385

BANGANARTISUDAN

BANGANARTITHE 2006 SEASON

Bogdan ¯urawski

The Banganarti expedition completed a season of fieldwork at the site between 17 Januaryand 20 March 2006.1

The excavation and clearing work, which proceeded in five main areas [cf. Fig. 1] wasparalleled by a restoration program focused on converting the Upper Church into a sitemuseum (the official opening is scheduled for 2009). This involved extensive cleaning andconservation, as well as building restoration activities. A number of discoveries of previouslyunknown murals and inscriptions were made in consequence.

Specialist study programs this season included a geodetic survey of the neighborhood ofBanganarti; calibration of the original Upper Church plan; aerial (kite) photography;documentation and revision of epigraphic material; iconological studies; potterydocumentation and studies (see report by D. Bagiñska in this volume); anthropological andarcheobotanical examination.

An ethnographic survey of the site's environs carried out the author, brought a collectionof oral testimony on the most recent history of the region, especially the early 20th centuryFunj/Shaiqiya settlers.

1 Banganarti mission staff in the 2006 season: Bogdan ¯urawski, Project Director; Adam £ajtar, epigraphist; Tomasz

P³óciennik, epigraphist; Magdalena £aptaœ, iconologist; Magdalena WoŸniak, archaeologist/iconologist; Anna B³asz-

czyk, Mariola Orzechowska, Agata Rak, archaeologists and draftspersons; Dobies³awa Bagiñska, archeologist/ceram-

ologist; Marta Momot, draftsperson; Mariusz Drzewiecki, Ada Oleœ-Niedzielska, archaeology students; Karol Piasecki,

physical anthropologist; Lisa Hildebrandt, palaeobotanist; Tadeusz Badowski, restorer; Dorota Moryto-Naumiuk,

restorer; Ryszard Szemraj, technical assistant, restorer and building engineering supervisor. The geodesic team

comprised Roman £opaciuk, Wies³aw Ma³kowski and £ukasz Moczulski. The NCAM was represented by senior

inspectors Ayasha and Fathiya Abder Rahman.

BANGANARTI

386

SUDAN

Fig. 2. Aerial view of Banganarti site after the 2006 season(Orthophotomap B. ¯urawski)

Fig. 1. Banganarti after the 2006 season, marking location of areas of excavation(Drawing R. £opaciuk)

387

BANGANARTISUDAN

Excavation and surface clearing proceededin five main areas [Figs 1, 2]:

– I and III covering the church complexand Western Building, where some ad-ditional clearing and documentation workwas completed;

– II, following the section of theenclosure wall northwest of the church,from about the middle of its length to thenorthwestern corner including the cornertower;

– IV, habitation district inside theenclosure wall to the east (clearing the topsof walls for the sake of kite photography andgeodetic surveying);

– V, outside the girdle wall, north of thesoutheastern corner tower (planned as a pot-tery trench for stratigraphic study)

AREAS I AND IIIThe late blockages between the supports ofthe Western Portico were pulled down[Fig. 3], exposing two longer Greekinscriptions of a dedicatory nature and anabundance of graffiti, as well as scratched –some not without artistic mastery –representations of animal and humanfigures. Exceptional among these drawingsis a portrait scratched on the southern face ofthe western portico pillar, executedpresumably at some time in the 12thcentury. It depicts a man in profile, feelinghis way with a short, pastoral staff (see Fig. 1on p. 399 and A. £ajtar's report in thisvolume). A graffito in Arabic, but writtenin Greek letters, identifies this figure as Ali,a def or guest, meaning visitor to the church.

EXCAVATION AND CLEARING WORK

Fig. 3. Reconstruction drawing of the original appearance of the main western façade of theRaphaelion (=Upper Church) (Interpretation B. ¯urawski, drawing M. Momot)

Ali must have been a Muslim drawn to theChristian sanctuary by its thaumathurgicalrenown, not unlikely in the hope of findinga cure for his illness.

Three trial pits were dug from theground level of the Upper Church, cuttingthrough the Lower Church deposits down toculturally sterile sand (which has beentested for at least 0.50 m for the sake of

certainty) [Fig. 4] The first was a northwardextension of a trial pit dug already in 2004(¯urawski 2005: 300) in front of theblocked door to the prothesis of the LowerChurch. It now exposed the east wall of thenorthern aisle and part of the north wall towhere the northern extension begins. Fainttraces of painted decoration (unidentifiable,possibly standing figures?) in yellow ocher,

BANGANARTI

388

SUDAN

Fig. 4. Outline of the Lower Church marked on a plan of the Upper Church, indicating position oftrial pits dug in 2006 (Drawing R. £opaciuk)

389

BANGANARTISUDAN

applied generously on a weak lime/mudplaster ground were detected on the latterwall. The plastering was treated and the testpit walls reinforced with corrugated-ironsheets and piping in order to displaya fragment of Nubian history in a naturalstratigraphical setting.

Overlying the sterile sand at the bottomof the trial pit was a layer of broken 7th-century amphorae and jars with flaringnecks, possibly imported from Alwa (seebelow, Banganarti pottery report byS. Bagiñska in this volume). This layer isa stratigraphic unit that has been found atmatching depth in all of the test pitsexcavated within the walls of the LowerChurch. The sherds must have accumulatedconcurrently with the existence of the firstLower Church and were used as a levelinglayer under the second church when it wasbuilt on top of the earlier walls (replacingamong others the original stone columnswith masonry piers). One theory for sucha magnitude of broken Early Christianamphorae being found in the deepest layersis that water for the baptismal ceremonieswhich took place in the baptistery of theLower Church I (see below) may have beenbrought in earthenware vessels, which weredumped afterwards outside the church walls.

The second trial pit was dug in thesouthern part of the haikal area of theLower Church, exposing also the southernhalf of the apse. On the west, the testingreached the southeastern pier of the LowerChurch and exposed the southeasterncorner of the southern aisle on the south,thus joining the 2005 trial pit (S.3) inwhich the Anastasis mural was found(¯urawski 2007: 309-312). The excavationrevealed evidence of brick foundationsunder the stone walls of the Lower Churchand a specific manner of construction of thefour central piers on a substructure ofdismantled stone voussoirs originating

from the original columns of the LC 1building. A similar technique (apparentlyof Meroitic origin as the Ammon temple inBegrawiya might suggest, was encounteredalso in an early church on Kom E inDongola, near el-Ghaddar (¯urawski 1995;1997: 182). The stones turned out to bearmurals, some of the earliest ever recorded inUpper Nubia (painted probably shortlyafter the first of the Banganarti churcheswas built). The wall paintings wereexecuted on a hard lime mortar backgroundwith extensive use of purple, white andyellow [Fig. 5]. An examination of a dozenor so painted voussoirs identified equestriansaints and archangels among the subjects.

Traces of a mural were also discovered onthe east wall of the southern aisle, to theright of the apse and left of a door leading tothe southern sacristy (diakonikon) in theLower Church. It was an illusionary ren-dering of ashlar masonry in purple paint,applied on a lime-plastered stone wall[Fig. 6].

The last trial pit dug inside the southernsacristy and in the passage behind the apse ofthe Lower Church exposed a baptistery[Fig. 7]. The lower one of the two masonrybasins, a typical Nubian “keyhole” fontprovided with two approaches steppingdown into the basin from east and west, wasfound with no base at all. Baptism musthave been accomplished by aspersion ratherthan immersion and the baptismal waterpoured over neophytes standing on a layer ofpure sand was inevitably soaked into it. Thistank was in use when the Lower Church wasstill covered with a flat wooden roofsupported on the four stone columnsdescribed above. A baptismal basin of redbrick bonded and plastered with a hydrauliclime mortar was built on the original floor inthe corner of the earlier baptistery [Fig. 8].The lower tank may have still been used atthis time.

BANGANARTI

390

SUDAN

Fig. 6. Imitation ashlar masonry painted on the east wall of the south aisle of the Lower Church(Photo B. ¯urawski)

Fig. 5. Holy rider (?) painted on lime plaster on a stone voussoir from a column belonging to thefirst Lower Church (Photo B. ¯urawski)

391

BANGANARTISUDAN

The baptistery was lavishly decoratedwith murals representing elongated, staticsaintly figures wearing slender pointedboots similar to the earliest portraits fromthe Faras cathedral and the Bawit chapels.As a point of interest, the smallestrepresentation – a figure of an archangel[cf. Fig. 9] – was found on the southern wallof the baptistery. Simultaneously, thefigures (of apostles?) on the east wall,reconstructed as at least 2.50 m high,belong among the biggest discovered so farin the church.

The eastern annex, entered from thepassage behind the apse, could have servedas a consignatorium (where confirmation withthe chrism took place as an integral part ofthe baptism). It has been extensively testedfor burials and other diagnostic features, butwith no result.

Fig. 8. Latitudinal section (looking south) through the baptistery; walls of the Lower Churchmarked in grey; on left, grave explored in 2003 (Drawing M. Momot & B. ¯urawski)

Fig. 7. Axonometric view of the two baptismalfonts (Drawing B. ¯urawski)

BANGANARTI

392

SUDAN

Following up on a comprehensiveinvestigation of the Western Building inArea III (¯urawski 2007: 304-307), archa-eological testing confirmed the existingassumptions concerning building phases inthis area. Excavations revealed an elaboratestaircase in the southern part of the WesternBuilding and drainage holes cut in the wallsfor removing excess water from the centralcourtyard and the rooms west of it.

AREA IIFieldwork on the site of the enclosure wallnorthwest of the church complex uncovereda stretch of the fortifications, running fromabout midpoint of this side to and includingthe corner tower. A 3 m wide trench wasdug (and a huge sand dune accumulatednorth of the wall removed), uncovering themud-brick structure of the wall, as well ashabitation units on either side, datedsecurely by a collection of 11th-12thcentury ceramics. A chronology of subse-

quent reconstructions and extensionsbeyond the walls was established, althoughwith gaps due to the heavily disturbedstratigraphy (for a discussion of thefortifications, see below, contribution byM. Drzewiecki in this volume). TheBanganarti fortifications have thus beenconfirmed as a peculiar form of combineddefensive and domestic function, the latterexpanding with time. In the 13th/14thcentury, the northern peribolos was turnedinto a living quarter, its defensive functiontotally forgotten. The huge corner towerwas dismantled and a square enclosure wasbuilt instead, its walls no wider that halfa meter.

A necklace of carnelian and rock crystalbeads [Fig. 10, see also Fig. 3 on p. 406] wasdiscovered west of the tower, already outsidethe enclosure wall.2 It was found next toa cooking pot with damaged bottom, a stonebasin and a fragmented ceramic toilet seat,in context with broken kitchen potterydated to the 11th /12th century.

2 Puzzling evidence of a "global" economy in this period in the form of a virtually indistinguishable rock crystal and

carnelian bead necklace from an 11th century grave in Dziekanowice in Great Poland, has been brought to the author's

attention by Jacek Wrzesiñski, to whom I am indebted for this information.

Fig. 9. Miniature painting of an archangel(Drawing B. ¯urawski, W. Chmiel& A. Rak)

Fig. 10. Carnelian and rock crystal beads froma necklace, 11th/ 12th century context (Photo B. ¯urawski)

393

BANGANARTISUDAN

AREA IVThe combined surface cleaning andsurveying of this area have uncovered andrecorded a group of mud-brick dwellingunits lining the enclosure wall on the insideof a huge semicircular feature, also of mudbrick, located on the outside of the wall[cf. Figs 1, 2]. The stratigraphy in this areais much disturbed due to modern maroq-digging and the finds from the surfacelayers are a mixed bag indeed.

AREA VDigging in the 'pocket' outside theenclosure wall, just west of the southeasterncorner tower, was expected to provide a pot-tery sequence covering a period from the 8thto the 11th century. The medieval practice ofdumping kitchen waste, ceramics included,outside the fortifications had already beentested as an informative source on ceramicstratigraphy and site chronology in earliertrenches of the kind dug in 2003 and 2005(¯urawski 2007: 307-309). The discovery ofa bread oven [Fig. 11] with contents intact,including grains, charcoal and kitchenwastes interrupted the work until flotationcould be done.

Fig. 11. Plan and section looking souththrough a bread oven found in Area V (Drawing M. Momot and B. ¯u-rawski)

A detailed contour plan of Banganarti andvicinity, now cultivated, was produced by ateam of surveyors using reflectorless total-station technology.3 It was integrated withthe previous documentation of thechurches, the rasterized traditionaldrawing of the Upper Church beingcalibrated with the digital layer by the 130references points measured inside thechurches. Correlation was completed using

AutoCad Map software with a RasterDesign extension module. The traditionaldrawing was imported to the digital planas an image (bitonal jpg with 300 dpiresolution) and scaled to the archaeologicalobject boundary, calculating total correlateerror. Each of the 130 points from thetraditional drawing was defined as a sourcepoint and redefined as a destination pointmeasured with a total station.

STUDY PROGRAMS

3 Remarks contributed by W. Ma³kowski.

BANGANARTI

394

SUDAN

Low-altitude photographs of thechurch and the fortifications were taken aspart of the aerial (kite) photographyprogram successfully implemented inBanganarti since 1998. Some 500 aerialphotographs were digitally assembled tocreate an orthophotomap of the site [cf.Fig. 2].

The pottery study program included,beside current documentation, also the

drawing, photographing and analysis ofceramics collected during previousMDASP operations (in 2004 and 2005) onSaffi Island and in Dar el-Arab.

Iconologists continued their study ofthe murals, including documentation ofnewly uncovered examples, e.g. the muralin Room 20 [Fig. 12]. The same was donefor the inscriptions, the epigraphic'revisiting' resulting in the uncovering of

Fig. 12.Three captives led on a rope by a horseman, next to a standing figure of a saint (?) on thewest wall of Room 20 of the Upper Church (Drawing M. Momot & B. ¯urawski)

395

BANGANARTISUDAN

REFERENCES

¯urawski, B.

1995 Old Dongola. The Mosaic Church (Kom E), 1993/94, PAM VI [=Reports 1994], 98-1081997 Early church on Kom E near Old Dongola. Excavations of 1993 and 1994 seasons,

CRIPEL 17/2, 181-1932005 Banganarti. 2004 season including activities at the fortress of ed-Deiga, PAM XVI

[=Reports 2004], 297-3082007 Banganarti 2004/2005, PAM XVII [=Reports 2005], 301-321

much new data, especially in the LowerChurch and on the columns and piers ofthe western portico of the Upper Church.The new inscriptions were ink copied andphotographed, and preliminary trans-lations provided (for a study of new finds,see below, contribution by A. £ajtar in thisvolume).

Studies and conservation of the muralsin the westernmost room of the Lower

Church with the mural of St Mercuriusspearing Julian Apostate were completed,following which the area was backfilledwith sand in order not to endanger the eastwall of the Western Building which partlyoverlaps the west wall of this room. TheMaria Orans mural was detached its placeon the wall and taken to Khartoum forconservation (for this work, see report byD. Moryto-Naumiuk in this volume).