The First Dynasty of Islam: The Umayyad Caliphate ad 661-750

diwan
. The origins of both the word and the institution are rather
obscure. Tradition tells us that the first 
diwan
 was instituted by the
caliph ‘Umar I (634–44) and that at that time the word referred to
the register of soldiers and pensions due to them which ‘Umar had
compiled. However, the word seems to have taken on its meaning of
‘government department’ already by the time of the first Umayyad,
Mu‘awiya, who is said to have had a 
diwan
 for the collection of
taxes and to have introduced two for the government chancery—one
for the writing of documents and one for the sealing of documents
(the 
d
i
w
a
n al-ras
a
’il
 and the 
d
i
w
a
n al-kh
a
tam
). However, the main
business of the administration remained the assessment and
collection of taxes, and in a general sense the 
diwan
 refers to the
administration. ‘Abd al-Malik is generally credited with changing
the official language of the 
diwan
 to Arabic.
After the Arab conquest of the Middle East, as is to be expected,
the previously existing administrative systems in the various
provinces were left as far as possible intact: not only did officials who
had served the Byzantines and Sasanids continue to serve the Arabs,
the administration continued to use the languages which had been in
use before the conquest, Greek, Coptic and Persian (Pahlevi). The
change to Arabic as the sole official administrative language, the
arabisation of the 
diwan,
 is generally associated with ‘Abd al-Malik
and al-Hajjaj although there is some obscurity about dates and
circumstances. Some sources, for instance, attribute the measure to al-
Walid I rather than to his father, and the report which says that the
change of language was introduced in response to the boorish
behaviour of a Greek clerk who had urinated in an inkwell should be
seen as an example of the anecdotal explanation of major gradual
changes which mediaeval sources delight in. At any rate the evidence
of the papyri does bear out the claim that Arabic began to be the
official language of administration from about the beginning of the
eighth century AD although the change was not made overnight and it
was not until almost the end of the Umayyad period that Arabic
became the language of administration in border provinces like

---

64
‘Abd al-Malik and al-Hajjaj
Khurasan. What effect the change had or was intended to have on the
ethnic origin of the personnel of the administration is difficult to say.
It seems that the bureaucrats continued to be overwhelmingly of non-
Arab descent, that is 
mawali,
 although as time passed the distinction
between Arab and non-Arab became increasingly less clear-cut.
Equally, it is possible that the move to Arabic was intended to
encourage the acquisition of the language by the subject peoples but,
on the other hand, the very fact that the changeover could begin
necessarily indicates that already by this time there must have been a
considerable number of potential bureaucrats with at least sufficient
Arabic for the requirements of the administration.
7
One of the 
diwans
 of the mediaeval Islamic administration
occupied itself with the running of the 
barid
. The 
barid
 was a sort of
communications system, consisting of routes linking the main centres
of the empire along which there were stations with horses at the ready
so that messengers could come and go quickly between the provinces
and the metropolis. Although theoretically a postal system, in effect it
was an instrument for keeping the government informed about
developments in the provinces, and the provincial controllers of the
barid
 were local spies on behalf of the central government. Here again
Muslim tradition gives to ‘Abd al-Malik an important role in the
organisation of the 
barid
 system, although the possibly Greek or Latin
etymology of the word suggests the continuation of a former
Byzantine institution, and one often feels that Muslim tradition finds
figures like ‘Umar and ‘Abd al-Malik convenient personalities with
which to associate institutions or developments which it considers
must have a decisive beginning but about which precise details are
lacking.
8
Another important development, again focusing on ‘Abd al-Malik
and al-Hajjaj, is the introduction, for the first time, of a specifically
Muslim coinage. As with the languages of administration, so with the
coinage: the Arab conquerors had taken over and only slightly
adapted the Byzantine and Sasanid coins which were in circulation,
and the mints which had produced these coins continued to do so for
the Arabs. The minting of gold coins was a Byzantine imperial
prerogative, and the Arabs continued to import gold coins from
Byzantium. In this way the pre-conquest gold denarius, silver
drachma and copper follis became the Arab 
dinar,
 
dirham and fils
.
Some experiments with a new type of coinage made by the Sufyanid
rulers proved unsuccessful, and it was not until the 690s, both in Syria
and in Iraq, that ‘Abd al-Malik and al-Hajjaj began to mint coins of a

---

‘Abd al-Malik and al-Hajjaj 
65
decisively new type, allegedly in response to a threat by the Byzantine
ruler to stamp the gold coins exported to the Arabs with anti-Muslim
formulae.
The most important characteristic of the new coinage was the fact
that it was purely epigraphic. The faces of the coins were inscribed
only with Muslim religious formulae, not with the portraits of rulers
or other pictorial representations which had marked the Byzantine and
Sasanid as well as some of the earlier Arab coins. This was a decisive
break with numismatic tradition, and provided the model which
Muslim coins have generally, but not always, followed since.
9
The lack of pictorial imagery is also a striking characteristic of the
Dome of the Rock and other early Islamic religious and public
buildings. Opposition to the figural representation of human beings
and animals is a marked feature of the Muslim religious tradition (as it
is in Judaism), but this has not prevented a flourishing tradition of
representational art at a popular or private level where the influence of
the religious scholars was more remote. There are vigorous and even
beautiful representations of human beings and animals, for example,
in the lodges and palaces which the Umayyads built for themselves
outside the towns. How far opposition to this sort of pictorial
representation was a feature of early Islam, and the sources of Muslim
hostility to such sculpture and painting, are questions which have
received considerable discussion. There is some evidence to indicate
that the iconoclastic movement in Byzantium, which came to the fore
under Leo III (717–41), was in part a response to developments in the
Muslim world, and the caliph Yazid II (720–4) is known to have
undertaken attacks on the images and statues of his Christian subjects
in Syria. (For further details, see O.Grabar, 
The formation of Islamic
art,
 75–103; P.Crone, ‘Islam, Judeo-Christianity, and Byzantine
iconoclasm’; and G.R. D.King, ‘Islam, iconoclasm and the
declaration of doctrine’, 
BSOAS,
 48 (1985).)
Taken together, the innovations of the early Marwanid period in the
field of administration and coinage help to strengthen the impression
of an administration becoming more centralised and uniform.
Furthermore, they add to the evidence provided by the new
monumental buildings—not only the Dome of the Rock but also the
mosque of the Prophet in Medina and the mosque in Damascus which
incorporated the former church of St John, both built by al-Walid
10
 —
of the emergence of a new and distinctive Arab Muslim state and
culture from what had begun as, in some ways, a Byzantine or Sasanid
successor state.

---

66
‘Abd al-Malik and al-Hajjaj