Ken Dark and Jan Kostenec
The Hagia Sophia Project, Istanbul: report on the 2009 season

The Hagia Sophia project, co-directed by Ken Dark and Jan Kostenec, aims at studying the church of Hagia Sophia as a cathedral complex. As such, this archaeological project aims to investigate not only on the impressive - and, of course, already extensively studied - church structure, but also the associated Byzantine buildings that once stood around it. A summary of the results of our work undertaken between 2004 and 2008 has been published in BBBS 351.

As in the previous seasons, we were allowed to examine the entire surroundings of the Byzantine church (within the boundaries of the Ayasofya Müzesi) and most of the, usually inaccessible, areas within the building itself (Figure 1). The purpose of this brief report is to present only highlights of the 2009 season, rather than a complete catalogue of all of the information recorded by us this year. Here, for the sake of brevity, initials are used to stand for points of the compass (eg. S = south, N = north), and C followed by a figure for centuries AD (eg. C6 = sixth century). We use ‘Hagia Sophia’ for the whole area within present Ayasofya Müzesi and ‘the church’ to refer to the Justinianic building, unless otherwise qualified.

In 2009 we continued our survey of the so-called ‘Baptistery’ and the adjacent SW outer buttress, and of the SW access ramp of the church. We examined the NE access ramp, the Skeuophylakion, and the area above the hypogeum (immediately N of the N middle door of the church). Among the ‘new’ material from these areas recorded in 2009, was previously unpublished evidence for the Byzantine-period decoration in the SW outer buttress. Fragments of a, hitherto unrecorded, decorative painting cover the vaulting of the ground-floor space in the E part of the buttress. This shows a regular pattern consisting of squares, quatrefoils and stylized leaves, painted with a limited colour palette (olive green, purple and orange)2. In addition, we made further advances in the study of the (poorly preserved) frescoes in the rectangular room on the top of the buttress. The third (and latest) layer of frescoes in this room can now be dated to the C12 on stylistic grounds - especially on the basis of the dynamic style of drapery and expressive rendering of faces of the saints. Moreover, a fragment of the central figure on the E wall of the room may be identified as enthroned Christ holding a Gospel book in His left hand while blessing with His right.  The position of Christ over the large round-headed opening in the E wall suggests that the fresco room might have served as an antechamber or narthex for the quatrefoil chapel to the E3.  This evidence may assist in understanding how the butress functioned as part of the church, suggesting that this area had a more active role in the liturgical life of the building, at least during the Middle Byzantine period, than the term  ‘buttress’ usually implies.

Figure 1: General plan of Hagia Sophia (after W. Müller-Wiener, Bildlexikon zur Topographie Istanbuls (Tübingen 1977) 90, figure 75), showing the areas surveyed in 2004-9.

Nevertheless, the most important material recorded in 2009 is from the rooms at the SW corner of the church. This area may be reasonably identified as part of the patriarchal palace, and included a large vaulted hall (Figure 2)4.  To the N of the hall (the ‘Large Hall’) there is another high, barrel vaulted, space. This latter room has been identified as the Byzantine-period Horologion in most previous studies of Hagia Sophia (Figures 1 and 2)5.  However, the Large Hall renders it more plausible that spaces formed originally one architectural unit, with the preserved high-barrel vaulted room being an antechamber to the Large Hall, into which it opened through a high triple arcade. We propose, therefore, to re-name this barrel-vaulted space the ‘Antechamber’, and that the term ‘Horologion’ is no longer used for it. The triple arcade may have been a monumentalised entrance to the Large Hall from the church. 

Figure 2: Remains of the Byzantine patriarchal palace at the SW corner of the church, view from SW: 1 – E wall of the Large Hall, 2 – N wall of the Large Hall, showing the blocked triple arcade originally opening to the Antechamber N of it, 3 – rooms with water tanks for the Ottoman ablution fountains, 4 – long vaulted room above the atrium.

Pure brick masonry with occasional levelling courses of greenstone suggests a broadly C6 date for both the Antechamber and Large Hall. However, it may be possible to be more precise about their dating: several in situ Byzantine bricks (visible in three holes in the N wall of the Antechamber) bear two-line stamps reading: +KONCTANS and +KOCTATINOY. Bricks with the same stamps have been recorded by earlier scholars in Hagia Sophia, including in the church itself, and in the 1940s excavations S of Hagia Eirene. These stamps may have been associated with Justinian’s rebuilding of the city- centre of Constantinople after the Nika riot in 5326.  The dimensions of the bricks used in the Antechamber (36-38 x 4.5-5.6cm) are the same as those in the original structure of the neighbouring Justinianic church, and it is, therefore, possible that the Antechamber and the Large Hall were both constructed by Justinian. This Justinianic dating may be supported by the presence of masons’ marks O? and KOZ on marble doorframes inserted into the arched opening through which the Antechamber communicates with a small room leading to the SW porch of the church. Numerous examples of such masons’ marks are known from inside the church7.

The Justinianic dating for the Large Hall and its Antechamber presented here, revises our previous dating (first published in Architectura in 2006) for the Large Hall. Although the archaeological interpretation of the structure as probably a reception-hall or purpose-built meeting place is unaltered, this re-opens the question of its identification among written records for sixth-century Hagia Sophia. One possibility may be the anonymous hall in the Patriarchate where 168 bishops gathered during the Second Council of Constantinople in 5538.

However, it must be remembered that the date of the bricks provides only a Terminus Post Quem for the date of the Large Hall. Thus, it remains possible (although perhaps in this context much less likely) that - if the bricks of the Antechamber came from old stock or were reused from a demolished Justinianic structure - the Large Hall and the Antechamber could be associated with the textually-attested work of the patriarch John III in the patriarchal palace9.

In addition, this year we examined in detail a cross-vaulted substructure beneath the Antechamber (Figure 4) Despite being known since the 1940s, this has been published only briefly by Dirimtekin in the 1960s - when he identified it as part of the convent of St. Olympias, built when St. John Chrysostom was the patriarch of Constantinople (398-404)10.  However, our investigation in 2009 revealed that the substructure is not earlier but later than the C6 Antechamber! It appears that after the Large Hall had collapsed, or was damaged, its triple arcade was bricked up (and the piers or columns of the arcade removed) and the cross-vaulted substructure was built inside the Antechamber, raising its floor level by about 1.5m (Figure 3). This plainly indicates an alteration in the function of this part of the building within the Byzantine period.

Figure 3: Antechamber, view from W. In the foreground: modern stairs and access to the substructure beneath the Antechamber. In the background: marble doorframes in the E arched opening.

Figure 4: Vaulted substructure beneath the Antechamber, view from SE.

The S substructure’s wall was built of large C6 bricks (resembling those in the Antechamber and elsewhere in the church) as high as a course of stone, which seems to have originally served as a foundation for the triple arcade of the Antechamber. However, the upper part of the same wall (above the course of stone) and also the cross-vaults employed much smaller bricks (mostly 33.5-35 x 4-4.5 cm). These smaller bricks may suggest a structure later than the C611.  This clearly shows that the brickwork of the wall above the stone course is part of the existing S wall of the Antechamber which replaced the original triple arcade, and that the vaulted substructure postdates the Antechamber. 

The observation that the substructure is later than the Antechamber would also account for the awkward manner in which marble doorframes were inserted into the E arched opening of the Antechamber. The marble door in question was certainly re-used: it was stripped of its massive moulded lintel and threshold in order to fit the dimensions of the arched opening. However, if the door was originally positioned on the original level of the floor of the Antechamber it would have fitted well into the available space.

Two stone piers topped with impost blocks carrying the vaulting of the substructure are probably C6 spolia, bearing the masons’ marks ?I, KOZ and O? (in ligature). These masons’ marks are attested in Justinian’s church and elsewhere in C6 contexts12.  It is possible that the impost blocks in the substructure were originally employed in the triple arcade leading from the Antechamber to the Large Hall.  If so, this would support the interpretations both that the substructure post-dated the Large Hall, and that it re-used C6 building materials.
 
A previously (poorly) published fresco decoration in the small room to the E of the Antechamber may also date to the time when this part of the patriarchal palace was modified. This is less well-preserved than when photographed by Ramazanoğlu in the 1950s. The ivy leaves that it once showed have almost disappeared, but the imitation marble revetment in the soffit of the E arch survives13.

To the N of the Antechamber are smaller C6 rooms on two levels. These were partly demolished when the minaret was built and water tanks for the Ottoman ablution fountains constructed14.  However, there is a well-preserved long vaulted room above the modified E end of the S wing of the C6 atrium of the church, and examining this enabled us to record further evidence for the Byzantine building (Figures 1, 2 and 5)15.  Although the long room appears to have been built in C6, like the other areas mentioned above it was later extensively modified. The S wall of the room is of double thickness – its outer part built of typical C6 brick (greatest length c.38 cm); its inner part of smaller bricks (mostly 32-34 x 3.5-4 cm). In view of this, the room’s vaulting and its secondary (raised) floor level may be associated with post-C6 rebuilding of this part of the patriarchal palace, as already identified in relation to the Antechamber. This may suggest that this whole area of Hagia Sophia was re-modelled when the C6 Large Hall and its antechamber went out of use.

Figure 5: Long vaulted room above the E end of the S wing of the atrium, view from W.

Thus, work in 2009 clarified further the architecture, plan and decoration of the patriarchal palace. It provided helpful dating evidence for the surviving features, especially for the Large Hall - which may have been, at least partly, contemporary with the Justinianic church, rather than later in date.  Evidence recorded this year suggests re-design of the south-west of Hagia Sophia after the Large Hall was disused, and this may have further implications for the re-use of other structures in this area, including those previously identified by us in relation to the so-called ‘Baptistery’ and South West Porch. One possibility is that these areas were re-modelled at the same time, perhaps contemporary with the insertion of the famous C10 mosaic above the porch to monumentalise a new principal passageway between the Patriarchate and church after the disuse of the C6 triple arcade and Antechamber.    

Acknowledgements
The co-directors of the project would like to thank the Turkish Ministry of Culture and Tourism for their permit to carry out the survey in Hagia Sophia and the director and staff of Ayasofya Müzesi for their invaluable help. Thanks are due to the Late Antiquity Research Group, Andrew W. Mellon Foundation and the American Research Institute in Turkey for their kind support, and to Dr T. Alušík, D. Tatar and A.E. Topaloğlu for their assistance during the survey. Dr Ida Toth (University of Oxford) provided assistance regarding the graffiti found in previous seasons.

Notes
1.  K.R. Dark and J. Kostenec, ‘The Hagia Sophia Project, Istanbul, 2004-8’, Bulletin of British Byzantine Studies 35 (2009) 56-68.
2.  Publication of photographs of the decorative fresco in black-and-white is impossible due to its bad state of preservation, so we are unable to show it here.
3.  The Enthroned Christ, executed in mosaic or fresco, was commonly depicted over the door leading from the narthex to the naos in the Middle and Late Byzantine period churches. In Istanbul, the best examples can be found in Hagia Sophia and the Chora (Kariye Camii).
4.  K.R. Dark and J. Kostenec, ‘The Byzantine Patriarchate in Constantinople and the Baptistery of the Church of Hagia Sophia’, Architectura 36 (2006) 113-130 (esp. 120-123); cf. Dark and Kostenec (as n. 1) 62.
5.  First proposed by A.M. Schneider, Die Grabung im Westhof der Sophienkirche zu Istanbul (Berlin 1941) 41-42, 44-45.
6.  See J. Bardill, Brickstamps of Constantinople (Oxford 2004) 2 vols., I, 64-67, 116 and the catalogue, 259 no. 663, 268 no. 675.1-2, 269f. no. 676.1-6, 272 no. 685.
7.  A. Paribeni, ‘Le sigle dei marmorari e l’organizzazione del cantiere’ in A.G. Guidobaldi and C. Barsanti, eds., Santa Sofia di Constantinopoli. L’arredo marmoreo della Grande Chiesa giustinianea (Vatican 2004)653-734, esp. 724 nos. 51.b, 51.c, 51.j, 51.k., 51.o. (KOZ) and 728 no. 69.d. (O?)
8.  G.D. Mansi, ed., Sacrorum conciliorum nova et amplissima collectio IX (Florence 1763) 173.
9.  According to John of Ephesus, the Patriarchate was reconstructed in a ‘magnificent manner’ by the patriarch John III Scholasticus during the reign of Justin II – see: E.W. Brooks (ed. and trans.), Iohannis Ephesini Historiae Ecclesiasticae, Pars Tertia, Corpus Scriptorum Christianorum Orientalium, Scriptores Syri, ser. 3, iii (Louvain 1935-36) 3.2.31; R. Payne Smith, The Third Part of the Ecclesiastical History of John, Bishop of Ephesus (Oxford 1860) 145.
10.  F. Dirimtekin, ‘The Olympiados Convent’, Ayasofya Müzesi Yıllıği 4 (1962) 47-8.
11.  On the decline of the dimensions of Constantinopolitan bricks after Justinian’s reign, see Bardill (as n. 6) 106; W.J. Mitchell, B. Aran and R.S. Liggett, ‘The application of statistical pattern-recognition techniques in dating ancient architectural constructions’, Environment and Planning B 9 (1982) 467-490; V. Ruggieri, Byzantine Religious Architecture (582-867): Its History and Structural Elements (Rome 1991) 153-161.
12.  Paribeni (as n. 7) 724 nos. 51.b, 51.c, 51.j, 51.k., 51.o. (KOZ),  723 nos. 45 and 45.a (?I and I?); L. Butler, The Nave Cornices of Hagia Sophia in Istanbul (diss. University of Pennsylvania 1989) 150 (?I and I?); J. Crow, J. Bardill and R. Bayliss,  Water Supply of Byzantine Constantinople (Journal of Roman Studies Monograph 11, London 2008) 200 (O? in ligature).
13.  M. Ramazanoğlu, ‘Die Baugeschichte der Sophien-Kirche Justinians’ in Atti dello VIII Congresso internazionale di studi bizantini. Palermo, 3-10 aprile 1951 (Rome 1953) II, 224-231 (esp. 225-6), pl. LIII figs. 7 and 8.
14.  See Dark and Kostenec (as n. 4) 123; F. Dirimtekin, ‘Le Local de Patriarcat à Sainte Sophie’, Istanbuler Mitteilungen 13/14 (1963/4) 113-127 (esp. 119-123).
15.  Schneider (as n. 5) 40-41; F. Dirimtekin, ‘The room on the south-west of the atrium’, Ayasofya Müzesi Yıllıği 2 (1960) 39-41; Dirimtekin (as n. 14) 123.