My attempt to provoke a discussion on the
issues of the 5000 year old World System
(or the Central [first order] world-system
[intersocietal network] - it does not seem so
important how we would call it) and the pre-
modern world-systems/intersocietal networks
in general does not appear to have been successful.
I tried to do this by mentioning some South Arabian
data, which to my mind did not fit into any of the
proposed world[-]system models. However, I think,
I should have taken into consideration the simple
fact that these data are completely unknown outside
a very narrow circle of specialists in this field - e.g.
I have never got across any references to any of
such Sabaeologist classics as Christian Robin,
Muhammad `Abd al-Qa:dir Ba:faqi:h, Avraam
Lundin, or A.F.L.Beeston in any of the WS papers
(including the ones dealing with the WS of the 1st mil.
BC - 1st mil.AD). Hence, it appears necessary to
spell out the above-mentione WS-relevant South Arabian
data in a bit more detail.
Andrey Korotayev
SOUTH ARABIA AS AN ELEMENT
OF THE LONG AXIAL AGE WORLD SYSTEM
(Some Theses)
Andrey Korotayev
0.1. I think, the data mentioned below could be reasonably
interpreted in two ways (which do not exclude each
other): South Arabia could be reasonably described
either as "core-exploiting" periphery of the Long Axial
Age Circum-Mediterranean world-system, or as a second
(or third?)-order centre of the Long Axial Age World System.
0.2. The first interpretation could be possible if we
consider a periphery not as a system element necessarily
exploited by the system core, but rather as such an
element whose evolution is influenced by the evolution
of the core to a qualitatively greater extent than its
own evolution influences the evolution of the core.
0.3. The other interpretation also appears rather
reasonable. It seems appropriate to speak within the
pre-Modern (but not necessarily only pre-Modern) World
System model about the centres of different orders (a
hierarchy of centres?). In some sense it appears
possible to say that within the LAA World System the
South Arabians managed to create their own second (or third?)
order world-system (of the world-economy rather than world-
empire type) whose periphery at some periods included
considerable parts of Eastern, Central and even Northern
Arabia, Ethiopia and East Africa.
1. South Arabia appears to have been included into
the world system in the late second 2nd - early 1st
mil.BC (Old Testament, Bet Shemesh tablet, Suhu texts,
Assyrian royal inscriptions, glottochronological data,
Russian-Yemeni excavations at Raybu:n and other
archaeological [French, Italian, American, German]
materials, including radiocarbon datings] &c). This
appears to have been connected with a significant (by
influence, but not necessarily by numbers) migration
from the North-West Arabia and seems to have gone hand
in hand with the development of the Transarabian incense
(but not only incense) trade and the emergence of the
South Arabian civilization.
2. This trade appears to have been monopolized from
the very beginning by the Sabaean Kingdom (which was
also the South Arabian hegemon in the 1st half of the
1st mil.BC), though at the end of the 8th cent.BC. this
monopoly seems to have been challenged by Awsa:n which
apparently acquired control over the main Arabian
centres of the frankincense production, which in its
turn appears to have led to the Sabaean colonization of
Ethiopia, the alternative source of the frankincense.
The crushing defeat of the Awsanian Kingdom by the
Sabaeans in the 1st half of the 7th century with the
consequent reacquisition of their control over the main
Arabian frankincense sources appears to have led to the
end of the Sabaean interest in their Ethiopian colonies
(as has been suggested most recently by Frantsouzoff).
3. By the end of the 1st half of the 1st mil.BC the
Sabaean hegemony in the Arabian South comes to its end.
On the other hand, one could observe the strengthening
of the main rival kingdoms - Minaean, Qatabanian and
Hadrami. In the second half of the 1st mil.BC the
Transarabian incense trade appears to have been
dominated by the Minaeans and Hadramis (as is evidenced
by the Classical sources and Minaean inscriptions [found
from South Arabia through Arabian North and Egypt to
Delos]) who seem to have developed a direct way through
the gravel passage in the Sayhad desert to bypass the
competitor kingdoms. The Minaean-Hadrami frankincense
monopoly appears to have been sometimes challenged by
the Sabaeans, though without any stable success. The
Qatabanians seem to have counteracted by the subjugating
of the myrrh-producing areas of the Arabian South West
corner and apparently by resuming the exploitation of
the very reach East African frankincense sources.
4. At the end of the 1st mil.BC the main incense
trade routes appear to have been transferred from land
to sea (in direct connection with the development of the
Greek navigation in the Red Sea and the Indian Ocean).
The Hadramis established their colony SMHRm at the most
reach frankincense region on the far-away Dhofar coast
(evidenced both by the Classical sources and Hadrami
inscriptions). Greeks moved beyond Aden (where they
previously met with the Indian merchants) and
established direct connections with India (though
continuing the active use of the South Arabian ports).
At the end of the first century BC the newly founded
(Himyarite?) Kingdom of Saba' and dhu:-Rayda:n subjugate
most of the South Arabian West. As it could not be
bypassed within the sea-borne pattern, it starts to play
an important role in the emergent Mediterranean -
Arabian - Indian transocean network. For the late 1st
mil.BC - 1st century CE we have the best evidence for
the patterns of the South Arabian WS participation (both
through the land and sea communication lines) - mainly
from the Classical sources (Agatharkhides, Strabo,
Periplus Maris Erythraei, Pliny the Elder &c) covering
both land-borne and sea-borne patterns. The most
important points which these data elucidate:
4.a. The land-borne incense trade being controlled
on the last stage by an extraterritorial merchant league
led by a Minaean merchant house of the Gebbanites (=
GB'N?) with the headquarters in the Qatabanian capital
Timna` (according to Beeston's interpretation of Pliny
which still remains to be finally proved).
4.b. The Hadrami state frankincense monopoly.
4.c. Extensive use of coerced labour in the main
frankincense producing region in Dhofar (not attested
for any other parts of South Arabia).
4.d. Introduction of the second harvest of
frankincense by the Hadramis to increase the yields of
this cash-crop to be sold on the Mediterranean markets.
4.e. The massive passive balance of trade of the
"Mediterraneans" with South Arabia.
4.f. The "Mediterraneans" being sure that they are
"exploited" by the South Arabians.
4.g. A failed attempt to counteract this by the
direct conquest of South Arabia (Aelius Gallus campaign
in 26 BC). This incidentally shows again that the
hypothesis that the pre-1500 world-economies are
necessarily transformed into world-empires is not true.
4.h. Important role of South Arabia in the Indian-
Mediterranean trade (Periplus even evidences the own
transocean trade activity of the merchants of the South
Arabian West in India).
4.i. Heavy involvement of the inhabitants of the
South Arabian West in the trade with East Africa where
an important South Arabian trading colony (in the
territory of the present-day Tanzania) under the supreme
jurisdiction of the kings of Saba' and dhu:-Rayda:n is
attested for the 1st cent.CE. &c.
5. The data of the literary sources is usefully
supplemented by the data of other sources -
archaeological, numismatic &c, such as:
5.a. The South Arabian coinage emerged in the 2nd
half of the 1st mil.BC under the direct Mediterranean
influence, the first SA coins being replicas of the
Athenian tetradrakhma, to which ancient Yemenites soon
time later started adding South Arabian letters (such
coins are even found in Asia Minor; incidently South
Arabians might have been the first to invent coin
nominal, as it was suggested that the SA letters which
the Yemenites placed on the Athena's cheek might have
been used to distinguish the coins with different
weights-values). An extremely strong Roman influence on
the coinage of the kings of Saba' and dhu:-Rayda:n of
the 1st cent.CE.
5.b. Archaeological material shows a strong
hellenistic influence on the South Arabian arts by the
end of the 1st mil.BC.
5.c. The recent Russian-Yemenite excavations (led by
Alexander Sedov) on the site of the main ancient Hadrami
port of Qana' (in the South Arabian East) provide
abundant coin findings (which is also true of the
Hadrami colony SMHRm in the main frankincense region of
Dhofa:r excavated by the American Albright expedition),
whereas coins are very rarely found in the internal
Hadrami areas &c.
6. A bit (but only apparently) surprisingly the
numerous South Arabian (mainly epigraphical) documents
themselves provide almost no evidence on the long-
distance trade (with the only important exception of the
Minaean inscriptions).
P.S. The above-traced pattern of the SA involvement into
the world system seems to have come to its end in the
2nd century CE apparently due to the occupation of the
Red Sea cost by the Ethiopians (incidently this
coincides with a dramatic decline of the coinage in the
SA West). This, however, does not seem to have effected
the Hadrami-controlled trade of the SA East, this rather
even stimulated its further development. According to
the data of the Qana' expedition the 3rd century layers
evidence a significant growth of the port, the
appearance of the Indian artefacts not found in the 1st
- 2nd cent. layers, the growth of the presence of the
Mesopotamian and Gulf products. Surprisingly, the South
Arabian - Mediterranean trade does not appear to have
been disrupted by the 3rd cent. crisis of the Roman
Empire. It might have resulted in the disappearance of
the imports from the Western Mediterranean, which is,
however, effectively counted by the growth of volume and
diversity of imports from the Mediterranean East. Even
more surprisingly there is even no direct evidence that
Qana' trade has been negatively affected in the 4th
century by the Christianization of the Empire (which led
to a dramatic fall in the incense consumption), though
the Hadrami colony SMHRm was abandoned in the 4th
century after the Himyarite conquest of Hadramawt. Some
decline of Qana' (and Qana' based trade) might have
occurred in the 5th century which evidences some
contraction of the port territory, though there no firm
data for the actual contraction of the trade, the data
rather indicate its restructuring with emerging very
strong links with Southern Palestine, the Gulf and
possibly Mesopotamia, which might be somehow connected
with some evidence for the "judaization" of the town in
the 5th century. The 6th century "ethiopization" seems
to go hand in hand with the growth of the importance of
the East African trade links (all of which fits quite
well in the political history of the area known from the
written sources). Yet, the economic role of SA in the WS
of the 3rd-6th centuries AD remains unclear due to the
general scarcity of data (both written and
archaeological with the only exception of Qana') on the
South Arabian trade for this period and their
contradictory character. However, there is no doubt that
South Arabia being a part of the WS influenced greatly
its evolution - take e.g. the rapid total transition of
South Arabia from the polytheistic religions to the
monotheistic ones at the end of the 4th century CE, or
the spectacular political events in South Arabia of the
6th century caused to a considerable extent by the SA
involvement into the WS political-diplomatic networks of
the time.
Andrey Korotayev, Senior Research Fellow
Oriental Institute of the Russian Academy of Sciences
12 Rozhdestvenka, Moscow 103753, Russia
ANDREI@RSUH.RU