political history of the Achaemenid empire. Translated into English by W.J. Vogelsang. Leiden: brill.
dittenberger, W.
—. 1903. Orientis Graeci Inscriptiones Selectae. I. Leipzig: hirzel.
donner, f.
—. 1986. Xenophon’s Arabia. Iraq 48: 1–14.
Ephʿal, I.
—. 1982. The Ancient Arabs. Nomads on the Borders of the Fertile Crescent 9th–5th Centuries
B.C. Jerusalem: Magnes.
gatier, P-L.
—. 1988. Philadelphie et Gerasa du royaume nabatéen à la Province d’Arabie. Pages 159–170 in
P-L. gatier, b. helly, J-P. rey-Coquais (eds), Géographie historique au Proche-Orient: Syrie,
Phénicie, Arabie, grecques, romaines, byzantines. Paris: Éditions du Cnrs.
—. 1995. La présence arabe à Gérasa et en Décapole. Pages 109–118 in H. Lozachmeur (ed.),
Présence arabe dans le Croissant fertile avant l’Hégire. Paris: Éditions recherche sur les
Civilisations.
—. 1996. Des girafes pour l’empereur. Topoi 6: 903–941.
Högemann, P.
—. 1985. Alexander der Große und Arabien. (Zetemata, 82). Munich: Beck.
Jacobson, D.M.
—. 2005. The Hellenistic Paintings of Marissa. (Palestine Exploration Fund Annual, 7). London:
Palestine exploration fund.
Jacoby, f.
—. 1923. Die Fragmente der griechischen Historiker. I. Genealogie und Mythographie. berlin:
Weidmann.
Lancaster, W. & Lancaster, f.
—. 1988. Thoughts on the Bedouinisation of Arabia. Proceedings of the Seminar for Arabian
Studies 18: 51–62.
Lesquier, J.
—. 1917. L’Arabarchès d’Égypte. Revue archéologique 5ème série, 6: 95–103.
Lobel, e. & Page, d. (eds)
—. 1963. Poetarum Lesbiorum Fragmenta. reprinted from the corrected sheets of the first edition (1955). Oxford: Clarendon.
32
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
Macdonald, M.C.A.
—. 1995a. Quelques réflexions sur les Saracènes, l’inscription de Rawwāfa et l’armée romaine.
Pages 93–101 in H. Lozachmeur (ed.), Présence arabe dans le Croissant fertile avant l’Hégire.
Paris: Éditions Recherche sur les Civilisations. [A revised version in English is printed as
Number VIII here]
—. 1995b. North Arabia in the First Millennium BCE. Pages 1355–1369 in J.M. Sasson (ed.),
Civilizations of the Ancient Near East. II. New York: Scribners.
— 1997. Trade Routes and Trade Goods at the Northern End of the “Incense Road” in the First
Millennium B.C. Pages 333–349 in A. Avanzini (ed.), Profumi d’Arabia. Atti del convegno.
(Saggi di Storia Antica, 11). Roma: “L’Erma” di Bretschneider. [= Number IX here]
—. 1998. Some Reflections on Epigraphy and Ethnicity in the Roman Near East, in G. Clarke
& d. harrison (eds), Identities in the Eastern Mediterranean in Antiquity. Proceedings of a
Conference held at the Humanities Research Centre in Canberra 10–12 November, 1997.
Mediterranean Archaeology 11: 177–190. [= Number IV here]
—. 2000. reflections on the linguistic map of pre-Islamic Arabia. Arabian archaeology and
epigraphy 11: 28–79. [= Number III here]
—. 2003. « Les Arabes en Syrie » or « La pénétration des Arabes en Syrie ». A question of perceptions. Pages 303–318 in La Syrie hellénistique. (Topoi Supplément, 4). Paris: de Boccard.
[= Number VI here].
—. forthcoming Arabs, Arabias, and Arabic before Late Antiquity. Topoi.
Macdonald, M.C.A., Muʾazzin, A. & Nehmé, L.
—. 1996. Les inscriptions safaïtiques de Syrie, cent quarante ans après leur découverte. Comptes
rendus de l’Académie des Inscriptions & Belles-Lettres: 435–494.
Malbran-Labat, F.
—. 1980. Éléments pour une recherche sur le nomadisme en Mésopotamie au premier millénaire
av. J.-C. I: “L’image du nomade”. Journal asiatique 268: 11–33.
Müller, C.
—. 1855. Geographi Graeci Minores. Paris: firmin didot.
Müller, W.W.
—. 1978. Weihrauch. Ein arabisches Produkt und seine Bedeutung in der Antike. Sonderdruck
des Artikels Weihrauch [Coll. 700–777] aus Paulys Realencyclopädie der Classischen Altertumswissenschaften. Supplement-Band 15. München: Druckenmüller.
—. 1992. Havilah. Page 82 in The Anchor Bible Dictionary. 3. New York: Doubleday.
— 1997. Namen von Aromata im antiken Südarabien. Pages 193–210 in A. Avanzini (ed.),
Profumi d’Arabia. Atti del convegno. (saggi di storia Antica, 11). roma: “L’erma” di bretschneider.
Pearson, L.
—. 1939. Early Ionian Historians. Oxford: Clarendon.
Peters, J.P. & thiersch, h.
—. 1905. Painted Tombs in the necropolis of Marissa (Marêshah). London: Palestine exploration fund.
Piccirillo, M.
—. 1991. I mosaici di Giordania. Catalogue of an exhibition at Castello di Malpaga, CavernagoBergamo, Marzo-Giugno 1991. Bergamo: Centro Culturale «Nicolò Rezzara» / Jerusalem:
Custodia di terra santa.
Potts, d.t.
—. 1990. The Arabian Gulf in Antiquity. (2 volumes) Oxford: Clarendon.
ArAbIAns, ArAbIAs, And the greeks: COntACt And PerCePtIOns
33
Powell, J.u.
—. 1925. Collectanea Alexandrina. Reliquiae minores Poetarum Graecorum Aetatis Ptolemaicae
323–146 A.C. .... Oxford: Clarendon.
Retsö, J.
—. 1997. The Arab Connection. Political Implications of Frankincense in Early Greece. Pages
473–480 in A. Avanzini (ed.), Profumi d’Arabia. Atti del convegno. (saggi di storia Antica,
11). roma: “L’erma” di bretschneider.
robin, C.
—. 1974. Monnaies provenant de l’Arabie du nord-est. Semitica 24: 83–125.
—. 1994. L’Égypte dans les inscriptions de l’Arabie méridionale préislamique. Pages 285–301 in
C. berger, g. Clerc, n. grimal (eds), Hommages à Jean Leclant. 4. Le Caire: Institut français
d’Archéologie Orientale.
—. 1995. Les graffiti «thamūdéens»: 6A et C. Pages 110, 112-116, 124 (= fig 12) in A. BülowJacobsen, et al., Les inscriptions d’Al-Muwayḥ. Bulletin de l’Institut Français d’Archéologie
Orientale 45: 103–124.
salles, J-f.
—. 1988. La circumnavigation de l’Arabie dans l’antiquité classique. Pages 75–102 in J-F. Salles
(ed.), L’Arabie et ses mers bordières. I. Itinéraires et voisinages. (Travaux de la Maison de
l’Orient, 16). Lyon: Maison de l’Orient.
sappho see Lobel & Page 1963.
seigne, J.
—. 1997. Les limites orientale et méridionale du territoire de Gerasa. Syria 74: 121–138, with an
appendix by M. Sartre (pp. 139–140).
sima, A.
—. 2000. Tiere, Planzen, Steine und Metalle in den altsüdarabischen Inschriften. Eine lexicalische
und realienkundliche Untersuchung. (Veröffentlichungen der orientalischen Kommission der
Akademie der Wissenschaften und der Literatur – Mainz, 46) Wiesbaden: Harrassowitz.
tidrick, k.
—. 1981. Heart-beguiling Araby. Cambridge: Cambridge university Press.
Zaccagnini, C.
—. 1982. The Enemy in the Neo-Assyrian Royal Inscriptions: The “Ethnographic” Description.
Pages 409–423 in H.J. Nissen & J. Renger (eds), Mesopotamien and seine Nachbarn. II.
berlin: reimer.
Zambrini, A.
—. 1997. Notes sur l’Arabie dans les sources classiques. Pages 481–494 in A. Avanzini (ed.),
Profumi d’Arabia. Atti del convegno. (saggi di storia Antica, 11). roma: “L’erma” di bretschneider.