Because Bell had met Koldewey during her visit to Babylon in 1911, it is conceivable that the subject of the Neo-Hittite bit hilani came up in their talks. Bell's diary entry from her visit to Assur in 1911 reveals, however, that it was probably Walter Andrae, Assur's excavation director, who first drew her attention to the bit hilani scheme and to Koldewey's discussion of its origins and development.188 The notion that this form passed into Achaemenid architecture seems to have found the support of Ernst Herzfeld, with whom Bell had an active correspondence at the time of her research on Ukhaidir and whom she cites as having suggested that the bit hilani was transmitted to the Achaemenids through Media.189
Andrae appears again as the brainchild for the next stage in the evolution of the iwan, as outlined by Bell in her chapter. This development appears in the art and architecture of the Parthians, and it is manifested clearly at Hatra, another site that Andrae had excavated and which Bell herself had taken the time to visit in 1911 (as described above). The spread of Hellenism and Roman expansion into the Near East brought Classical artistic conceptions to the art and architecture of the Parthians. Thus, Parthian buildings often bear Ionic-style columns and capitals, and Greek-inspired geometric mosaics, plaster ornaments and stucco wall fragments, not to mention Greek architectural units, like the square peristyle court.190 As Bell explains, however, certain architectural features in this period continue to bear a Near Eastern imprint, and nowhere is this continuity more apparent than with the iwan arrangement, which she perceives as the Parthian interpretation of the bit hilani scheme.191 In Parthian architecture, the portico and audience chamber are transformed into one room, the iwan. It became a longitudinal rectangular chamber, walled on three sides, with the fourth distinguished by an arched opening taking up most or all of its width.192 According to Bell, the columns of the former bit hilani portico simply decorate the walls that flank the archway of the iwan.193 Most significantly, the Parthian iwan became roofed with a barrel vault. The original scheme of the vault was a Mesopotamian innovation executed in mudbrick, which can be traced back, for example, to Assyrian palace corridors and gateways.194 Nevertheless, by the time the vault was incorporated into the Parthian architecture at Hatra in the first century CE, it had passed from Greece and Rome to the west, where it had been transformed from brick into stone.195 Moreover, it was from the west that the vaulted room was elevated from its place in minor side passageways and shallow chambers to its use in regal reception halls, given its ability to emphasize the room's height and longitudinal axis.196 At Hatra, Bell illustrates the Parthian vaulted iwan arrangement in the central Temple of the Great Iwans, believed in her time to be a royal palace. The building was characterized by two large central iwans, 21 metres in width, each of which was roofed with barrel vaults and flanked by a row of smaller iwans (Fig. 5.18).197
The next phase of the evolution of the eastern palace takes Bell to Persia, where she begins her investigation of the architecture of the Sasanians, whom she argues had adopted the iwan arrangement from the Parthians or the Achaemenids for their own palatial buildings. At Ardashir's third-century palace at Firuzabad – the earliest of the buildings from the Sasanian period known in Bell's time – one finds a deep, vaulted iwan, providing access to a domed audience hall in behind.198 To counteract the main thrust of the vaulted iwan, the side chambers were placed at right angles to it instead of running parallel, as at Parthian Hatra, for example.199 Bell also describes the architecture of the well-preserved building at Sarvistan, believed to date to the fifth century CE, with an arched-entrance iwan that provided access to a domed audience hall behind.200 Turning to the sixth-century Palace of Khosrow at Qasr-i-Shirin, Bell draws attention to the central part of the palace, with its large, open court in front and its monumental porch (Room 1). The iwan took the form of a closed antechamber (Room 2), this leading to a lavish hall of audience in behind with its own deep iwan (Rooms 3 and 4).201 This was the latest realization of the bit hilani, already recognizable at earlier Firuzabad, although less magnificently, and providing the general scheme that would be adopted by the early Islamic architects in their planning of Ukhaidir's ceremonial core, with its open court, iwan and square hall of audience in behind. As has already been noted, Bell was also struck by the so-called iwan groups in the Palace of Khosrow, these taking on the same appearance and arrangement within the palace as the baits within the palace of Ukhaidir.202 To Bell, Ukhaidir was closely linked to this remarkable palace at Qasr-i-Shirin in many ways, the latter providing much of the inspiration for her later Islamic complex. She must have felt that her efforts to visit this place in person and to carefully plan it were among her most worthwhile endeavours.
In Iraq itself, the iwan also appeared to have been adopted into elite architecture during the Sasanian period. Bell was particularly intrigued by reports of the desert hiras of the Lakhmid princes, Arab allies of the Sasanians, who had lived in the Mesopotamian desert between the third and seventh centuries.203 Indeed, it was the mention of these elusive palaces, where princes could escape the confines of their urban courts and return to the simpler lifestyles of their desert nomadic forebears,204 that had drawn her to the region west of the Euphrates River in the first place and led to her discovery of Ukhaidir. Although none of these Lakhmid hira complexes was very well known in Bell's time, having never been systematically explored – or even positively located, in some cases – later Islamic historians had written about them, describing them as palaces comprising a central audience chamber for the king (the centre or ‘breast’) and two wings that lay to the right and left, in which were lodged the king's courtiers and the special stores of the royal wardrobe and wine.205 Bell believed she could see some resemblance of this layout at Ukhaidir, with its central iwan being the principal audience chamber for the princely occupant, flanked by private apartments. An additional and tantalizing parallel could also be found within the early Islamic palace of Balkuwara at Samarra, excavated by Ernst Herzfeld in 1911, with its central block of monumental gates, cruciform domed reception hall and opposing iwans, flanked by wings on either side for residential quarters, storage facilities, parade grounds and stables.206 Both cases further bolstered the notion that important architectural features in early Islamic palatial architecture had earlier Sasanian-period antecedents, including those that had developed in Mesopotamia.
Finally, Bell could not ignore the monumental use of the iwan in the Sasanian Taq-i Kisra at Ctesiphon in central Iraq. The massive iwan, open to its full width on one side, occupied pride of place in the centre of the palace, where it served as the king's ceremonial audience hall. It was flanked on either side by five vaulted chambers and was covered by an enormous pitched-brick barrel vault, the largest of its kind in any pre-modern brick building. While this large audience hall is not particularly akin to the comparatively modest scale of the iwan at Ukhaidir, it could still be regarded as having developed from earlier palatial schemes with their open-ended audience halls, realized now through a massive vaulted roof and vast space underneath.
Overall, in describing these architectural examples from the pre-Islamic periods, Bell effectively underlined the strength of Ukhaidir's pre-Islamic influences, these having emanated largely from ancient Mesopotamia and Persia. Presumably, these earlier constructions, with their frequent – and often lavish – use of the iwan within palatial contexts, would have been well known to the architects of the early Islamic period who were constructing their own sumptuous palaces, such as Ukhaidir, in the same regions. Much of Ukhaidir's distinctive interior arrangement, and the centrality of its own iwan, then, was clearly situated within a long line of palatial schemes firmly at home within eastern traditions.
Commentary on Bell's Scholarship on the Development of the Iwan
Bell's tracing of the iwan back to its Hittite origins, and her following it through its Assyrian, Achaemenid, Parthian and Sasanian manifestations, was an ambitious endeavour, and few scholars today would carry out such a bold scheme, given the span of
centuries, the cultural groups and the morphological transformations undergone by this particular type of reception hall. Particularly problematic today is the place of the palaces at Sarvistan and Qasr-i-Shirin in her ambitious evolutionary scheme. Bell regarded these as representing notable Persian Sasanian antecedents that inspired later Islamic architecture, as at Ukhaidir. Recent scholarship has argued convincingly, however, that these edifices are probably not Sasanian at all, but rather are early Islamic. Thus, the points of similarity among these structures can be explained by their contemporaneity. On the one hand, Bell may be credited for properly recognizing these parallels, but on the other hand, because of their revised date, her evolutionary scheme, in which Sasanian architecture played such an important role in the transmission of architectural principles, is weakened.
Fig. 5.18 Andrae’s reconstruction of the Temple of the Great Iwans at Hatra, Parthian Era, highlighting the open-ended North and South Iwans; these were believed to have inspired similarly planned reception halls in the later Sasanian and Islamic periods, and they are well represented in the palace at Ukhaidir.
Most scholars today, nevertheless, would agree that the Islamic iwan does hearken back to Parthian times, where it was well known at Hatra in the first century CE but also recognized in, among other places, the Parthian palace complex at Assur and the north quadrant of the Parthian fortress at the site of Nippur.207 Interesting, the latter complexes are characterized by a four-iwan arrangement, with the halls grouped around a central courtyard.208
A continuing debate concerns, however, the origins of the iwan. Some argue that its open-ended form with mudbrick barrel vault was an eastern take on the Hellenistic flat-roofed, columned portico, this replacement most readily evidenced in first-century architecture at the Mesopotamian site of Seleucia-on-the-Tigris, where Parthian exchanges with the cultures of Greece and Rome were particularly strong.209 Alternatively, it might have been an adaptation of the Roman house's tablinum in the architecture of Iran and Mesopotamia.210 Still others have preferred an entirely Eastern inspiration for the iwan, positing Iranian origins or even venturing that it was an adaptation of the huts of the marsh inhabitants of southern Mesopotamia, in which the barrel-like roofs were built out of bent bundles of reeds covered with matting.211
Few scholars of Parthian, Sasanian and Islamic antiquity today seem to accept that the Neo-Hittite bit hilani lies at the root of the much later iwan, although interestingly, this notion has had its proponents. F. Oelmann took up Koldewey's idea in a lengthy article in 1922.212 Reuther, in his discussion of the existence of the iwan in Parthian architecture, refers back to Oelmann's article and treats as a possibility the resemblance between the iwan and an element of the Hittite palaces at Zinjirli.213 In his discussion of Ukhaidir in the more recent past, R. Hillenbrand proposes that in addition to the palace's Syrian Umayyad character, it is overlain with features rooted in native Mesopotamia, like the bit hilani of Syro-Hittite temple architecture, although Hillenbrand does not pursue this identification further.214
In defining what is meant precisely by the Neo-Hittite bit hilani and its adoption by Neo-Assyrian kings in their palaces, the Near Eastern art historian Irene Winter considers the possibility that the iwan might be the latest manifestation of this ancient form. Particularly compelling to her is the apparent multifaceted nature of the bit hilani, whose form in Neo-Hittite and Neo-Assyrian contexts was associated with a gate complex, a palatial reception hall or a private suite, much like the later Parthian, Sasanian and Islamic iwans functioned within this variety of contexts.215 She finds further tantalizing the fact that the Neo-Hittite bit hilani suites of rooms are sometimes grouped around a central courtyard, as at the site of Zinjirli or in the Neo-Assyrian palace of Sennacherib at Nineveh, just as one finds three or four iwans arranged around a central courtyard in various building complexes in the later periods.216 While incomplete evidence for a direct continuity of the bit hilani to the iwan over such a long period of time prevents us from confirming such a relationship between the two forms, Winter's compelling parallels do urge one to consider more seriously the impact of architectural forms from the pre-Parthian-Sasanian and pre-Islamic Near Eastern worlds into the later periods.217 Moreover, her observations, which essentially fall in line with Bell's own arguments concerning the origins of the iwan, discourage one from quickly dismissing Bell's bold scheme.
While the development of the palatial iwan constituted the principal architectural feature investigated by Bell in her lengthy and complicated chapter on the genesis of the early Islamic palace, she did not ignore other architectural elements and cultural influences that made their way into palaces like Ukhaidir. Of particular note is her consideration of Ukhaidir's fortified exterior appearance, with its high walls and rounded towers, and her argument that such defensive architecture could ultimately be traced back to the fortified camps established by the Romans along their Arabian desert frontier or limes.218 She consequently showed how these fortified camps provided the essential blueprint for the defences of elite desert residences of the early Islamic Umayyad period (660–750 BCE). Bell particularly emphasized two Umayyad castles in present-day Jordan with which she was most familiar, Qasr Kharana (Bell's Kharaneh) – which she would end up visiting and recording in 1914 (Fig. 5.19)219 – and Mshatta, lying only a few more kilometres to the west of Qasr Kharana, in the western desert (Fig. 5.20).220 Both of these castles’ defensive character, characterized by their high walls and rounded towers, recalled the early Roman forts and at the same time provided direct inspiration for the fortified appearance of the slightly later Abbasid castle at Ukhaidir, on the eastern side of the Syrian desert.
Through her discussion of these features and their diverse origins, Bell was ultimately emphasizing the unique, hybrid character of early Islamic architecture. While the interior arrangements of palaces such as Ukhaidir had clearly been affected by traditions emanating from the East – with their central iwan schemes and flanking apartments, not to mention some of their technological elements and materials, which were born of local traditions – other architectural features could often be traced back to Rome and the West. The key to early Islamic architecture, therefore, was understanding its unique blending of Eastern and Western traditions. Bell's research, so comprehensively reported in Palace and Mosque at Ukhaidir, astutely traced the multidirectional nature of Ukhaidir's influences. With such observations, Bell moved beyond the more simplistic assertions of scholars such as Strzygowski, who with their polemicizing stance were determined to locate and isolate only one vital source of inspiration that defined the essence of an artistic or architectural monument, be it from the East or the West. Bell, in her scholarly maturity, accepted the complexity with which ideas and influences had been exchanged and intermingled in the early years of Islam, when old traditions were combined with new elements to give rise to a novel and distinctive cultural style.221
Palace and Mosque at Ukhaidir: Flutter in the Dovecotes?
Palace and Mosque at Ukhaidir, the work into which Bell poured all of her archaeological field investigations, her correspondences and discussions with other scholars, and her own extensive research, was lavishly published by Oxford's Clarendon Press in 1914. It featured oversized pages, fold-out plans, maps and an abundance of cleanly reproduced black-and-white photographs. The book's sumptuous form was an appropriate medium for this ambitious project, with its detailed, well-illustrated treatment of Ukhaidir, not to mention its review of all of the architectural forms, through the ages, that had provided the inspiration for Ukhaidir's palatial layout and mosque.
At long last, Bell had completed her most ambitious and complicated work that had continued to absorb her attentions from the time she had first laid eyes on the spectacular castle of Ukhaidir in early 1909. But did it ultimately meet her expectations as a researcher and scholar? When she had first excitedly announced her discovery of Ukhaidir back in 1909, she thought she had found the most ‘important building of its period’ and vo
wed that she would ‘publish it in a big monograph all to itself’ that would make ‘a flutter in the dovecotes’. But in the end, was this the significant scholarly contribution she had striven for, and did it achieve the recognition for which she might have hoped?
A definitive ‘yes’ or ‘no’ cannot be offered. The reception of Bell's work was very mixed and remains so. Many of the reviews that appeared back in 1914 were not overly effusive in their praise. While most were clearly impressed with Bell's scholarship, many disliked the ponderous writing style, and her adoption of what one called the ‘German method of throwing crude note-books undigested at the head of your readers’.222 It must be admitted that ‘few will take the pains to wade beyond the first ten pages’,223 as Bell is ‘throughout severely technical’.224 As was rightly observed, if one is looking forward to the ‘romance of travel’, with the ‘vivid descriptions of manners and records of Eastern chat which made “Desert and the Sown” so fascinating’, one will be disappointed by the dense, scholarly content of this monograph. Even still, it must be admitted that the patient reader will find ‘a great harvest of information’.225
Such drawbacks aside, it is impossible to ignore the prodigious learning that Bell's book evinces, particularly its comparative materials, which Massignon's and Reuther's earlier reports of Ukhaidir do not present so exhaustively. Bell was also praised for her ability to support her arguments with ‘a solid mass of monumental evidence’.226 Few scholars criticized the content of the work, the exception being Marcel Dieulafoy, who in a lengthy review still disagreed with Bell's proposed Islamic date for the palace, believing that her identification of a mosque at Ukhaidir was unconvincing; he remained firm in his belief in the pre-Islamic date of the complex.227 But even he admired Bell's clear writing, her rich evidentiary documentation and the abundance of comparative sources she had been able to collect.228
Britain and the Arab Middle East Page 25