Jabal Harun - Petra

Transliterated Name	Language	Name
Jabal Harun	Arabic	جابال هارون‎

Introduction

Jabal Harun (Mount Harun) is located ~5 km. southwest of the main site (cardo) of Petra and has traditionally been recognized by Muslims, Christians, and Jews as the place where Moses' brother Aaron was buried (Frosen et al, 2002). As such, it may have remained as an ecclesiastical and pilgrimage site after Petra's decline in the 7th century CE. About 150 m from the peak of Jabal Harun lies the remains of what is thought to have been a Byzantine monastery/pilgrimage center dedicated to Aaron.

Petra - Introduction Webpage

from Petra - Introduction - click link to open new tab

Maps, Aerial Views, Plans, and Sections

Maps

Fig. 1 - Location Map

Fig. 1

The Jabal Harun area

Kouki and Lavento (2013)

from Kouki and Lavento (2013)

Aerial Views

Jabal Harun in Google Earth

Jabal Harun in Google Earth

click on image to explore this site on a new tab in Google Earth
Fig. 1 - Aerial View after excavations

Figure 1

The FJHP site following the end of excavations in 2007 (by Z. T. Fiema).

Fiema (2013)

from Fiema (2013)

Plans

Site Plans

Normal Size

Fig. 2 - Plan of entire site

Fig. 2

Plan of the FJHP site

(by K. Koistenan and V. Putkonen)

Fiema (2013)

from Fiema (2013)
Fig. 2 - Plan of the monastery

Fig. 2 Chapter 5

The plan of the monastery following the 2005 field season.

Fiema (2008)

with walls and trench locations marked from Fiema (2008)

Magnified

Fig. 2 - Plan of

Fig. 2

Plan of the FJHP site

(by K. Koistenan and V. Putkonen)

Fiema (2013)

entire site from Fiema (2013)
Fig. 2 - Plan of

Fig. 2 Chapter 5

The plan of the monastery following the 2005 field season.

Fiema (2008)

the monastery with walls and trench locations marked from Fiema (2008)

Phase specific plans

Fig. 8 - Plan of Church and Chapel

Fig. 8 Chapter 6

The church and the chapel marking in Phase 2.

Mikkola et al (2008)

in Phase 2 from Mikkola et al (2008)
Fig. 10 - Isometric reconstruction

Fig. 10

Isometric reconstruction of the church and the chapel in Phase 2.

Mikkola et al (2008)

of the church and the chapel in Phase 2 from Mikkola et al (2008)
Fig. 56 - Plan of Church and Chapel

Fig. 56 Chapter 6

The church and the chapel marking in Phase 5.

Mikkola et al (2008)

in Phase 5 from Mikkola et al (2008)
Fig. 37 - Isometric reconstruction

Fig. 37

Isometric reconstruction of the church and the chapel in Phases 4-5.

Mikkola et al (2008)

of the church and the chapel in Phases 4-5 from Mikkola et al (2008)
Fig. 59 - Plan of Church and Chapel

Fig. 59 Chapter 6

The church and the chapel marking in Phase 7.

Mikkola et al (2008)

in Phase 7 from Mikkola et al (2008)
Fig. 67 - Isometric reconstruction

Fig. 67

Isometric reconstruction of the church and the chapel in Phase 7.

Mikkola et al (2008)

of the church and the chapel in Phase 7 from Mikkola et al (2008)
Fig. 81 - Plan of Church and Chapel

Fig. 81 Chapter 6

The church and the chapel marking in Phase 9.

Mikkola et al (2008)

in Phase 9 from Mikkola et al (2008)
Fig. 88 - Reconstruction of the

Fig. 88

Tentative reconstruction of the appearance of the church and the chapel in Phase 9.

Mikkola et al (2008)

appearance of the church and the chapel in Phase 9 from Mikkola et al (2008)
Fig. 107 - Plan of Church and Chapel

Fig. 107 Chapter 6

The church and the chapel marking in Phase 11.

Mikkola et al (2008)

in Phase 11 from Mikkola et al (2008)

Area Plans

Church and Chapel

Normal Size

Fig. 1 - Plan of

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

Church and Chapel with all walls and trench locations marked from Fiema (2008)
Labeled Plan of

Plan of the church and the chapel with parts of the structure labeled in red

modified by JW from Fig. 1 of Fiema (2008)

Church and Chapel - modified from Fiema (2008)

Magnified

Fig. 1 - Plan of

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

Church and Chapel with all walls and trench locations marked from Fiema (2008)
Labeled Plan of

Plan of the church and the chapel with parts of the structure labeled in red

modified by JW from Fig. 1 of Fiema (2008)

Church and Chapel - modified from Fiema (2008)

Stone by Stone plans of parts of the Church and Chapel

Normal Size

Fig. 17 - Stone by stone

Fig. 17

The plan of the eastern half of the church

Mikkola et al (2008)

plan of the eastern half of the church from Mikkola et. al. (2008)
Fig. 33 - Stone by stone

Fig. 33

The plan of the northwestern part of the church.

Mikkola et al (2008)

plan of the northwestern part of the church from Mikkola et al (2008)
Fig. 36 - Stone by stone

Fig. 36

The plan of the southwestern part of the church.

Mikkola et al (2008)

plan of the southwestern part of the church from Mikkola et al (2008)
Fig. 50 - Stone by stone

Fig. 50

Plan of the narthrex

Mikkola et al (2008)

plan of the narthex from Mikkola et al (2008)
Fig. 25 - Stone by stone

Fig. 25

The plan of the eastern part of the chapel.

Mikkola et al (2008)

plan of the eastern part of the chapel from Mikkola et al (2008)
Fig. 29 - Stone by stone

Fig. 29

The plan of the western part of the chapel.

Mikkola et al (2008)

plan of the western part of the chapel from Mikkola et al (2008)

Magnified

Fig. 17 - Stone by stone

Fig. 17

The plan of the eastern half of the church

Mikkola et al (2008)

plan of the eastern half of the church from Mikkola et. al. (2008)
Fig. 33 - Stone by stone

Fig. 33

The plan of the northwestern part of the church.

Mikkola et al (2008)

plan of the northwestern part of the church from Mikkola et al (2008)
Fig. 36 - Stone by stone

Fig. 36

The plan of the southwestern part of the church.

Mikkola et al (2008)

plan of the southwestern part of the church from Mikkola et al (2008)
Fig. 50 - Stone by stone

Fig. 50

Plan of the narthrex

Mikkola et al (2008)

plan of the narthex from Mikkola et al (2008)
Fig. 25 - Stone by stone

Fig. 25

The plan of the eastern part of the chapel.

Mikkola et al (2008)

plan of the eastern part of the chapel from Mikkola et al (2008)
Fig. 29 - Stone by stone

Fig. 29

The plan of the western part of the chapel.

Mikkola et al (2008)

plan of the western part of the chapel from Mikkola et al (2008)

Sections

Fig. 4 - Trench E, western baulk

Fig. 4

Trench E, western baulk

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 13 - Trench F, eastern baulk

Fig. 13

Trench F, eastern baulk. Also showing the façade wall (W. AA, loci F.30 and 31) of the southern pastophorion, the pilaster F.32 of the southern colonnade, and the stone tumbles/soil layers inside the apse.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 19 - Trench E, southern baulk

Fig. 19

Trench E, southern baulk. Showing the profile of the synthronon, the layers of collapse inside the apse and the deposit of mosaic fragments in Phase 13.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 46 - Trench L, northern baulk

Fig. 46

Trench L, northern baulk. The stratigraphy of the atrium showing the pavement, location of columns, and subsequent stone collapses.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 58 - Trench B, northern baulk

Fig. 58

Trench B, northern baulk. Featuring the stratigraphy in the atrium showing the pavement, the water channel. locus B.12, the buttress, locus B.02 from Phase 9, and the stone tumbles.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 76 - Trench C, northern baulk

Fig. 76

Trench C, northern baulk. The stratigraphy inside the apse and the bema of the chapel, featuring the masonry altar pedestal, locus C.16, and the layers of stone tumble.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 94 - Trench F, western baulk

Fig. 94

Trench F, western baulk, featuring the bench, locus F.15, and the stone and soil deposits inside the southern aisle and the nave.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 114 - Trench F, northern baulk

Fig. 114

Trench F, northern baulk, featuring stone tumbles and soil deposits mentioned in the text.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 117 - Trench G, eastern baulk

Fig. 117

Trench G, eastern baulk, featuring the stone tumbles mentioned in the text.

Mikkola et al (2008)

from Mikkola et al (2008)

Phasing

Phasing Summary and Seismic Events

from Mikkola et al (2008) in Chapter 6 of Petra - the mountain of Aaron : the Finnish archaeological project in Jordan.

Following seven field seasons of excavation (1998-2005), the obtained stratigraphic information and the associated finds allows for the recognition of fourteen consecutive phases of occupation, destruction, rebuilding and disuse in the area of the church and the chapel.¹ Of these, Phase 1 represents the pre-ecclesiastical occupation of the high plateau, Phases 2-8, the period of continuous monastic occupation interspersed with episodes of destruction, and Phases 9-14, the later occupation for which the ecclesiastical function of the church can no longer be supported, as well as the eventual abandonment of the church and the chapel of Jabal Harun. Specifically, Phases 3, 6, 8, 10 and 12 represent phases of destruction. The most likely explanation for most of these destructions is seismic events, and in some cases the evidence for an earthquake seems clear. However, in other cases, especially for Phase 6, alternative explanations will be considered as well. Notably, the multiple episodes of destruction and restoration seem well attested by the evidence of changes in the glass repertoire in the church and the chapel throughout the existence of these structures.

Footnotes

1 For the criteria in recognition of a phase, see Chapter 3. The site formation process and the impact of natural phenomena, such as earthquakes, are discussed in Chapter 5

Phasing Tables

Brief Phasing Table

from Fiema and Frosen (2008 Appendix C:34)

Concordance between the phasing of the Church and Chapel of the site

Fiema and Frosen (2008 Appendix C)

Detailed Phasing Table

from Mikkola et al (2008)

Nabatean Fineware Pottery Dating at Ez-Zantur

from Schmid (1995)
Ez-Zantur Excavations utilized Nabatean fineware chronology of Schmid (2000) - which I don't currently have access to
reference was made to Schmid's Nabatean fineware chronology in section(s) of the excavation reports dealing with pre-monastic phasing at Jabal Harun

Left

Chronology of Nabatean finewares

Right

Typology and chronology of the Nabataean fine ware

Both from Schmid (1995)

Pre-Monastic Phase IV Destruction Event - 363 CE or an earthquake from around that time

Discussion

Fiema and Frosen (2008 Appendix C:34) report that evidence was found in Room 25 of an early destruction, followed by a period of decay before the monastery at Jabal Harun was constructed. Archaeoseismic evidence included a shattered second story floor (O.41) some of which was protruding from a Wall G (e.g. Fig. 8). They also surmised that the core of the western building partially collapsed and the 2nd story floor was destroyed entirely. Remains of that floor were incorporated in the Byzantine structures. Finally, they suggest that the superstructure and arches of the southern cistern (Room 36) may also have collapsed.

References

Appendix C of Fiema and Frosen (2008)

Phase IV The 363 Earthquake

from Fiema and Frosen (2008 Appendix C:34)

The structures and soundings made in Room 25 provided evidence of an early destruction and the following period of decay that apparently preceded the building of the monastery. A dramatic piece of evidence the shattered second story floor (O.41), some remains of which are still protruding from Wall G (e.g. Fig. 8). The core of Western Building must have partially collapsed and the second story was entirely destroyed, as remains of its floor were incorporated in the Byzantine structures. The superstructure and arches of the southern cistern (Room 36) may also have collapsed. All of this may well be related to the famous earthquake of May 19, 363 CE⁴⁶ [JW: The southern Cyril Quake struck on the night of May 18, 363 CE] which is archaeologically well-evidenced by excavations in central Petra at sites such the Temple of Winged lions, the Colonnaded Street, the so-called Great Temple, and the residential complex at es-Zantur⁴⁷. According to a contemporary literary source (Bishop, Cyril of Jerusalem), the earthquake destroyed more than half of Petra⁴⁸. Given the fact that the earthquake severely damaged a host of other cities as well, it stems very unlikely that Jabal Harun, located less than five kilometers from downtown Petra, was left unharmed.

Footnotes

JW: I don't have access yet to the full appendix so I could not record the footnotes. What follows are links to pages in this catalog discussing the specific subjects footnoted.

46 363 CE Cyril Quakes

47 Petra - Ez-Zantur

48 ?

Mikkola et al (2008)

Figures

Fig. 2 - Cistern interior

Fig. 2

The interior of the cistern at the foot of the summit of Jabal Harun.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 3 - Re-used building element

Fig. 3

An architectural element in Wall Y

Mikkola et al (2008)

from pre-monastic times in Wall Y from Mikkola et al (2008)
Fig. 4 - Trench E, western baulk

Fig. 4

Trench E, western baulk

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 5 - Grey tesserae found

Fig. 5

Large quantities of grey tesserae found in Trench J.

Mikkola et al (2008)

in Trench J from Mikkola et al (2008)
Fig. 6 - Water channels

Fig. 6

Four water channels in the area of the cistern. The settling tank is visible in the center.

Mikkola et al (2008)

around the cistern from Mikkola et al (2008)
Fig. 7 - Interior of the settling tank

Fig. 7

Interior of the settling tank featuring the hydraulic mortar in the center.

Mikkola et al (2008)

from Mikkola et al (2008)

Discussion

Phase 1. Pre-Monastic Remains

Introduction

from Mikkola et al (2008:83-84)

This phase includes all evidence and elements that pre-date the construction of the church and the chapel in the central space of the monastic complex. Generally, the pre-ecclesiastical occupation of the high plateau is well attested. Within the perimeter of the later monastery, the presence of the so-called Western Building is particularly important. It is a building complex featuring very different construction materials and techniques from all the other buildings at the site, including a different kind of mortar and well-dressed ashlars that are two or three times larger than the average size of building stones on the site. That the Western Building is not contemporary with the church is also suggested by fact that its orientation differs markedly from the east-west -orientation of the rest of the complex. Apparently, the Western Building was built as a part of a Nabataean architectural complex at the site,⁴ but was incorporated into the monastery once it was established.

That the Western Building and the associated, pre-ecclesiastical structures are of a Nabataean date, is also indicated by the stratified finds of Nabataean pottery there and everywhere at the site as well as on the surrounding plateau. Additional evidence of sedentary human presence on plateau is provided by the existence of a large cistern (size ca. 18.0 x 5.0 m), still in use, at the foot of the summit of Jabal Harun, ca. 200 m ENE of the monastery (Fig. 2). The cistern is partially hewn into the bedrock, but has masonry-built walls and is spanned by fifteen arches built in typical Nabataean fashion.⁵ Similarly, the evidence for a substantial water management system built around the mountain suggests a Nabataean presence. As Jabal Harun is the highest mountain in the vicinity of the city of Petra, with a commanding view over the Wadi ‘Araba rift valley, it seems only natural for the Nabataeans to have established a presence on the mountain.

Only a few features in the area of the church can be securely associated with this early phase. These mainly consist of reused stone blocks that seem to derive from buildings other than those extant at the site. No structures in the area of the church or the chapel can be assigned to a pre-monastic phase with certainty. This is partly due to the fact that archaeological exposure was limited to a relatively small area, because the sandstone bedrock was encountered close to the floor levels in many of the trenches excavated so far. Therefore, it seems possible that if the buildings predating the church and the chapel had masonry-built foundations, these had been removed in the course of the subsequent building activities. The only surviving elements are those that have been carved into the bedrock.

Footnotes

4 The second volume of the FJHP series will fully describe the Nabataean constructions at the site. Significant discoveries related to the Nabataean phases at the site were made only during the 2007 fieldwork season.

5 For descriptions, see Wiegand 1920: 141; Lindner 2003: 187, Abb. 22. Compare also with Dar al-Birka (Site Wadi Musa 18A) – an arched Nabataean chamber built at a spring in the Wadi Musa area (‘Amr et al. 1998: 522).

Pre-Byzantine Elements in the Area of the Church

from Mikkola et al (2008:84-85)

The evidence for the Phase 1 occupation in the area of the church mostly consists of a few building blocks in reused positions, as well as early pottery and glass finds from the strata in soundings made under the floor of the church. The building blocks that do not appear to belong to the church may originate from a building that predates the construction of the Byzantine basilica. These include a decorated architectural stone (fragment of a cornice?) used as an ashlar in Wall Y (Fig. 3) and a limestone fragment of a cornice (0.63 x 0.45 x 0.25 m). The latter has been reused as a building stone in a pilaster (locus U.28) located between the nave and the northern aisle. A similar fragment, ca. 0.60 m wide and 0.34 m long, was used in the southern section of the masonry-built chancel screen (locus F.05h) of Phase 7. This pre-Byzantine, apparently monumental, building appears to have included columns, for some fragments of column drums have been incorporated in the Phase 2 bench (locus B.10) along Wall J, still visible in the area of the later Atrium. Furthermore, a column base (part of locus V.24) carved into the bedrock and probably also of this phase, was found west of the chapel.

Soundings made under the floor of the church yielded some material that predates the construction of the church, including fragments of pottery, bone and glass. Glass fragments found in a sounding (loci E.26a-c) made in the Apse of the church can be dated to the first half of the 5th century at the latest (Fig. 4, also Fig. 19).⁶ Similarly, a few soundings (including locus B.25e) made under the floor of the church yielded pottery dated exclusively to the Nabataean and Roman periods. This pottery is no doubt related to the pre-monastic presence on the mountain, and was deposited in the area of the church when foundations for the building were laid with soil gathered from the plateau.

Footnotes

6 All dates in A.D. unless otherwise indicated.

Pre-Byzantine Elements in the Area of the Chapel

from Mikkola et al (2008:85)

The evidence for pre-Byzantine activity in the area of the chapel is similarly sketchy, consisting of reused building blocks and Nabataean pottery found in soundings. The lowermost layer (locus C.24e, a buildup for the original chapel floor; elevation 44.12 m⁷) in a sounding made in the sanctuary of the chapel contained Nabataean fine ware and common ware, datable to the 2nd-4th centuries. A reused fragment of a limestone cornice (0.55 x 0.30 x 0.15 m) in front of the southern cupboard (locus C.30) is evidently of pre-Byzantine date and it originates from some unknown, yet monumental, structure. A small but interesting element, possibly also relating to such a building, is the piece of opus sectile floor, reused in the Phase 2 step (locus Y.32) in the doorway between the church and the chapel. The outer wall (Wall V) of the building complex also features rather unusual blocks with regard to their size, being much larger than the average blocks used in construction of the church.⁸ Similar, massive ashlars have been used in the Western Building, and thus appear to relate to some pre-Byzantine building activity on the plateau.

Footnotes

7 Whenever the height of an object or feature is provided in absolute terms, the first two digits are omitted. Thus an object located at 1244.12 m above sea level will be described here as being at 44.12 m asl, as all tacheometer measurements at the site are within the range of 1200-1300 m asl.

8 An example is a block in locus Y.12 measuring 1.65 x 0.35 x 0.20 m

The Area West of the Chapel

from Mikkola et al (2008:85)

Some indications of possible pre-Byzantine activity were also discovered in the area west of the chapel. In particular, the sandstone bedrock there preserves traces of what may well have been a monumental building predating the structures visible now. A column base (part of locus V.24, diameter 0.58 m, height 0.04 m) hewn out of bedrock was encountered west of Wall OO. In theory, it might also have been a feature of the Phase 2 chapel, and if so, associated with a hypothetical doorway in the original western wall of the chapel (infra). There is little evidence for such a doorway, however, and the location of the Phase 2 baptismal font seems incompatible with a doorway. Therefore, it seems more likely that the column base precedes the church and the chapel altogether. Perhaps associated with the column base, the bedrock in the surrounding area had been levelled (loci V.22, V.23).

Scattered Finds

from Mikkola et al (2008:85)

A number of finds probably deriving from Phase 1 buildings were found in the excavated soil layers. These consist mainly of building blocks reused in the construction of the upper courses of the walls, and of fragments of Nabataean pottery from the soil used as fill for the walls. A few fragments of Nabataean-type stucco decoration were also found in the soil. These may originate from the Western Building, where traces of similar stucco have been found.

Large tesserae (size ca. 0.03 x 0.03 x 0.03 m), made of light grey or bluish limestone (Fig. 5), form an intriguing and problematic group of finds, possibly associated with this phase. The tesserae were found either single or still forming small mosaic chunks. However, even though tens of thousands of such tesserae have been found practically everywhere at the site, the location of a mosaic floor(s) from which they might originate remains unknown. One possibility is that a Phase 1 building had a floor made of such tesserae. However, large limestone tesserae have been used in later phases as well, as is evidenced, e.g., by the repairs made in the Narthex mosaic.⁹

Footnotes

9 For the description of repairs to the mosaic, see Chapter 9

The Water Channels

from Mikkola et al (2008:85-86)

The use of the large cistern (size ca. 6.00 x 1.00 m) in the central courtyard evidently predates the construction of the church and the chapel. The cistern remains unexcavated, but is apparently partially formed by a natural cavity in the bedrock that had been modified and possibly expanded. Its northwestern edges are eroded and rounded, but the southeastern edge is quite well preserved. Inside the cistern, there appears to be a small subterranean tunnel-like trough entering the cistern from its northeastern end. The clear evidence for the use of the cistern is, however, the presence of four rock-cut water channels on its eastern side (Fig. 6). The channels emerge from under the northern part of the entrance porch to the church complex, and all but the southernmost channel probably predate the construction of the church.¹⁰ The two northernmost channels – locus H.34 (0.28 m wide, 0.18 deep), and locus H.35 (0.25 wide, 0.10 m deep) – resemble each other in both shape and size. They run to the northeastern end of the cistern, nearly joining each other close to the edge of the cistern, and both of them end in a flat spout. The third channel from the north, locus H.33 (0.25 m wide, 0.20 m deep), however, differs from the others in that it ends in the finely constructed settling tank (locus H.33b), carved into the bedrock close to the eastern edge of the cistern. The tank is a round (diameter 1.00 m), barrel-shaped space with a flat bottom. The interior of the structure is plastered with typical Nabataean hydraulic mortar (Fig. 7).¹¹ A good parallel for this installation is the settling tank in the Atrium of the Petra church (Phase V), although that was built in the Byzantine period.¹²

As all the channels approach the cistern from the direction of the church, and as they emerge from under the mosaic floor, it is not possible to know for certain where they come from or with what structure or specific phase they are associated. It is clear, however, that rainwater was collected in the area east of the cistern over a long period of time, and the number of channels running from that direction suggests that the water management system has been modified and/or expanded several times.

Footnotes

10 The southernmost channel is associated with the raised floor of the Atrium and it belongs to Phase 5.

11 Ueli Bellwald, 2000, personal communication

12 Fiema 2001: 72-73, with references to examples of parallel installations found elsewhere.

Dating

from Mikkola et al (2008:86-87)

The absolute dating of the only certain pre-Byzantine structure at the site, i.e., the Western Building, has yet to be established with precision, but the method of construction strongly suggests a Nabataean date. Combined with the presence of a characteristically Nabataean cistern at the foot of the summit of Jabal Harun, these structures provide evidence for a significant Nabataean presence there. This presence may be as early as the 3rd century B.C., as suggested by a ¹⁴C date acquired from a ceramic lamp.¹³ But this date must be treated with extreme caution since such an early date is so far not confirmed by any other datable elements of material culture found at Jabal Harun.

Specifically in the area of the church and the chapel, further evidence of the early sedentary occupation is offered by the abundant finds of Nabataean fine ware pottery and reused pre-Byzantine architectural elements. The latter can be generally dated to the 1st century B.C./A.D. As for the painted Nabataean fine ware, the finds from the excavation seasons 1998-2001 have been fully analyzed and thus a representative sample achieved.¹⁴ The clear majority of altogether 685 painted sherds can be assigned to Phase 3 in Schmid’s typology (20 - early 2nd century).¹⁵ Particularly prominently represented were subphases 3b (70/80 - 100) comprising 63% of the sherds, and 3a (20-70/80), to which 24% of the sherds could be assigned. Individual sherds representing phases 2c (1-20) and 4 (later 2nd-3rd centuries) were also found.

As for common ware pottery found in some loci in the area of the chapel that might be associated with the pre-monastic occupation, these include types characteristic of the 4th and early 5th centuries and no later material. More specific information is provided by the glass finds. Glass lamps fragments found in the foundation layers of the Apse (loci E.26a-c) belong to the common glass types of the first half of the 5th century in Petra. This indicates that the church was preceded by an unknown building that was lit with glass lamps and was still occupied in the 5th century. A corroded Byzantine coin (Reg. no. 155), probably from the first half of the 5th century, found in locus E.26b, seems to confirm the dating suggested by the glass finds.¹⁶ All these finds offer a good terminus post quem for the construction of the basilica in Phase 2.

Finally, a fragment of an unidentified mammalian bone found in the foundation layers (locus E.26c) of the Apse of the church was dated to 80-220 cal. A.D. (1865 ± 35 BP, Hela-1061). This result indicates that mammalian bones were already scattered around the plateau centuries before the construction of the church. When the church was constructed, these bones, contained in the soil from the plateau, were redeposited as a part of the foundation layer of the Apse.

All these finds suggest that the initial phase of active Nabataean use of the plateau began relatively early and it may mirror parallel development at the site of Petra. Whether or not the occupation of the plateau in the area of the later church and the chapel continued without a break until the construction of the monastic site is unknown. The 363 earthquake which is both historically and archaeologically evidenced at Petra may have caused a destruction and thus a temporary disruption of occupation at Jabal Harun as well.¹⁷ However, the ceramic and glass finds indicate that occupation would presumably have resumed in the 5th century.

Based on finds of pottery, glass and coins, Phase 1 can thus be dated approximately between the 1st (3rd ?) century B.C. and the mid-to later 5th century A.D., with the most active period of use beginning sometime in the early 1st century A.D. The mid-to later 5th century date for the end of Phase 1 also coincides well with the postulated construction date of the church and the chapel in Phase 2.

Footnotes

13 An AMS-dating (Hela-968) was acquired from soot in the nozzle of a ceramic lamp found in the lowest stratum (locus O.40) in the northern room of the Western Building. The date of 390- 200 cal. B.C. (2225 ± 75 BP) is surprisingly early for Jabal Harun, and it does not provide a dating (apart from a terminus post quem) for the building itself. The lamp may have been deposited as a part of soil gathered from the plateau in order to lay the foundations for the building. All radiocarbon datings in this text have been calibrated using the OxCal v. 3.10 program, which employs atmospheric data published by Reimer et al. 2004. The calibrated dates are reported in “one sigma” or 68.2% probablility.

14 Silvonen 2003.

15 Schmid 2000: 147-152, and Fig. 98.

16 We are grateful to Daniel Keller for the initial assessment of the coin finds from the church and the chapel. The full presentation of the numismatic material will be included in Volume II. Notably, a Nabataean coin (Reg. no.255) found inside the “tomb” (locus M.33b) in the southern pastophorion of the church can be dated to the 1st century. This find must be similarly related to the pre-monastic activities on the plateau, but has been redeposited during the construction of the basilica.

17 For the impact of that earthquake on Petra, see Russell 1980, Hammond 1980, and Fiema 2001: 18 (in the area of the Petra church) and 2002.

Phase 3 Destruction Event - mid to late 6th century CE

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 8 - Plan of Church and Chapel

Fig. 8 Chapter 6

The church and the chapel marking in Phase 2.

Mikkola et al (2008)

in Phase 2 from Mikkola et al (2008)
Fig. 10 - Isometric reconstruction

Fig. 10

Isometric reconstruction of the church and the chapel in Phase 2.

Mikkola et al (2008)

of the church and the chapel in Phase 2 from Mikkola et al (2008)

Figures

Fig. 30 - Charcoal and molten glass

Fig. 30.

Locus V. 15 – the deposit of charcoal and molten glass.

Mikkola et al (2008)

from the Phase 3 fire from Mikkola et al (2008)
Fig. 34 - Collapsed Column

Fig. 34.

Locus T.14 — the collapsed column

Mikkola et al (2008)

found among Phase 13 stone tumble from Mikkola et al (2008)
Fig. 44 - Repaired marble floor

Fig. 44.

The appearance of the marble floor in the nave of the church.

Mikkola et al (2008)

in the nave of the church - from Mikkola et al (2008)

Discussion

Mikkola et al (2008) characterize the Phase 3 destruction event as "catastrophic" and suggest that an earthquake was the most likely cause. The shock appears to have caused the upper parts of the walls of the church to collapse which led to a fire when the burning oil lamps crashed into the floor. In many parts of the church, the arches, clerestory walls, columns and upper parts of the walls collapsed. The roof system also appears to have been damaged as it was rearranged in the following phase. Fallen ashlars shattered the marble floor and the furnishings of the church and the chapel. Numerous fragments of marble colonnettes, chancel screens, etc. were re-used as building elements in Phase 4. Repaired walls of Phase 4 contained a multitude of fragments of marble slabs from the floor of Phase 2, now used as chinking stones. The fills of the walls were full of debris, especially Wall I which was first constructed in Phase 4. The debris included broken marble furnishings, pottery, glass, nails and roof tiles

The chapel was also damaged - as indicated by extensive Phase 4 repairs. The roof supports were rearranged and Phase 2 columns collapsed and were re-used in Phase 4 despite the fact that some of the drums were broken. The western wall of the chapel appears to have collapsed entirely as it was replaced by a new one (Wall OO). Parts of Wall H also appear to have been damaged as evidenced by a rebuild of its upper courses in the ensuing phase. The pilasters in the chapel also appear to have been destroyed as their replacements, which were visible during excavations, were backed by wall plaster.

Although damage was extensive, not all the walls in the church appear to have collapsed. Some remained standing. The apse of the church also appears to have survived. The shattered marble floor in the church was re-used with fragments pieced together like a jigsaw puzzle (Fig. 44). Large quantities of debris, including charcoal, burnt tiles, broken and fire-damaged glass and ceramic sherds, and pieces of marble and other stones, were found in the midden located outside the monastery enclosure, excavated in Trench R. The uniformity of the debris led the excavators to conclude that they were refuse from a fire-related destruction - cleared from the Church and Chapel before re-building activity began in Phase 4.

Dating to the mid to late 6th century CE was based on glass finds complemented by ceramics and fits well with the late 6th century Inscription at Areopolis Quake.

References

Mikkola et al (2008)

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 8 - Plan of Church and Chapel

Fig. 8 Chapter 6

The church and the chapel marking in Phase 2.

Mikkola et al (2008)

in Phase 2 from Mikkola et al (2008)
Fig. 10 - Isometric reconstruction

Fig. 10

Isometric reconstruction of the church and the chapel in Phase 2.

Mikkola et al (2008)

of the church and the chapel in Phase 2 from Mikkola et al (2008)

Figures

Fig. 30 - Charcoal and molten glass

Fig. 30.

Locus V. 15 – the desposit of charcoal and molten glass.

Mikkola et al (2008)

from the Phase 3 fire from Mikkola et al (2008)
Fig. 34 - Collapsed Column

Fig. 34.

Locus T.14 — the collapsed column

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 44 - Repaired marble floor

Fig. 44.

The appearance of the marble floor in the nave of the church.

Mikkola et al (2008)

in the nave of the church - from Mikkola et al (2008)

Discussion

Phase 2 Roof Support System

from Mikkola et al (2008:97)

No clear trace of any of the roof supports related to this phase was found. However, it is postulated that in light of the width of the interior of the chapel – almost 6 m – one should envision the existence of transverse arches, rather than horizontal wooden cross-beams. Such arches probably sprung directly from Walls GG and H by means of arch-springers similar to those found in the pastophoria of the church. Alternatively, the arches might have been supported by pilasters, but no trace of such pilasters has been preserved. On top of the arches, a system of roof beams and trusses would have supported the actual roof. As in the church, glass lamps provided lighting, as is evidenced by lamp fragments found in a sounding made in the Bema (locus C.35a). Some of the lamp fragments found in a deposit (locus V.15) outside the western wall of the chapel, and resulting from the clearance after the destruction in Phase 3, may also derive from the chapel.

In spite of the fact that few roof tiles were found during the course of excavation, it is reasonable to assume that – like the early church – the roof of the early chapel was tiled. It is more difficult, however, to ascertain the form of the roof. As the church and the chapel shared the same E-W wall (Wall H), one could imagine that the chapel’s roof was slanted towards the north, basically following the incline of the roof over the northern aisle of the church. This proposition, however, faces several difficulties. First, such an arrangement would have deprived the northern aisle of the church of any window openings, thus the only means providing light to that space would have been through the hanging lamps. Secondly, the slanting roof of the chapel could have been much lower than the end of the slanting roof over the northern aisle, thus providing space for the windows in the upper part of Wall H. However, in such a case, the slanting incline of the chapel’s roof would end at an extremely low position at the point of junction with Wall GG, the northern wall of the chapel. Another possibility is the existence of a flat roof over the chapel in Phase 2. Although such a flat roof is postulated to have covered the church area in Phase 9, this solution for the chapel in Phase 2 does not seem likely.

The preferable solution to this issue is to envisage a pitched roof for the chapel in Phase 2 (and later). Such a roof would allow for the existence of windows in Wall H,⁸⁶ and it would facilitate the draining of rainwater from the area of the chapel. As such, there would be two buildings – the church and the chapel – standing side by side and connected to each other, each with a pitched roof, the roof of the chapel being lower than that of the church. The evidence for such an arrangement is scant but perhaps it can be reinforced by some iconographic representations. Although the details are unclear and difficult to interpret, the mosaic in the church in Tayyibat al-Imam (northern Syria) portrays a large basilican church, which seems to have a small building with a pitched roof attached to its long side. In another example, two churches, one large and one small (a chapel?), but both with pitched roofs are represented on the mosaic in the Church of Bishop Sergius (6th century) in Umm ar-Rasas. However, the interpretation of the perspective convention used there is difficult; these edifices might or might not have been meant to be viewed as standing side by side.⁸⁷

The pitched roof over the chapel, allowing for the collection of rainwater, would also find support in the existence of a late channel (locus V.04a) running west of the chapel. It seems to have collected water from the northern side of the chapel roof, suggesting that at least in Phase 9, the roof could not have been flat. Although highly hypothetical, it nonetheless offers a clue that a similar system of a water channel and a pitched roof may have already existed in Phase 2.

Footnotes

86 These either provided the light for the interior of the chapel or the interior of the northern aisle of the church. The isometric view presented here (Fig. 10) favors the former option. However, as the height of the church increased in Phase 4 (Fig. 37 ), the windows in Wall H could provide light for the northern aisle. At any rate, due to the lack of convincing evidence, the issue remains unresolved.

87 Both examples are discussed by Duval 2003b: 243-245, fig. 17c (two separate buildings or a five aisled basilica?) and 2003b: 272-274, fig. 27d.

Phase 3 The First Destruction

Damage Evidence

from Mikkola et al (2008:102)

This phase represents a catastrophic event that caused the first major destruction of the site. Judging by the totality of the damage, a major seismic event seems to be the most likely explanation for the destruction¹⁰². It appears that the seismic shock caused the collapse of the upper parts of walls, and the burning oil lamps, falling on the floor, caused the conflagration. The destruction was severe. In many parts of the church, the arches, clerestory walls, columns and upper parts of the walls collapsed. That the roof support system was severely damaged is indicated, among other ways, by the fact that it was completely rearranged in the following phase. The falling stones shattered the marble floor and the furnishings of the church and the chapel, and while the floor was haphazardly repaired in the following phase, much of the furnishings were apparently damaged beyond repair. This is evidenced by the numerous fragments of marble colonnettes, chancel screens, etc., found in reused positions in the structures of Phase 4.

The intensity of the event is also indicated by the evidence of repairs to the upper portions of the walls of the church and the chapel. The repaired walls of Phase 4 feature numerous fragments of marble slabs from the floor of Phase 2, now used as chinking stones. Various kinds of debris ended up in the fills of the walls, especially in Wall I which was constructed in Phase 4. In fact, a large portion of the finds of broken marble furnishing, pottery, glass, nails and roof tiles, found in the late layers of stone tumble, derive from the interior of the repaired walls and therefore predate Phase 3.

The totality of the destruction was ensured by a fire that raged in the buildings.¹⁰³ The fire was hot enough to melt the glass lamps of the polycandela and leave marks of burning on much of the marble decoration of the church. A deposit (locus V.15) of charcoal and broken, partly melted, lamp fragments, probably associated with Phase 3, was found west of the chapel (Fig. 30). This locus seems to represent burned material cleared from the church and the chapel immediately after the fire and the earthquake. Several of the large iron nails – probably originally used in the structures of the roof – found during the excavations, similarly exhibit signs of exposure to intense heat.

The chapel was also heavily affected. This is indicated by the extent of the repairs made in Phase 4, particularly by the complete rearrangement of the roof supports. The system of pilasters now visible in the chapel is not original, as is evidenced by the presence of wall plaster behind the pilasters, the use of marble slab fragments as chinking stones (in loci Y17 and Y20), and the different construction techniques used. The Phase 4 columns of the chapel, moreover, seem to derive from the collapsed columns of Phase 2 structures, as some of the drums used in them are broken. The original western wall of the chapel also seems to have collapsed to the extent that it was deemed easier to build a new wall (Wall OO). Finally, parts of Wall H also appear to have been badly damaged, as its upper courses were rebuilt in the following phase, using large quantities of recycled material.

Footnotes

102 For the impact of earthquakes on the area of Petra, see Chapter 5

103 For the fire that destroyed the Petra church in Phase VIII, see Fiema 2001: 91-94.

Dating

from Mikkola et al (2008:102-103)

The destruction in Phase 3 was a momentous event. Therefore, its dating postulated here would equal the date for the end of Phase 2 and, simultaneously, the beginning of Phase 4. As stated in relation to the dating of Phase 2, glass finds, especially from the deposit of burned material west of the chapel (locus V.15), span the 5th to mid-6th centuries, providing an excellent terminus post quem for Phase 3. The lamps are notably the only relics from the excavations, in addition to broken marble pieces from the walls’ interiors that feature the impact of fire. As such, the only destruction episode in the history of the site, which features both structural collapse and fire, must represent the destruction of the Phase 2 occupation. Similarly, the datable ceramics described above as related to Phase 2 indicate the mid- to later 6th century for the end of Phase 2.

Fragments of charcoal scraped from inside some melted glass lamps were radiocarbon-dated (Hela-998). The result of 1440 ±35 B.P. or ca. 594-650 cal. A.D. is problematic, as it apparently contradicts the dating of Phase 3 suggested above. Datings of wood charcoal are generally considered notoriously unreliable because of the possibility of the “old wood” effect¹⁰⁴ – making dates appear older than they should – but here, on the contrary, the result seems too late in date. This could be taken to indicate that the deposit of V.15 is, in fact, related to some later event (Phase 6?) rather than Phase 3, which would contradict the well-established glass chronology. However, the possibility of contamination is high – the charcoal in locus V.15 may have been mixed with the later material, either during excavation or already in antiquity.¹⁰⁵ Given the fact that the glass lamps have been exposed to intense heat, and that Phase 3 appears to be the only phase with evidence for a fire that could have caused the melting, it seems probable that the lamps (and the charcoal inside them) should be associated with Phase 3.

At any rate, it is safe to postulate that the Phase 3 destruction took place around the mid- to-later 6th century. Again, no historically known human-induced destruction can be proposed here and the seismic event, as postulated above, appears a much more plausible explanation. The only historically documented seismic events of consequence, which affected the Near East in the mid-6th century, are the earthquakes of 551 and 559. The famous earthquake of July 9, 551 struck the areas of coastal Lebanon and Egypt, as well as Arabia, Syria and Palestine.¹⁰⁶ It was often associated with the destruction of Petra, but no reliable archaeological evidence in the city can be marshalled to prove its impact.¹⁰⁷ The earthquake on Good Friday, 559, affected the areas of Galilee, Palestine and Arabia,¹⁰⁸ but again its possible impact on Petra is virtually unknown. Therefore, it must be assumed that the monastery fell victim to one of the less known, more localized but sufficiently destructive, seismic events, which might have had an epicenter in the Wadi ‘Araba area perhaps resulting from the aftershocks of the 551 or 559 earthquakes,.

Footnotes

104 E.g., Renfrew and Bahn 1991: 127.

105 A second attempt (Hela-1062) to obtain a radiocarbon-date for locus V.15 – this time from bone finds associated with the layer – proved to be equally problematic. The result – ca. 220-340 cal. A.D. (1765 ± 35 B.P.) – seems much too early for the deposit. Both results illustrate the problems associated with radiocarbon dates made from a material that is not well understood. The unexpectedly early dating of the bones from V.15 may relate to the fact that bones associated with a pre-monastic occupation at Jabal Harun appear to have been scattered throughout the plateau, and may thus have been redeposited in various later contexts. This is also indicated by the bones found in the foundation deposits of Phase 2 (loci E.26a-c), described in relation to Phase 1.

106 Guidoboni et al. 1994: 332-336. See Chapter 5 for details

107 Russell 1985: 45, and Fiema 1991: 214-217. For the recent reconsideration, see Fiema 2002: 234-235. There is also no indication of that destruction or its potential aftermath in the 6th century Petra Papyri (Frösén 2001: 490).

108 Di Segni 1999: 154-155, note 32.

Phase 4 The Basilica with the Atrium - excerpts

Excerpted from from Mikkola et al (2008:103-117)

... the walls of the structures [in the Church] did not entirely collapse in Phase 3.

...The height of the columns [of the Church] can be estimated to have been at minimum 3.85 m, since both columns were found collapsed among the stone tumble of Phase 3 (Fig. 34).

...The apse of the church appears to have survived the events of Phase 3 comparatively well.

... It is impossible to assess the extent of the damage inflicted on the original marble furnishing of the bema [of the Church] in Phase 3. It must have been considerable, judging from the quantities of broken marble included as fill in both new walls (e.g., Wall I) and the old, reconstructed walls (e.g., Wall H). However, some elements must have survived either intact or in pieces, which could have been reused after necessary modifications.

... The destruction of the fine marble pavement [of the Church] was amongst the more permanent damage caused by the event of Phase 3. The rebuilding in Phase 4 took great effort, using all resources available, and evidently the community of Jabal Harun could not afford to fully replace the broken marble floor with a new pavement. Instead, the broken pavers were painstakingly pieced together, like a huge jigsaw puzzle. The area of the nave (e.g., in locus E24) presents good examples of this (Fig. 44).

... extensive damage suffered by the original western wall of the chapel. ...

Area West of the Chapel

Large quantities of debris, including charcoal, burnt tiles, glass and ceramic sherds broken and fire-damaged, pieces of marble and other stones, were found in the midden located outside the monastery enclosure, excavated in Trench R. Due to the uniformity of these deposits and the clear indication that they originated from a fire-related destruction, it is probable that these represent Phase 3 debris cleared out from the area of the church and the chapel at the beginning of Phase 4.

Phase 6 Destruction Event (?) - 1st half of 7th century CE - inferred from remodelling and may not have been a response to an earthquake

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 56 - Plan of Church and Chapel

Fig. 56 Chapter 6

The church and the chapel marking in Phase 5.

Mikkola et al (2008)

in Phase 5 from Mikkola et al (2008)
Fig. 37 - Isometric reconstruction

Fig. 37

Isometric reconstruction of the church and the chapel in Phases 4-5.

Mikkola et al (2008)

of the church and the chapel in Phases 4-5 from Mikkola et al (2008)

Figures

Fig. 5 - Iconoclastic damage

Fig. 5

Iconoclastic damage to the narthex mosaic

(by J. Vihonen)

Fiema (2013)

from Fiema (2013)

Discussion

Mikkola et al (2008) inferred difficult to interpret seismic destruction in Phase 6 based on repairs to floors and pilasters, a major remodelling of the site, and a change in furnishings that took place in Phase 7. The new furnishings suggested a change in liturgy which could have also been partly or fully responsible for the remodel. While referring to an iconoclastic edict by the Caliph Yazid II in 723/724 CE, Fiema (2013:799) noted that Muslims initially used Christian edifices for prayer, with the result that these edifices had to conform to Islamic prescriptions (Bowersock 2006: 91-111). The primary prescription would likely have been against the Christian embrace of iconography which Muslims and Jews viewed as a violation of the ancient commandments against idolatry. Shared use of sites by Muslims and Christians can be seen, for example, in the Church of Kathisma between Jerusalem and Bethlehem. Jabal Harun, the supposed burial site of Moses' brother Aaron, would likely have been a site that Muslims visited. Moses is mentioned more frequently in the Quran than any other personage (136 times) and his life is narrated more often than any other prophet. Aaron is also frequently mentioned. In fact, the nearby summit of Jabal Harun currently houses a mosque dedicated to Aaron and iconoclastic damage was observed in following phase (7) in the narthex (see Fig. 5). Mikkola et al (2008) noted that, while difficult to date, it seems probable that the iconoclastic damage done to the narthex mosaic [of the Church - Fig. 5] can be assigned to this phase [phase 7]. Thus, the change in liturgy associated with the remodeling observed for Phase 7 could have been a reaction to increased Muslim visitation rather than seismic damage or some combination of structural damage/decay and accommodation of Muslim pilgrims. If an earthquake was wholly or partly responsible for the remodel, the excavators suggested it struck in the early to mid 7th century CE which might fit with the Sword in the Sky Quake (~634 CE) or the Jordan Valley Quakes (659/660 CE).

References

Mikkola et al (2008)

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 56 - Plan of Church and Chapel

Fig. 56 Chapter 6

The church and the chapel marking in Phase 5.

Mikkola et al (2008)

in Phase 5 from Mikkola et al (2008)
Fig. 37 - Isometric reconstruction

Fig. 37

Isometric reconstruction of the church and the chapel in Phases 4-5.

Mikkola et al (2008)

of the church and the chapel in Phases 4-5 from Mikkola et al (2008)

Figures

Fig. 5 - Iconoclastic damage

Fig. 5

Iconoclastic damage to the narthex mosaic

(by J. Vihonen)

Fiema (2013)

from Fiema (2013)

Discussion

Phase 4 Roofing and Roof Support System

The Church

from Mikkola et al (2008:105-107)

The complete rearrangement of the roof support system is one of the major new features of Phase 4. The colonnades of Phase 2 were replaced with a single pair of columns in the middle of the shortened church, supporting two pairs of arches (length 3.80 m) running E-W.115 The arches would have consisted of approximately 20 voussoirs.¹¹⁶ The two columns (loci T.14 and U.25) were erected closer to the central axis of the church than those of Phase 2, making the nave of Phase 4 narrower (ca. 5.70 m) and the aisles wider (ca. 3.75 m) than in Phase 2. The columns were made of reused drums (diameter 0.55 m, height varying between ca. 0.25 and 0.50 m) of Phase 2 and were topped with Nabataean-style capitals. The height of the columns can be estimated to have been at minimum 3.85 m, since both columns were found collapsed among the stone tumble of Phase 13 (Fig. 34).

Supporting the long E-W -running arches at their extremities, two pairs of pilasters were built against the ends of Wall T (the Apse wall) and against Wall I. The pilasters were carefully constructed of well-shaped limestone ashlars and mortar, using the header-stretcher method. In the Apse, the southern pilaster (locus F.06), 0.58 m wide and 0.63 m deep, survives to a height of 1.71 metres and the northern one (locus U.08), 0.52 m wide and 0.73 m deep, remains 1.90 m high. The new pilasters created a narrow 0.05 to 0.10 m space between them and the arch buttresses of Phase 2, and these spaces were filled with small stones and mud mortar – loci F.06a and U.08a on the southern and northern sides, respectively (Fig. 35, see also Fig. 13). The corresponding pilasters, which abut Wall I, were equally well built, using good quality building material. The southern pilaster (locus T.12), 0.58 m wide and 0.74 m deep, surviving to a height of 1.73 m (Fig. 36), and the northern pilaster (locus G.05), 0.52 m wide, 0.75 m deep and 1.71 m high (Fig. 33), show a quality workmanship such as is only rarely found at the site, clearly built with an eye for the unusual stress caused by the long arches. Incidentally, the northern pilaster was built on top of a marble slab, which bears an inscription. Unfortunately, further investigation was not possible and only a few letters could be identified.¹¹⁷

This apparent professional and expert reconstruction of the elements of the roof support in Phase 4 was fully justified. If the height of the two columns in the nave was more or less equal (or even only slightly different) than that of the columns in Phase 2, and the roof support in the aisles remained the same (i.e., beams and trusses), that posed a clear problem for the builders in selecting the proper form of the roof over the church. If the maximum height (ca. 1.90 m) of a semicircular, 3.80 m long arch is added to the known height of the columns (at least 3.85 m), the roof of the church can be estimated to have stood at a height of nearly six meters at least, not counting the height of the clerestory walls. As there is no evidence for N-S arches in the aisles in Phase 4, the horizontal supports for roofing in the aisles would have to be anchored in the space above the E-W arches in the nave, i.e., in the clerestory walls. Although the two arches in the nave appear exceedingly long, there is no reason to doubt that, if well-built, they could support the clerestory walls above them. As such, there are two possibilities for the overall roofing that could have existed in Phase 4. The reconstructed church of Phases 4-5 (Fig. 37) shows the church retaining a classic basilica form, i.e., with the clerestory walls, the pitched roof over the nave and the slanting roofs over the aisles. An alternative is a single pitched roof extending over the nave and the aisles. This would, however, still require the existence of some sort of clerestory, perhaps lower than in Phase 2. Otherwise, with the height of columns being around 4 m, the pitched roof would have, in fact, been close to a flat roof, and the roof supports (horizontal beams) for the aisles would not have had adequate structures for anchorage. Thus, with a composite basilica-type roof, or with a single pitched roof, the presence of the clerestory walls was necessary. Regardless of which option was selected by the Phase 4 builders, it is apparent that the church in this phase was an edifice of unusual proportions: its ground plan nearly square, the nave with a roof at least as high, or higher, than in Phase 2 and the roofs over the aisles definitely higher than in Phase 2.

Footnotes

116 Compare the arches in the vaulted cistern by the summit of Jabal Harun, which span a similar width (note 5).

117 For further details, see Chapter 10, inscription 11

The Chapel

from Mikkola et al (2008:114-115)

As noted above, the roof support system of Phase 2 in the chapel is unknown. Regardless of its design, it was now entirely reconstructed or replaced by a system of pilasters and columns serving as pilasters, and transverse arches (Fig. 52). This system is reminiscent of that found in some churches in Arabia Provincia – relatively narrow, monoapsidal edifices, with pilasters or piers against the long walls and the single nave spanned by the transverse arches – such as appear in some churches in Umm alJimal, Sabha, Khirbet es-Samra, and Nitl.¹⁴³

Two pairs of columns, made of reused drums (diameter 0.55 m) probably originating from the Phase 2 church, were erected in the eastern part of the chapel (Fig. 53, see also Fig. 25) directly against the walls to serve as pilasters. Of the eastern pair of columns, the northern column (locus Y.04) has six drums still standing, with a total height of 2.25 m (elevation 46.75 m). The uppermost drum is badly weathered, but appears to have been decorated with crosses. The southern column (locus C.13), with six drums remaining, is 2.20 m high (elevation 46.63 m). In the western pair of columns (loci Y.14, Y.15), the northern column has seven drums (height 2.05 m), while the southern column with its six drums is 2.07 m high. Both of these columns were topped by an undecorated, conical pilaster cap. Further to the west of the columns, the arches were supported by three pairs of pilasters. In their present appearance, only the westernmost pair (locus I.06, 0.70 m x 0.55, and locus I.11, 0.56 m x 0.55 m), built using the header-stretcher technique, appear to derive from Phase 4 (Fig. 54). The two remaining pairs of pilasters – the eastern pair (loci Y.17, Y.18) and the middle pair (loci I.07, I.09, Y.19, Y.20) – probably stood in the same positions as the current pilasters, but have been substantially repaired or completely rebuilt in Phase 7.

The E-W distances between the individual pilasters and columns are relatively uniform, ranging from 1.35 m to 1.45 m. The only exception to this is the distance between the western pair of columns and the easternmost pair of pilasters, which is some 0.30 m wider (1.75 m). This discrepancy has an obvious explanation in that the Phase 2 doorway is located between these roof supports. At any rate, the distance between all arches is sufficient to allow the installation of a superstructure typical of the region. Presumably wooden, but possibly also stone, horizontal E-W beams would have been placed over the arches’ walls, forming the ceiling of the chapel.¹⁴⁴ A series of wooden trusses above the ceiling would then support the pitched roof overlaid with tiles.

The roof support structures are preserved well enough to allow a rough estimate of the height of the roof inside the chapel. The span of the arches between columns Y.14 and Y.15 (which seem to have survived almost completely) was ca. 4.35 m. If the arches formed a perfect semicircle, the height of the room appears to have been ca. 4.20 m (2.05 m plus 2.18 m, or the height of the columns Y.14/Y.15 plus the radius of the arch). Assuming that the average thickness of a voussoir was 0.30 m, it can be estimated that the arches would have consisted of ca. 20 to 25 voussoirs.¹⁴⁵

Footnotes

143 E.g., at Umm al-Jimal, the Church of Masechos, the East Church (Butler 1929: 20-22, discussion p. 187) and the Chapel outside the East Wall (Michel 2001: 169-173); the “small” church at Sabha (Michel 2001: 184); the South Church at Umm alQuttayn (Michel 2001: 189), Churches nos. 20, 81, and 90 at Khirbet es-Samra (Desreumaux and Humbert 2003: 27, 29- 31; Michel 2001: 194-195) and both the North and South Churches at Nitl (Piccirillo 2001: 272, 276; Michel 2001: 365- 367). For the general characteristics of churches in the Hawran, see Shereshevski 1991: 114-119, 124-126, 129-133.

144 As in Phase 2, a pitched roof for the chapel is a preferable solution over a flat roof. For examples featuring flat roofs but also supported by transverse arches supported by pilasters, see the vernacular architecture of the region (Hirschfeld 1995: 125- 127, 241-243; Al-Azzawi et al. 1995: 325) but also Room I (Phase III) adjacent to the Petra church (Fiema 2001: 20-21; Kanellopoulos 2001: 153-157).

145 This estimate can be calculated using the formula p =pr/0.3m, where x is the number of voussoirs, r the radius of the semicircle (in this case, 2.175 m) and 0.3 m the average thickness of a voussoir.

Phase 6 The Second Destruction (?)

Damage Evidence

from Mikkola et al (2008:119-120)

Whereas the event of Phase 3 was almost certainly a massive earthquake coupled with a raging fire, it is much more difficult to interpret precisely what happened in Phase 6. The reason for distinguishing this phase at all is that something must have prompted the extensive rebuilding activities of Phase 7. However, whether it was an earthquake, a spontaneous collapse of the inside structures, or some less dramatic reason, is not immediately clear. The long arches and central columns of Phase 4 were a somewhat unusual feature in Byzantine church architecture, and may have turned out to be structurally too weak to support the building. A partial collapse or the immediate danger of such might have occurred or been anticipated. It is then possible that the old roof supports were dismantled and new ones built in a planned way. Exploiting such an opportunity, a thorough remodelling of the building for aesthetic or functional reasons might then have been executed.

However, there are some problems with this explanation. First of all, the changes made in Phase 7 were not limited to the roof supports, but several new installations were built in the altar areas of both the church and the chapel, and the entire church interior was plastered anew. These changes may or may not have been related to the prevailing liturgical requirements. It also seems curious that the sturdy pilasters of the Phase 4 church were abandoned and replaced by less well-built pillars in Phase 7. That all of this was done for aesthetic or functional reasons seems unlikely. Perhaps the most important clue to the nature of the event is offered by the finds of glass and marble elements. The church of Phase 7 no longer featured a marble chancel screen or ambo, and it was lit with new types of glass lamps. It is not easy to see why the marble decorations and old glass lamps would have been discarded if the building was simply remodelled in an orderly manner. Therefore, one must assume that the roof supports and lamps fell as a result of some event, either an earthquake or a spontaneous collapse due to the structural instability of the building. Such an event might have wrecked most of the church furnishings beyond repair. It should also be mentioned that the Byzantine shrine under the weli, which was studied by Wiegand, may well have been abandoned as a result of an earthquake, and the most convenient phase for this would be Phase 6.¹⁵⁶ However, the extent of the damage is uneven throughout the structures and, while some elements must have collapsed, others, such as the columns in the church, seem to have somehow survived. Therefore, it is possible that Phase 6 represents only a relatively minor seismic event, but still one that provided a convenient reason for the extensive remodelling of the church and the chapel in the following phase. The extent of building activities in Phase 7, therefore, need not directly reflect the intensity of the earthquake.

At least some of the arches of the Phase 4 church seem to have fallen, probably causing heavy damage to the floor and furnishings of the church, but, curiously, the two columns of Phase 4 were left standing. Judging by the methods and materials used, the marble floor has been damaged and repaired on several occasions, but it is difficult to attribute a particular series of repairs to the damage inflicted in Phase 6. The same is true of the walls, which have been damaged and repaired on several occasions, but there are few chronological indications to help to date a particular series of repairs. One may only conclude that some of the damage to which the floor and walls bear witness might have been caused by the Phase 6 event.

The Narthex mosaic also appears to have suffered damage, as it was repaired using large limestone tesserae in the following phase. In its northern quarter, an area of repairs (Repair I) running from east to west suggests that something heavy (a column?) fell on it. The extent of the damage may also be related to the relative instability of the Narthex floor, as the early water channels (the two northernmost and the central one) run under the mosaic. Some smaller repairs made with small tesserae may also relate to damage resulting in Phase 6.

The chapel was unlikely to have been unaffected by the event of Phase 6, but the extent of damage there is difficult to estimate. At least some of the pilasters are likely to have collapsed, as they appear to have been rebuilt in the following phase. The westernmost pair of pilasters may have withstood the possible tremor, as might the two pairs of columns in the eastern part of the chapel. Probably, the walls also suffered some damage, but this is difficult to distinguish from the damage and repairs of Phases 3 and 4. It is possible that the Phase 2 baptismal font was destroyed by stones falling from above, as it was abandoned and replaced by a new font in Phase 7. Alternatively, the font may have been replaced for functional reasons.

Footnotes

156 For the further ramifications of such a scenario, see Chapter 16.

Dating

from Mikkola et al (2008:120)

If the Phase 7 remodelling was indeed in response to a destruction, i.e., if Phase 6 should be treated as a momentous destructive event, there are several earthquakes which affected Palestine in the course of the first half of the 7th century. These include the historically documented events in 634, 637, 641, and in June 658 or 659.¹⁵⁷ It has been suggested that one of these earthquakes completed the demolition of the already derelict and abandoned Petra church in Phase X.¹⁵⁸ Also, recent archaeological work in the Wadi ‘Araba indicates the occurrence of several earthquakes there, some of them severe, throughout the 7th century.¹⁵⁹

Further information for a possible date of Phase 6 (i.e., the end of Phases 4-5) is provided by the datable evidence related to Phase 7. A terminus ante quem for Phase 6 is provided by the glass material found west of the chapel (in loci V.10, J.18-20 and J.21/24), probably associated with the church of Phase 7. This material should be dated to the 7th century, at any rate not earlier than the end of the 6th century. A rough 7th century date is also postulated for the ceramic deposits associated with Phase 7. Finally, the appearance of certain new liturgical installations, e.g., the masonry altar in the Phase 7 chapel, is dated to the 7th century, more specifically to its second half. All this indicates that Phase 7 began sometime around the mid-7th century. This, in turn, indicates a date of the early to mid-7th century for the end of Phases 4-5 and the assumed destruction in Phase 6.

Footnotes

157 Guidoboni et al. 1994: 355-358; Amiran, Arieh, and Turcotte 1994: 266. For a complete list and commentary, see Chapter 5.

158 Fiema 2001: 111.

159 Niemi 2007: 414.

Phase 8 Destruction Event - mid 8th century CE

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 59 - Plan of Church and Chapel

Fig. 59 Chapter 6

The church and the chapel marking in Phase 7.

Mikkola et al (2008)

in Phase 7 from Mikkola et al (2008)
Fig. 67 - Isometric reconstruction

Fig. 67

Isometric reconstruction of the church and the chapel in Phase 7.

Mikkola et al (2008)

of the church and the chapel in Phase 7 from Mikkola et al (2008)

Figures

Fig. 80 - Tilted Wall

Fig. 80.

Wall J tilting southward, following the earthquake of Phase 8.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 5 - Iconoclastic damage

Fig. 5

Iconoclastic damage to the narthex mosaic

(by J. Vihonen)

Fiema (2013)

from Fiema (2013)

Discussion

Although Mikkola et al (2008) noted that continuous re-building and structural damage had probably made the buildings weaker and more vulnerable to seismic events, they characterized the destruction observed in Phase 8 as major. Wall J was observed tilted to the south (Fig. 80) and was shored up by a buttress that was installed in the following phase. The arches of the southern aisle and those spanning the nave appear to have collapsed along with the semi-dome over the apse. The floor of the apse was damaged and the bema also suffered heavy damage. Although the stone tumble in this part of the church was cleared in Phase 9, the semidome was not rebuilt and the apse floor wasn't repaired.

Parts of the collonades of the atrium collapsed and the atrium floor was damaged but it was difficult to determine whether it was damaged in phase 8. The square pilaster (locus L.14) or pedestal in the eastern part of the atrium was also probably destroyed in Phase 8. The arch covering the southern pastophorion most likely collapsed in Phase 8, considering the fact that the entire southern wall of the basilica was severely affected by the destruction. Therefore, unlike the one in the northern pastophorion, the arch must have been rebuilt in Phase 9, as is evidenced by the discovery of the collapsed voussoirs of a fallen arch found among the stone tumble inside the room (locus M.04). The mosaic in the narthex (Fig. 5) was also damaged, especially in the central medallion, which was never repaired. The date of this damage, however, is uncertain - it may have been caused by the events of either Phase 8 or 10.

The northern part of the church fared better than the southern part and arches covering the northern aisle survived intact. Glass finds support the idea that some walls survived Phase 8 comparatively well, as at least some windowpanes used in Phase 7 appear to have remained in use in Phase 9. It was suggested that that the northern part of the church, as abutted by the structure of the chapel, was firmly buttressed by its compact form and thus could better withstand the earth tremor.

Iconoclastic damage may provide a terminus post quem for this destruction (see Fig. 5) as the first iconoclast period began around 730 CE (Wikipedia). Mikkola et al (2008) suggested that the Phase 8 destruction occurred soon after this iconoclastic damage.

References

Mikkola et al (2008)

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 59 - Plan of Church and Chapel

Fig. 59 Chapter 6

The church and the chapel marking in Phase 7.

Mikkola et al (2008)

in Phase 7 from Mikkola et al (2008)
Fig. 67 - Isometric reconstruction

Fig. 67

Isometric reconstruction of the church and the chapel in Phase 7.

Mikkola et al (2008)

of the church and the chapel in Phase 7 from Mikkola et al (2008)

Figures

Fig. 80 - Tilted Wall

Fig. 80.

Wall J tilting southward, following the earthquake of Phase 8.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 5 - Iconoclastic damage

Fig. 5

Iconoclastic damage to the narthex mosaic

(by J. Vihonen)

Fiema (2013)

from Fiema (2013)
Fig. 62 - southern “cross-column”

Fig. 62

The southern “cross-column” (on the left).

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 63 - Pilasters

Fig. 63

The appearance of the pilasters set against the northern “cross-column.”

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 64 - Three pilasters

Fig. 64.

The three pilasters (extreme left) of Phase 7 built against Wall H in the northern aisle.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 65 - The plastered element

Fig. 65

The plastered element, locus G.12 (in the center).

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 66 - cross scratched

Fig. 66

A cross scratched on the surface of the plastered element, locus T.11. (in the center).

Mikkola et al (2008)

on the surface of the plastered element from Mikkola et al (2008)
Fig. 103 - Buttress against

Fig. 103

The Phase 9 buttress, locus T.31, built against Wall J. (in the center).

Mikkola et al (2008)

Wall J from Mikkola et al (2008)
Fig. 99 - Phase 9 Buttress

Fig. 99

The Phase 9 buttress, loci B.02, B.16, B.18 and L.02, in the atrium (in the center).

Mikkola et al (2008)

in the atrium from Mikkola et al (2008)
Fig. 36 - Stone by stone plan

Fig. 36

The plan of the southwestern part of the church.

Mikkola et al (2008)

of SW part of church from Mikkola et al (2008)
Fig. 58 - Trench B section

Fig. 58

Trench B, northern baulk. Featuring the stratigraphy in the atrium, including the pavement, the water channel, locus B.12, the buttress, locus B.02, from Phase 9, and the stone tumbles.

Mikkola et al (2008)

from Mikkola et al (2008)

Discussion

Phase 7 Roof Support System

The Church

from Mikkola et al (2008:122-124)

The system of roof support in Phase 7 is complex. The long arches running E-W of Phase 4 were replaced by a system of pilasters and short arches, which now sprang from north to south. But the columns and pilasters of the Phase 4 church were not removed. Equipped with additional supporting features, they were all left in situ and integrated into the new system.

Four new pillars were constructed between the columns and the pilasters of Phase 4, thus creating a symmetrical E-W oriented system of supports, in which a pilaster standing against Wall I was followed by a freestanding pillar, then by a column, then again by a freestanding pillar, and finally by a pilaster abutting Wall T (Figs. 60 and 61). The pillars were constructed of three segments. They had a “core” made of yellow limestone blocks forming small square pillars ca. 0.40 x 0.40 m in horizontal section. Relatively thin (width ca. 0.15 m) additions serving as pilasters were built against their northern and southern sides, resulting in the case of each freestanding pillar in a structure that was 0.40 m wide (E-W) and 0.70 m long. The eastern pair of these pillars was integrated into the structure of the Bema. As such, the northern pillar (locus U.38/39/40e), ca. 0.38 x 0.70 and 1.20 m high, formed the NW corner of the new ambo (Fig. 17, Col. Fig. 25) and the southern one (locus F.05b/05j) served as the western end of the masonry chancel screen (locus F.05c/ h). In the western part of the church, a pillar (locus G.13/U.24/ 34), ca. 0.42 x 0.62 and 1.58 m high, was erected in the northern aisle and a second pillar (locus T.19), ca. 0.40 x 0.69 and 0.70 m high, was built in the southern aisle.

The construction of pilasters against already existing columns of Phase 4 created what might be termed “cross-columns,” as they are cruciform in horizontal section (Fig. 62, also Figs. 17 and 33). That this shape is intentional is shown by the fact that the exteriors of the “cross-columns” have been plastered with the fine, smooth plaster, characteristic of Phase 7. The northern “cross-column”(locus U.25) is abutted by four pilasters. These are the northern pilaster (locus U.27) ca. 0.40 m wide, 0.42 m deep and of surviving height 0.57 m, the eastern pilaster (locus U.28), ca. 0.45 x 0.30 x 1.00 m, the southern pilaster (locus U.26), ca. 0.40 x 0.49 x 0.57 m, and western pilaster (locus U.29), 0.45 x 0.30 x 1.30 m (Fig. 63). The southern “cross-column” (locus T.14) also had four abutting pilasters (locus T.32), three of which survive on the western, northern and eastern sides of the column. The average width of these pilasters is ca. 0.40 m, depth ca. 0.30 m and surviving height 0.57 m. On the southern part of the column, the original pilaster was rebuilt in Phase 9.

In the northern aisle, three pilasters were built along Wall H (Fig. 64). From west to east, these pilasters can be described as follows. Locus G.27/U.16, located on the western part of Wall H, is 0.43 m long, 0.37 m wide and survives only to the height of 0.60 m. Moving eastward is locus U.17 which is 0.47 m long, 0.35 wide and currently only 0.80 m high. Locus U.18 in the east is better preserved; in its present state, it is 0.46 m long, 0.31 m wide and 1.42 m high. From these pilasters, arches sprung over the aisle, in a N-S direction, to meet the pillars and the central column. From west to east, the spans of these arches were 3.00 m, 2.64 m and 3.05 m. The arching in the southern aisle no doubt mirrored that in the northern aisle, with three arches spanning the aisle. However, because of changes made in Phase 9, none of the Phase 7 pilasters built against Wall J have survived in their original form. Assuming that the pilasters were approximately the same size as those against Wall H, the spans of the arches would have varied between ca. 2.60 (central arch) and 3.10 m.

The nave was spanned by three arches running NS. The westernmost of these arches (length 4.88 m) covered the space between the western free-standing pillars, the central arch (length 4.70 m) sprang between the two “cross-columns,” and the eastern arch (length 4.52 m) between the two pillars in the east. Altogether, all the pillars and pilasters described above supported a series of three rows of arches running north-south, with three arches in each row, covering the nave and the two aisles of the church. The fourth arch running north-south (length 4.57 m), supported by two pilasters (locus F.05a, size 0.78 x 0.45 m, and locus U.37, size 0.70 x 0.38 m), was built immediately in front of the Apse. This arch, which sprang over the Bema, seems to have been a light structure that did not bear much weight. Possibly, it either reinforced the edge of the semidome, or just marked the entrance to the Apse area.

In addition to the arches running N-S, some kind of support system seems to have also run from east to west. Plastered elements (locus G.12, ca. 0.26 m x 0.47 m, height 1.71 m, Fig. 33 and locus T.11, ca. 0.50 x 0.30, height 1.75 m, Fig. 36) were added to the eastern faces of the Phase 4 pilasters abutting Wall I (Fig. 65). One of these – locus T.11 – has a cross carved in the plaster on its southern side (Fig. 66). Similar elements, loci F.05a and U.37, were added to the western faces of the pilasters abutting the Apse wall and to the western and eastern sides of the two “crosscolumns” (loci U.28, U.29 and T.32). The precise purpose of these plastered elements is not immediately clear. They seem too thin and weakly built to have served as pilasters supporting actual arches. The distance between them in an E-W direction is only 1.40 m on average – rather short for an arch, although short arches have been encountered at the site.¹⁶¹ But if arches were used, one wonders why the well-built pilasters of Phase 4 – preserved in situ – were not used instead. Probably, the intention was to further reduce the E-W space between the corresponding western pilasters, the columns and the eastern pilasters. For this reason, it seems more probable that the plastered elements supported wooden beams running E-W (or, less likely, stone architraves) rather than arches. With such wooden beams as supports, the system of freestanding pillars would have been made much more stable.¹⁶²

The most enigmatic element in this entire scheme is the form of the actual roof. With the elimination of the high E-W arches from Phases 4-5, and the employment of E-W wooden beams in the nave, the need for the cler-estory walls would have disappeared. The main support element for the entire structure would now be the series of N-S arches throughout the nave and the aisles and E-W beams. A flat roof is not impossible.¹⁶³ However, a somewhat flattened, pitched roof fully extending over the nave and the aisles is preferred here (Fig. 67).¹⁶⁴

Footnotes

161 E.g., in the Western Building.

162 E.g., the Roman basilica at Shaqqa, Syria – a building dated to the late 2nd-early 3rd century – which features a roof support system consisting of transverse (N-S) arches with longitudinal (E-W) arches (Butler 1929: 16-17, Ill. 9 and 10). Note that the longitudinal arches are not depicted in the plan (Ill. 9).

163 Butler (1929: 22-24 and Ill. 17-20) associates a flat roof and galleries over the aisles with all transverse-supported basilicas, e.g., the Shaqqa basilica (see previous note) and the basilicas in Tafha and Nimreh.

164 Such a church would externally somewhat resemble the modern reconstruction of the Memorial of Moses structure on Mt. Nebo, especially its roof.

The Chapel

from Mikkola et al (2008:128)

The roof continued to be supported by a system of N-S -running arches, as in Phase 4, but it seems that the two central pairs of pilasters, i.e., loci Y.17 and Y.19/I.07 abutting Wall H and loci Y.18 and Y.20/I.09 abutting Wall GG, had to be rebuilt in this phase, and the arches between them rebuilt. This hypothesis is based on the observation that the building technique used on these pilasters differs from the two westernmost pairs of pilasters: they are built less regularly and feature marble slab fragments used as chinking stones. The sizes of the pilasters are as follows. pilaster locus Y.17 is currently eight courses high (1.99 m), 0.53 m wide and 0.59 m long, Y.18 is 10 courses high (2.40 m), 0.54 m wide and 0.62 m long, Y.19/I.07 is nine courses high (2.20 m), 0.53 m wide and 0.68 m long, and Y.20/I.09 is nine courses high (2.22 m), 0.54 m wide and 0.62 m long. Presumably, the remaining roof supports were not affected by the calamity, as they do not show any obvious signs of rebuilding.

Inevitably, any wooden material, especially if easily recoverable, would have been promptly salvaged from the ruined structures of the church and the chapel during the latest occupation phases at the site, and used for fuel or other purposes. Furthermore, the site was still frequented by the pilgrims, even after the monastic occupation ceased. This needs to be carefully considered in light of fact that, apart from the use of arches, the form and the construction materials of the roof support in the chapel might have now become somewhat different from that of the church. For example, no traces of wooden beams, trusses or the “insulating” clayish soil otherwise found inside the church were discovered in the final collapse layers of the chapel (part of Phase 12). Instead, a layer of grey, ashy soil (loci I.12, Y.21) covered the fallen arches, and the upper surfaces of many of the voussoirs were stained grey by the ash. It seems that this ash might originate from a thick layer of ashy mortar (now completely dissolved), probably used as waterproofing. It is then postulated that the chapel in this phase might still have been roofed with a pitchedtype roof, which would have used some timber material as a frame. However, instead of the usual ceramic roof tiles, a thick layer of mortar now covered the chapel’s roof. Notably, the opinion has been expressed that the wooden/stone roof frame in the Negev churches was covered by waterproof mortar.¹⁷⁴

Footnotes

174 Colt 1962: 11. However, Shereshevski (1991: 174) maintains that the roofs of the Negev churches were either tiled or thatched

Phase 8 The Third Destruction

Damage Evidence

from Mikkola et al (2008:131)

Phase 8 represents yet another calamity which befell the site, probably another earthquake. As noted before, continuous re-building and structural damage caused by earlier destructions had probably made the buildings weaker and thus more vulnerable to seismic events, even relatively minor ones. However, this event seems to have been a major one, causing the collapse of the church's semidome and the columns of the atrium.

In particular, the earthquake caused Wall J to severely tilt towards the south (Fig. 80), causing the collapse of the arches in the southern aisle. The wall was left leaning towards the south and it had to be supported by a buttress in the following phase. In addition to the arches of the southern aisle, those spanning the nave appear to have collapsed. Such a pattern of collapse would indeed be expected. With the mutual supporting arch and beam system introduced in Phase 7, the collapse of one N-S arch in the aisle would have seriously impaired the stability of the corresponding N-S arch across the nave. However, the northern part of the church survived the disaster better. For example, it seems that the arches covering the northern aisle survived intact. The glass finds also support the idea that some walls survived Phase 8 comparatively well, as at least some windowpanes used in Phase 7 appear to have remained in use in Phase 9. All this may probably be explained by the fact that the northern part of the church, as abutted by the structure of the chapel, was firmly buttressed by its compact form and thus could better withstand the earth tremor.

The apse and bema also suffered heavy damage in Phase 8. The semidome covering the apse must have collapsed in the earthquake, destroying the floor of the apse beyond repair. The resulting tumble was cleared in the following phase, but the semidome and the apse floor were never repaired. The arch supporting the roof of the northern pastophorion probably fell too. In the southern pastophorion, falling stones caused severe damage to the floor due the presence of hollow compartments underneath. The part of floor that covered the southern compartment was destroyed and never repaired. It is uncertain if the arch there collapsed as well. It may have been left standing, but the roof was nonetheless severely damaged.

In the atrium, parts of the collonades collapsed. The atrium floor shows damage, but it is again difficult to determine whether it was damaged in this phase. The square pilaster (locus L.14) or pedestal in the eastern part of the atrium was also probably destroyed then. The mosaic in the narthex shows damage, especially in the central medallion, which was never repaired. Dating of the damage is uncertain - it may have been caused by the events of either Phase 8 or 10.

Probably for the reason given above, the chapel seems to have been much less affected than the church. The pavement and the installations inside the chapel apparently were left relatively intact. The only sign of damage is found outside Wall GG, where a supporting buttress (part of locus Y.13) was built against the northern side of the wall in the following phase. There is no evidence suggesting that the arches or the roof collapsed. The arch support system in the chapel was simple yet apparently adequate to withstand a tremor. Nevertheless, the construction of the buttress against Wall GG in the following phase indicates that either some damage was inflicted in Phase 8, or it was a preventive measure.

Dating

from Mikkola et al (2008:131-132)

As the iconoclastic activities have been postulated to have taken place at Jabal Harun in the early 8th century, and still within the duration of Phase 7, the momentous yet disastrous event, i.e., the destruction in Phase 8, may, in fact, have occurred soon afterwards, effectively ending Phase 7. The best candidate for such event is the major earthquake on January 18, 749. The inflicted destruction was well-evidenced and widespread and it affected Lebanon, Syria, Palestine (e.g., Jerusalem, Jericho and Beth Shan) and Transjordan (the regions of Amman, Madaba and Mt. Nebo).¹⁹¹ Again, its impact on the Petra area is historically unknown, but considering the documented magnitude and intensity of destruction it is not unlikely that at least the series of aftershocks, tremors and surface faulting could have reached as far as Petra. Some destruction layers found in Petra were associated with a major seismic event of roughly 8th century date, which, according to Peter Parr, effectively ended occupation in the city.¹⁹² Furthermore, it has recently been claimed that one of the ecclesiastical edifices in Petra – the Blue Chapel – was destroyed in this earthquake.¹⁹³ Considering that the likelihood of this hypothesis is high, it is probable that the 749 earthquake is also responsible for the end of Phase 7 at Jabal Harun

Footnotes

191 Tsafrir and Foerster 1992: 234; Amiran, Arieh, and Turcotte 1994: 266-267; Guidoboni et al. 1994: 366-370. Dated to 748 by Russell (1985: 47-49). It has been suggested that some ecclesiastical structures in Transjordan were destroyed by this earthquake, among them, the basilica at Umm Qays/Gadara (Weber 1998: 449) and the chapel at Yajuz (Khalil 1998: 463).

192 Parr 1959: 107-108.

193 Bikai 2002b: 2. Initially dated by the author to 749/750 (Bikai 2002a: 451), then corrected to 748/749 (Bikai 2004: 63).

Phase 9 Domestic Occupation - excerpts

Excerpted from from Mikkola et al (2008:132-143)

... The arch covering the southern pastophorion most likely collapsed in Phase 8, considering the fact that the entire southern wall of the basilica was severely affected by the destruction. Therefore, unlike the one in the northern pastophorion, the arch must have been rebuilt in Phase 9, as is evidenced by the discovery of the collapsed voussoirs of a fallen arch found among the stone tumble inside the room (locus M.04).

... As the iconoclastic activities have been postulated to have taken place at Jabal Harun in the early 8th century, and still within the duration of Phase 7, the destruction in Phase 8 may, have occurred soon afterwards. The best candidate for such event is the major earthquake on January 18, 749. ... it's impact on the Petra area is historically unknown ... Some destruction layers found in Petra were associated with a major seismic event of roughly 8th century date, which, according to Peter Parr, effectively ended occupation in the city (Parr 1959:107-108). Furthermore, it has recently been claimed that one of the ecclesiastical edifices in Petra - the Blue Chapel - was destroyed in this earthquake (2002a:451, 2002b.2004:63).

Phase 9 reconstruction

The fallen columns of the atrium were not re-erected, but were cleared away and used elsewhere. The damaged floor was repaired, and a section of Wall H in the atrium (loci V.06, X.13) was rebuilt.

... The most significant element of Phase 9 in the atrium is, however, the construction of a massive platform or buttress (loci B.02, B.16 [fill], B.18 [facade], and L.02) in the southeastern corner of the atrium, against Wall I (Fig. 99, also Figs. 36 and 58).

... A number of structures located outside the church were investigated in the course of excavation. The largest and perhaps most significant of these is the long buttress (locus T.31), built against Wall J (Fig. 103). The assignment of this buttress to Phase 9 is certain; it was clearly built after the wall tilted south in Phase 8. Therefore, it is likely that the buttress was built to support the wall against potential earth tremors. ²¹⁹

... The walls of the chapel seem to have withstood the event of Phase 8, in spite of the fact that it caused so much damage to the church. However, the walls probably suffered some structural damage. This is suggested by the construction of stone buttresses outside and against Wall GG.

Footnotes

219 Compare with a sloping revetment (talus), a reinforcing installation, known at the Byzantine sites in the Negev, e.g., at the North Church in Rehovot (Tsafrir 1988a: 26-27, 77) and the one built against the northern annex of the church at Horvat Karkur `Illit (Figueras 2004b: 33). Similar external reinforcements are also known from the North Church at Sobata and the citadel complex at Oboda."

Phase 10 Destruction Event - late 8th or early 9th century CE

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 81 - Plan of Church and Chapel

Fig. 81 Chapter 6

The church and the chapel marking in Phase 9.

Mikkola et al (2008)

in Phase 9 from Mikkola et al (2008)
Fig. 88 - Reconstruction of the

Fig. 88

Tentative reconstruction of the appearance of the church and the chapel in Phase 9.

Mikkola et al (2008)

appearance of the church and the chapel in Phase 9 from Mikkola et al (2008)

Discussion

Mikkola et al (2008) report that continuous sedentary occupation, at least in the area of the church and the chapel, probably ended in this phase whose destruction was probably of seismic character. Because much of the stone tumble in the church and the chapel created by this event was cleared in the following phase, it was difficult to securely associate any of the excavated strata with the collapse in Phase 10. Destruction was described as follows:

The most obvious evidence of this destruction consists of craters left in the church floor by tumbling stones. The marble floor was badly damaged in especially in the western part of the nave and the northern aisle, where much of the floor was removed in the following phase. It seems probable that the long N-S arch running between pilasters T.04 and G.06 collapsed in this phase. Several depressions left in the floor (locus T.29) of the nave mark the places hit by the falling stones. The stones that caused the depressions were, however, removed in Phase 11. Indirect evidence also exists for the collapse of the westernmost arch in the northern aisle and the one that spanned the eastern-most part of the nave, for in these areas the marble floor was removed in Phase 11. It seems reasonable to assume that the removal of the floors was related to the damage caused by stones falling from the arches and other structures of the roof, whereas the floor was left untouched in those parts of the church where the arches did not collapse.

As the walls and columns of the atrium and the narthex had been badly damaged and already partially removed in Phases 8 and 9, they probably were not heavily affected by the destruction of Phase 10. However, some of the stone tumble (lowest parts of locus H.02) in the area of the narthex may have been caused by this event.

Mikkola et al (2008) stated that it was impossible to provide any reasonably accurate date for this disaster however since ceramic deposits associated with Phase 11 provide a very rough date of the 9th century for that phase, a prior destruction [i.e. the Phase 10 destruction] would have to have occurred sometime in the later 8th or early 9th century.

References

Mikkola et al (2008)

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 81 - Plan of Church and Chapel

Fig. 81 Chapter 6

The church and the chapel marking in Phase 9.

Mikkola et al (2008)

in Phase 9 from Mikkola et al (2008)
Fig. 88 - Reconstruction of the

Fig. 88

Tentative reconstruction of the appearance of the church and the chapel in Phase 9.

Mikkola et al (2008)

appearance of the church and the chapel in Phase 9 from Mikkola et al (2008)
Fig. 107 - Plan of Church and Chapel

Fig. 107 Chapter 6

The church and the chapel marking in Phase 11.

Mikkola et al (2008)

in Phase 11 from Mikkola et al (2008)

Figures

Fig. 85 - Phase 9 arch supports

Fig. 85

The Phase 9 arch supports (two columns and a pilaster) and the wide bench built against Wall J in the southern aisle.

(by J. Vihonen)

Fiema (2013)

in S Aisle of Church from Fiema (2013)
Fig. 86 - Phase 9 arch supports

Fig. 86

The Phase 9 arch supports on each side of the southern aisle.

(by J. Vihonen)

Fiema (2013)

in S Aisle of Church from Fiema (2013)
Fig. 87 - Phase 9 pilaster

Fig. 87

The Phase 9 pilaster (left side), locus F.05d.

(by J. Vihonen)

Fiema (2013)

from Fiema (2013)
Fig. 76 - Trench C, northern baulk

Fig. 76

Trench C, northern baulk. The stratigraphy inside the apse and the bema of the chapel, featuring the masonry altar pedestal, locus C.16, and the layers of stone tumble.

Mikkola et al (2008)

from Mikkola et al (2008)

Discussion

Phase 9 Roof Support and Roofing

The Church

from Mikkola et al (2008:134-135)

Due to the collapse of the arches of the southern aisle and the nave in the course of Phase 8, the roof support system was again rebuilt. This may also be indicated by the glass finds, as new types of glass lamps now appear among the finds. Indeed, as opposed to the chapel, the lamp types used in the church of Phase 9 represent exclusively new types, confirming the idea that the church suffered more damage in Phase 8 than the chapel. Thus some of the Phase 7 arch or beam supports were now replaced by the new ones, i.e., the pilasters and columns serving in that capacity. Abutting Wall J, two columns and a pilaster were erected (Fig. 85). The column drums have a diameter of 0.55 m and probably derive from the collapsed colonnades of the Atrium. In its present shape, the western column (locus T.16) consists of seven drums, with a total height of 1.97 m. The middle column (locus T.15) has a height of 1.80 m, but only four of its column drums can be seen, the rest being covered by a bench (locus T.23). The pilaster (locus F.07), 0.53 m wide, 0.50 m long, in the eastern part of the aisle has six courses surviving, with a height of 1.80 m.

Corresponding supports for the arches were built on the northern side of the southern aisle (Fig. 86). A column (locus T.17, diameter 0.55 m) was erected next to a freestanding pillar (locus T.19) of Phase 7, and a pilaster (locus T.30), ca. 0.52 x 0.48 m, was built abutting the Phase 4 column (locus T.14). Finally, a second pilaster (locus F.05d) was built abutting an originally freestanding pillar (locus F.05b/j) in the eastern part of the aisle, which now was integrated with Wall Y (Fig. 87). As the new supports for the westernmost arch spanning the nave form a part of Walls FFF and BB, they have already been described in the section above. That arch was approximately 3.90 m long. The next (to the east) arch spanning the nave was in exactly the same location and used the same structural supports as in Phase 7, so it was either unaffected by the calamity of Phase 8, or rebuilt without modification. Finally, the arch covering the central part of the nave, i.e., the easternmost arch in the nave, was reerected, supported by a new pilaster (locus U.26, 0.40 x 0.49 x 0.57 m) built abutting the southern side of the column between the nave and the northern aisle. In the southern part of the nave, the arch was supported by a pilaster of Phase 7, which was integrated with Wall Y.

The northern aisle’s arch support system, consisting of masonry elements ca. 0.42 cm wide probably derives from Phase 7. It did not seem to have collapsed in the Phase 8 destruction since there were no signs of any rebuilding related to this or any later phase.²⁰⁹ It therefore seems that it was mainly the southern wall of the church (Wall J) that buckled in during the Phase 8 earthquake, thereby necessitating the reerection of the arches in the southern aisle only. With the new masonry supports for the arches in the southern aisle being ca. 0.55 m wide, and the passageway from the southern aisle to the nave being asymmetrical to that between the nave and the northern aisle, the entire arrangement within the building appears somewhat awkward but presumably indicates the different functions of these spaces within the building.

To summarize, the Phase 9 roof over the church structure would have been borne by a somewhat haphazard, asymmetrical system of N-S arches with horizontal beams laid on top of them, and some partitioning walls, the latter probably rising all the way up to the ceiling. Some new and old (Phase 7) pilasters and columns served as arch supports. There were three N-S arches in the southern and northern aisles each and three arches in the nave but not in a symmetrical arrangement as in Phase 7. Specifically, in the nave, the central and the easternmost arches corresponded to the westernmost and central ones in the aisles. The westernmost arch in the nave did not have corresponding arches in the aisles, but instead, it was supported by the L-shaped walls. The two easternmost arches in the aisles probably did not have corresponding arches in the nave, unless, as mentioned above, there was an arch close by the Bema line. This remains unproven but if there was one, it must have survived from Phase 7. In any case, it is probable that a simple flat roof over the entire structure was introduced in this phase (Fig. 88). It would have been made of wooden beams over the arches, and with branches in the gaps, all covered by layers of clayey soil²¹⁰ and probably topped by some sort of mortar (?) layer.²¹¹

Footnotes

209 In the course of excavation, no collapsed rows of arches were discovered in the central and western parts of the northern aisle. This indicates that the collapsed arches (and other roof construction remains, like wooden beams, etc.) were cleared away at some point, most likely in Phase 11, for the purpose of removing the pavement of the aisle. The other option, that the northern aisle was destroyed in Phase 8 and never rebuilt in Phase 9, seems very unlikely considering the care taken in the rebuilding of the southern aisle.

210 For similar deposits in the Lower Church (C101) at Humeima, see Schick 1995: 326.

211 For such constructions, still encountered in the vernacular architecture in the region, see Hirschfeld 1995: 123-127. The clayey soil is well represented in the later collapse deposits inside the church. See also description of Phase 12.

The Chapel

from Mikkola et al (2008:142)

No major changes were made to the roofing in this phase, as all pilasters and arches seem to have survived the event of Phase 8. The roof was probably still of the pitched type although it might also have been replaced by one slanting down toward the north where the water-collecting installations described above are located. It is evident that the roof was no longer tiled in this phase, because very few fragments of roof tiles were found in the collapse layers of Phase 12. Instead, the roof seems to have been covered with a thick layer of mortar, at least some of which was mixed with a bluish-grey ash, presumably in order to make it more waterproof.²²² Large amounts of disintegrated ashy mortar were found in the collapse layers of Phase 12, i.e., loci C.07, C.08, C.09, I.02, I.08, I.10, I.12, I.15, I.16, Y.05, Y.08, Y.21, and Y.24 (Fig. 76).

Footnotes

222 See note 174 for a similar roofing system known from the Negev.

Phase 10 Fourth Destruction

Damage Evidence

from Mikkola et al (2008:143)

A disaster in Phase 10, probably of seismic character, probably did end the continuous, sedentary occupation at least in the area of the church and the chapel. From this point on, limited occupation continued in these two partially ruined buildings. However, it is not entirely clear whether or not the disaster in Phase 10 also spelled the end of the permanent monastic presence on the plateau. Most probably, some kind of limited occupation would have continued and the Western Building seems to be the best candidate for such occupation. Furthermore, much of the stone tumble in the church and the chapel created by this event had been cleared in the following phase. This makes it difficult to securely associate any of the excavated strata with the collapse in Phase 10.

The most obvious evidence of this destruction consists of craters left in the church floor by tumbling stones. The marble floor was badly damaged in especially in the western part of the nave and the northern aisle, where much of the floor was removed in the following phase. It seems probable that the long N-S arch running between pilasters T.04 and G.06 collapsed in this phase. Several depressions left in the floor (locus T.29) of the nave mark the places hit by the falling stones. The stones that caused the depressions were, however, removed in Phase 11. Indirect evidence also exists for the collapse of the westernmost arch in the northern aisle and the one that spanned the easternmost part of the nave, for in these areas the marble floor was removed in Phase 11. It seems reasonable to assume that the removal of the floors was related to the damage caused by stones falling from the arches and other structures of the roof, whereas the floor was left untouched in those parts of the church where the arches did not collapse.

As the walls and columns of the Atrium and the Narthex had been badly damaged and already partially removed in Phases 8 and 9, they probably were not heavily affected by the destruction of Phase 10. However, some of the stone tumble (lowest parts of locus H.02) in the area of the Narthex may have been caused by this event.

As before, the chapel was less affected by the calamity than the church. It seems that none of the arches supporting the roof of the chapel fell, and consequently, all the walls and other structures of the chapel seem to have survived reasonably well. However, it is possible that some elements fell from the roof or the walls, since the floor of the chapel was partially removed in the next phase. This may also be related to the lack of permanent occupation and resultant collecting or looting activities in Phase 11, but as much of the floor was made of sandstone and of little value, it is probable that the floor already suffered some damage in this phase.

Dating

from Mikkola et al (2008:143)

It is impossible to provide any reasonably accurate date for this disaster. Considering the fact that the ceramic deposits associated with Phase 11 provide a very rough date of the 9th century for that phase, a prior destruction would have to have occurred sometime in the later 8th or early 9th century. Some major earthquakes that affected Palestine during that time include the ones dated to 756, 808 and 859.²²³ However, none of them can be associated with Phase 10 destruction with certainty.

Footnotes

223 Amiran, Arieh, and Turcotte 1994: 267-268.

Phase 11 The Collecting Activities

Excerpted from from Mikkola et al (2008:143-)

The continual wreckage caused by numerous earthquakes, in spite of periods of repair and rebuilding, eventually reduced the ecclesiastical complex to ruins. Again, the chapel had apparently survived the calamity of Phase 10 much better, yet by this phase, it seems that both the church and the chapel ceased to have any ecclesiastic function. The partially ruined structures were not abandoned, however, as is evidenced by the fact that a considerable effort was made to clear the buildings of stone tumble (Fig. 107). This may mean a limited but possibly not permanent domestic use of both buildings, although a clear attempt was made to recover some useful elements of their construction and furnishing. The evidence from other parts of the monastic site indicates that pilgrims would undoubtedly still have visited the mountain. Therefore, the existence of a small religious community at Jabal Harun in this phase should be considered plausible. Such a community would have provided some services and shelter for the visitors, especially as some parts of the complex, such as the chapel, were still in a reasonable state to provide shelter. The ceramics associated with this phase no longer feature any Byzantine residuals, only the Umayyad-Abbasid types.

Most importantly, however, Phase 11 is characterized by the removal and collection of reusable material – glass sherds, mosaic tesserae (glass and stone) and marble fragments – into small piles and caches throughout the church and the chapel. Such activities are well evidenced in the context of derelict churches in Palestine and Transjordan. For example, caches of collected glass sherds and/or glass tesserae have been found in the Petra church, the Blue Chapel in Petra, the church at Khirbet al-Kerak, and in some churches in Jerash.²²⁴ Marble and fragments of multicolored frescoes were collected in the North Church at Rehovot.²²⁵ Many fragments of wall mosaic were found in a dump pit outside the basilica at Umm Qays/Gadara,²²⁶ either for temporary storage or permanent disposal. Marble tiles and panelling were removed during the Umayyad period from the East Church at Pella.²²⁷ The glass piles at Jabal Harun should be considered as material collected for further use rather than a means of ritual disposal of the ecclesiastical glass.²²⁸ However, whether this material was intended for reuse at the site or to be transported elsewhere is unknown, but both options are equally plausible.²²⁹

A second characteristic feature is the occurrence of small fireplaces made directly on the floors of the church and the chapel, apparently with no concern for the damage done to the floors or the soot staining the walls. Such fireplaces, associated with finds of flint and bones, are found everywhere in the area of the church and the chapel, both indoors and in the open air. While these fireplaces and finds are somewhat difficult to associate with the monastic community, pilgrims and other visitors might have been responsible for these deposits. Notable are simple charcoal drawings and Arabic graffiti (infra) made on some marble slab fragments found near the fireplaces and probably associated with this or the following phases.

Footnotes

224 Petra church (Fiema 2001: 96-98); Blue Chapel in Petra (Bikai 2002a: 451); Khirbet al-Kerak (Delougaz 1960: 49); Jerash (Baur 1938: 514-515).

225 Tsafrir 1988a: 66-68.

226 Weber 1998: 451.

227 McNicoll, Smith, and Hennessy 1982: 113.

228 For glass intentionally deposited in favissae, see the Petra church (Fiema 2001: 75, 95-69).

229 For the evidence of recycling of glass, see Keller, forthcoming. See also comments in Chapter 12

Phase 12 destruction event - very late 9th or 10th centuries CE

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 107 - Plan of Church and Chapel

Fig. 107 Chapter 6

The church and the chapel marking in Phase 11.

Mikkola et al (2008)

in Phase 11 from Mikkola et al (2008)

Figures

Fig. 114 - Section drawing

Fig. 114.

Trench F, northern baulk, featuring stone tumbles and soil deposits mentioned in the text.

Mikkola et al (2008)

showing stone tumble from Mikkola et al (2008)
Fig. 115 - Collapsed arches

Fig. 115.

Collapsed arches in locus F.05.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 116 - Collapsed arch

Fig. 116

Collapsed arch in locus T.05.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 117 - Trench G, eastern baulk

Fig. 117

Trench G, eastern baulk, featuring the stone tumbles mentioned in the text.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 118 - Collapsed arches

Fig. 118

Collapsed arches in locus G.04

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 34 - Collapsed Column

Fig. 34.

Locus T.14 — the collapsed column

Mikkola et al (2008)

found among Phase 13 stone tumble from Mikkola et al (2008)
Fig. 30 - Charcoal and molten glass

Fig. 30.

Locus V. 15 – the desposit of charcoal and molten glass.

Mikkola et al (2008)

from the Phase 3 fire from Mikkola et al (2008)

Discussion

Mikkola et al (2008) report that all remaining roof structures now collapsed, forming the lowest layer of stone tumble which included remains of wooden roof beams, branches and clayey soil from the structures of the Phase 9 roofs. The thickness of this stone tumble varied significantly from one trench to another, but the average thickness of the layer in the church was ca. 1.5 m and in the chapel as much as 1.8 m. Although gradual decay and periodic earthquakes caused caused more stone falls and soil deposition, the overlying layers of stone tumble were less intensive. Several rows of voussoirs from fallen arches were found among the [Phase 12] tumble in both the church and the chapel.

Underneath the lowermost deposits of stone tumble was a hard-packed, clayey soil which contained relatively few finds and probably represents material fallen from the structures of the roof. The wooden roof beams found within this soil deposit were thought to form the main part of the roof construction while the branches, covered by a thick layer of clayey soil, filled the gaps and helped to create an even surface for the roof. Apparently these roof materials fell first as they were overlain by the arches and other stone elements of the walls. The beams and branches were in a poor state of preservation and heavily carbonized, apparently because of natural decay rather than burning.

Mikkola et al (2008) also found the remains of fallen arches which they described as follows:

Remains of two fallen arches were found in the layer of stone tumble (loci F.04, F.09, F.10, F.ll) in the eastern part of the nave (Fig. 114), one running N-S between the pilasters loci F.07 and F.05d, and one apparently running E-W between the same pilaster (F.05d) to pilaster F.06 (Fig. 115). Clear remains of fallen arches were found in the stone tumble (loci T.05, T.08, T.10) in the western part of the aisle (Fig. 116).

Phase 12 stone tumble and destruction layers in various parts of the Church and Chapel are described by Mikkola et al (2008) in the collapsible panels below. Phase 12 destruction (aka final destruction) was dated via ceramics to the very late 9th or 10th centuries CE.

The Church

Eastern part of the nave

In the eastern part of the nave, the stone tumble (loci G.03 [lower part], G.16, G.17, T.05, T.10, U.03 [lower part], U.10) included a row of voussoirs running from the southern column (locus T.14) towards a pilaster (locus G.06) in the north (Fig. 117). However, as the two supports are not in the same line, the arch cannot have sprung between them. It seems that the force of the earthquake had thrown the northernmost voussoirs towards the west, and that fallen arch originally sprang between the southern column and the pilaster (locus U.26) abutting the northern column.

Central part of the nave

The tumble in the central part of the nave included some drums fallen from the northern column (locus U.25), but it is probable that the entire column did not collapse as some drums were found very close to the surface in the nave.

Northern Aisle

In the stone tumble (loci G.04, G.04a, G.10, G.11, G.14 [top], U.03 [lower part], U.09) above the clayey soil, two rows of voussoirs dearly resulting from fallen arches running N-S were discovered (Fig. 118, also Fig. 117). The first of these - between the column (locus U.25) and pilaster (locus U.17) — was scattered over a large area, testifying to the force of the earthquake. A second row of voussoirs was found between the pilasters (loci U.18 and U.39) in the eastern part of the nave. No remains of fallen arches were discovered in the western part of the northern aisle.

Apse and Bema

Inside the apse, the earthquake of Phase 12 created a layer of stone tumble consisting mainly of crushed, yellowish limestone (loci E.16, F.02, F.10 M.14, U.11).

... The northern pastophorion [of the Church] was filled with a layer of stone tumble (locus E.08 and the lower part of locus E.05). This deposit did not contain any evidence of a fallen arch, only a couple of long voussoirs, which may have been part of the Phase 9 steps (locus E.12) leading up to Wall T. A thick layer of stone tumble (loci M.13, M.15) also fell inside the southern pastophorion where, however, the voussoirs of an arch running N-S were found among the tumble.

Atrium and Narthrex

The stone tumble (loci B.07, L.05, L.06, L.06a, L.08, L.09, X.02, X.04, and X.05; Figs. 46, 58) resulting from Phase 12 destruction is concentrated along the edges of the walls and is not exceedingly heavy. The atrium walls were possibly already much reduced in height, following the previous earthquakes, and the resulting debris cleared in the meanwhile. In the northern part of the atrium, two fallen columns were found among the stone tumble (part of locus X.05). The column standing in the northeastern corner of the atrium has fallen towards the NW. Six drums originally part of this column were found in the tumble. The column to the west of this column had been taller when it collapsed; ten drums in a row running towards the NE were found among the tumble. It is possible that the latter column fell later, sometime in Phase 14, as it appears to have fallen on top of the first column. Most of the stone tumble (locus H.02) in the area of the narthex was caused by this destruction (Col. Fig. 30).

The Chapel

The Phase 12 destruction caused a major collapse in the chapel, resulting in a stone tumble (loci I.02, I.08, I.10, I.15, I.16, Y.05 [lower part], Y.08, Y.24) especially in the western and central parts of the chapel. The four central and western arches of the chapel fell, all the voussoirs belonging to these arches were found in neat rows, resting on the soil of loci Y.09 and I.10. The easternmost arch, however, apparently did not collapse at this point. In addition to the arches, the semidome of the chapel must also have collapsed now. The exterior of Wall S suffered extensive damage and much of the apse wall tumbled towards the east (loci C.3a, C.11). A tangible piece of evidence of collapsing stones in the apse area can be found in the northern cupboard, where the lower shelf (locus Y.10c) had been smashed into pieces. The stones that broke the shelf were removed in the following phase, but the pieces of the broken shelf was left in place.

References

Mikkola et al (2008)

from Mikkola et al (2008)

Plans and Figures

Plans

Fig. 1 - Plan of Church and Chapel

Fig. 1 Chapter 5

The plan of the church and the chapel marking all walls and trench locations.

Fiema (2008)

with all walls and trench locations marked from Fiema (2008)
Fig. 107 - Plan of Church and Chapel

Fig. 107 Chapter 6

The church and the chapel marking in Phase 11.

Mikkola et al (2008)

in Phase 11 from Mikkola et al (2008)
Fig. 29 - plan of the western

Fig. 29

The plan of the western part of the chapel.

Mikkola et al (2008)

part of the chapel from Mikkola et al (2008)

Figures

Fig. 114 - Section drawing

Fig. 114.

Trench F, northern baulk, featuring stone tumbles and soil deposits mentioned in the text.

Mikkola et al (2008)

showing stone tumble from Mikkola et al (2008)
Fig. 115 - Collapsed arches

Fig. 115.

Collapsed arches in locus F.05.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 116 - Collapsed arch

Fig. 116

Collapsed arch in locus T.05.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 117 - Trench G, eastern baulk

Fig. 117

Trench G, eastern baulk, featuring the stone tumbles mentioned in the text.

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 118 - Collapsed arches

Fig. 118

Collapsed arches in locus G.04

Mikkola et al (2008)

from Mikkola et al (2008)
Fig. 30 - Charcoal and molten glass

Fig. 30.

Locus V. 15 – the desposit of charcoal and molten glass.

Mikkola et al (2008)

from the Phase 3 fire from Mikkola et al (2008)

Discussion

Phase 12 Final Destruction

Introduction

from Mikkola et al (2008:148)

The fifth destruction at the site can also be called the final one, at least as far as the church and the chapel are concerned. The event damaged both buildings so badly that after this point, no effort was made to repair or adapt them to any kind of use. There was no attempt to clear the resulting stone tumble, apparently with the exception of the Apse area in the chapel. This observation may be significant, especially as the current state of preservation of the chapel’s structure is better than that of the church. In other words, despite the damage, the chapel could still, theoretically, provide a shelter.²³⁹

Admittedly, and as specified above, some stone tumbles encountered during the excavations may represent the Phase 10 destruction, cleared in Phase 11. The distinguishing of Phase 10 deposits from those related to Phase 12 could not always be fully accomplished. However, the majority of the lowest, heavy layers of stone tumble inside both the church and the chapel may reasonably be assigned to this phase. All remaining roof structures now collapsed, forming the lowest layer of stone tumble. Several rows of the voussoirs from fallen arches were found among the tumble in both the church and the chapel. This lowest layer also includes remains of wooden roof beams, branches and clayey soil from the structures of the Phase 9 roofs. The thickness of the stone tumble varied significantly from one trench to another, but the average thickness of the layer in the church was ca. 1.5 m and in the chapel as much as 1.8 m. As a result of gradual decay and periodic earthquakes, stones continued to fall and soil continued to accumulate inside the ruins even after Phase 12, but this resulted in much less intensive layers of stone tumble.

Footnotes

239 For the further ramifications of this observation, see Chapter 16.

The Church

Introduction

from Mikkola et al (2008:148-149)

Throughout the church interior, the floor was covered with a layer of hard-packed, clayey soil directly under the lowermost deposits of stone tumble. This layer, which contained relatively few finds, probably represents material fallen from the structures of the roof. This is supported by the fact that in the soil were also found some remains of wooden roof beams and branches. The beams no doubt formed the main part of the roof construction while the branches, covered by a thick layer of clayey soil, filled the gaps and helped to create an even surface for the roof. Apparently, the branches, beams and clayey soil were the first part of the roof structure to fall in the earthquake of Phase 12, and were only then followed by the arches and other stone elements of the walls. The beams and branches were in a poor state of preservation and heavily carbonized, apparently because of natural decay rather than burning. Bits of charcoal were found commonly throughout the layer of hard-packed, clayey soil on top of the floor. Like the beams, these are likely to represent carbonized remains of roof supports, probably almost completely disintegrated branches. No distinct layers of fire-related ashy earth were encountered, and it does not seem that there was a fire associated with the final destruction.

Southern Aisle

from Mikkola et al (2008:149)

In the southern aisle, the layer of clayey soil was almost 0.8 m thick in its eastern part (loci F.14, F.12) and ca. 0.6 m thick in the west (locus T.13). A few large stones were also present in the soil, and the layer included a fair amount of marble fragments and crumbled plaster probably originating from the walls. In the western part of the aisle, a badly disintegrated roof beam (length 2.10 m, elevation 45.81 m) was found lying in a SE-NW direction (Fig. 113), and a number of equally poorly preserved branches were on the same level. In the eastern part of the aisle, a roof beam (length 2.35 m, width 0.21 m, thickness 0.08 m) was found almost immediately on top of the marble floor (in locus F.14), with only 0.02 m of sand separating the beam from the floor. Remains of three more roof beams, ca. 2.0 m long and resting in a N-S direction, were found higher up, in locus F.12, (elevation 45.33 m).

Remains of two fallen arches were found in the layer of stone tumble (loci F.04, F.09, F.10, F.11) in the eastern part of the nave (Fig. 114), one running N-S between the pilasters loci F.07 and F.05d, and one apparently running E-W between the same pilaster (F.05d) to pilaster F.06 (Fig. 115). Clear remains of fallen arches were found in the stone tumble (loci T.05, T.08, T.10) in the western part of the aisle (Fig. 116), and in the central part were the ten drums and the capital of the collapsed Phase 4 column in locus T.14. Under the drums, furthermore, was found a fallen Phase 7 pilaster, originally a part of locus T.32, toppled over by the falling column.

Nave

from Mikkola et al (2008:149-150)

The layer of clayey soil (loci T.18, U.21, U.23 [partly]) fallen from the roof structures in the nave was ca. 0.35 m thick. In the western part of the nave, the remains of one carbonized roof beam resting in a N-S position were found together with some branches resting in an E-W position (in the upper part of locus T.21), relatively close to floor level (elevation 45.19). Two more wooden beams were found in the central part of the nave, also resting approximately in a N-S position, ca. 0.10 m above the marble floor (part of locus U.21, elevation 45.22, Col. Fig. 29).

In the eastern part of the nave, the stone tumble (loci G.03 [lower part], G.16, G.17, T.05, T.10, U.03 [lower part], U.10) included a row of voussoirs running from the southern column (locus T.14) towards a pilaster (locus G.06) in the north (Fig. 117). However, as the two supports are not in the same line, the arch cannot have sprung between them. It seems that the force of the earthquake had thrown the northernmost voussoirs towards the west, and that fallen arch originally sprang between the southern column and the pilaster (locus U.26) abutting the northern column. The tumble in the central part of the nave included some drums fallen from the northern column (locus U.25), but it is probable that the entire column did not collapse as some drums were found very close to the surface in the nave.²⁴⁰

Footnotes

240 These drums collapsed later on, probably as a result of natural deterioration and decay of the ruins.

Northern Aisle

from Mikkola et al (2008:150)

The layer of clayey soil (loci U.13, U.32) in the northern aisle was on average ca. 0.30 m thick. Its compactness varied, being much more hard-packed in the eastern part of the aisle, where the roof had been preserved until this point, than in the west, where the westernmost arch, and associated beams, branches and clayey soil, had apparently already collapsed in Phase 10. The soil did not include any well-preserved roof beams, only small fragments. The best preserved of these was a fragment of a roof beam found 1.0 m west of the doorway of the northern pastophorion, ca. 0.20 m above the floor level. No remains of beams were discovered in the western parts of the northern aisle.

In the stone tumble (loci G.04, G.04a, G.10, G.11, G.14 [top], U.03 [lower part], U.09) above the clayey soil, two rows of voussoirs clearly resulting from fallen arches running N-S were discovered (Fig. 118, also Fig. 117). The first of these – between the column (locus U.25) and pilaster (locus U.17) – was scattered over a large area, testifying to the force of the earthquake. A second row of voussoirs was found between the pilasters (loci U.18 and U.39) in the eastern part of the nave. No remains of fallen arches were discovered in the western part of the northern aisle.

Apse and Bema

from Mikkola et al (2008:150)

Inside the Apse, the earthquake of Phase 12 created a layer of stone tumble consisting mainly of crushed, yellowish limestone (loci E.16, F.02, F.10, M.14, U.11). Although the surviving (lower) part of Wall T is mainly made of sandstone blocks, the upper courses of it were apparently constructed of limestone, a few blocks of which can still be seen in situ. When they collapsed inside the Apse, the tumble formed a pit, in the middle of which the enigmatic tessera pile was deposited in the following phase.

Pastophoria

from Mikkola et al (2008:150)

The strata excavated inside the pastophoria did not include a comparable layer of hard-packed clayey soil as found deposited on the aisles and the nave of the church. It is probable that these rooms were already partially deprived of roofing when the disaster struck in Phase 11. The northern pastophorion was filled with a layer of stone tumble (locus E.08 and the lower part of locus E.05). This deposit did not contain any evidence of a fallen arch, only a couple of long voussoirs, which may have been part of the Phase 9 steps (locus E.12) leading up to Wall T. A thick layer of stone tumble (loci M.13, M.15) also fell inside the southern pastophorion where, however, the voussoirs of an arch running N-S were found among the tumble

Atrium and Narthex

from Mikkola et al (2008:150-151)

The stone tumble (loci B.07, L.05, L.06, L.06a, L.08, L.09, X.02, X.04, and X.05; Figs. 46, 58) resulting from Phase 11 destruction is concentrated along the edges of the walls and is not exceedingly heavy. The Atrium walls were possibly already much reduced in height, following the previous earthquakes, and the resulting debris cleared in the meanwhile. In the central part of the Atrium, the strata above the floor consisted only of windblown sand. In the northern part of the Atrium, two fallen columns were found among the stone tumble (part of locus X.05).

The column standing in the northeastern corner of the Atrium has fallen towards the NW. Six drums originally part of this column were found in the tumble. The column to the west of this column had been taller when it collapsed; ten drums in a row running towards NE were found among the tumble. It is possible that the latter column fell later, sometime in Phase 14, as it appears to have fallen on top of the first column.

Most of the stone tumble (locus H.02) in the area of the Narthex was caused by this destruction (Col. Fig. 30). The tumble included some column drums fallen from the columns of the stylobate, and a row of four doorjambs west of the northern doorway. The northern doorway was filled with cobblestones and sand (locus H.04).

The Chapel

from Mikkola et al (2008:151)

The Phase 12 destruction caused a major collapse in the chapel, resulting in a stone tumble (loci I.02, I.08, I.10, I.15, I.16, Y.05 [lower part], Y.08, Y.24) especially in the western and central parts of the chapel. The four central and western arches of the chapel fell, all the voussoirs belonging to these arches were found in neat rows, resting on the soil of loci Y.09 and I.10. The easternmost arch, however, apparently did not collapse at this point. In addition to the arches, the semidome of the chapel must also have collapsed now. However, as relatively little stone tumble was found inside the Apse, the tumble associated with the collapse was probably cleared in the following phase. The exterior of Wall S suffered extensive damage and much of the Apse wall tumbled towards the east (loci C.3a, C.11). A tangible piece of evidence of collapsing stones in the Apse area can be found in the northern cupboard, where the lower shelf (locus Y.10c) had been smashed into pieces. The stones that broke the shelf were removed in the following phase, but the pieces of the broken shelf was left in place.

Dating

from Mikkola et al (2008:151)

As mentioned above, the ceramics associated with the destruction in Phase 12 indicate that this event should have happened sometime in the very late 9th or in the course of the 10th century.

Orientation of presumed seismic damage

from Mikkola et al (2008)

In general, the E-W running walls are better preserved than those running N-S. This fact is probably explained by the seismic characteristics prevalent in the Wadi Araba rift valley, which mainly result in earthquakes exhibiting E-W movement. These are likely to cause more damage to walls running in a N-S direction than to those running E-W.

Pre-Monastic Phase IV Destruction Event - 363 CE or an earthquake from around that time

Effect	Location	Image(s)	Description
Shattered floor	second story floor O.41 in Room 25		A dramatic piece of evidence the shattered second story floor (O.41), some remains of which are still protruding from Wall (e.g. Fig. 8) - Appendix C of Fiema and Frosen (2008)
Collapsed walls	core of Western Building		The core of Western Building must have partially collapsed and the second story was entirely destroyed, as remains of its floor were incorporated in the Byzantine structures. - Appendix C of Fiema and Frosen (2008)
Collapsed arches	southern cistern (Room 36)	Fig. 2 The interior of the cistern at the foot of the summit of Jabal Harun. JW: Unsure if this is the southern cistern Mikkola et al (2008)	The superstructure and arches of the southern cistern (Room 36) may also have collapsed - Appendix C of Fiema and Frosen (2008)

Phase 3 Destruction Event - mid to late 6th century CE

Church

**Church**
Effect	Location	Image(s)	Description
Roof collapse	Church Roof Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		the roof support system was severely damaged is indicated, among other ways, by the fact that it was completely rearranged in the following phase - Mikkola et. al. (2008) Various kinds of debris ended up in the fills of the walls, especially in Wall I which was constructed in Phase 4. In fact, a large portion of the finds of broken marble furnishing, pottery, glass, nails and roof tiles, found in the late layers of stone tumble, derive from the interior of the repaired walls and therefore predate Phase 3. - Mikkola et. al. (2008)
Wall collapse	Upper parts of walls in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		collapse of the upper parts of walls - Mikkola et. al. (2008) In many parts of the church, ... clerestory walls ... collapsed - Mikkola et. al. (2008) The intensity of the event is also indicated by the evidence of repairs to the upper portions of the walls of the church - Mikkola et. al. (2008)
Arch collapse	Arches of the church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		In many parts of the church, the arches ... collapsed - Mikkola et. al. (2008)
Fallen Columns	columns of the church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		In many parts of the church, ... columns... collapsed - Mikkola et. al. (2008)
Shattered Floors	Marble floors of the church - particularly locus F.24 in the nave Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	Fig. 44. The appearance of the marble floor in the nave of the church. JW: This is after this floor was haphazardly repaired in the following phase [i.e. after phase 3]. Mikkola et al (2008)	The falling stones shattered the marble floor ... of the church ... and ... the floor was haphazardly repaired in the following phase - Mikkola et. al. (2008) The repaired walls of Phase 4 feature numerous fragments of marble slabs from the floor of Phase 2, now used as chinking stones. - Mikkola et. al. (2008)
Broken Furnishings	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		The falling stones shattered the ... furnishings of the church ... and ... much of the furnishings were apparently damaged beyond repair. This is evidenced by the numerous fragments of marble colonnettes, chancel screens, etc., found in reused positions in the structures of Phase 4. - Mikkola et. al. (2008) Various kinds of debris ended up in the fills of the walls, especially in Wall I which was constructed in Phase 4. In fact, a large portion of the finds of broken marble furnishing, pottery, glass, nails and roof tiles, found in the late layers of stone tumble, derive from the interior of the repaired walls and therefore predate Phase 3. - Mikkola et. al. (2008)
Fire	Possibly the entire site Fig. 2 Chapter 5 The plan of the monastery following the 2005 field season. Fiema (2008)	Fig. 30. Locus V. 15 – the deposit of charcoal and molten glass. Mikkola et al (2008)	It appears that the seismic shock caused the collapse of the upper parts of walls, and the burning oil lamps, falling on the floor, caused the conflagration. - Mikkola et. al. (2008) The totality of the destruction was ensured by a fire that raged in the buildings. The fire was hot enough to melt the glass lamps of the polycandela and leave marks of burning on much of the marble decoration of the church. A deposit (locus V.15) of charcoal and broken, partly melted, lamp fragments, probably associated with Phase 3, was found west of the chapel (Fig. 30). This locus seems to represent burned material cleared from the church and the chapel immediately after the fire and the earthquake. Several of the large iron nails – probably originally used in the structures of the roof – found during the excavations, similarly exhibit signs of exposure to intense heat. - Mikkola et. al. (2008)
Debris	from various parts of the site but deposited in Trench R Fig. 2 Chapter 5 The plan of the monastery following the 2005 field season. JW: Trench R is in the lower left corner of drawing Fiema (2008)		Mikkola et. al. (2008) reports that large quantities of debris, including charcoal, burnt tiles, broken and fire-damaged glass and ceramic sherds, and pieces of marble and other stones, were found in the midden located outside the monastery enclosure, excavated in Trench R. The uniformity of the debris led the excavators to conclude that they were refuse from a fire-related destruction - cleared from the Church and Chapel before re-building activity began in Phase 4.

Chapel

**Chapel**
Effect	Location	Image(s)	Description
Roof collapse	Chapel Roof Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		This is indicated by the extent of the repairs made in Phase 4, particularly by the complete rearrangement of the roof supports - Mikkola et. al. (2008) Various kinds of debris ended up in the fills of the walls, especially in Wall I which was constructed in Phase 4. In fact, a large portion of the finds of broken marble furnishing, pottery, glass, nails and roof tiles, found in the late layers of stone tumble, derive from the interior of the repaired walls and therefore predate Phase 3. - Mikkola et. al. (2008)
Wall collapse	Upper parts of walls in the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		collapse of the upper parts of walls - Mikkola et. al. (2008) The intensity of the event is also indicated by the evidence of repairs to the upper portions of the walls of ... the chapel. - Mikkola et. al. (2008)
Wall collapse	Western Wall of the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		The original western wall of the chapel also seems to have collapsed to the extent that it was deemed easier to build a new wall (Wall OO) - Mikkola et. al. (2008)
Wall collapse	Wall H of the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		parts of Wall H also appear to have been badly damaged, as its upper courses were rebuilt in the following phase, using large quantities of recycled material. - Mikkola et. al. (2008)
Fallen Columns	the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		The Phase 4 columns of the chapel, moreover, seem to derive from the collapsed columns of Phase 2 structures, as some of the drums used in them are broken. - Mikkola et. al. (2008)
Shattered Floors	Marble floors of the chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		The falling stones shattered the marble floor ... of ... the chapel, and ... the floor was haphazardly repaired in the following phase - Mikkola et. al. (2008) The repaired walls of Phase 4 feature numerous fragments of marble slabs from the floor of Phase 2, now used as chinking stones. - Mikkola et. al. (2008)
Broken Furnishings	The Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		The falling stones shattered the ... furnishings of ... the chapel ... and ... much of the furnishings were apparently damaged beyond repair. This is evidenced by the numerous fragments of marble colonnettes, chancel screens, etc., found in reused positions in the structures of Phase 4. - Mikkola et. al. (2008) Various kinds of debris ended up in the fills of the walls, especially in Wall I which was constructed in Phase 4. In fact, a large portion of the finds of broken marble furnishing, pottery, glass, nails and roof tiles, found in the late layers of stone tumble, derive from the interior of the repaired walls and therefore predate Phase 3. - Mikkola et. al. (2008)
Broken Pilasters	The chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)		The system of pilasters now visible in the chapel is not original, as is evidenced by the presence of wall plaster behind the pilasters - Mikkola et. al. (2008)
Fire	Possibly the entire site Fig. 2 Chapter 5 The plan of the monastery following the 2005 field season. Fiema (2008)	Fig. 30. Locus V. 15 – the deposit of charcoal and molten glass. Mikkola et al (2008)	It appears that the seismic shock caused the collapse of the upper parts of walls, and the burning oil lamps, falling on the floor, caused the conflagration. - Mikkola et. al. (2008) The totality of the destruction was ensured by a fire that raged in the buildings. The fire was hot enough to melt the glass lamps of the polycandela and leave marks of burning on much of the marble decoration of the church. A deposit (locus V.15) of charcoal and broken, partly melted, lamp fragments, probably associated with Phase 3, was found west of the chapel (Fig. 30). This locus seems to represent burned material cleared from the church and the chapel immediately after the fire and the earthquake. Several of the large iron nails – probably originally used in the structures of the roof – found during the excavations, similarly exhibit signs of exposure to intense heat. - Mikkola et. al. (2008)
Debris	from various parts of the site but deposited in Trench R Fig. 2 Chapter 5 The plan of the monastery following the 2005 field season. JW: Trench R is in the lower left corner of drawing Fiema (2008)		Mikkola et. al. (2008) reports that large quantities of debris, including charcoal, burnt tiles, broken and fire-damaged glass and ceramic sherds, and pieces of marble and other stones, were found in the midden located outside the monastery enclosure, excavated in Trench R. The uniformity of the debris led the excavators to conclude that they were refuse from a fire-related destruction - cleared from the Church and Chapel before re-building activity began in Phase 4.

Phase 6 Destruction Event (?) - 1st half of 7th century CE - inferred from remodelling and may not have been a response to an earthquake

only better attested damage is listed

Effect	Location	Description
Arch Collapse	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008)	At least some of the arches of the Phase 4 church seem to have fallen, probably causing heavy damage to the floor and furnishings of the church, but, curiously, the two columns of Phase 4 were left standing. - Mikkola et. al. (2008)
Possible Wall damage	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008)	Judging by the methods and materials used, the marble floor has been damaged and repaired on several occasions, but it is difficult to attribute a particular series of repairs to the damage inflicted in Phase 6. The same is true of the walls, which have been damaged and repaired on several occasions, but there are few chronological indications to help to date a particular series of repairs. One may only conclude that some of the damage to which the floor and walls bear witness might have been caused by the Phase 6 event. - Mikkola et. al. (2008)
Possible Floor damage	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008)	Judging by the methods and materials used, the marble floor has been damaged and repaired on several occasions, but it is difficult to attribute a particular series of repairs to the damage inflicted in Phase 6. The same is true of the walls, which have been damaged and repaired on several occasions, but there are few chronological indications to help to date a particular series of repairs. One may only conclude that some of the damage to which the floor and walls bear witness might have been caused by the Phase 6 event. - Mikkola et. al. (2008)

Phase 8 Destruction Event - mid 8th century CE

Effect	Location	Image(s)	Description
Vault collapse	Semidome over apse in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		collapse of the church's semidome - Mikkola et. al. (2008) The semidome covering the apse must have collapsed in the earthquake, destroying the floor of the apse beyond repair. - Mikkola et. al. (2008) The resulting tumble was cleared in the following phase but the semidome and the apse floor were never repaired. - Mikkola et. al. (2008)
Tilted Wall	Wall J in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)	Fig. 80. Wall J tilting southward, following the earthquake of Phase 8. Mikkola et al (2008) Fig. 103 The Phase 9 buttress, locus T.31, built against Wall J. (in the center). Mikkola et al (2008)	the earthquake caused Wall J to severely tilt towards the south - Mikkola et. al. (2008) The wall was left leaning towards the south and it had to be supported by a buttress in the following phase - Mikkola et. al. (2008) A number of structures located outside the church were investigated in the course of excavation. The largest and perhaps most significant of these is the long buttress (locus T.31), built against Wall J (Fig. 103). The assignment of this buttress to Phase 9 is certain; it was clearly built after the wall tilted south in Phase 8. Therefore, it is likely that the buttress was built to support the wall against potential earth tremors. - Mikkola et. al. (2008)
Tilted Wall	Wall GG in the Chapel (part of locus Y.13) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The only sign of damage [in the Chapel] is found outside Wall GG, where a supporting buttress (part of locus Y.13) was built against the northern side of the wall in the following phase. - Mikkola et. al. (2008) the construction of the buttress against Wall GG in the following phase indicates that either some damage was inflicted in Phase 8, or it was a preventive measure. - Mikkola et. al. (2008) The walls of the chapel seem to have withstood the event of Phase 8, in spite of the fact that it caused so much damage to the church. However, the walls probably suffered some structural damage. This is suggested by the construction of stone buttresses outside and against Wall GG. - Mikkola et. al. (2008)
Tilted Walls ?	southeastern corner of the atrium, against Wall I Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)	Fig. 99 The Phase 9 buttress, loci B.02, B.16, B.18 and L.02, in the atrium (in the center). Mikkola et al (2008) Fig. 36 The plan of the southwestern part of the church. Mikkola et al (2008) Fig. 58 Trench B, northern baulk. Featuring the stratigraphy in the atrium, including the pavement, the water channel, locus B.12, the buttress, locus B.02, from Phase 9, and the stone tumbles. Mikkola et al (2008)	The most significant element of Phase 9 in the atrium is, however, the construction of a massive platform or buttress (loci B.02, B.16 [fill], B.18 [facade], and L.02) in the southeastern corner of the atrium, against Wall I (Fig. 99, also Figs. 36 and 58). - Mikkola et. al. (2008)
Limited Upper Wall collapse and roof collapse ?	southern pastophorion in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the southern pastophorion, falling stones caused severe damage to the floor due the presence of hollow compartments underneath. The part of floor that covered the southern compartment was destroyed and never repaired. It is uncertain if the arch there collapsed as well. It may have been left standing, but the roof was nonetheless severely damaged. - Mikkola et. al. (2008)
Displaced, Folded, or Collapsed Walls	section of Wall H in the atrium (loci V.06, X.13) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		a section of Wall H in the atrium (loci V.06, X.13) was rebuilt. - Mikkola et. al. (2008)
Folded or Displaced Walls ? (suggested by possible pilaster damage	atrium in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The square pilaster (locus L.14) or pedestal in the eastern part of the atrium was also probably destroyed - Mikkola et. al. (2008)
Arch collapse	N-S arch spanning the Southern Aisle in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		the earthquake caused Wall J to severely tilt towards the south (Fig. 80), causing the collapse of the arches in the southern aisle - Mikkola et. al. (2008) collapse of one N-S arch in the aisle - Mikkola et. al. (2008)
Arch collapse	N-S arches spanning the nave in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In addition to the arches of the southern aisle, those spanning the nave appear to have collapsed. Such a pattern of collapse would indeed be expected. With the mutual supporting arch and beam system introduced in Phase 7, the collapse of one N-S arch in the aisle would have seriously impaired the stability of the corresponding N-S arch across the nave - Mikkola et. al. (2008)
Arch collapse	northern pastophorion in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The arch supporting the roof of the northern pastophorion probably fell too. - Mikkola et. al. (2008)
Arch collapse ?	southern pastophorion in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the southern pastophorion, falling stones caused severe damage to the floor due the presence of hollow compartments underneath. The part of floor that covered the southern compartment was destroyed and never repaired. It is uncertain if the arch there collapsed as well. It may have been left standing, but the roof was nonetheless severely damaged. - Mikkola et. al. (2008) The arch covering the southern pastophorion most likely collapsed in Phase 8, considering the fact that the entire southern wall of the basilica was severely affected by the destruction. Therefore, unlike the one in the northern pastophorion, the arch must have been rebuilt in Phase 9, as is evidenced by the discovery of the collapsed voussoirs of a fallen arch found among the stone tumble inside the room (locus M.04). - Mikkola et. al. (2008)
collonade and Column Collapse	atrium in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the atrium, parts of the collonades collapsed. - Mikkola et. al. (2008) The fallen columns of the atrium were not re-erected, but were cleared away and used elsewhere. - Mikkola et. al. (2008) collapse of ... the columns of the atrium - Mikkola et. al. (2008)
Shattered Floor	Floor of the apse in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The semidome covering the apse must have collapsed in the earthquake, destroying the floor of the apse beyond repair. - Mikkola et. al. (2008) The resulting tumble was cleared in the following phase - Mikkola et. al. (2008)
Shattered Floor	Floor of the apse in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the southern pastophorion, falling stones caused severe damage to the floor due the presence of hollow compartments underneath. The part of floor that covered the southern compartment was destroyed and never repaired. - Mikkola et. al. (2008)
Shattered Floors ?	atrium in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the atrium, parts of the collonades collapsed. The atrium floor shows damage, but it is again difficult to determine whether it was damaged in this phase. - Mikkola et. al. (2008)
Shattered Floor ?	narthex in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The mosaic in the narthex shows damage, especially in the central medallion, which was never repaired. Dating of the damage is uncertain - it may have been caused by the events of either Phase 8 or 10 - Mikkola et. al. (2008)
Unspecified damage (probably from falling objects)	bema in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The apse and bema also suffered heavy damage in Phase 8. - Mikkola et. al. (2008) The resulting tumble was cleared in the following phase - Mikkola et. al. (2008)

Phase 10 Destruction Event - late 8th or early 9th century CE

Effect	Location	Description
Cratered Floor	western part of the nave and the northern aisle in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	The most obvious evidence of this destruction consists of craters left in the church floor by tumbling stones. The marble floor was badly damaged in especially in the western part of the nave and the northern aisle, where much of the floor was removed in the following phase. - Mikkola et. al. (2008) Several depressions left in the floor (locus T.29) of the nave mark the places hit by the falling stones. The stones that caused the depressions were, however, removed in Phase 11. - Mikkola et. al. (2008)
Collapsed Arch	N-S arch running between pilasters T.04 and G.06 in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	It seems probable that the long N-S arch running between pilasters T.04 and G.06 collapsed in this phase. - Mikkola et. al. (2008)
Collapsed Arch	westernmost arch in the northern aisle and the one that spanned the easternmost part of the nave Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Indirect evidence also exists for the collapse of the westernmost arch in the northern aisle and the one that spanned the easternmost part of the nave, for in these areas the marble floor was removed in Phase 11. It seems reasonable to assume that the removal of the floors was related to the damage caused by stones falling from the arches and other structures of the roof, whereas the floor was left untouched in those parts of the church where the arches did not collapse. - Mikkola et. al. (2008)
Collapse	lowest parts of locus H.02 in the area of the Narthex Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	As the walls and columns of the Atrium and the Narthex had been badly damaged and already partially removed in Phases 8 and 9, they probably were not heavily affected by the destruction of Phase 10. However, some of the stone tumble (lowest parts of locus H.02) in the area of the Narthex may have been caused by this event. - Mikkola et. al. (2008)
Roof Collapse ?	floor of the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	As before, the chapel was less affected by the calamity than the church. It seems that none of the arches supporting the roof of the chapel fell, and consequently, all the walls and other structures of the chapel seem to have survived reasonably well. However, it is possible that some elements fell from the roof or the walls, since the floor of the chapel was partially removed in the next phase. This may also be related to the lack of permanent occupation and resultant collecting or looting activities in Phase 11, but as much of the floor was made of sandstone and of little value, it is probable that the floor already suffered some damage in this phase. - Mikkola et. al. (2008)

Phase 12 destruction event - very late 9th or 10th centuries CE

Church

Effect	Location	Image(s)	Description
Roof collapse	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		Throughout the church interior, the floor was covered with a layer of hard-packed, clayey soil directly under the lowermost deposits of stone tumble. This layer, which contained relatively few finds, probably represents material fallen from the structures of the roof. This is supported by the fact that in the soil were also found some remains of wooden roof beams and branches. The beams no doubt formed the main part of the roof construction while the branches, covered by a thick layer of clayey soil, filled the gaps and helped to create an even surface for the roof. Apparently, the branches, beams and clayey soil were the first part of the roof structure to fall in the earthquake of Phase 12, and were only then followed by the arches and other stone elements of the walls. The beams and branches were in a poor state of preservation and heavily carbonized, apparently because of natural decay rather than burning. Bits of charcoal were found commonly throughout the layer of hard-packed, clayey soil on top of the floor. Like the beams, these are likely to represent carbonized remains of roof supports, probably almost completely disintegrated branches. No distinct layers of fire-related ashy earth were encountered, and it does not seem that there was a fire associated with the final destruction. - Mikkola et. al. (2008) All remaining roof structures now collapsed, forming the lowest layer of stone tumble - Mikkola et. al. (2008) This lowest layer also includes remains of wooden roof beams, branches and clayey soil from the structures of the Phase 9 roofs. - Mikkola et. al. (2008)
Roof collapse	nave - loci T.18, U.21, U.23 [partly], T.21 (upper part of locus relatively close to floor level (elevation 45.19) in the western part of nave), part of locus U.21, elevation 45.22 in the central part of the nave ca. 0.10 m above the marble floor Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 29 The plan of the western part of the chapel. Mikkola et al (2008)	The layer of clayey soil (loci T.18, U.21, U.23 [partly]) fallen from the roof structures in the nave was ca. 0.35 m thick. In the western part of the nave, the remains of one carbonized roof beam resting in a N-S position were found together with some branches resting in an E-W position (in the upper part of locus T.21), relatively close to floor level (elevation 45.19). Two more wooden beams were found in the central part of the nave, also resting approximately in a N-S position, ca. 0.10 m above the marble floor (part of locus U.21, elevation 45.22, Col. Fig. 29). - Mikkola et. al. (2008) All remaining roof structures now collapsed, forming the lowest layer of stone tumble - Mikkola et. al. (2008) This lowest layer also includes remains of wooden roof beams, branches and clayey soil from the structures of the Phase 9 roofs. - Mikkola et. al. (2008)
Roof collapse	Southern Aisle - eastern and western parts and locus F.12 (elevation 45.33 m) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		In the western part of the aisle, a badly disintegrated roof beam (length 2.10 m, elevation 45.81 m) was found lying in a SE-NW direction (Fig. 113), and a number of equally poorly preserved branches were on the same level. In the eastern part of the aisle, a roof beam (length 2.35 m, width 0.21 m, thickness 0.08 m) was found almost immediately on top of the marble floor (in locus F.14), with only 0.02 m of sand separating the beam from the floor. Remains of three more roof beams, ca. 2.0 m long and resting in a N-S direction, were found higher up, in locus F.12, (elevation 45.33 m). - Mikkola et. al. (2008) All remaining roof structures now collapsed, forming the lowest layer of stone tumble - Mikkola et. al. (2008) This lowest layer also includes remains of wooden roof beams, branches and clayey soil from the structures of the Phase 9 roofs. - Mikkola et. al. (2008)
Roof collapse	Northern Aisle - eastern parts Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The layer of clayey soil (loci U.13, U.32) in the northern aisle was on average ca. 0.30 m thick. Its compactness varied, being much more hard-packed in the eastern part of the aisle, where the roof had been preserved until this point, than in the west, where the westernmost arch, and associated beams, branches and clayey soil, had apparently already collapsed in Phase 10. The soil did not include any well-preserved roof beams, only small fragments. The best preserved of these was a fragment of a roof beam found 1.0 m west of the doorway of the northern pastophorion, ca. 0.20 m above the floor level. No remains of beams were discovered in the western parts of the northern aisle. - Mikkola et. al. (2008)
Collapsed Walls	southern aisle - loci F.14, F.12 in the east and locus T.13 in the west Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		In the southern aisle, the layer of clayey soil was almost 0.8 m thick in its eastern part (loci F.14, F.12) and ca. 0.6 m thick in the west (locus T.13). A few large stones were also present in the soil, and the layer included a fair amount of marble fragments and crumbled plaster probably originating from the walls - Mikkola et. al. (2008)
Collapsed Walls	Wall T inside the Apse - loci E.16, F.02, F.10, M.14, and U.11 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		Inside the Apse, the earthquake of Phase 12 created a layer of stone tumble consisting mainly of crushed, yellowish limestone (loci E.16, F.02, F.10, M.14, U.11). Although the surviving (lower) part of Wall T is mainly made of sandstone blocks, the upper courses of it were apparently constructed of limestone, a few blocks of which can still be seen in situ. When they collapsed inside the Apse, the tumble formed a pit, in the middle of which the enigmatic tessera pile was deposited in the following phase. - Mikkola et. al. (2008)
Arch collapse	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		Several rows of the voussoirs from fallen arches were found among the tumble in both the church and the chapel - Mikkola et. al. (2008)
Arch collapse	eastern part of the nave one running N-S between the pilasters loci F.07 and F.05d, and one apparently running E-W between the same pilaster (F.05d) to pilaster F.06 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 114. Trench F, northern baulk, featuring stone tumbles and soil deposits mentioned in the text. Mikkola et al (2008) Fig. 115. Collapsed arches in locus F.05. Mikkola et al (2008)	Remains of two fallen arches were found in the layer of stone tumble (loci F.04, F.09, F.10, F.11) in the eastern part of the nave (Fig. 114), one running N-S between the pilasters loci F.07 and F.05d, and one apparently running E-W between the same pilaster (F.05d) to pilaster F.06 (Fig. 115). - Mikkola et. al. (2008)
Arch collapse	eastern part of the nave - fallen arch originally sprang between the southern column and the pilaster (locus U.26) abutting the northern column Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 117 Trench G, eastern baulk, featuring the stone tumbles mentioned in the text. Mikkola et al (2008)	In the eastern part of the nave, the stone tumble (loci G.03 [lower part], G.16, G.17, T.05, T.10, U.03 [lower part], U.10) included a row of voussoirs running from the southern column (locus T.14) towards a pilaster (locus G.06) in the north (Fig. 117). However, as the two supports are not in the same line, the arch cannot have sprung between them. It seems that the force of the earthquake had thrown the northernmost voussoirs towards the west, and that fallen arch originally sprang between the southern column and the pilaster (locus U.26) abutting the northern column. - Mikkola et. al. (2008)
Arch collapse	Southern Aisle - western part Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 116 Collapsed arch in locus T.05. Mikkola et al (2008)	Clear remains of fallen arches were found in the stone tumble (loci T.05, T.08, T.10) in the western part of the aisle (Fig. 116), and in the central part were the ten drums and the capital of the collapsed Phase 4 column in locus T.14. Under the drums, furthermore, was found a fallen Phase 7 pilaster, originally a part of locus T.32, toppled over by the falling column. - Mikkola et. al. (2008)
Arch collapse	Northern Aisle - first one between the column (locus U.25) and pilaster (locus U.17) and second one between the pilasters (loci U.18 and U.39) in the eastern part of the nave Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 118 Collapsed arches in locus G.04 Mikkola et al (2008) Fig. 117 Trench G, eastern baulk, featuring the stone tumbles mentioned in the text. Mikkola et al (2008)	In the stone tumble (loci G.04, G.04a, G.10, G.11, G.14 [top], U.03 [lower part], U.09) above the clayey soil, two rows of voussoirs clearly resulting from fallen arches running N-S were discovered (Fig. 118, also Fig. 117). The first of these – between the column (locus U.25) and pilaster (locus U.17) – was scattered over a large area, testifying to the force of the earthquake. A second row of voussoirs was found between the pilasters (loci U.18 and U.39) in the eastern part of the nave. No remains of fallen arches were discovered in the western part of the northern aisle. - Mikkola et. al. (2008)
Arch collapse	southern pastophorion - loci M.13, M.15 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The strata excavated inside the pastophoria did not include a comparable layer of hard-packed clayey soil as found deposited on the aisles and the nave of the church. It is probable that these rooms were already partially deprived of roofing when the disaster struck in Phase 11. The northern pastophorion was filled with a layer of stone tumble (locus E.08 and the lower part of locus E.05). This deposit did not contain any evidence of a fallen arch, only a couple of long voussoirs, which may have been part of the Phase 9 steps (locus E.12) leading up to Wall T. A thick layer of stone tumble (loci M.13, M.15) also fell inside the southern pastophorion where, however, the voussoirs of an arch running N-S were found among the tumble - Mikkola et. al. (2008)
Fallen Columns	central part of the nave included some drums fallen from the northern column (locus U.25) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The tumble in the central part of the nave included some drums fallen from the northern column (locus U.25), but it is probable that the entire column did not collapse as some drums were found very close to the surface in the nave. - Mikkola et. al. (2008)
Fallen Columns	northern part of the Atrium - part of locus X.05) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The stone tumble (loci B.07, L.05, L.06, L.06a, L.08, L.09, X.02, X.04, and X.05; Figs. 46, 58) resulting from Phase 11 destruction is concentrated along the edges of the walls and is not exceedingly heavy. The Atrium walls were possibly already much reduced in height, following the previous earthquakes, and the resulting debris cleared in the meanwhile. In the central part of the Atrium, the strata above the floor consisted only of windblown sand. In the northern part of the Atrium, two fallen columns were found among the stone tumble (part of locus X.05). - Mikkola et. al. (2008)
Fallen Columns	locus H.02 in the area of the Narthex Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 30. Locus V. 15 – the desposit of charcoal and molten glass. Mikkola et al (2008)	Most of the stone tumble (locus H.02) in the area of the Narthex was caused by this destruction (Col. Fig. 30). The tumble included some column drums fallen from the columns of the stylobate, and a row of four doorjambs west of the northern doorway. The northern doorway was filled with cobblestones and sand (locus H.04). - Mikkola et. al. (2008)
Fallen Columns ?	northeastern corner of the Atrium Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The column standing in the northeastern corner of the Atrium has fallen towards the NW. Six drums originally part of this column were found in the tumble. The column to the west of this column had been taller when it collapsed; ten drums in a row running towards NE were found among the tumble. It is possible that the latter column fell later, sometime in Phase 14, as it appears to have fallen on top of the first column. - Mikkola et. al. (2008)
Broken Doorjambs	west of the northern doorway of the Narthex Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 30. Locus V. 15 – the desposit of charcoal and molten glass. Mikkola et al (2008)	Most of the stone tumble (locus H.02) in the area of the Narthex was caused by this destruction (Col. Fig. 30). The tumble included some column drums fallen from the columns of the stylobate, and a row of four doorjambs west of the northern doorway. The northern doorway was filled with cobblestones and sand (locus H.04). - Mikkola et. al. (2008)

Chapel

Effect	Location	Description
Roof collapse	Church and CHapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	All remaining roof structures now collapsed, forming the lowest layer of stone tumble - Mikkola et. al. (2008) This lowest layer also includes remains of wooden roof beams, branches and clayey soil from the structures of the Phase 9 roofs. - Mikkola et. al. (2008)
Vault collapse	inside the Apse Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	the semidome of the chapel must also have collapsed now. However, as relatively little stone tumble was found inside the Apse, the tumble associated with the collapse was probably cleared in the following phase - Mikkola et. al. (2008)
Collapsed Walls	Wall S and Apse wall Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	The exterior of Wall S suffered extensive damage and much of the Apse wall tumbled towards the east (loci C.3a, C.11). A tangible piece of evidence of collapsing stones in the Apse area can be found in the northern cupboard, where the lower shelf (locus Y.10c) had been smashed into pieces. The stones that broke the shelf were removed in the following phase, but the pieces of the broken shelf was left in place. - Mikkola et. al. (2008)
Arch collapse	The four central and western arches of the chapel - collapsed arches resting on the soil of loci Y.09 and I.10 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Several rows of the voussoirs from fallen arches were found among the tumble in both the church and the chapel - Mikkola et. al. (2008) The four central and western arches of the chapel fell, all the voussoirs belonging to these arches were found in neat rows, resting on the soil of loci Y.09 and I.10. The easternmost arch, however, apparently did not collapse at this point. - Mikkola et. al. (2008)
Stone tumble	Chapel - (loci I.02, I.08, I.10, I.15, I.16, Y.05 [lower part], Y.08, Y.24) especially in the western and central parts of the chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	The Phase 12 destruction caused a major collapse in the chapel, resulting in a stone tumble (loci I.02, I.08, I.10, I.15, I.16, Y.05 [lower part], Y.08, Y.24) especially in the western and central parts of the chapel - Mikkola et. al. (2008)

Phase 8 Destruction Event - mid 8th century CE

Modified by JW from Mikkola et al (2008)

Deformation Map

modified by JW from Fig. 1 of Mikkola et al (2008)

Phase 10 Destruction Event - late 8th or early 9th century CE

Modified by JW from Mikkola et al (2008)

Deformation Map

modified by JW from Fig. 1 of Mikkola et al (2008)

Phase 12 destruction event - very late 9th or 10th centuries CE

Modified by JW from Mikkola et al (2008)

Deformation Map

modified by JW from Fig. 1 of Mikkola et al (2008)

Pre-Monastic Phase IV Destruction Event - 363 CE or an earthquake from around that time

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Description	Intensity
Fractures, fold, and popups on pavements	second story floor O.41 in Room 25	A dramatic piece of evidence the shattered second story floor (O.41), some remains of which are still protruding from Wall (e.g. Fig. 8) - Appendix C of Fiema and Frosen (2008)	VI + (probably higher since a different type of damage is indicated)
Collapsed walls	core of Western Building	The core of Western Building must have partially collapsed and the second story was entirely destroyed, as remains of its floor were incorporated in the Byzantine structures. - Appendix C of Fiema and Frosen (2008)	VIII +
Collapsed arches	southern cistern (Room 36)	The superstructure and arches of the southern cistern (Room 36) may also have collapsed - Appendix C of Fiema and Frosen (2008)	VI +

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Phase 3 Destruction Event - mid to late 6th century CE

Church

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

**Church**
Effect	Location	Description	Intensity
Wall collapse	Upper parts of walls in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	collapse of the upper parts of walls - Mikkola et. al. (2008) In many parts of the church, ... clerestory walls ... collapsed - Mikkola et. al. (2008) The intensity of the event is also indicated by the evidence of repairs to the upper portions of the walls of the church - Mikkola et. al. (2008)	VIII+
Arch collapse	Arches of the church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	In many parts of the church, the arches ... collapsed - Mikkola et. al. (2008)	VI+
Fallen Columns	columns of the church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	In many parts of the church, ... columns... collapsed - Mikkola et. al. (2008)	V+

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Chapel

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

**Chapel**
Effect	Location	Description	Intensity
Wall collapse	Upper parts of walls in the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	collapse of the upper parts of walls - Mikkola et. al. (2008) The intensity of the event is also indicated by the evidence of repairs to the upper portions of the walls of ... the chapel. - Mikkola et. al. (2008)	VIII+
Wall collapse	Western Wall of the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	The original western wall of the chapel also seems to have collapsed to the extent that it was deemed easier to build a new wall (Wall OO) - Mikkola et. al. (2008)	VIII+
Wall collapse	Wall H of the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	parts of Wall H also appear to have been badly damaged, as its upper courses were rebuilt in the following phase, using large quantities of recycled material. - Mikkola et. al. (2008)	VIII+
Rotated and displaced masonry blocks in walls and drums in columns	the Chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	The Phase 4 columns of the chapel, moreover, seem to derive from the collapsed columns of Phase 2 structures, as some of the drums used in them are broken. - Mikkola et. al. (2008)	VIII+
Displaced or Folded walls leading to broken Pilasters	The chapel Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 8 Chapter 6 The church and the chapel marking in Phase 2. Mikkola et al (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008)	The system of pilasters now visible in the chapel is not original, as is evidenced by the presence of wall plaster behind the pilasters - Mikkola et. al. (2008)	VII+

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Phase 6 Destruction Event (?) - 1st half of 7th century CE - inferred from remodelling and may not have been a response to an earthquake

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Image(s)	Description	Intensity
Arch Collapse	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 56 Chapter 6 The church and the chapel marking in Phase 5. Mikkola et al (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008)		At least some of the arches of the Phase 4 church seem to have fallen, probably causing heavy damage to the floor and furnishings of the church, but, curiously, the two columns of Phase 4 were left standing. - Mikkola et. al. (2008)	VI +

The archeoseismic evidence requires a minimum Intensity of VI (6) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Phase 8 Destruction Event - mid 8th century CE

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Image(s)	Description	Intensity
Vault collapse	Semidome over apse in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		collapse of the church's semidome - Mikkola et. al. (2008) The semidome covering the apse must have collapsed in the earthquake, destroying the floor of the apse beyond repair. - Mikkola et. al. (2008) The resulting tumble was cleared in the following phase but the semidome and the apse floor were never repaired. - Mikkola et. al. (2008)	VIII +
Tilted Wall	Wall J in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)	Fig. 80. Wall J tilting southward, following the earthquake of Phase 8. Mikkola et al (2008) Fig. 103 The Phase 9 buttress, locus T.31, built against Wall J. (in the center). Mikkola et al (2008)	the earthquake caused Wall J to severely tilt towards the south - Mikkola et. al. (2008) The wall was left leaning towards the south and it had to be supported by a buttress in the following phase - Mikkola et. al. (2008) A number of structures located outside the church were investigated in the course of excavation. The largest and perhaps most significant of these is the long buttress (locus T.31), built against Wall J (Fig. 103). The assignment of this buttress to Phase 9 is certain; it was clearly built after the wall tilted south in Phase 8. Therefore, it is likely that the buttress was built to support the wall against potential earth tremors. - Mikkola et. al. (2008)	VI +
Tilted Wall	Wall GG in the Chapel (part of locus Y.13) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The only sign of damage [in the Chapel] is found outside Wall GG, where a supporting buttress (part of locus Y.13) was built against the northern side of the wall in the following phase. - Mikkola et. al. (2008) the construction of the buttress against Wall GG in the following phase indicates that either some damage was inflicted in Phase 8, or it was a preventive measure. - Mikkola et. al. (2008) The walls of the chapel seem to have withstood the event of Phase 8, in spite of the fact that it caused so much damage to the church. However, the walls probably suffered some structural damage. This is suggested by the construction of stone buttresses outside and against Wall GG. - Mikkola et. al. (2008)	VI +
Tilted Walls ?	southeastern corner of the atrium, against Wall I Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)	Fig. 99 The Phase 9 buttress, loci B.02, B.16, B.18 and L.02, in the atrium (in the center). Mikkola et al (2008) Fig. 36 The plan of the southwestern part of the church. Mikkola et al (2008) Fig. 58 Trench B, northern baulk. Featuring the stratigraphy in the atrium, including the pavement, the water channel, locus B.12, the buttress, locus B.02, from Phase 9, and the stone tumbles. Mikkola et al (2008)	The most significant element of Phase 9 in the atrium is, however, the construction of a massive platform or buttress (loci B.02, B.16 [fill], B.18 [facade], and L.02) in the southeastern corner of the atrium, against Wall I (Fig. 99, also Figs. 36 and 58). - Mikkola et. al. (2008)	VI +
Limited Upper Wall collapse and roof collapse ?	southern pastophorion in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the southern pastophorion, falling stones caused severe damage to the floor due the presence of hollow compartments underneath. The part of floor that covered the southern compartment was destroyed and never repaired. It is uncertain if the arch there collapsed as well. It may have been left standing, but the roof was nonetheless severely damaged. - Mikkola et. al. (2008)	VIII+
Displaced, Folded, or Collapsed Walls	section of Wall H in the atrium (loci V.06, X.13) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		a section of Wall H in the atrium (loci V.06, X.13) was rebuilt. - Mikkola et. al. (2008)	VII-VIII +
Folded or Displaced Walls ? (suggested by possible pilaster damage	atrium in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The square pilaster (locus L.14) or pedestal in the eastern part of the atrium was also probably destroyed - Mikkola et. al. (2008)	VII +
Arch collapse	N-S arch spanning the Southern Aisle in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		the earthquake caused Wall J to severely tilt towards the south (Fig. 80), causing the collapse of the arches in the southern aisle - Mikkola et. al. (2008) collapse of one N-S arch in the aisle - Mikkola et. al. (2008)	VI +
Arch collapse	N-S arches spanning the nave in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In addition to the arches of the southern aisle, those spanning the nave appear to have collapsed. Such a pattern of collapse would indeed be expected. With the mutual supporting arch and beam system introduced in Phase 7, the collapse of one N-S arch in the aisle would have seriously impaired the stability of the corresponding N-S arch across the nave - Mikkola et. al. (2008)	VI +
Arch collapse	northern pastophorion in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The arch supporting the roof of the northern pastophorion probably fell too. - Mikkola et. al. (2008)	VI +
Arch collapse ?	southern pastophorion in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the southern pastophorion, falling stones caused severe damage to the floor due the presence of hollow compartments underneath. The part of floor that covered the southern compartment was destroyed and never repaired. It is uncertain if the arch there collapsed as well. It may have been left standing, but the roof was nonetheless severely damaged. - Mikkola et. al. (2008) The arch covering the southern pastophorion most likely collapsed in Phase 8, considering the fact that the entire southern wall of the basilica was severely affected by the destruction. Therefore, unlike the one in the northern pastophorion, the arch must have been rebuilt in Phase 9, as is evidenced by the discovery of the collapsed voussoirs of a fallen arch found among the stone tumble inside the room (locus M.04). - Mikkola et. al. (2008)	VI +
collonade and Column Collapse	atrium in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the atrium, parts of the collonades collapsed. - Mikkola et. al. (2008) The fallen columns of the atrium were not re-erected, but were cleared away and used elsewhere. - Mikkola et. al. (2008) collapse of ... the columns of the atrium - Mikkola et. al. (2008)	V +
Fractures, folds, and popups on regular pavements - Shattered Floor	Floor of the apse in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The semidome covering the apse must have collapsed in the earthquake, destroying the floor of the apse beyond repair. - Mikkola et. al. (2008) The resulting tumble was cleared in the following phase - Mikkola et. al. (2008)	VI + (probably higher due to falling stones)
Fractures, folds, and popups on regular pavements - Shattered Floor	Floor of the apse in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the southern pastophorion, falling stones caused severe damage to the floor due the presence of hollow compartments underneath. The part of floor that covered the southern compartment was destroyed and never repaired. - Mikkola et. al. (2008)	VI + (probably higher due to falling stones)
Fractures, folds, and popups on regular pavements - Shattered Floors ?	atrium in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		In the atrium, parts of the collonades collapsed. The atrium floor shows damage, but it is again difficult to determine whether it was damaged in this phase. - Mikkola et. al. (2008)	VI + (probably higher due to falling stones)
Fractures, folds, and popups on regular pavements - Shattered Floor ?	narthex in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 59 Chapter 6 The church and the chapel marking in Phase 7. Mikkola et al (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008)		The mosaic in the narthex shows damage, especially in the central medallion, which was never repaired. Dating of the damage is uncertain - it may have been caused by the events of either Phase 8 or 10 - Mikkola et. al. (2008)	VI + (probably higher due to falling stones)

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archaeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Phase 10 Destruction Event - late 8th or early 9th century CE

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Description	Intensity
Cratered Floor due to some Wall Displacement	western part of the nave and the northern aisle in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	The most obvious evidence of this destruction consists of craters left in the church floor by tumbling stones. The marble floor was badly damaged in especially in the western part of the nave and the northern aisle, where much of the floor was removed in the following phase. - Mikkola et. al. (2008) Several depressions left in the floor (locus T.29) of the nave mark the places hit by the falling stones. The stones that caused the depressions were, however, removed in Phase 11. - Mikkola et. al. (2008)	VII+
Collapsed Arch	N-S arch running between pilasters T.04 and G.06 in the Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	It seems probable that the long N-S arch running between pilasters T.04 and G.06 collapsed in this phase. - Mikkola et. al. (2008)	VI+
Collapsed Arch	westernmost arch in the northern aisle and the one that spanned the easternmost part of the nave Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Indirect evidence also exists for the collapse of the westernmost arch in the northern aisle and the one that spanned the easternmost part of the nave, for in these areas the marble floor was removed in Phase 11. It seems reasonable to assume that the removal of the floors was related to the damage caused by stones falling from the arches and other structures of the roof, whereas the floor was left untouched in those parts of the church where the arches did not collapse. - Mikkola et. al. (2008)	VI+
Collapse - possibly due to limited wall displacement	lowest parts of locus H.02 in the area of the Narthex Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 81 Chapter 6 The church and the chapel marking in Phase 9. Mikkola et al (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	As the walls and columns of the Atrium and the Narthex had been badly damaged and already partially removed in Phases 8 and 9, they probably were not heavily affected by the destruction of Phase 10. However, some of the stone tumble (lowest parts of locus H.02) in the area of the Narthex may have been caused by this event. - Mikkola et. al. (2008)	VII+

The archeoseismic evidence requires a minimum Intensity of VII (7) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Phase 12 destruction event - very late 9th or 10th centuries CE

Church

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Image(s)	Description	Intensity
Collapsed Walls	southern aisle - loci F.14, F.12 in the east and locus T.13 in the west Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		In the southern aisle, the layer of clayey soil was almost 0.8 m thick in its eastern part (loci F.14, F.12) and ca. 0.6 m thick in the west (locus T.13). A few large stones were also present in the soil, and the layer included a fair amount of marble fragments and crumbled plaster probably originating from the walls - Mikkola et. al. (2008)	VIII+
Collapsed Walls	Wall T inside the Apse - loci E.16, F.02, F.10, M.14, and U.11 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		Inside the Apse, the earthquake of Phase 12 created a layer of stone tumble consisting mainly of crushed, yellowish limestone (loci E.16, F.02, F.10, M.14, U.11). Although the surviving (lower) part of Wall T is mainly made of sandstone blocks, the upper courses of it were apparently constructed of limestone, a few blocks of which can still be seen in situ. When they collapsed inside the Apse, the tumble formed a pit, in the middle of which the enigmatic tessera pile was deposited in the following phase. - Mikkola et. al. (2008)	VIII+
Arch collapse	Church Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		Several rows of the voussoirs from fallen arches were found among the tumble in both the church and the chapel - Mikkola et. al. (2008)	VI+
Arch collapse	eastern part of the nave one running N-S between the pilasters loci F.07 and F.05d, and one apparently running E-W between the same pilaster (F.05d) to pilaster F.06 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 114. Trench F, northern baulk, featuring stone tumbles and soil deposits mentioned in the text. Mikkola et al (2008) Fig. 115. Collapsed arches in locus F.05. Mikkola et al (2008)	Remains of two fallen arches were found in the layer of stone tumble (loci F.04, F.09, F.10, F.11) in the eastern part of the nave (Fig. 114), one running N-S between the pilasters loci F.07 and F.05d, and one apparently running E-W between the same pilaster (F.05d) to pilaster F.06 (Fig. 115). - Mikkola et. al. (2008)	VI+
Arch collapse	eastern part of the nave - fallen arch originally sprang between the southern column and the pilaster (locus U.26) abutting the northern column Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 117 Trench G, eastern baulk, featuring the stone tumbles mentioned in the text. Mikkola et al (2008)	In the eastern part of the nave, the stone tumble (loci G.03 [lower part], G.16, G.17, T.05, T.10, U.03 [lower part], U.10) included a row of voussoirs running from the southern column (locus T.14) towards a pilaster (locus G.06) in the north (Fig. 117). However, as the two supports are not in the same line, the arch cannot have sprung between them. It seems that the force of the earthquake had thrown the northernmost voussoirs towards the west, and that fallen arch originally sprang between the southern column and the pilaster (locus U.26) abutting the northern column. - Mikkola et. al. (2008)	VI+
Arch collapse	Southern Aisle - western part Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 116 Collapsed arch in locus T.05. Mikkola et al (2008)	Clear remains of fallen arches were found in the stone tumble (loci T.05, T.08, T.10) in the western part of the aisle (Fig. 116), and in the central part were the ten drums and the capital of the collapsed Phase 4 column in locus T.14. Under the drums, furthermore, was found a fallen Phase 7 pilaster, originally a part of locus T.32, toppled over by the falling column. - Mikkola et. al. (2008)	VI+
Arch collapse	Northern Aisle - first one between the column (locus U.25) and pilaster (locus U.17) and second one between the pilasters (loci U.18 and U.39) in the eastern part of the nave Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 118 Collapsed arches in locus G.04 Mikkola et al (2008) Fig. 117 Trench G, eastern baulk, featuring the stone tumbles mentioned in the text. Mikkola et al (2008)	In the stone tumble (loci G.04, G.04a, G.10, G.11, G.14 [top], U.03 [lower part], U.09) above the clayey soil, two rows of voussoirs clearly resulting from fallen arches running N-S were discovered (Fig. 118, also Fig. 117). The first of these – between the column (locus U.25) and pilaster (locus U.17) – was scattered over a large area, testifying to the force of the earthquake. A second row of voussoirs was found between the pilasters (loci U.18 and U.39) in the eastern part of the nave. No remains of fallen arches were discovered in the western part of the northern aisle. - Mikkola et. al. (2008)	VI+
Arch collapse	southern pastophorion - loci M.13, M.15 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The strata excavated inside the pastophoria did not include a comparable layer of hard-packed clayey soil as found deposited on the aisles and the nave of the church. It is probable that these rooms were already partially deprived of roofing when the disaster struck in Phase 11. The northern pastophorion was filled with a layer of stone tumble (locus E.08 and the lower part of locus E.05). This deposit did not contain any evidence of a fallen arch, only a couple of long voussoirs, which may have been part of the Phase 9 steps (locus E.12) leading up to Wall T. A thick layer of stone tumble (loci M.13, M.15) also fell inside the southern pastophorion where, however, the voussoirs of an arch running N-S were found among the tumble - Mikkola et. al. (2008)	VI+
Fallen Columns	central part of the nave included some drums fallen from the northern column (locus U.25) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The tumble in the central part of the nave included some drums fallen from the northern column (locus U.25), but it is probable that the entire column did not collapse as some drums were found very close to the surface in the nave. - Mikkola et. al. (2008)	V+
Fallen Columns	northern part of the Atrium - part of locus X.05) Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The stone tumble (loci B.07, L.05, L.06, L.06a, L.08, L.09, X.02, X.04, and X.05; Figs. 46, 58) resulting from Phase 11 destruction is concentrated along the edges of the walls and is not exceedingly heavy. The Atrium walls were possibly already much reduced in height, following the previous earthquakes, and the resulting debris cleared in the meanwhile. In the central part of the Atrium, the strata above the floor consisted only of windblown sand. In the northern part of the Atrium, two fallen columns were found among the stone tumble (part of locus X.05). - Mikkola et. al. (2008)	V+
Fallen Columns	locus H.02 in the area of the Narthex Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Fig. 30. Locus V. 15 – the desposit of charcoal and molten glass. Mikkola et al (2008)	Most of the stone tumble (locus H.02) in the area of the Narthex was caused by this destruction (Col. Fig. 30). The tumble included some column drums fallen from the columns of the stylobate, and a row of four doorjambs west of the northern doorway. The northern doorway was filled with cobblestones and sand (locus H.04). - Mikkola et. al. (2008)	V+
Fallen Columns ?	northeastern corner of the Atrium Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)		The column standing in the northeastern corner of the Atrium has fallen towards the NW. Six drums originally part of this column were found in the tumble. The column to the west of this column had been taller when it collapsed; ten drums in a row running towards NE were found among the tumble. It is possible that the latter column fell later, sometime in Phase 14, as it appears to have fallen on top of the first column. - Mikkola et. al. (2008)	V+

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Chapel

Earthquake Archeological Effects chart

Earthquake Archeological Effects (EAE)

Rodríguez-Pascua et al (2013: 221-224)

of Rodríguez-Pascua et al (2013: 221-224)

Effect	Location	Description	Intensity
Vault collapse	inside the Apse Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	the semidome of the chapel must also have collapsed now. However, as relatively little stone tumble was found inside the Apse, the tumble associated with the collapse was probably cleared in the following phase - Mikkola et. al. (2008)	VIII+
Collapsed Walls	Wall S and Apse wall Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	The exterior of Wall S suffered extensive damage and much of the Apse wall tumbled towards the east (loci C.3a, C.11). A tangible piece of evidence of collapsing stones in the Apse area can be found in the northern cupboard, where the lower shelf (locus Y.10c) had been smashed into pieces. The stones that broke the shelf were removed in the following phase, but the pieces of the broken shelf was left in place. - Mikkola et. al. (2008)	VIII+
Arch collapse	The four central and western arches of the chapel - collapsed arches resting on the soil of loci Y.09 and I.10 Fig. 1 Chapter 5 The plan of the church and the chapel marking all walls and trench locations. Fiema (2008) Fig. 107 Chapter 6 The church and the chapel marking in Phase 11. Mikkola et al (2008)	Several rows of the voussoirs from fallen arches were found among the tumble in both the church and the chapel - Mikkola et. al. (2008) The four central and western arches of the chapel fell, all the voussoirs belonging to these arches were found in neat rows, resting on the soil of loci Y.09 and I.10. The easternmost arch, however, apparently did not collapse at this point. - Mikkola et. al. (2008)	VI+

The archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

References

Articles and Books

Bikai, P. M. 1996 Petra, Ridge Church. P. 531 in Archaeology in Jordan section. Patricia M. Bikai and Virginia Egan, eds. American Journal of Archaeology 100, no. 3, pp. 507-536.

Bikai, P. and M. Perry (2001). "Petra North Ridge Tombs 1 and 2: Preliminary Report." Bulletin of the American Schools of Oriental Research 324: 59 - 78.

Bikai, P. M. 2002a Petra. North Ridge Project. Pp. 450-51 in Archaeology in Jordan section. St. H. Savage, K. Zamora and D. R. Keller, eds. American Journal of Archaeology 106: 435-458.

Bikai, P. M. 2002b North Ridge Project. ACOR Newsletter vol 14.1. Summer, pp. 1-3.

Bikai, P. M. (2002). The churches of Byzantine Petra, in Petra. Near Eastern Archeology, 116, 555-571

Bikai, P. M. 2004 Petra: North Ridge Project. Pp. 59-63 in Studies in the History and Archaeology of Jordan VIII. F. al-Kraysheh ed. Amman. Bikai, Patricia M., and Megan Perry

Bikai, P., et al. (2020). Petra: The North Ridge, American Center of Oriental Research.

Bowersock, G. W. (2006). Mosaics as History: The Near East from Late Antiquity to Islam (Revealing Antiquity). Cambridge: Harvard University Press. - can be borrowed with a free account from archive.org

Eklund, S. (2008). Stone Weathering in the Monastic Building Complex on Mountain of St Aaron in Petra, Jordan.

Fiema, Z. T. (2002). "The Byzantine monastic / pilgrimage center of St. Aaron near Petra, Jordan." Arkeologipäivät.

Fiema, Z. T., et al. (2001). The Petra Church, American Center of Oriental Research.

Fiema, Z. T. (2013). "Visiting the sacred : continuity and change at Jabal Hārūn " Studies in the history and archaeology of Jordan. Department of Antiquities, Amman, Hashemite Kingdom of Jordan-Amman. Vol. 4 11.

Frosen et al. (2000). "The 1999 Finnish Jabal Harun Project: A Preliminary Report " Annual of the Department of Antiquities of Jordan 44.

Hertel et al. (2013) Chapter 19. Summary and Final Remarks. in Kouki, P., & Lavento, M. (2013). Petra - The Mountain of Aaron III. The Archaeological Survey. Societas Scientiarum Fennica.

Mikkola et al (2008) Chapter 6 The Church and the Chapel: Analysis and Phasing in Fiema, Z. T. and J. Frösén (2008). Petra - the mountain of Aaron I : the Finnish archaeological project in Jordan. Helsinki, Societas Scientiarum Fennica.

Parr, Peter 1959 Rock Engravings from Petra. Palestine Exploration Quarterly 91, pp. 106-108.

Excavation Reports

Fiema, Z. T. and J. Frösén (2008). Petra - the mountain of Aaron I : the Finnish archaeological project in Jordan. Helsinki, Societas Scientiarum Fennica.

Petra - The Mountain of Aaron II

Kouki, P., & Lavento, M. (2013). Petra - The Mountain of Aaron III. The Archaeological Survey. Societas Scientiarum Fennica.

Websites

Finnish Jabal Harun Project

Finnish Jabal Harun Project - Publications

Petra North Ridge Project

Petra: The North Ridge at ACOR

Wikipedia page for Tomb of Aaron

from Wikipedia

Petra - Jabal Harun