Pella

Transliterated Name	Language	Name
Pella	Greek	Πέλλα
Fahl	Hebrew	פחל
Fāhl or Fihl	Arabic	فاهل or فيهل
Khīrbīt Fāhl	Arabic	كهيربيت فاهل
Tabaqat Fāhl	Arabic	تاباقات فاهل
Pihil(um)	Ancient Semitic
Berenike
Philippeia

Introduction

Pella, aka Fahl, is located in the foothills east of the Jordan Valley ~30 km. south of the Sea of Galilee. It has been accepted as ancient Pella of the Decapolis (Smith in Stern et al, 1993). Occupation of the site started in the Neolithic and continued until modern times with relatively few gaps - most notably the late Iron IIB/C (c. 730-540 BC) and Persian (539-332 BC) periods ( Tidmarsh, 2024:7-8). In addition to the main mound (known as Khirbet Fahl), occupation has been discovered on Tell Husn to the south and in several of the valleys in the area.

Identification

from Robert Houston Smith in Stern et al (1993 v.3)

Since its identification in 1852 by E. Robinson, Khirbet (or Tabaqat) Fahl has been accepted as ancient Pella of the Decapolis. The site lies approximately at sea level amid the foothills of the eastern side of the Jordan Valley, fewer than 30 km (19 mi.) south of the Sea of Galilee (map reference 2075.2065). The word Fahl (or Fihl, as it appears in early Arabic texts) is the linguistic equivalent of the ancient Semitic place name Pihil(um), which occurs as early as 1800 BCE in Egyptian texts. The hellenized name Pella came into use after the conquests of Alexander the Great, who was born in Pella in Macedonia, as a phonetic approximation of the Semitic name.

Abundant water and mild winters made Pella one of the most desirable sites in the valley for habitation in antiquity. Although hot during the summer months, Pella is still cooler than most sites in the Jordan Valley. During the winter it is free from frost, unlike some places in the northern part of the valley. Annual precipitation today, which may not differ greatly from that at times in the past, is 345 mm and, occurring mainly in the months from December to February, is sufficient to permit spring crops. The site also has the important advantage of a powerful perennial spring that flows out of the gravel hill on which the city was built, as it has done for at least eight thousand years and probably much longer.

The central feature at the site is a 400-m-long ovoid mound, where the major amount of habitation has taken place through the centuries. Immediately west of the mound is a small tableland (in Arabic tabaqah, an element in the name Tabaqat Fahl) stretching westward I km (0.6 mi.) to the scarp overlooking the Jordan Valley. Low hills that flank the mound on the north and east contain tombs that range in date from the Early Bronze Age through the Byzantine period. South of the central mound, across Wadi Jirm, rises a large natural hill called Tell el-I:Iu~n, much of which was utilized as a cemetery from the Bronze Age onward. Both the crest and the lower slopes of the hill also display occupational remains from late periods, particularly the Byzantine, when the city reached its greatest size and population.

History

from Robert Houston Smith in Stern et al (1993 v.3)

Although it does not appear in the biblical record, Pella is mentioned in about a hundred early historical documents, ranging from Egyptian execration texts through late medieval references. An Egyptian papyrus from the thirteenth century BCE indicates that in the Late Bronze Age Pella supplied chariot parts to Egypt. Josephus relates that Pella was destroyed by Alexander Jannaeus in 83-82 BCE and in 63 BCE was brought under Roman control by Pompey, who is generally credited with having forged Pella and other hellenized cities in southern Syria and northern Transjordan into the federation known as the Decapolis (Josephus, Antiq. XII, 397; XIV, 75; War I, 104, !56; Pliny, NHV, 74). Eusebius, doubtless relying on an early tradition, states that early Christians, seeking to escape the Roman siege of Jerusalem in 70 CE, fled to Pella (HE 5, 2-3). How long these refugees stayed at Pella is not recorded, but a late first or early second century sarcophagus found beneath the paving of the north apse of the west church may be a relic of their sojourn. By the mid second century, Christianity was firmly enough established at Pella that the city was home to the early Christian apologist Aristo (Eus., HE IV, 6, 3). Epiphanius, a Church father writing not long after Eusebius, reports that these Christians subsequently returned to Jerusalem, but that a heretical form of Christianity subsequently flourished in the vicinity of Pella (Haer. 29, 7). The city's warm baths are mentioned in a third to fourth century rabbinic text (J.T., Shevi'it 6, I, 36c). The city fell under Arab domination in 635 CE, following a major battle with Byzantine forces that is reported in Islamic histories as the "battle of Pella." In 747 CE [JW: Actually 749], Pella was destroyed by a massive earthquake that was recorded by Arab chronographers [JW: not just Arab chronographers]. Although the city is mentioned in some accounts in the Middle Ages, maps of the period show that the location of the city was forgotten with the passing of centuries.

Exploration and Excavations

Early Exploration and Excavation up until 1985

from Robert Houston Smith in Stern et al (1993 v.3)

Pella was described and mapped by G. Schumacher in 1887; the report was published in the following year by the Palestine Exploration Fund. In 1933, J. Richmond, of the Mandatory Department of Antiquities, surveyed the site and subsequently published a description and a map of the central ruins. In 1958, R. W. Funk and H. N. Richardson, under the auspices of the American Schools of Oriental Research, conducted two weeks of excavation in two places on the mound. In 1964, a representative of the Jordan Department of Antiquities excavated at least eleven tombs at the site, chiefly on and around Tell el-Husn.

In 1967, The College of Wooster in Wooster, Ohio, mounted a major expedition under the direction of R. H. Smith. Excavations had been carried out in only two areas (I and II) when work was interrupted by the Six-Day War. Field operations were not resumed until 1979, when the college was joined by The University of Sydney (Australia), with J. B. Hennessy and A. W. McNicoll as co-directors of the Sydney contingent. Excavations conducted in subsequent years produced a large amount of new information about the site's history and material culture. Between 1979 and 1991, excavations and related investigations were conducted in thirty-four areas at Pella and in the vicinity. The College of Wooster completed its field activities in 1985.

University of Sydney Excavations starting in 1979

from Stern et al (2008)

Since 1979, the University of Sydney has conducted more then 20 seasons of excavations at Pella (see Vol. 3, pp. 1174–1180; and below, The Islamic Period, in this entry), under the direction of J. B. Hennessy and (since 1992) S. J. Bourke. The excavations have so far revealed remains from various periods ranging from the Neolithic period to late Umayyad times.

The Cathedral, also known as the Civil Complex Church, was partially excavated in 1979–1985 by a team from Wooster College in Ohio, under the direction of R. H. Smith. In 1994, the University of Sydney team renewed excavations within it, in order to completely expose the structure and partially reconstruct it. The Cathedral cella was covered by extensive earthquake destruction debris in 748/749 CE. Below the debris, a red and white tiled pavement was exposed along the entire northern aisle, while an elaborate mosaic pavement was encountered in the southern. All dated material is consistent with a sixth-century CE construction date for the church, and a final destruction at the very end of the Umayyad period. North of the Cathedral, remains of Umayyad stables, probably for camels, were exposed. This suits the understanding that in the Umayyad period the area was used as a caravanserai.

Excavations below the eastern slopes of the main mound exposed a large rectangular room, perhaps a warehouse or a shop, dating to the fourth century CE. After the destruction of this room, a new structure with a cobbled reinforced doorway was built over it. Several hundred small bronze coins have been found in and around the structure. Late Byzantine houses were exposed nearby.

Excavations Summary

from Tidmarsh (2024:7-8)

While Pella had been visited by the English travellers Charles Leonard Irby and James Mangles in the early nineteenth century (Irby and Mangles 1823: 92-3), identified by Edward Robinson during his visit to the site in 1852 (Robinson 1856: 320-4), and its remains extensively surveyed by the German scholar Gottlieb Schumacher in 1887 (Schumacher 1888), the first archaeological soundings were not carried out until 1958 by the American Schools of Oriental Research under R.W. Funk and H.N. Richardson. These soundings were confined to limited excavations in two squares on the main mound and only a brief summary of the results was published (Funk and Richardson 1958).²

The first major archaeological excavations at the site were commenced in the spring of 1967 by the College of Wooster under the direction of Robert Smith, who published his results in Pella of the Decapolis 1 (1973). Unfortunately the excavation program was curtailed in its first season by the outbreak of the Arab-Israeli war in June, and work was not resumed until 1979 as a joint project between Wooster and the University of Sydney under the direction of Smith, the late J. Basil Hennessy and the late Anthony McNico11.³ In 1985 the College of Wooster ceased its involvement with the site, the excavations continuing under the auspices of the University of Sydney, at first under the overall direction of Basil Hennessy and, subsequently, of Stephen Bourke.

Besides the excavation reports in the Annual of the Department of Antiquities of Jordan, Bulletin of the American Schools of Oriental Research and numerous more specialised articles, the results so far have appeared in detail in Archaeology of Jordan 11.2 (Homes-Fredericq and Hennessy 1989), the College of Wooster publications Pella of the Decapolis 1 (Smith 1973), Pella of the Decapolis 2 (Smith and L.P. Day 1989), the University of Sydney's Pella in Jordan I (McNicoll et al. 1982) and Pella in Jordan 2 (McNicoll et al. 1992). More recently, the coins from the excavations of 1979-90 have been published (Sheedy et al. 2001), as have the results of excavations at the Natufian site of of Wadi al-Hammeh, a westward-flowing tributary of the Jordan River located some 2 kilometres to the north of the tell (Edwards 1992, 2007, 2013; McNicoll et al. 1982:17-27). Numerous Palaeolithic, Kebaran and Natufian artefacts have also been unearthed from scattered sites on the interfluvial ridge and terrace sections adjacent to the wall and close to the shores of the ancestral Like Lisan (Misnumber 1992).

Excavations at Pella Itself have demonstrated that the main mound (Khirbet Fahl) was settled by at least Pre-Pottery Neolithic times (Bourke 1997: 96-7, 2008 2015/16) with relatively few gaps - most notably the late Iron IIB/C (c. 730-540 BC) and Persian (539-332 BC) periods - in its occupation sequence until modern times when, in 1967, the small village of Tabaqat Fahl was moved from the tell to a more westerly position. Work on Tell Husn - the largely natural hill immediately to the south of the Wadi Jirm al-Moz separating it from the main mound - has indicated intermittent occupation since the Chalcolithic period (Bourke 2014; Bourke et al. 1999; Watson and Tidmarsh 1996).

Hellenistic (332-63 BC) architectural remains and artefacts have now been unearthed from many of the plots on the main mound (Figure 1.1) in Areas III, IV, VIII, XXIII, XXVIII, XXXII; in the Wall Jirm al-Moz (Area IX) and on Tell Husn (Areas XI, XXXIV). Furthermore, two fortresses from this period have been located on nearby Jebel Sartaba (Area XIII) and Jebel Hammeh (Area XXX). The fort on Sartaba has been planned and investigated by limited soundings by the College of Wooster inside the structure (McNicoll et al. 1982: 65-7) while the defensive wall on Jebel Hammeh has been partially traced by a University of Sydney team but no excavations undertaken (McNicoll et al. 1992:103-5).

Whereas Hellenistic artefacts have been wowed from numerous plots on the main mound and elsewhere. Early Roman (63 BC -135 AD) material, represented chiefly by 'Roman tonnes of Eastern Sigillata A (ESA), is found only In small quantities and in disturbed deposits on the main mound in plots IIIP, IIIQ and IVL (Table 1.1). The structure(s) associated with this latter material were, however, completely obliterated by later Byzantine and early Islamic (mainly Ummayad) overbuilding. Thus, architectural evidence for the Early Roman period is essentially restricted to the Civic Complex in the Wadi Jirm al-Moz (Area IX), Tell Husn (Areas XI and XXXIV) and the tombs of Areas VI and VII. From Area VI to the south-west of Husn, an unplundered tomb (Tomb 54) dating to the late first or early second century AD was cleared In 1983 (McNicoll et al. 1992:124-333). Much of the organic material from this tomb - including cedar beams, pine planking, and a pair of leather soles - was well preserved with many intact glass vessels also recovered. The tomb contained little in the way of pottery although worth noting was the presence of a wheelmade knife-pared ("Herodian") lamp (McNicoll et al. 1992:pl. 87.4) and three fragments of a 'Galilean" bowl of Kfar Hananya Form IB (Adan-Bayewitz 1993: 91-7).⁴ A further tomb (Tomb 40) from the same area of late first century at or early first century AD was also explored (McNicoll et al. 1982: 87-8). Five other tombs of Early Roman date (first to second centuries AD), of which at least four had been robbed, were unearthed by the Wooster team in the South Cemetery (Area VII) in 1979 (McNicoll et al., 1980:75-6).⁴

Footnotes

2 For a full discussion of the early exploration of Pella by Western travellers and archaeologists, see Smith 1973: 10-14.

3 See Smith 1973: 20-2, for a vivid account of the problems faced by the expedition, which was in the field at the outbreak of the 1967 war.

4 With the exception of Areas VIII. IX and XIII (College of Wooster), the other areas from which Hellenistic and Early Roman material has been recovered were excavated by University of Sydney teams. The results of the College of Wooster excavations in these three areas are dealt with briefly In this volume for the sake of completion but for a fuller description (although much remains unpublished) see the relevant sections in Smith (1973), Smith and Day (1989) and McNicoll et al. (1982, 1992) as the excavation reports in Annual of the Department of Antiquities of Jordan, and Bulletin of the American Schools of Oriental Research.

5 I am grateful to Sandra Gordon (pers. com.) for further information regarding the Early Roman tombs from Areas VI and VII. For the later reuse of an Early Roman sarcophagus in the West Church complex (Area I), see Smith 1973:143-9.

Maps, Aerial Views, Plans, Drawings, Sections, and Photos

Maps

Normal Size

Fig. 0.1 Main sites

Figure 0.1

The southern Levant, showing main sites mentioned in the text.

Tidmarsh (2024)

in the southern Levant from Tidmarsh (2024)
Fig. 4.7 Pre and Post 749 features
Figure 4.7

Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level).

Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m

courtyard house, damaged in the 659 earthquake, -43 m

central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m

East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m

eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m

extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m

buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m

west residential quarter (excavation back-filled),destroyed in 749, -51 m

West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m

policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt

-2 m (Base image: Google Earth, 20 August 2021).

Blanke and Walmsley (2022)
of Islamic Pella (Fihl) from Blanke and Walmsley (2022)
Map of Pella and

Map of Pella and environs

click on image to open a magnifiable version in a new tab

Schumacher (1888)

environs from Schumacher (1888)
Fig. 24 Topographic

Figure 24

Topographical Map of a part of the Jordan Valley, showing Pella, Beth-shan, and Gadara

Smith (1973)

Map of the Jordan Valley surrounding Pella from Smith (1973)
Fig. 25 Geologic

Figure 25

Geological Map of the northern Jordan Valley at contour intervals of 300 m

Smith (1973)

Map of the Jordan Valley surrounding Pella from Smith (1973)

Magnified

Fig. 0.1 Main sites

Figure 0.1

The southern Levant, showing main sites mentioned in the text.

Tidmarsh (2024)

in the southern Levant from Tidmarsh (2024)
Fig. 4.7 Pre and Post 749
Figure 4.7

Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level).

Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m

courtyard house, damaged in the 659 earthquake, -43 m

central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m

East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m

eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m

extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m

buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m

west residential quarter (excavation back-filled),destroyed in 749, -51 m

West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m

policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt

-2 m (Base image: Google Earth, 20 August 2021).

Blanke and Walmsley (2022)
features of Islamic Pella (Fihl) from Blanke and Walmsley (2022)
Map of Pella and

Map of Pella and environs

click on image to open a magnifiable version in a new tab

Schumacher (1888)

environs from Schumacher (1888)
Fig. 24 Topographic

Figure 24

Topographical Map of a part of the Jordan Valley, showing Pella, Beth-shan, and Gadara

Smith (1973)

Map of the Jordan Valley surrounding Pella from Smith (1973)
Fig. 25 Geologic

Figure 25

Geological Map of the northern Jordan Valley at contour intervals of 300 m

Smith (1973)

Map of the Jordan Valley surrounding Pella from Smith (1973)

Aerial Views

Pella and Tell el-Husn in Google Earth

Pella and Tell el-Husn in Google Earth

click on image to explore this site on a new tab in Google Earth
Oblique Aerial View of

Oblique Aerial View of Areas III, IV, and IX

Pella Project - University of Sydney

Areas III, IV, and IX from the Pella Project - University of Sydney
Oblique Aerial View of

Oblique Aerial View of Areas III and IV

Pella Project - University of Sydney

Areas III and IV from the Pella Project - University of Sydney
Fig. 2 - View of Area XXXII temple precinct

Fig. 2

View of Area XXXII temple precinct.

Bourke (2004)

from Bourke (2004)
Plate 1B The Civic Complex

Plate 1B

The Civic Complex in the spring of 1983, near the end of four seasons of excavation.

Smith et al. (1989)

in the spring of 1983, near the end of four seasons of excavation from Smith et al. (1989)
Plate 13B Aerial View of the

Plate 13B

The western approach to the [Area IX Civic Complex] Church during excavation (aerial view). The paving at the foot of the staircase has not yet been exposed. In the upper part of the photograph is the atrium. The arch stones lying on the paving of the atrium have been reassembled; most of them were found in the same part of the atrium, but in greater disorder (see Fig. 10)

Smith et al. (1989)

Area IX Civic Complex Church during excavation from Smith et al. (1989)
Fig. 4 - Photo of the Pella Houses

Fig. 4

View of the Pella Houses A–B in Area IV, after excavation. Part of the section shown in figure 5 remains on the left; note the fallen column (Walmsley).

Walmsley (2008)

A–B in Area IV from Walmsley (2008)
Fig. 2 - Photo of Mamluk Mosque

Fig. 2

General view to the north-west of the main archaeological mound at Fahl (Pella), showing the Mamlûk mosque (centre-left), the location of the village (upper centre-right) and cemetery (upper far right, where the earlier Byzantine–early Islamic housing is exposed)

Photo: S.J. Bourke.

Walmsley (1997

and other structures on main mound at Pella from Walmsley (1997)

Plans and Drawings

Site Plans

Normal Size

Map of Pella and environs

Pella and its environs: map of the region and excavation areas. Area IV is at middle top.

Smith in Stern et al (1993)

from Walmsley and Smith in McNicoll et al (1992)
Fig. 1 - Map of archaeological

Fig. 1

Map of archaeological areas at Pella

Pella Project, modified Walmsley

Walmsley (2008)

areas at Pella from Walmsley (2008)
Pella contour plan showing

Pella contour plan showing excavation Areas, 2019

Pella Project - University of Sydney

excavation Areas, 2019 from the Pella Project - University of Sydney

Magnified

Map of Pella and environs

Pella and its environs: map of the region and excavation areas. Area IV is at middle top.

Smith in Stern et al (1993)

from Walmsley and Smith in McNicoll et al (1992)
Fig. 1 - Map of archaeological

Fig. 1

Map of archaeological areas at Pella

Pella Project, modified Walmsley

Walmsley (2008)

areas at Pella from Walmsley (2008)
Pella contour plan showing

Pella contour plan showing excavation Areas, 2019

Pella Project - University of Sydney

excavation Areas, 2019 from the Pella Project - University of Sydney

Area Plans and Drawings

Area IV - Umayyad Housing

Normal Size

Fig. 3 - General plan of

Fig. 3

General plan of the Umayyad housing in Area IV (Walmsley)

Walmsley (2008)

the Umayyad housing in Area IV from Walmsley (2008)
Seismic destruction in

Endplate - Area IV - Plan of House - House G

Walmsley and Smith in McNicoll et al (1992)

Rooms 13, 14, and 15 of House G in Area IV from Walmsley and Smith in McNicoll et al (1992)

Magnified

Fig. 3 - General plan of

Fig. 3

General plan of the Umayyad housing in Area IV (Walmsley)

Walmsley (2008)

the Umayyad housing in Area IV from Walmsley (2008)
Seismic destruction in

Endplate - Area IV - Plan of House - House G

Walmsley and Smith in McNicoll et al (1992)

Rooms 13, 14, and 15 of House G in Area IV from Walmsley and Smith in McNicoll et al (1992)

Iron and Bronze Age Temples

Normal Size

Sequential phases of six

Sequential phases of six superimposed temples showing approximate position of one above the other from Middle Bronze I to Iron Age II

Bourke (2013)

superimposed Temples at Pella (MBI-Iron II) from Bourke (2013)
Artist’s impression of Pella

An artist’s impression of how the Pella Migdol Temple would have looked during the Middle Bronze Age (1600 BC).

JW: Bourke (2013) identified the 4th phase of the Temples as having the so-called ‘Migdol’ or Fortress Temple form

Ben Churcher at Astarte Resources

"Migdol" Temple in the Middle Bronze Age (1600 BC) from Ben Churcher at Astarte Resources
Fig. 3 - Schematic plans of

Fig. 3

Schematic plans of three main phases of temple construction.

Bourke (2004)

three main phases of temple construction from Bourke (2004)

Magnified

Sequential phases of six

Sequential phases of six superimposed temples showing approximate position of one above the other from Middle Bronze I to Iron Age II

Bourke (2013)

superimposed Temples at Pella (MBI-Iron II) from Bourke (2013)

Double Caravanserai Complex

Normal Size

Fig. 23 - double caravanserai

Fig. 23

Plan of the excavated section of the double caravansery complex at Pella, second half of the eighth century

Walmsley (2007)

complex at Pella, 2nd half of the 8th century from Walmsley (2007)

Magnified

Fig. 23 - double caravanserai

Fig. 23

Plan of the excavated section of the double caravansery complex at Pella, second half of the eighth century

Walmsley (2007)

complex at Pella, 2nd half of the 8th century from Walmsley (2007)

Area I - West Church Complex

Normal Size

Plan of West Church

Plan of West Church Complex in Area I

Smith (1973)

Complex in Area I from Smith (1973)
Fig. 50 Reconstruction

Figure 50

Reconstruction of the West Church Sanctuary and atrium. The placement of windows is conjectural.

Smith (1973)

drawing of the West Church Sanctuary and atrium from Smith (1973)

Magnified

Plan of West Church

Plan of West Church Complex in Area I

Smith (1973)

Complex in Area I from Smith (1973)

Area IX - Civic Complex and Church

Normal Size

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 18 Detailed Plan

Fig. 18

Detailed plan of part of the north dependency and Parvis. Locus 50, the "Chamber of the Camels," contains the skeletons of seven camels, a camel foetus, an ass, and Human 2. The small chamber, Locus 43, in the northeast corner of Locus 50, held the skeleton of Human 1. Locus 66, also a chamber, held the skeletons of two cows; the outline of stones beneath the skeletons is somewhat uncertain. Locus 67, a part of the Parvis, had the skeleton of two cows, and Locus 22, an extension of the north dependency, had fragile remains of a horse and a foal. In Locus 23 is the stile across the entrance to the north portal of the atrium, which prevented animals from wandering into the atrium and Church. Excavated portions of drainage channels, declining in a generally westward direction beneath the paving, are indicated by bold straight lines; hypothetical continuations, based on paving patterns, are indicated by rows of dots.

Smith et al. (1989)

of the "Chamber of Camels" in Area IX with earthquake crushed human and animal skeletons from Smith et al. (1989)
Fig. 26 Reconstructed

Fig. 26

Reconstructed plan of the church in Phase 3, around A.D. 614 until 658-660. The bema has been extended further west. The reconstruction of the west portico of the atrium is purely conjectural, particularly the walls shown at the top of the small staircase. The precise positions of the columns in the west collonade is not known, and the wall at the north end of the monumental staircase is inferred.

Smith et al. (1989)

plan of the Area IX church in Phase 3 from Smith et al. (1989)
Fig. 27 Reconstructed

Fig. 27

Reconstructed perspective drawing of the church in Phase 3. Details of the way in which the arch was installed in the west wall of the atrium are conjectural. The bases and shafts of the collonade at the top of the monumental staircase are shown as being more uniform than they probably were.

Smith et al. (1989)

perspective drawing of the Area IX church in Phase 3 from Smith et al. (1989)
Fig. 28 Reconstructed

Fig. 28

Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis.

Smith et al. (1989)

plan of the Area IX church in Phase 4 from Smith et al. (1989)

Magnified

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 18 Detailed Plan

Fig. 18

Detailed plan of part of the north dependency and Parvis. Locus 50, the "Chamber of the Camels," contains the skeletons of seven camels, a camel foetus, an ass, and Human 2. The small chamber, Locus 43, in the northeast corner of Locus 50, held the skeleton of Human 1. Locus 66, also a chamber, held the skeletons of two cows; the outline of stones beneath the skeletons is somewhat uncertain. Locus 67, a part of the Parvis, had the skeleton of two cows, and Locus 22, an extension of the north dependency, had fragile remains of a horse and a foal. In Locus 23 is the stile across the entrance to the north portal of the atrium, which prevented animals from wandering into the atrium and Church. Excavated portions of drainage channels, declining in a generally westward direction beneath the paving, are indicated by bold straight lines; hypothetical continuations, based on paving patterns, are indicated by rows of dots.

Smith et al. (1989)

of the "Chamber of Camels" in Area IX with earthquake crushed human and animal skeletons from Smith et al. (1989)
Fig. 26 Reconstructed

Fig. 26

Reconstructed plan of the church in Phase 3, around A.D. 614 until 658-660. The bema has been extended further west. The reconstruction of the west portico of the atrium is purely conjectural, particularly the walls shown at the top of the small staircase. The precise positions of the columns in the west collonade is not known, and the wall at the north end of the monumental staircase is inferred.

Smith et al. (1989)

plan of the Area IX church in Phase 3 from Smith et al. (1989)
Fig. 28 Reconstructed

Fig. 28

Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis.

Smith et al. (1989)

plan of the Area IX church in Phase 4 from Smith et al. (1989)

Sections

Fig. 5 - Representative

Fig. 5

Representative section through Houses A–B, showing collapse lines and location of objects.

Walmsley (2008)

section through Houses A–B from Walmsley (2008)
Fig. 49 Section thru

Figure 49

Section of II S-R along the east-west elevation line through the center of the church and atrium.

Smith (1973)

the east-west elevation line of the West Church and atrium from Smith (1973)
Fig. 13 Stratigraphic
Fig. 13

Stratigraphic sections through A-A' in the sanctuary of the Church, and B-B' in the Parvis (for locations, see plan in Fig. to). Five levels are identified. Note that Level 2 is not present in section B-B'.

Level 1: Gray topsoil containing small stones and pockets of Marnluk and post-Mamluk sherds.

Level 2: Hard-packed brown soil, with some gravel; largely sterile but containing some sherds of various periods washed from upslope. (This level is not present in section B-B'.)

Level 3: Dense, yellowish-brown loess, largely sterile but containing a few Abbasid-Fatimid sherds. At the beginning of this level a robber-pit was cut into level 4.

Level 4: Debris from the earthquake of A.D. 747, loose, grayish-brown soil with rubble and architectural stones.

Level 5: Brown, clayey soil containing trash accumulation from the earthquake of A.D. 717, including rubble, roof tile fragments, bits of mosaic, chunks of cement and artifacts (see Pls. 56—57).

Smith et al. (1989)
sections through Area IX Church and Parvis from Smith et al. (1989)
Fig. 19 Reconstruction

Fig. 19

Reconstruction of the gallery that was constructed in the north dependency during Phase 3 (excavated as Loci 43 and 50), viewed schematically from the north. On the right, behind the upper and lower Ionic colonnade, is the large, reused column of Ajlun limestone that formed the porch of the north portal of the atrium.

(See also the perspective drawing in Fig. 27.)

Smith et al. (1989)

of Phase 3 gallery in the north dependency of the Area IX church from Smith et al. (1989)

Photos

Destruction layer in Trench XXXIVF 200

XXXIV F 200, Destruction layer 224.2 pottery smash on surface also showing ashy layer 224.1 to south

Pella Project - University of Sydney

showing pottery smash and ashy layer from the Pella Project - University of Sydney
Umayyad Collapse in Area IV

Area IV Umayyad Collapse

Pella Project - University of Sydney

from the Pella Project - University of Sydney
Marble paving slabs shattered

Plate LXXVIII.1

Marble paving slabs in the sanctuary of the Civic Complex Church shattered in the earthquake of A.D. 746/47. The fragments were preserved by a column which collapsed on the slabs.

JW: Actually that earthquake struck in 749

Smith in ADAJ Plates 1983

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Smith in ADAJ Plates 1983
Skeletons of camels killed

Plate LXXXI.2

Skeletons found in the courtyard of the "hall of camels" just north of the Civic Complex Church. The animals are lying as they were killed in the earthquake of A.D. 746/7

JW: Actually that earthquake struck in 749

Smith in ADAJ Plates 1983

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Smith in ADAJ Plates 1983
Skeletons of humans killed

Plate LXXXI.2

Area IV Plot P (extension). Skeletons of two charred adult human beings with covering textiles, found in the AD 746/7 destruction deposit.

JW: Actually that earthquake struck in 749

Walmsley and Smith in McNicoll et al (1992)

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Walmsley and Smith in McNicoll et al (1992)
Plate 10A Collapse

Plate 10A

I N: view toward the south. One the left is a surviving course of stones of the west front wall of the sanctuary, with the threshold and drain visible. On the right is paving of the atrium colonnade: fallen on the paving is a column drum, and on top of the drum four courses of the collapsed west wall.

Smith (1973)

in the Western Church Complex (Area I) from Smith (1973)
Plate 12 Phase 4 Collapse

Plate 12

I U: view toward the east. The massive block of fallen masonry on the left and the toppled column drum at the rear are the results of the earthquake which ended phase 4 of the [Western] church's history. The capital in the foreground may belong to the same column.

Smith (1973)

in the Western Church Complex (Area I) from Smith (1973)
Plate 13B Buttress against

Plate 13B

I G: view toward the east. Buttress against east wall of the sanctuary, with phase 3 plaster visible.

Smith (1973)

east wall of the sanctuary in the Western Church Complex (Area I) from Smith (1973)
Plate 16A Wall 30 (tilted ?)

Plate 16A

I E: view toward the east. Wall 30 is on the left. Several original pavers are in situ: above them, in shadow, is a tabun.

Smith (1973)

from Smith (1973)
Plate 18B Dented opus sectile

Plate 18B

Extant portion of mosaic I surviving beneath later raised paving in the apse [of the Area IX Civic Complex Church]. East is at the top of the photograph. The cylindrical depression in the opus sectile paving was made by a falling column drum.

Smith et al. (1989)

paving in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 19B Spalled column base

Plate 19B

Base of one of the large columns of the atrium porch [in the Area IX Civic Complex Church].

Smith et al. (1989)

from the atrium porch in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 19C Remains of the

Plate 19C

Remains of the lunate bema in the sanctuary of the [Area IX Civic Complex] Church, with evidence of subsequent modification. Assorted architectural stones were incorporated in the fill around the bema as the floor was raised and enlarged.

Smith et al. (1989)

lunate bema in the sanctuary of the Area IX Civic Complex Church from Smith et al. (1989)
Plate 26A Segment of

Plate 26A

Segment of the collapsed wall of the central apse of the [Area IX Civic Complex] Church between the central and southern windows, resting against the soil of the hillside that rises steeply behind it. on each side, lying on the wide lower portion of the semi-circular apse wall, are small rectangular stones that were part of the windows sills.

Smith et al. (1989)

collapsed wall of the central apse of the Area IX Civic Complex Church from Smith et al. (1989)
Plate 37A Skeleton of

Plate 37A

Three of the five skeletons excavated in the [Area IX] Civic Complex. A. Remains found in the southwest corner of the sanctuary

Smith et al. (1989)

a victim of the Phase 5 749 CE earthquake from the Area IX Civic Complex Church from Smith et al. (1989)
Plate 37B Skeleton of

Plate 37B

Three of the five skeletons excavated in the [Area IX] Civic Complex. B. Skeleton found in the drainage channel in the triangular chamber.

Smith et al. (1989)

a victim of the Phase 5 749 CE earthquake from the Area IX Civic Complex Church from Smith et al. (1989)
Plate 37C Skeleton of

Plate 37C

Three of the five skeletons excavated in the [Area IX] Civic Complex. C. Skeleton found inside the entrance to Locus 50, the Chamber of the Camels.

Smith et al. (1989)

a victim of the Phase 5 749 CE earthquake from Locus 50 in the "Chamber of the Camels" in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 37D Nose or ear

Plate 37D

Gold ear or nose ring (45134) found with the skeleton in Locus 50 [the "Chamber of the Camels" in the north dependency of the [Area IX] Civic Complex Church; length 18 mm

Smith et al. (1989)

ring if a a victim of the Phase 5 749 CE earthquake from Locus 50 of the "Chamber of the Camels" in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 39A Skeletons of

Plate 39A

View westward into Locus 50, the Chamber of the Camels [in the Area IX Civic Complex], showing several camel skeletons as excavated. The drums of the large columns in the background are in a temporary arrangement, and were subsequently re positioned properly.

Smith et al. (1989)

camel victims of the Phase 5 749 CE earthquake from Locus 50 in the "Chamber of the Camels" in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 39B Skeletons of

Plate 39B

The intertwined skeletons of Camels B and C [in Locus 50, the "Chamber of Camels" in the Area IX Civic Complex], viewed from the east.

Smith et al. (1989)

camel victims of the Phase 5 749 CE earthquake from Locus 50 in the "Chamber of the Camels" in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 40A Skeleton of

Plate 40A

Camel A skeleton in Locus 50 [the "Chamber of Camels" in the Area IX Civic Complex]

Smith et al. (1989)

Camel victim A of the Phase 5 749 CE earthquake from Locus 50 in the "Chamber of the Camels" in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 40C Skeleton of

Plate 40C

Skeleton of Camel F [in the Area IX Civic Complex]

Smith et al. (1989)

Camel victim F of the Phase 5 749 CE earthquake in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 40D Skeleton of

Plate 40D

Skeleton of foetus of Camel B in Locus 50 [the "Chamber of Camels" in the Area IX Civic Complex]

Smith et al. (1989)

the foetus of Camel victim B of the Phase 5 749 CE earthquake from Locus 50 in the "Chamber of the Camels" in the Area IX Civic Complex Church from Smith et al. (1989)
Plate 40E Skeletons of

Plate 40E

Ass Skeleton, with legs of Camel E at upper left and the skeleton of Camel D at right [in the Area IX Civic Complex]

Smith et al. (1989)

Camel victims D and E and an Ass from the Phase 5 749 CE earthquake in the Area IX Civic Complex Church from Smith et al. (1989)

Phasing

Middle and Late Bronze Age Stratigraphy of Area III on Tabaqat Fahl (Pella)

from Bourke, Sparks, and Schroder (2006)

Table I

Area III: Stratigraphic Summary

Bourke, Sparks, and Schroder (2006)

Hellenistic and Early Roman Occupation Sequence

from Tidmarsh (2024)

Table 1.1

Pella: Summary of Hellenistic and Early Roman Occupation Sequence

College of Wooster and University of Sydney excavations.

Tidmarsh (2024)

Byzantine/Early Islamic

Discussion

Walmsley (2007b:327) discussed redating of some of the levels at Pella

The pottery from Watson’s Phase 5 (initially dated ca. 600–640, later up to 659–60) is of primary interest here, although the preceding Phase 4 probably spanned into the early seventh century.²³ Phase 5 ceramics came from particularly good deposits with little rubbish survival, making the corpus unusually clean and representative.

Footnotes

23 Watson, “Change,” 234, in which Phase 4 contains imports datable up to 660, and Watson proposes a possible corpus date to 620.

Table

from Watson (1992:234)

Table 1

Preliminary Chronology for Pella Byzantine/Early Islamic Phases

Watson (1992)

Time periods from Stern et al (1993)

Age	Dates	Comments
Early Bronze IA-B	3300-3000 BCE	‎
Early Bronze II	3000-2700 BCE	‎
Early Bronze III	2700-2200 BCE	‎
Middle Bronze I	2200-2000 BCE	‎EB IV - Intermediate Bronze
Middle Bronze IIA	2000-1750 BCE	‎
Middle Bronze IIB	1750-1550 BCE	‎
Late Bronze I	1550-1400 BCE	‎
Late Bronze IIA	1400-1300 BCE	‎
Late Bronze IIB	1300-1200 BCE	‎
Iron IA	1200-1150 BCE	‎
Iron IB	1150-1100 BCE	‎
Iron IIA	1000-900 BCE	‎
Iron IIB	900-700 BCE	‎
Iron IIC	700-586 BCE	‎
Babylonian & Persian	586-332 BCE	‎
Early Hellenistic	332-167 BCE	‎
Late Hellenistic	167-37 BCE	‎
Early Roman	37 BCE - 132 CE	‎
Herodian	37 BCE - 70 CE	‎
Late Roman	132-324 CE	‎
Byzantine	324-638 CE	‎
Early Arab	638-1099 CE	Umayyad & Abbasid‎
Crusader & Ayyubid	1099-1291 CE	‎
Late Arab	1291-1516 CE	Fatimid & Mameluke‎
Ottoman	1516-1917 CE	‎

Time periods from Meyers et al (1997)

Phase	Dates	Variants
Early Bronze IA-B	3400-3100 BCE	‎
Early Bronze II	3100-2650 BCE	‎
Early Bronze III	2650-2300 BCE	‎
Early Bronze IVA-C	2300-2000 BCE	Intermediate Early-Middle Bronze, Middle Bronze I‎
Middle Bronze I	2000-1800 BCE	‎Middle Bronze IIA
Middle Bronze II	1800-1650 BCE	‎Middle Bronze IIB‎
Middle Bronze III	1650-1500 BCE	‎‎Middle Bronze IIC
Late Bronze IA	1500-1450 BCE	‎
Late Bronze IIB	1450-1400 BCE	‎
Late Bronze IIA	1400-1300 BCE	‎
Late Bronze IIB	1300-1200 BCE	‎
Iron IA	1200-1125 BCE	‎
Iron IB	1125-1000 BCE	‎
Iron IC	1000-925 BCE	Iron IIA‎
Iron IIA	925-722 BCE	Iron IIB‎
Iron IIB	722-586 BCE	‎Iron IIC
Iron III	586-520 BCE	Neo-Babylonian‎
Early Persian	520-450 BCE	‎
Late Persian	450-332 BCE	‎
Early Hellenistic	332-200 BCE	‎
Late Hellenistic	200-63 BCE	‎
Early Roman	63 BCE - 135 CE	‎
Middle Roman	135-250 CE	‎
Late Roman	250-363 CE	‎
Early Byzantine	363-460 CE	‎
Late Byzantine	460-638 CE	‎
Early Arab	638-1099 CE	‎
Crusader & Ayyubid	1099-1291 CE	‎
Late Arab	1291-1516 CE	‎
Ottoman	1516-1917 CE	‎

The Iron Age in the Southern Levant

Early Bronze II Earthquake - c. 2800 BCE

Discussion

Bourke et. al. (2009) report that EB II [Early Bronze II] occupational strata are sealed by a thick layer of destruction debris, associated with quantities of burnt mudbrick and stone. Immediately above these layers is an ephemeral “squatter phase” which dates from around 2900/2800 cal BC. Bourke et. al. (2009) characterize the EB II destruction layer as "site-wide" and "of some severity" and indicate it may have been caused by an earthquake. Raphael and Agnon (2018:773), citing Bourke (2000:235) and Bourke et. al. (2009), date this EB II destruction to around 2800 BCE and list the destruction of domestic architecture, public buildings and defense wall.

References

Bourke et. al. (2009)

from Bourke et. al. (2009)

Abstract

This article reports on 10 new accelerator mass spectrometry (AMS) radiocarbon dates from early phases of the Early Bronze Age at the long-lived settlement of Pella (modern Tabaqat Fahl) in the north Jordan Valley. The new AMS dates fall between 3400 and 2800 cal BC, and support a recent suggestion that all Chalcolithic period occupation had ceased by 3800/3700 cal BC at the latest (Bourke et al. 2004b). Other recently published Early Bronze Age ¹⁴C data strongly supports this revisionist scenario, suggesting that the earliest phase of the Early Bronze Age (EBA I) occupied much of the 4th millennium cal BC (3800/3700 to 3100/3000 cal BC). As this EB I period in the Jordan Valley is generally viewed as the key precursor phase in the development of urbanism (Joffe 1993), this revisionist chronology has potentially radical significance for understanding both the nature and speed of the move from village settlement towards a complex urban lifeway.

Excerpt

EB II occupational strata are sealed by a thick layer of destruction debris, associated with quantities of burnt mudbrick and stone. Immediately above these layers, in shallow, much-disturbed deposits suggestive of an ephemeral short-term re-occupation (often characterized as a “squatter phase” in much regional literature), a single date (OZG878) documents occupation which could potentially date anywhere between 2900–2600 cal BC. However, given the immediate post-destruction stratigraphic context, the transitional Early Bronze Age II/Early Bronze Age III makeup of the ceramic assemblage (Bourke 2000), and the absence of even a single sherd of the Khirbet Kerak ceramic characteristic of mature Early Bronze Age III assemblages (Philip and Millard 2000), it seems probable that this ephemeral “squatter phase” dates from around 2900/2800 cal BC.

... The end of EB II occupation at Pella seems to be associated with a site-wide destruction of some severity. Cultural assemblage analysis suggests that this EB II assemblage at Pella is to be dated towards the end of the EB II period in general (Bourke 2000:252), as isolated ceramic elements that look forward to later EB III assemblages support this reconstruction. This would imply an end date for the EB II period in general somewhere between 2900/2800 cal BC. The Pella scenario is paralleled with closely similar archaeological circumstances (a site-wide destruction perhaps earthquake-related, ending all EB II occupation) and like radiometric data sets at nearby Tell Abu Kharaz (Fischer 2000:224; Philip and Millard 2000:282–3) and at Tell es-Saidiyeh (Tubb 1998:41–8; Philip and Millard 2000:283–4), further to the south in the east Jordan Valley.

Culturally, the beginning of the EB III period is generally correlated with the rise of the Egyptian Old Kingdom, and set at around 2700 BC (Stager 1992; Philip 2008). However, associating the rise of Dynasty 3 in Egypt with the beginning of the EB III period has always been tentative, with insufficient exports in either direction to knit the cultural assemblages together securely (Philip and Millard 2000:281). Placing the start date of the EB III period at around 2800 cal BC, perhaps a century earlier than traditionally posited, is consistent with EB III assays from Tell esh-Shuna (Philip and Millard 2000:284) and Tell es-Saidiyeh (Ambers and Bowman 1998:430).

Raphael and Agnon (2018)

from Raphael and Agnon (2018:773)

EB II (3000-2700 BCE)

earthquake destruction of domestic architecture, public buildings and defense wall (the latter remained in use since the EB I), c. 2800 BCE (Bourke 2000: 235; Bourke et. al. 2009).

Phase VIII Earthquake - Middle Bronze IIB - ca. 1700 BCE or slightly earlier

Discussion

In Plot IIIC of Area III on Tabaqat Fahl (Pella) Walmsley et. al. (1993:178-180) noted that mudbrick and stone debris from Walls 41 and 47 suggest that Phase VIII was rendered uninhabitable through earthquake activity. East-West Wall 47 was preserved to a height of 50 cm. and most of the eastern metre of the wall had collapsed to the south in antiquity. Wall 41, the main city wall, runs north-south. Raphael and Agnon (2018:774) mentioned that the archaeologists have some reservations (S. Bourke, pers. comm.) about whether an earthquake is responsible and report a date of ca. 1700 BCE or slightly earlier.

References

Walmsley et. al. (1993)

from Walmsley et. al. (1993:178-180)

Figures

Fig. 1 - Pella contour plan/area locations

Fig. 1

Pella contour plan/area locations

Bourke (2004)

from Bourke (2004)
Fig. 7 - Plot IIIC.

Fig. 7

Plot IIIC. Plan of Phase VIII architecture

Walmsley et. al. (1993:179)

Plan of Phase VIII architecture from Walmsley et. al. (1993:179)

Excerpts

AREA III AND IV EXCAVATIONS

...

Middle Bronze Age IIA-IIA: Plot IIIC (East Cut Phases VIII-X)²¹

Phase VIII: MBIIB

At the beginning of the Tenth season, Plot IIIC was divided in half some four metres south of the north section line, and only the southern half of the trench was excavated.

Slightly over 1.25 m of deposit was removed in what was called IIIC South, exposing the very fragmentary remains of Phase VIII architecture (Fig. 7; pottery, Fig. 12:9-11).²² This architecture consisted of east-west Wall 47, and bonded into it, north-south Wall 49. Both were fairly insubstantial 40-60 cm wide walls, made up of small and medium fieldstone foundations, topped with patches of yellow and brown mudbricks. Both walls were preserved to a maximum height of 50 cm. Linking the walls together was a good rammed clay floor, 47.5. Set into this surface, in the northwest comer of the room formed by Walls 47/49, was a tabun, F.97. The probable stone paved doorway leading north from the centre of Wall 47 was disturbed by the Phase VI Pit F.75. The eastern end of Wall 47 was difficult to determine with precision, as most of the eastern metre of the wall had collapsed to the south in antiquity. However, the line itself was not in doubt, nor was the fact that it abutted the main city wall, Wall 41. In this collapse, a quantity of mudbrick debris (45.7/ 45.10) had been deposited on occupation buildup (47.2), on the floor 47.5. In colour and texture, the debris so resembled the city-wall that it seems likely that it fell from the wall's western face. The mudbrick and stone debris from Walls 41 and 47 suggest that Phase VIII was rendered uninhabitable through earthquake activity.

Intermediate Phase: MBIIB/C Burials

At some time after the abandonment of the Phase VIII architecture (Walls 47/49), and in a still obscure period before the extensive levelling associated with the construction of the Phase VII architecture (Walls 40/42/43), this area of the tell was used as a burial ground. A first intramural burial (F.98) was discovered in the northwestern margins of IIIC South, during the Tenth season (Fig. 8). This contained the body of an adult male, portions of a goat, and some sixteen ceramic vessels (ten juglets, three jugs and three bowls; for a selection see Fig. 13), a dagger and pommel, and flint (firelighter?).

Although it was the intention to leave the northern half of IIIC unexcavated for the present, erosion at the beginning of the Twelfth season revealed an elaborate multi intramural burial (F.106) several centimetres below the limit of excavation in 1986 (Fig. 9). This, too, is to be dated to the "interphase" between VIII/VII, and with more confidence, as the foundation trench for Phase VII Wall 42 cuts the northwestern margins of this second grave. It contained the articulated skeletons of at least three individuals, two adult females and an infant child. The easternmost adult is only partially preserved, as the margins of the grave were disturbed by Phase VII levelling, Phase VI Pit feature 75, and modern erosion. The grave also contained some twenty-two ceramic vessels (six jugs, three dipper juglets, six juglets, five carinated bowls, two platter bowls), two of which - a platter bowl and a globular juglet - were of the well known Tell al-Yahudiyeh Ware (Fig. 14). As the grave lay barely 20 cm below the Phase VII walls and surfaces, it seems likely that extensive levelling of the area occurred before Phase VII was constructed. The difference in time, based on ceramic evidence, does not seem to have been great, however, as the grave goods are best seen as Late MBIIB, and Phase VII ceramics placed early in MBIIC.

Probably to be associated with this "intermediate" phase is a third burial (F.107) cut through, and badly disturbed by a robber (?) pit feature 96. Feature 107 was a cist burial, lined with a single row of well formed yellow mudbricks on its north and south sides. Only the northern half of the burial, containing the articulated upper body of what is likely to be an adult female, was left relatively undisturbed. Fragments remained of a second adult burial (sex indeterminate) in the southern margins of the cist. Grave goods consisted of a jug and a juglet.

Footnotes

21. Plot supervisors: Sheldon Gosline and Helga Fiedler (1988); Kate da Costa and Amanda Parrish (1990).

22. For discussion of Phase VI and Phase VII, see T.F. Potts, S.J. Bourke et al., 'Preliminary Report on the University of Sydney's Eighth and Ninth Seasons of Excavation at Pella in Jordan', ADAJ 32 (1988), pp.1 30-131.

Raphael and Agnon (2018)

from Raphael and Agnon (2018:774)

MB IIB (1750-1550 BCE)

ca. 1700 BCE or slightly earlier. The archaeologists have some reservations (S. Bourke, pers. comm.).

Phase VIC Earthquake (?) - Middle Bronze III - c. 1600 BCE

Discussion

Bourke, Sparks, and Schroder (2006:26) described possible seismic destruction in Phase VIC of Area III on Tabaqat Fahl (Pella)

At some stage during the life of these structures, but after the re-laying of several floors, a severe earthquake (?) destruction resulted in significant damage to the entire complex. Both Wall 27 and the city wall (Wall 41) suffered major structural damage. Several large pieces of the inner face of the city wall fractured and collapsed onto floor surfaces. The three small cubicles built against the inner face of the city wall suffered a fiery destruction, with clear evidence of wall and floor fracturing, and much broken pottery and other objects sealed by a thick brick-filled debris layer.

Wall 41 runs N-S and Wall 27 runs E-W.

References

Bourke, Sparks, and Schroder (2006)

from Bourke, Sparks, and Schroder (2006:22, 26)

Figures

Table I Area III:

Table I

Area III: Stratigraphic Summary

Bourke, Sparks, and Schroder (2006)

Stratigraphic Summary from Bourke, Sparks, and Schroder (2006)
Fig. 7 Plans of five major

Fig. 7

Schematic plans of five major MBA Phases in Area III. Excavation Area 20 x 15 metres in extent

Bourke, Sparks, and Schroder (2006)

Middle Bronze Age Phases in Area III from Bourke, Sparks, and Schroder (2006)
Fig. 7 Closeup - Plan of Phase VI

Fig. 7 Closeup

Schematic plan of Phase VI in Area III. Excavation Area 20 x 15 metres in extent

Bourke, Sparks, and Schroder (2006)

in Area III from Bourke, Sparks, and Schroder (2006)
Fig. 17 City Wall 41 Phase VII repair

Fig. 17

View of City Wall 41 Phase VII repair, built against main inner wall face in centre background, looking east

Bourke, Sparks, and Schroder (2006)

in Area III from Bourke, Sparks, and Schroder (2006)

Excerpts

3. AREA III: THE MIDDLE BRONZE AGE STRATIGRAPHIC SEQUENCE

... Phase VII (Figs. 17, 20, 25)

Wall 27 (Early), Walls 42-44, Oven F.88, Brick Wall F.84 and Associated Deposits

Phase VII constructions were very badly disturbed by numerous deep foundation trenches resulting from the multiple phases of substantial building activity in the succeeding Phase VI. Phase VII is made up of north-south Wall 42 and east-west Wall 44, forming part of a small room in the north-west corner of the trench. Associated with this phase is a very large circular pebble-paved oven (F.88), which takes up much of the area east of Wall 42. To the north of the oven, an east-west yellow mudbrick wall (F.84) is preserved in the north baulk, and to the south fragments of a second east-west Wall 43 make up the north and south sides of the `oven room'. A series of good plaster surfaces (43.10, 43.7, 43.2) provide a well separated occupational sequence, with 43.10 probably constructional for all features, and 43.2 the last good surface in the area. On the southern margins of the trench, a large e/w mudbrick wall, Wall 27, marks the southern limits of the Phase VII complex outlined above (Fig. 19). It seems possible that an east-west passageway ran along the outer southern face of this wall. All Phase VII deposits south of this passageway were lost to erosion.

Phase VI (Figs. 19, 21-25)

Phase VI has at least three subphases. Two represent double-wall reconstructions cut through earlier sub-phases. Several walls have substantial foundation trenches, and a number of deep pits complicate interpretation. Consequently, there are many small patches of floor surfaces associated with each of the several major and many minor alterations to the basic Phase VI groundplan. To complicate matters further, the even more massive stone foundations for the subsequent Phase V "Governor's Residence" complex fragment further an already complex stratigraphic sequence.

MAGNESS-GARDINER's (1997: 312) recent review of the Pella Phase VI stratigraphy is in error in all major points. Phase VI is not MB I in date, and is not associated with the primary construction of the city fortifications, rather with their end. Magness-Gardiner also confuses the main Phase VI architecture with later Phase V deposits. Confusion is heightened by her association of the Phase V Lion Box and cuneiform tablet deposits (Porrs 1987) with Phase VI architecture.

Phase VI C

Walls 10, 11 and F.34; Wall 27 (Late) and Associated Deposits

The monumental east-west mudbrick Wall 27, which delimited the southern complex wall in the preceding Phase VII, is thickened and its superstructure re-bricked at the beginning of Phase VI. It is possible that this re-bricking includes an east-west mudbrick staircase up to a north-south walkway, running along the inner line of the city fortification wall. A series of small cubicles (Walls 10 and 11, and F.34) were constructed in the northern margins of the walkway, probably during this phase, or just possibly in the subsequent rebuild. The foundations of the Phase V "Governor's Residence" complicates interpretation at this point.

At some stage during the life of these structures, but after the re-laying of several floors, a severe earthquake (?) destruction resulted in significant damage to the entire complex. Both Wall 27 and the city wall (Wall 41) suffered major structural damage. Several large pieces of the inner face of the city wall fractured and collapsed onto floor surfaces. The three small cubicles built against the inner face of the city wall suffered a fiery destruction, with clear evidence of wall and floor fracturing, and much broken pottery and other objects sealed by a thick brick-filled debris layer.

Phase VI B (Figs. 22-24)

III C Walls 36-40; III D Walls 6 and 7; III F Wall 5; Features 71, 83 (Fig. 18) and 85, and Associated Deposits

Following the earthquake damage, the whole complex was rebuilt. Internal structural mudbrick wall patch (F.85) and buttress (F.83) were added to the inner face of the city wall (Wall 41), and the east-west passageway along the inner face of III F Wall 7 was probably blocked. Wall 27 was cut down and rebuilt, with stone buttresses along its north (III C F.71) and south (III D Wall 6) faces shoring up the original mudbrick wall. In the central area of the trench west of the city wall and north of the rebuilt Wall 27, a substantial square double-walled room (III C Walls 36, 37 and 40 and III D Wall 7) was constructed in stone and mudbrick, faced with thick mudplaster, and provided with benches along the north (Wall 38) and east (Wall 39) interiors. No more than small scraps of a build-up of surfaces were preserved against the south face of III D Wall 6. All other deposits were lost to erosion.

The walkway and the cubicle rooms along the inner face of the north-south city wall (Wall 41) were never repaired. It seems probable that the entire region east of Wall 41 was rebuilt along different lines. A mudbrick buttress wall (III F Wall 5) was cut down against the inner northern face of III F Wall 7. Along with the III C corner buttress F.83, both constructions probably aimed at strengthening the east-west circuit wall, III F Wall 7. The north baulk intervenes at this point, so it remains unclear whether III C Buttress F.83 and III F Bench Wall 5 merely narrowed the east-west passageway, or sealed it off completely. However, subsequent deep fill layers devoid of occupational surfaces suggest infilling with Middle/Late Bronze Age tip/rubbish deposits, making it likely that the area went out of use at this point.

LB IIA Earthquake - towards the end of the 14th century BCE

Discussion

Bourke, Sparks, and Mairs (1999:53-57) report that an entire Bronze Age complex in Area XXIV on Tell el-Husn consisting of a single main constructional phase, and at least two phases of rebuilding was destroyed in a massive earthquake, probably dating towards the end of the LBIIA period [1400-1300 BCE], based on pottery sealed in destruction layers. A hoard of eleven leaf-shaped copper alloy arrowheads and stunning bolts all in an excellent state of preservation were found in the destruction debris. Such arrowheads first appear during the LBA, probably in response to the contemporary developments in scale armour. Several burials of infants and young children were uncovered, sealed beneath the later rebuilding phases of the complex, which dated to the 17th century BCE (MB/LB) based on ceramics.

While discussing evidence due to the same event in Area XXII on Tabaqat Fahl (Pella), Bourke (2004:8-9) noted that there was a sharp warping of the underlying foundations in the north temple area [in Area XXII], still clear today from aerial photographs. Raphael and Agnon (2018:775), while citing Bourke (2012), report that the same earthquake was evident due to a major change in the design of the temple [Pella Migdol Temple ?], around 1350-1300 BCE which was probably due to a severe earthquake where the western wall of the temple revealed stress-twisting and shattering.

Bourke (2004:8-9) noted that similar earthquake-related damage is found throughout the city and in buildings on nearby Tell Husn (Bourke et al. 1999).

References

Bourke (2004)

from Bourke (2004)

Figures

Fig. 1 - Pella contour plan/area locations

Fig. 1

Pella contour plan/area locations

Bourke (2004)

from Bourke (2004)
Fig. 2 - View of Area XXXII temple precinct

Fig. 2

View of Area XXXII temple precinct.

Bourke (2004)

from Bourke (2004)
Fig. 3 - Schematic plans of

Fig. 3

Schematic plans of three main phases of temple construction.

Bourke (2004)

three main phases of temple construction from Bourke (2004)

Excerpts

Introduction

from Bourke (2004:1)

This article summarises key discoveries of the University of Sydney excavations at Pella in Jordan between 1996–2001. Work centred on the excavation of a three-phase fortress temple complex on the south side of the city mound, and in the study of changing cult practice in the temple precinct over the 800 years of its occupation (ca 1600–800 BC).

... Investigations began in 1994 when chance discoveries in the South Field (Area XXXII) revealed the presence of a massive stone building, the largest pre-Classical structure discovered at the site (Bourke et al. 2003).

ARCHITECTURAL HISTORY OF THE TEMPLE PRECINCT

First Constructional Phase: Middle Bronze Age (ca 1650–1450 BC)

from Bourke (2004:4)

There are three distinct phases in the constructional history of the temple proper. The first phase of construction consisted of a thick-walled “hollow-box” rectangular structure. The key external features are two projecting square stone buttress/piers flanking a wide entrance way through the east wall. The interior consists of an open rectangular space provided with a neatly paved mud brick floor, but otherwise containing no internal dividing walls and apparently no ritual paraphernalia (cult statue, cult vessels or offerings) of any kind. There is no good evidence for any cult practice or offering deposits located within the temple during this early phase. The few offering deposits that have been detected are located outside the temple, to the south and east.

Cult paraphernalia favours the worship of a male deity, and the simplicity of architectural design (an empty box) would favour a numinous aniconic deity. We suggest that El, father of the gods and head of the Canaanite pantheon, best fits this description of the deity worshipped in this first phase of the Pella temple. We acknowledge that there is no consensus as to the specific architectural, archaeological or iconographic paraphernalia to be associated with the worship of each male deity in the Canaanite pantheon (Dever 1983). The majority of Levantine archaeological literature on the subject of Bronze Age gods generally opts for Baal, Hadad or Dagan as the three most likely candidates to have been worshipped in Bronze Age Canaanite temples (Mazar 1992).

It is curious how little archaeological presence is accredited to El, given that he was the head of the Canaanite pantheon and ruler over all the gods. Contemporary mythological texts make clear the dominance of El in the Middle Bronze Age Canaanite pantheon (Lewis 1996; Pitard 2002), and yet few of the many Canaanite temples discovered over the last hundred years of excavation in the region has ever been specifically attributed to his worship. We believe this to be in error, and would like to suggest that the massive rectangular “empty-box” temple form, as represented by the Middle Bronze Age temples from Shechem, Megiddo, Hazor (Area A), Tel Kittan, Tell Hayyat and Pella be associated with the worship of Canaanite El. Given the geographical proximity of most of the abovementioned sites to each other, it seems probable that a specific inland central Levantine aspect of Canaanite El was being venerated (Albright 1968).

First Refurbishment: Late Bronze I (ca 1450 – 1350 BC)

from Bourke (2004:7-8)

At some stage in the early Late Bronze Age (ca 1450 BC) several alterations to the temple fabric took place. The first and potentially most important change was the construction of a cross-wall in the western quarter of the original “hollow-box” cella. This had the effect of defining a formal Holy of Holies for the first time. The floor area west of the cross wall (within the newly created Holy of Holies) was removed down to a depth of 1.5 metres, re-filled with multiple layers of medium-sized fieldstones, and sealed with a thick, yellow-white, lime plaster floor surface. The cross-wall also formed the foundation course for a formal threshold and entranceway into the Holy of Holies, although the exact format of this original threshold was obscured by later reconstructions.

The eastern facade of the temple was also remodelled, with two massive 5 x 5 metre hollow square towers built upon the projecting solid stone buttresses that flanked the original entrance to the temple. The change from the original stone and (presumably) solid brick pier superstructure to a hollow tower format may have been designed to reduce weight-stress on the abutting temple facade, or perhaps to facilitate the construction of high flanking towers.

It is no easy matter to evaluate the significance of these architectural changes for cult practice and religious belief. The changed architectural form of the early Late Bronze Age temple need not reflect any significant change in cult, but we suggest that it does. The action of dividing off a Holy of Holies for the first time is a significant departure from previous practice, and bespeaks an altered view of the relationship between man and god.

The Ugaritic religious epics (Pitard 2002) contain legends that document the triumph of Baal in a war between the gods, and is generally interpreted as recording the spread of Baal worship in early Late Bronze Age Canaan. This assumed pre-eminence seems to have led to the attribution of virtually all Late Bronze Age Canaanite temples to Baal, even though very few have any inscriptional evidence to favour such an association (Mazar 1992). Male iconography does predominate (figurines, cult statues, incense burners) so the worship of some male deity is not easily disputed. When Late Bronze Age texts do identify individual temples, they name Baal, Reshep, Hadad and Dagan as titular deities (van der Toorn et al. 1999).

As well, there is an undoubted presence (if not pre-eminence) of Baal worship in the southern Levantine Late Bronze Age, more specifically at Pella where ruling prince Mut-Balu proclaims his loyalty to Baal by his very name (Hess 1989). Thus the broad association of Baal, Hadad or Dagan worship with Late Bronze Age temples is not unreasonable. Explaining the apparent change from El to Baal worship is more of a problem, although the spread of Hurrian peoples and their distinctive religious beliefs into Canaan and southern Anatolia at this time is well documented (Na’aman 1994; Hess 1997). It may be that the spread of Baal worship into the southern Levant is broadly connected with the arrival of Hurrian immigrants and the rise of the Hurrian Mitannian empire (Klengel 1992). Indeed, it was this new presence of Hurrian Mitannians in the southern Levant that Thutmosis III claimed to have provoked his first military campaigns, which ultimately brought much of Canaan (including western Jordan) under Egyptian control for the first time (Redford 1992).

South Levantine Late Bronze Age temple architecture changes from the simple “empty-box” form of the Middle Bronze Age temples at Shechem, Megiddo and Hazor (Area A), to more architecturally complex internally subdivided structures such as those at Hazor (Area H), Lachish (Acropolis Temple) and Beth Shan (Mekal). Architectural change need not reflect change in cult practice, but when these architectural changes occur across the south Levantine landscape at the same time as new North Syrian Hurrian cultural traditions appear, there is more strength to arguments that seek to link changed architectural forms with changing religious beliefs (Hess 1989). We view the early Late Bronze Age changes to the original Middle Bronze Age temple form at Pella in this context of widespread change in cultural and religious beliefs, largely attributable to the influence of North Syrian Hurrian religious forms.

Second Constructional Phase: Late Bronze Age II (ca 1350 – 1150 BC)

from Bourke (2004:8-9)

A major change to the re-modelled temple design occurred around 1350 BC, probably as a result of severe earthquake damage. Similar earthquake-related damage is found throughout the city and in buildings on nearby Tell Husn (Bourke et al. 1999). Two key alterations occurred thereafter. The entire temple structure was narrowed and the cella was provided with a colonnade.

The entire structure was levelled down to the stone foundations and new (much less massive) stone and mudbrick walls were built along the outer edge of the original east, south and western wall lines. However, a new north wall line was created five metres to the south of the Middle Bronze Age original, resulting in a significant narrowing of the entire structure. This was probably brought about by the sharp warping of the underlying foundations in the north temple area, still clear today from aerial photographs. At this time the original wide entrance to the Holy of Holies was narrowed and re-centred, and rebuilt using roughly dressed limestone and more carefully dressed (and drilled) basalt orthostat blocks; the latter were probably reused from earlier structures. Two small basalt column bases now flanked the re-configured entrance to the Holy of Holies. The floor of the Holy of Holies was re-laid, with new small stone foundational layers sealed by a thick, yellow, plaster floor surface. A number of distinct Egyptian-style foundation deposits were placed in shallow pits below this re-laid floor.

The new (much narrower) rectangular cella to the east of the rebuilt Holy of Holies was provided with a central colonnade at this time, indicated by the presence of three pillar bases. The western and eastern column bases had relatively small sub-structural foundations, but the central column base was provided with a massive limestone sub-structure, implying that it was designed to be the major weight-bearing support. All three column base foundations were cut into the mud brick paving of the original cella floor. Traces of burnt wooden columns were found in direct association with both of the smaller column bases. Thin, off-white plaster floors were laid across the narrowed cella area and the lower regions of the interior wall surfaces were sealed with a thick, monochrome, pale brown mud plaster.

The eastern facade of the temple was also remodelled, although subsequent Iron Age re-use in this area has made the exact form of the Late Bronze Age structure difficult to reconstruct with any confidence. However, it seems probable that the two hollow-square towers flanking the early Late Bronze Age temple entrance collapsed in the earthquake and were not rebuilt. If this interpretation is correct, then the area to the east of the reconstructed east wall would have been an open pebble-paved plaza.

These alterations to the early Late Bronze Age temple form could be interpreted as a simple structural response to severe earthquake damage, in that virtually all changes could be seen as a necessary strengthening of the original structurally unsound “hollow-box” design. However, the construction of a pillared hall, the addition of flanking columns at the entrance to the Holy of Holies, and the presence of “Egyptianising” foundation deposits may all reflect a new cultural influence at work.

While we have no reason to posit major change in local religious beliefs in the Late Bronze Age II, the architectural remodelling of the Pella temple coincides closely with the first presence of the Egyptian Nineteenth Dynasty pharaohs in the region (Redford 1992). This Late New Kingdom dynasty profoundly changed the ways in which the Canaanite empire had been administered previously, being far more inclined to interfere directly in the running of vassal states (Weinstein 1981). From this time (ca 1300 BC) an accelerated “Egyptianisation” of local elite culture can be observed, as can direct Egyptian influence on local Canaanite architectural modes (Wimmer 1990; Higginbottom 1996). With this in mind, it may be that the Egyptianising foundation deposits and the pillared hall at Pella provide evidence for an increasingly pervasive Egyptianisation of local elite culture east of the Jordan during the later New Kingdom.

The remodelled temple remained in use until the end of the Bronze Age (ca 1150 BC), when the entire site of Pella suffered a major destruction. This may also have been due to earthquake activity, although human agency remains possible, as this is the time of the enigmatic Sea People descent on Egypt, generally (if not always reliably) associated with a widespread destruction horizon throughout the region at this time (Sandars 1978).

Iron Age I Temple Use (ca 1150 – 950 BC)

from Bourke (2004:9)

The post-destruction Iron Age I (ca 1150– 950 BC) temple deposits were badly disturbed by later building activities. There is some meagre archaeological evidence for activity in and about the Holy of Holies during the two hundred years of the Iron I period, but the pillared hall and eastern facade seems to have collapsed into ruin. All areas surrounding the Holy of Holies were given over to domestic use, and much disturbed by numerous rubbish pits (Bourke et al. 2003). The immediate area in and about the Holy of Holies seems to have retained some measure of cultic function, although the mixture of cultic and domestic practice renders the precise nature of cult practice obscure

Third Constructional Phase: Iron Age II (ca 950 – 800 BC)

from Bourke (2004:10-11)

The third constructional phase marked a profound change to the form and (arguably) the function of the Pella temple. After the two hundred years of decline represented by the Iron Age I (ca 1150–950 BC), the area of the Late Bronze Age Holy of Holies was completely rebuilt as two separate storage and cultic rooms, with access to the cult room via an indirect entranceway in the southeast corner. The cult room was provided with benches around its west and north sides, and what appears to be a stepped mud brick podium was built against the eastern wall, perhaps for the display of cult paraphernalia. The northern room was filled with baskets of lentils and bags of grain, all burnt in the final destruction. Most cultic items, favissae and offering debris were located in the open courtyard area immediately to the east of the cult room. The roughly square courtyard area was dominated by a massive stone altar, positioned roughly in the centre of the courtyard. The major cult items, which included the ceramic “Cow Box”, and associated incense cups and a chalice, were found in destruction debris beside the stone altar.

Direct architectural parallels for the Iron Age II temple form are elusive. At Shechem (Stager 1999) and Tel Kittan (Eisenberg 1977), Iron Age II structures were reconstructed directly on top of Late Bronze Age originals, although the architectural forms are not particularly close to those at Pella. However, reasonably close parallels are found with the Iron II temples from Tel Qasile (Mazar 1980) on the Palestinian coast, and some individual design elements are paralleled in Iron Age temples at nearby Beth Shan (Rowe 1940). The contemporary material culture (and cult practice) at these two sites display an eclectic mixture of local Canaanite and “Aegean-Cypriot” influences, which many researchers equate (rather shakily) with the “Sea Peoples” (Tubb 2001), or more specifically with the better-known Biblical Philistines (Dothan 1982).

Architectural parallels are consistent with a significant change in cult practice, and offering vessels and figurines display relatively unambiguous links with the Palestinian coast for the first time during these Iron I-II horizons. Whilst it is probably unwise to equate specific politico-historical events with changing archaeological circumstances, the sharp change in cult practice at Pella does seem to indicate the presence of a major new influence in the region, with all archaeological indicators favouring a source on the Palestinian coastal plain (Singer 1994). It is difficult not to view these purely archaeological circumstances as consistent with Biblical testimony relating to the penetration of the originally coastal Philistine peoples into the eastern Jezreel Valley, which many regard as occurring at precisely this time (Raban 1991; Singer 1994).

The remodelled Iron Age II temple precinct at Pella was in use for perhaps 150 years (ca 950–800 BC) before the temple and the entire settlement was destroyed in an extensive conflagration (Bourke et al. 2003). While earthquake activity has been suggested as the likely cause for similarly dated destruction horizons at Deir ’Alla (Franken 1992), the same horizon of destruction at nearby Tell Hammeh (Cahill et al. 1987) and Tel Rehov (Mazar 1999) has been attributed to the military activities of either Egyptian (or just possibly) Aramaean invaders. At Pella, while earthquake destruction is still considered the most probable cause, significant militaria (specifically iron arrowheads and scale armour) are consistently associated with this destruction horizon. Whatever the ultimate cause, this destruction proved to be catastrophic to the long-term well-being of the city of Pella, as settlement ceased across the site for the next 500 years, only reviving with the Seleucid occupation of the region after 200 BC (Bourke 1997)

ARTEFACTUAL DISCOVERIES AND CULTIC FUNCTION

The Late Bronze Age II Temple: Foundation, Offering and Destruction Deposits

from Bourke (2004:13-14, 20)

... Finally, there are materials found in situ within the destruction deposits which mark the end of the Late Bronze Age temple (ca 1150 BC). These materials are normally found upon the floor surfaces, and therefore may well instruct on the spatial patterning of cult practices, although much material is found within a thick destruction deposit that sealed the entire area. Materials found within this deposit could theoretically derive from upper storey and rooftop locations, as well as lower storey floor levels.

... Destruction Deposits; Ceramic Objects

A variety of offering utensils were recovered, sometimes in pieces, sometimes more or less intact, either contained within or sealed below a thick layer of ash and brick debris.

...

The Iron Age II Temple: Offering and Destruction Deposits

from Bourke (2004:13-14, 20)

... Destruction Deposits

The destruction of the Iron II temple (ca 800 BC) ended significant occupation in the area for more than five hundred years. Thick deposits of ash and brick debris sealed the temple proper and most nearby areas. Interpretation of the final Iron II destruction horizons is complicated by the large and intrusive Late Antique (ca 550 AD) foundation trenches that cut through much of the area, largely frustrating attempts to study the spatial patterning of objects found in situ below destruction horizons. The Iron II temple proper suffered quite severely from later constructional activity. However, the area of the eastern courtyard surrounding the central altar was largely undisturbed, and it was here that many cult objects were identified. These include the ceramic model shrine (the “Cow-Box”), perforated ceramic cups, used for incense offerings, and the painted ceramic chalice, used in libation offerings.

...

Bourke, Sparks, and Mairs (1999)

from Bourke, Sparks, and Mairs (1999:53-57)

Figures

Fig. 2 - Sketch plan of

Fig. 2

Sketch plan of Husn trenches: Bronze Age deposits highlighted (not to scale).

Bourke, Sparks, and Mairs (1999)

Husn trenches: Bronze Age deposits highlighted from Bourke, Sparks, and Mairs (1999)
Fig. 3 - Sketch plan of

Fig. 3

Sketch plan of platform and gatehouse complex (not to scale).

Bourke, Sparks, and Mairs (1999)

platform and gatehouse complex from Bourke, Sparks, and Mairs (1999)
Fig. 4 - Plan of Middle

Fig. 4

Trench XXXIVF. Plan of Middle and Late Bronze Age structures (1:100).

Bourke, Sparks, and Mairs (1999)

and Late Bronze Age structures in Trench XXXIVF from Bourke, Sparks, and Mairs (1999)
Plate 7.2 - MBA Stone terracing

Plate 7.2

Tell Husn (Pella). Trench XXXIVF. MBA Stone terracing and LBA Silo.

Bourke, Sparks, and Mairs (1999)

and LBA Silo in Trench XXXIVF from Bourke, Sparks, and Mairs (1999)
Plate 8.1 - Trench XXXIVF

Plate 8.1

Tell Husn (Pella). Trench XXXIVF. LBA Structures canted over (to left) in earthquake.

Bourke, Sparks, and Mairs (1999)

LBA Structures canted over (to left) in earthquake from Bourke, Sparks, and Mairs (1999)

Excerpt

1. TRENCH XXXIVF: THE MIDDLE AND LATE BRONZE AGE FORTRESS

A 4 x 11 m trench (XXXIVF) was opened north of the Byzantine fortress wall in 1993 to explore promising Hellenistic deposits outside the fort. Whilst the Hellenistic materials proved to be a thin rubbish deposit, a dense complex of stone-lined bins and a thoroughfare were uncovered and dated to the MB/LB period.⁴ In 1994, trench XXXIVF was expanded eleven metres to the east, thereby forming a 15 x 11 m exposure exploring virtually all of the remaining summit north and east of the massive Byzantine fortress (fig. 3). As well, a 2 x 4 m extension in the north-west corner was opened to investigate the very large Bronze Age border wall (Wall 1) detected in 1993.

Excavations in the western half of the extended XXXIVF quickly isolated a series of walls, bins, and stone pavements of MB/LB date, whilst excavations in the the eastern half exposed EBA mud-brick debris and ashy deposits immediately below the thin Late Byzantine/Hellenistic debris layers (fig. 4).

The western complex (fig. 4; pls. 7:2; 8:1) consists of a single main constructional phase, and at least two phases of rebuilding. Between the construction and last rebuilding phases, and probably associated with the first rebuilding phase, a series of single inhumation burials were discovered, dug into the constructional material, and in at least two cases sealed by walls of the final rebuilding phase. The constructional phase dates within the MBIIC period, the first major rebuilding dates from the MB/LB period, and the second (relatively minor) rebuilding to the LBIIA period. The entire complex was destroyed in a massive earthquake, probably dating towards the end of the LBIIA period, based on pottery sealed in destruction layers (fig. 5).

A hoard, consisting of eleven leaf-shaped arrowheads of copper alloy and three stunning bolts, all in an excellent state of preservation, (fig. 6) was recovered from the lip of one of two large stone-lined storage bins (Feature 6: fig. 4; pl. 7: 2). The uniformity of the collection suggests that they may have been from a quiver of weapons. The hoard dates to the period of the final destruction of the Husn complex by earthquake, based on ceramic evidence (fig. 5). The quiver probably fell into the storage bin during the destruction of the building.

Thickened leaf-shaped arrowheads with a central rib first appear during the LBA, probably in response to the contemporary developments in scale armour. They are frequently found in association with blunt stunning bolts, although these are usually present in smaller quantities. Parallels for both types of projectile are found throughout Syro-Palestine, including Kamid el-Loz,⁵ the Amman Airport Temple,⁶ Beth Shan,⁷ Akko,⁸ Gezer,⁹ Lachish,¹⁰ Tell Jedur,¹¹ and Tell el-'Ajjul.¹²

Sealed beneath the later rebuilding phases of the fortress, several burials of infants and young children were uncovered. They were cut into the fabric of the MBA stone foundations, and were sealed by earliest MBA floors. Ceramics from the burials date them to the 17th century BC (MB/LB period).

Footnotes

4 Watson—Tidmarsh art. cit. 308-11.

5 R. Miron, Kamid el-Loz 10, 1990, pls. 13-14, 19; M. Metzger—U. Barthel, Kamid el-Loz 8, 1993, pls. 32-5.

6 V. Hankey in: S. Bourke—J.-P. Descceudres (eds.), Trade, Contact, and the Movement of Peoples in the Eastern Mediterranean: Studies in Honour of J. Basil Hennessy, Meditarch Suppl. 3 (1995) 180-1 figs. 8-9.

7 F. James—P. McGovern, The Late Bronze Egyptian Garrison at Beth Shan: A Study of Levels VII and VIII (1993) figs. 156: 3-7; 157: 33-4.

8 S. Ben-Arie—G. Edelstein, Tombs near the Persian Gardens (1977) figs. 20: 3-13; 21: 1-17; 22: 1-17.

9 J. Seger—H. Lance, Gezer V (1988) pls. 27: 3-14; 28: 4-8, 14.

10 O. Tufnell, Lachish IV: The Bronze Age (1958) pl. 25: 2-3, 17-18, 19-22, 26-27, 47.

11 Ben-Arieh—Edelstein op. cit. fig. 6: 6-18

12 W. Petrie, Ancient Gaza I (1931) pl. 21: 106; W. Petrie, Ancient Gaza II (1932) pls. 16: 114, 129-41; 17: 142-3, 147-59, 161—8.

Raphael and Agnon (2018)

from Raphael and Agnon (2018:775)

LB II (1400-1200 BCE)

major change in the design of the temple, around 1350-1300 BCE, probably as a result of a severe earthquake. Western wall of the temple revealed stress-twisting and shattering (Bourke, 2012).

Iron I Earthquake (?) - ca. 1150 BCE

Discussion

Bourke (2004:8-9) described a destruction layer dated to ca. 1150 BCE in Iron I at the Temple Complex in Area XXII on Tabaqat Fahl (Pella) and throughout the entire site of Pella.

The remodelled temple remained in use until the end of the Bronze Age (ca 1150 BC), when the entire site of Pella suffered a major destruction. This may also have been due to earthquake activity, although human agency remains possible, as this is the time of the enigmatic Sea People descent on Egypt, generally (if not always reliably) associated with a widespread destruction horizon throughout the region at this time (Sandars 1978).

Bourke (2004:9) noted that post-destruction Iron Age I (ca 1150– 950 BC) temple deposits were badly disturbed by later building activities.

References

Bourke (2004)

from Bourke (2004)

Figures

Fig. 1 - Pella contour plan/area locations

Fig. 1

Pella contour plan/area locations

Bourke (2004)

from Bourke (2004)
Fig. 2 - View of Area XXXII temple precinct

Fig. 2

View of Area XXXII temple precinct.

Bourke (2004)

from Bourke (2004)
Fig. 3 - Schematic plans of

Fig. 3

Schematic plans of three main phases of temple construction.

Bourke (2004)

three main phases of temple construction from Bourke (2004)

Excerpts

Introduction

from Bourke (2004:1)

This article summarises key discoveries of the University of Sydney excavations at Pella in Jordan between 1996–2001. Work centred on the excavation of a three-phase fortress temple complex on the south side of the city mound, and in the study of changing cult practice in the temple precinct over the 800 years of its occupation (ca 1600–800 BC).

... Investigations began in 1994 when chance discoveries in the South Field (Area XXXII) revealed the presence of a massive stone building, the largest pre-Classical structure discovered at the site (Bourke et al. 2003).

ARCHITECTURAL HISTORY OF THE TEMPLE PRECINCT

First Constructional Phase: Middle Bronze Age (ca 1650–1450 BC)

from Bourke (2004:4)

There are three distinct phases in the constructional history of the temple proper. The first phase of construction consisted of a thick-walled “hollow-box” rectangular structure. The key external features are two projecting square stone buttress/piers flanking a wide entrance way through the east wall. The interior consists of an open rectangular space provided with a neatly paved mud brick floor, but otherwise containing no internal dividing walls and apparently no ritual paraphernalia (cult statue, cult vessels or offerings) of any kind. There is no good evidence for any cult practice or offering deposits located within the temple during this early phase. The few offering deposits that have been detected are located outside the temple, to the south and east.

Cult paraphernalia favours the worship of a male deity, and the simplicity of architectural design (an empty box) would favour a numinous aniconic deity. We suggest that El, father of the gods and head of the Canaanite pantheon, best fits this description of the deity worshipped in this first phase of the Pella temple. We acknowledge that there is no consensus as to the specific architectural, archaeological or iconographic paraphernalia to be associated with the worship of each male deity in the Canaanite pantheon (Dever 1983). The majority of Levantine archaeological literature on the subject of Bronze Age gods generally opts for Baal, Hadad or Dagan as the three most likely candidates to have been worshipped in Bronze Age Canaanite temples (Mazar 1992).

It is curious how little archaeological presence is accredited to El, given that he was the head of the Canaanite pantheon and ruler over all the gods. Contemporary mythological texts make clear the dominance of El in the Middle Bronze Age Canaanite pantheon (Lewis 1996; Pitard 2002), and yet few of the many Canaanite temples discovered over the last hundred years of excavation in the region has ever been specifically attributed to his worship. We believe this to be in error, and would like to suggest that the massive rectangular “empty-box” temple form, as represented by the Middle Bronze Age temples from Shechem, Megiddo, Hazor (Area A), Tel Kittan, Tell Hayyat and Pella be associated with the worship of Canaanite El. Given the geographical proximity of most of the abovementioned sites to each other, it seems probable that a specific inland central Levantine aspect of Canaanite El was being venerated (Albright 1968).

First Refurbishment: Late Bronze I (ca 1450 – 1350 BC)

from Bourke (2004:7-8)

At some stage in the early Late Bronze Age (ca 1450 BC) several alterations to the temple fabric took place. The first and potentially most important change was the construction of a cross-wall in the western quarter of the original “hollow-box” cella. This had the effect of defining a formal Holy of Holies for the first time. The floor area west of the cross wall (within the newly created Holy of Holies) was removed down to a depth of 1.5 metres, re-filled with multiple layers of medium-sized fieldstones, and sealed with a thick, yellow-white, lime plaster floor surface. The cross-wall also formed the foundation course for a formal threshold and entranceway into the Holy of Holies, although the exact format of this original threshold was obscured by later reconstructions.

The eastern facade of the temple was also remodelled, with two massive 5 x 5 metre hollow square towers built upon the projecting solid stone buttresses that flanked the original entrance to the temple. The change from the original stone and (presumably) solid brick pier superstructure to a hollow tower format may have been designed to reduce weight-stress on the abutting temple facade, or perhaps to facilitate the construction of high flanking towers.

It is no easy matter to evaluate the significance of these architectural changes for cult practice and religious belief. The changed architectural form of the early Late Bronze Age temple need not reflect any significant change in cult, but we suggest that it does. The action of dividing off a Holy of Holies for the first time is a significant departure from previous practice, and bespeaks an altered view of the relationship between man and god.

The Ugaritic religious epics (Pitard 2002) contain legends that document the triumph of Baal in a war between the gods, and is generally interpreted as recording the spread of Baal worship in early Late Bronze Age Canaan. This assumed pre-eminence seems to have led to the attribution of virtually all Late Bronze Age Canaanite temples to Baal, even though very few have any inscriptional evidence to favour such an association (Mazar 1992). Male iconography does predominate (figurines, cult statues, incense burners) so the worship of some male deity is not easily disputed. When Late Bronze Age texts do identify individual temples, they name Baal, Reshep, Hadad and Dagan as titular deities (van der Toorn et al. 1999).

As well, there is an undoubted presence (if not pre-eminence) of Baal worship in the southern Levantine Late Bronze Age, more specifically at Pella where ruling prince Mut-Balu proclaims his loyalty to Baal by his very name (Hess 1989). Thus the broad association of Baal, Hadad or Dagan worship with Late Bronze Age temples is not unreasonable. Explaining the apparent change from El to Baal worship is more of a problem, although the spread of Hurrian peoples and their distinctive religious beliefs into Canaan and southern Anatolia at this time is well documented (Na’aman 1994; Hess 1997). It may be that the spread of Baal worship into the southern Levant is broadly connected with the arrival of Hurrian immigrants and the rise of the Hurrian Mitannian empire (Klengel 1992). Indeed, it was this new presence of Hurrian Mitannians in the southern Levant that Thutmosis III claimed to have provoked his first military campaigns, which ultimately brought much of Canaan (including western Jordan) under Egyptian control for the first time (Redford 1992).

South Levantine Late Bronze Age temple architecture changes from the simple “empty-box” form of the Middle Bronze Age temples at Shechem, Megiddo and Hazor (Area A), to more architecturally complex internally subdivided structures such as those at Hazor (Area H), Lachish (Acropolis Temple) and Beth Shan (Mekal). Architectural change need not reflect change in cult practice, but when these architectural changes occur across the south Levantine landscape at the same time as new North Syrian Hurrian cultural traditions appear, there is more strength to arguments that seek to link changed architectural forms with changing religious beliefs (Hess 1989). We view the early Late Bronze Age changes to the original Middle Bronze Age temple form at Pella in this context of widespread change in cultural and religious beliefs, largely attributable to the influence of North Syrian Hurrian religious forms.

Second Constructional Phase: Late Bronze Age II (ca 1350 – 1150 BC)

from Bourke (2004:8-9)

A major change to the re-modelled temple design occurred around 1350 BC, probably as a result of severe earthquake damage. Similar earthquake-related damage is found throughout the city and in buildings on nearby Tell Husn (Bourke et al. 1999). Two key alterations occurred thereafter. The entire temple structure was narrowed and the cella was provided with a colonnade.

The entire structure was levelled down to the stone foundations and new (much less massive) stone and mudbrick walls were built along the outer edge of the original east, south and western wall lines. However, a new north wall line was created five metres to the south of the Middle Bronze Age original, resulting in a significant narrowing of the entire structure. This was probably brought about by the sharp warping of the underlying foundations in the north temple area, still clear today from aerial photographs. At this time the original wide entrance to the Holy of Holies was narrowed and re-centred, and rebuilt using roughly dressed limestone and more carefully dressed (and drilled) basalt orthostat blocks; the latter were probably reused from earlier structures. Two small basalt column bases now flanked the re-configured entrance to the Holy of Holies. The floor of the Holy of Holies was re-laid, with new small stone foundational layers sealed by a thick, yellow, plaster floor surface. A number of distinct Egyptian-style foundation deposits were placed in shallow pits below this re-laid floor.

The new (much narrower) rectangular cella to the east of the rebuilt Holy of Holies was provided with a central colonnade at this time, indicated by the presence of three pillar bases. The western and eastern column bases had relatively small sub-structural foundations, but the central column base was provided with a massive limestone sub-structure, implying that it was designed to be the major weight-bearing support. All three column base foundations were cut into the mud brick paving of the original cella floor. Traces of burnt wooden columns were found in direct association with both of the smaller column bases. Thin, off-white plaster floors were laid across the narrowed cella area and the lower regions of the interior wall surfaces were sealed with a thick, monochrome, pale brown mud plaster.

The eastern facade of the temple was also remodelled, although subsequent Iron Age re-use in this area has made the exact form of the Late Bronze Age structure difficult to reconstruct with any confidence. However, it seems probable that the two hollow-square towers flanking the early Late Bronze Age temple entrance collapsed in the earthquake and were not rebuilt. If this interpretation is correct, then the area to the east of the reconstructed east wall would have been an open pebble-paved plaza.

These alterations to the early Late Bronze Age temple form could be interpreted as a simple structural response to severe earthquake damage, in that virtually all changes could be seen as a necessary strengthening of the original structurally unsound “hollow-box” design. However, the construction of a pillared hall, the addition of flanking columns at the entrance to the Holy of Holies, and the presence of “Egyptianising” foundation deposits may all reflect a new cultural influence at work.

While we have no reason to posit major change in local religious beliefs in the Late Bronze Age II, the architectural remodelling of the Pella temple coincides closely with the first presence of the Egyptian Nineteenth Dynasty pharaohs in the region (Redford 1992). This Late New Kingdom dynasty profoundly changed the ways in which the Canaanite empire had been administered previously, being far more inclined to interfere directly in the running of vassal states (Weinstein 1981). From this time (ca 1300 BC) an accelerated “Egyptianisation” of local elite culture can be observed, as can direct Egyptian influence on local Canaanite architectural modes (Wimmer 1990; Higginbottom 1996). With this in mind, it may be that the Egyptianising foundation deposits and the pillared hall at Pella provide evidence for an increasingly pervasive Egyptianisation of local elite culture east of the Jordan during the later New Kingdom.

The remodelled temple remained in use until the end of the Bronze Age (ca 1150 BC), when the entire site of Pella suffered a major destruction. This may also have been due to earthquake activity, although human agency remains possible, as this is the time of the enigmatic Sea People descent on Egypt, generally (if not always reliably) associated with a widespread destruction horizon throughout the region at this time (Sandars 1978).

Iron Age I Temple Use (ca 1150 – 950 BC)

from Bourke (2004:9)

The post-destruction Iron Age I (ca 1150– 950 BC) temple deposits were badly disturbed by later building activities. There is some meagre archaeological evidence for activity in and about the Holy of Holies during the two hundred years of the Iron I period, but the pillared hall and eastern facade seems to have collapsed into ruin. All areas surrounding the Holy of Holies were given over to domestic use, and much disturbed by numerous rubbish pits (Bourke et al. 2003). The immediate area in and about the Holy of Holies seems to have retained some measure of cultic function, although the mixture of cultic and domestic practice renders the precise nature of cult practice obscure

Third Constructional Phase: Iron Age II (ca 950 – 800 BC)

from Bourke (2004:10-11)

The third constructional phase marked a profound change to the form and (arguably) the function of the Pella temple. After the two hundred years of decline represented by the Iron Age I (ca 1150–950 BC), the area of the Late Bronze Age Holy of Holies was completely rebuilt as two separate storage and cultic rooms, with access to the cult room via an indirect entranceway in the southeast corner. The cult room was provided with benches around its west and north sides, and what appears to be a stepped mud brick podium was built against the eastern wall, perhaps for the display of cult paraphernalia. The northern room was filled with baskets of lentils and bags of grain, all burnt in the final destruction. Most cultic items, favissae and offering debris were located in the open courtyard area immediately to the east of the cult room. The roughly square courtyard area was dominated by a massive stone altar, positioned roughly in the centre of the courtyard. The major cult items, which included the ceramic “Cow Box”, and associated incense cups and a chalice, were found in destruction debris beside the stone altar.

Direct architectural parallels for the Iron Age II temple form are elusive. At Shechem (Stager 1999) and Tel Kittan (Eisenberg 1977), Iron Age II structures were reconstructed directly on top of Late Bronze Age originals, although the architectural forms are not particularly close to those at Pella. However, reasonably close parallels are found with the Iron II temples from Tel Qasile (Mazar 1980) on the Palestinian coast, and some individual design elements are paralleled in Iron Age temples at nearby Beth Shan (Rowe 1940). The contemporary material culture (and cult practice) at these two sites display an eclectic mixture of local Canaanite and “Aegean-Cypriot” influences, which many researchers equate (rather shakily) with the “Sea Peoples” (Tubb 2001), or more specifically with the better-known Biblical Philistines (Dothan 1982).

Architectural parallels are consistent with a significant change in cult practice, and offering vessels and figurines display relatively unambiguous links with the Palestinian coast for the first time during these Iron I-II horizons. Whilst it is probably unwise to equate specific politico-historical events with changing archaeological circumstances, the sharp change in cult practice at Pella does seem to indicate the presence of a major new influence in the region, with all archaeological indicators favouring a source on the Palestinian coastal plain (Singer 1994). It is difficult not to view these purely archaeological circumstances as consistent with Biblical testimony relating to the penetration of the originally coastal Philistine peoples into the eastern Jezreel Valley, which many regard as occurring at precisely this time (Raban 1991; Singer 1994).

The remodelled Iron Age II temple precinct at Pella was in use for perhaps 150 years (ca 950–800 BC) before the temple and the entire settlement was destroyed in an extensive conflagration (Bourke et al. 2003). While earthquake activity has been suggested as the likely cause for similarly dated destruction horizons at Deir ’Alla (Franken 1992), the same horizon of destruction at nearby Tell Hammeh (Cahill et al. 1987) and Tel Rehov (Mazar 1999) has been attributed to the military activities of either Egyptian (or just possibly) Aramaean invaders. At Pella, while earthquake destruction is still considered the most probable cause, significant militaria (specifically iron arrowheads and scale armour) are consistently associated with this destruction horizon. Whatever the ultimate cause, this destruction proved to be catastrophic to the long-term well-being of the city of Pella, as settlement ceased across the site for the next 500 years, only reviving with the Seleucid occupation of the region after 200 BC (Bourke 1997)

ARTEFACTUAL DISCOVERIES AND CULTIC FUNCTION

The Late Bronze Age II Temple: Foundation, Offering and Destruction Deposits

from Bourke (2004:13-14, 20)

... Finally, there are materials found in situ within the destruction deposits which mark the end of the Late Bronze Age temple (ca 1150 BC). These materials are normally found upon the floor surfaces, and therefore may well instruct on the spatial patterning of cult practices, although much material is found within a thick destruction deposit that sealed the entire area. Materials found within this deposit could theoretically derive from upper storey and rooftop locations, as well as lower storey floor levels.

... Destruction Deposits; Ceramic Objects

A variety of offering utensils were recovered, sometimes in pieces, sometimes more or less intact, either contained within or sealed below a thick layer of ash and brick debris.

...

The Iron Age II Temple: Offering and Destruction Deposits

from Bourke (2004:13-14, 20)

... Destruction Deposits

The destruction of the Iron II temple (ca 800 BC) ended significant occupation in the area for more than five hundred years. Thick deposits of ash and brick debris sealed the temple proper and most nearby areas. Interpretation of the final Iron II destruction horizons is complicated by the large and intrusive Late Antique (ca 550 AD) foundation trenches that cut through much of the area, largely frustrating attempts to study the spatial patterning of objects found in situ below destruction horizons. The Iron II temple proper suffered quite severely from later constructional activity. However, the area of the eastern courtyard surrounding the central altar was largely undisturbed, and it was here that many cult objects were identified. These include the ceramic model shrine (the “Cow-Box”), perforated ceramic cups, used for incense offerings, and the painted ceramic chalice, used in libation offerings.

...

Walmsley et. al. (1993)

from Walmsley et. al. (1993:189)

AREA XXXII EXCAVATIONS

... Iron Age (Area XXXII Phases C-E)²⁹

Excavation of the well-constructed architecture of Phase D was completed this season. In plot XXXIIA this consisted of three small square rooms which abut each other from north to south, a partially covered courtyard to their east, and a small alleyway, to the north. In plot XXXIIB, to the west, the continuation of these three north south rooms was exposed, whilst a small north-south passageway was found to separate these kitchen structures from a large open courtyard further to the west.

There is nothing about these structures to suggest anything other than a well constructed domestic housing unit. Extensive ceramic material has been recovered off the floors (Fig. 15: 1-3). The complex underwent a fiery destruction, probably at some stage during the tenth century BC.

Although ceramic analysis is still at a stage, considerably more cataloguing has reinforced the impression that there is but a small interval between the destructions of Phase D (late tenth century BC?) and Phase C (early ninth century BC?).

Canadian excavations at Tell al Hammeh continue to produce material close to our own in circumstance and kind, and the extensive Iron Age destruction at Tell as-Sa'idiyeh would seem to parallel our Phase D horizon.³⁰ Together they suggest some form of linked circumstance in the destructions attested to in the region. Whilst the latter destruction (Phase C) may well have a natural cause-and recent controversy over the extent and impact of Shishak's razzia notwithstanding-the case for the earlier (Phase D) destruction being attributed to Shishak should be considered. Whilst Ussishkin is right to point out that there has never been any compelling evidence to attribute the extensive early Iron Age II destruction horizon in Palestine to Shishak, Dever is equally correct to observe that the extensive destruction horizons do exist, Shishak or no, and some, most probably linked, explanation for the extensive disruption must be attempted.³¹

Whether it be through earthquake or human action, it seems certain now that Pella shares in the misfortunes of its western neighbours in the tenth century BC.

Footnotes

29. Supervisors: Erin Crumlin (XXXIIA); Jodie Benton (XXXIIB); Graham Phillip (XXXIIC).

30. J. Cahill, G. Lipton and D. Tarler, 'Tell el Hammah, 1988', /El 38 (1988), pp.191-194; J.N. Tubb, 'Preliminary Report on the Fourth Season of Excavations at Tell es-Sa'idiyeh in the Jordan Valley', Levant 22 (1 990) pp. 21-38.

31. D. Ussishkin, 'Notes on Megiddo, Gezer, Ashdod, and Tel Balash in the Tenth to Ninth Centuries b.c.', BASOR 277-8 (1990), pp. 71-91.

Raphael and Agnon (2018)

from Raphael and Agnon (2018:775)

Iron I (1200-1000 BCE)

c. 1150, the entire site suffered destruction, possibly due to an earthquake though human activity remains an option (Bourke 2004: 9).

Iron IIB Earthquake (?) - ~800 BCE

Discussion

Bourke (2004:13-14, 20) described a destruction layer dated to ~800 BCE in Iron IIb at the Temple Complex in Area XXII on Tabaqat Fahl (Pella) and throughout the area.

The remodelled Iron Age II temple precinct at Pella was in use for perhaps 150 years (ca 950–800 BC) before the temple and the entire settlement was destroyed in an extensive conflagration (Bourke et al. 2003). While earthquake activity has been suggested as the likely cause for similarly dated destruction horizons at Deir ’Alla (Franken 1992), the same horizon of destruction at nearby Tell Hammeh (Cahill et al. 1987) and Tel Rehov (Mazar 1999) has been attributed to the military activities of either Egyptian (or just possibly) Aramaean invaders. At Pella, while earthquake destruction is still considered the most probable cause, significant militaria (specifically iron arrowheads and scale armour) are consistently associated with this destruction horizon. Whatever the ultimate cause, this destruction proved to be catastrophic to the long-term well-being of the city of Pella, as settlement ceased across the site for the next 500 years, only reviving with the Seleucid occupation of the region after 200 BC (Bourke 1997)

Bourke (2004:13-14, 20) discussed the destruction layer and challenges in interpreting it.

Thick deposits of ash and brick debris sealed the temple proper and most nearby areas. Interpretation of the final Iron II destruction horizons is complicated by the large and intrusive Late Antique (ca 550 AD) foundation trenches that cut through much of the area, largely frustrating attempts to study the spatial patterning of objects found in situ below destruction horizons. The Iron II temple proper suffered quite severely from later constructional activity. However, the area of the eastern courtyard surrounding the central altar was largely undisturbed, and it was here that many cult objects were identified.

References

Bourke (2004)

from Bourke (2004)

Figures

Fig. 1 - Pella contour plan/area locations

Fig. 1

Pella contour plan/area locations

Bourke (2004)

from Bourke (2004)
Fig. 2 - View of Area XXXII temple precinct

Fig. 2

View of Area XXXII temple precinct.

Bourke (2004)

from Bourke (2004)
Fig. 3 - Schematic plans of

Fig. 3

Schematic plans of three main phases of temple construction.

Bourke (2004)

three main phases of temple construction from Bourke (2004)

Excerpts

Introduction

from Bourke (2004:1)

This article summarises key discoveries of the University of Sydney excavations at Pella in Jordan between 1996–2001. Work centred on the excavation of a three-phase fortress temple complex on the south side of the city mound, and in the study of changing cult practice in the temple precinct over the 800 years of its occupation (ca 1600–800 BC).

... Investigations began in 1994 when chance discoveries in the South Field (Area XXXII) revealed the presence of a massive stone building, the largest pre-Classical structure discovered at the site (Bourke et al. 2003).

ARCHITECTURAL HISTORY OF THE TEMPLE PRECINCT

First Constructional Phase: Middle Bronze Age (ca 1650–1450 BC)

from Bourke (2004:4)

There are three distinct phases in the constructional history of the temple proper. The first phase of construction consisted of a thick-walled “hollow-box” rectangular structure. The key external features are two projecting square stone buttress/piers flanking a wide entrance way through the east wall. The interior consists of an open rectangular space provided with a neatly paved mud brick floor, but otherwise containing no internal dividing walls and apparently no ritual paraphernalia (cult statue, cult vessels or offerings) of any kind. There is no good evidence for any cult practice or offering deposits located within the temple during this early phase. The few offering deposits that have been detected are located outside the temple, to the south and east.

Cult paraphernalia favours the worship of a male deity, and the simplicity of architectural design (an empty box) would favour a numinous aniconic deity. We suggest that El, father of the gods and head of the Canaanite pantheon, best fits this description of the deity worshipped in this first phase of the Pella temple. We acknowledge that there is no consensus as to the specific architectural, archaeological or iconographic paraphernalia to be associated with the worship of each male deity in the Canaanite pantheon (Dever 1983). The majority of Levantine archaeological literature on the subject of Bronze Age gods generally opts for Baal, Hadad or Dagan as the three most likely candidates to have been worshipped in Bronze Age Canaanite temples (Mazar 1992).

It is curious how little archaeological presence is accredited to El, given that he was the head of the Canaanite pantheon and ruler over all the gods. Contemporary mythological texts make clear the dominance of El in the Middle Bronze Age Canaanite pantheon (Lewis 1996; Pitard 2002), and yet few of the many Canaanite temples discovered over the last hundred years of excavation in the region has ever been specifically attributed to his worship. We believe this to be in error, and would like to suggest that the massive rectangular “empty-box” temple form, as represented by the Middle Bronze Age temples from Shechem, Megiddo, Hazor (Area A), Tel Kittan, Tell Hayyat and Pella be associated with the worship of Canaanite El. Given the geographical proximity of most of the abovementioned sites to each other, it seems probable that a specific inland central Levantine aspect of Canaanite El was being venerated (Albright 1968).

First Refurbishment: Late Bronze I (ca 1450 – 1350 BC)

from Bourke (2004:7-8)

At some stage in the early Late Bronze Age (ca 1450 BC) several alterations to the temple fabric took place. The first and potentially most important change was the construction of a cross-wall in the western quarter of the original “hollow-box” cella. This had the effect of defining a formal Holy of Holies for the first time. The floor area west of the cross wall (within the newly created Holy of Holies) was removed down to a depth of 1.5 metres, re-filled with multiple layers of medium-sized fieldstones, and sealed with a thick, yellow-white, lime plaster floor surface. The cross-wall also formed the foundation course for a formal threshold and entranceway into the Holy of Holies, although the exact format of this original threshold was obscured by later reconstructions.

The eastern facade of the temple was also remodelled, with two massive 5 x 5 metre hollow square towers built upon the projecting solid stone buttresses that flanked the original entrance to the temple. The change from the original stone and (presumably) solid brick pier superstructure to a hollow tower format may have been designed to reduce weight-stress on the abutting temple facade, or perhaps to facilitate the construction of high flanking towers.

It is no easy matter to evaluate the significance of these architectural changes for cult practice and religious belief. The changed architectural form of the early Late Bronze Age temple need not reflect any significant change in cult, but we suggest that it does. The action of dividing off a Holy of Holies for the first time is a significant departure from previous practice, and bespeaks an altered view of the relationship between man and god.

The Ugaritic religious epics (Pitard 2002) contain legends that document the triumph of Baal in a war between the gods, and is generally interpreted as recording the spread of Baal worship in early Late Bronze Age Canaan. This assumed pre-eminence seems to have led to the attribution of virtually all Late Bronze Age Canaanite temples to Baal, even though very few have any inscriptional evidence to favour such an association (Mazar 1992). Male iconography does predominate (figurines, cult statues, incense burners) so the worship of some male deity is not easily disputed. When Late Bronze Age texts do identify individual temples, they name Baal, Reshep, Hadad and Dagan as titular deities (van der Toorn et al. 1999).

As well, there is an undoubted presence (if not pre-eminence) of Baal worship in the southern Levantine Late Bronze Age, more specifically at Pella where ruling prince Mut-Balu proclaims his loyalty to Baal by his very name (Hess 1989). Thus the broad association of Baal, Hadad or Dagan worship with Late Bronze Age temples is not unreasonable. Explaining the apparent change from El to Baal worship is more of a problem, although the spread of Hurrian peoples and their distinctive religious beliefs into Canaan and southern Anatolia at this time is well documented (Na’aman 1994; Hess 1997). It may be that the spread of Baal worship into the southern Levant is broadly connected with the arrival of Hurrian immigrants and the rise of the Hurrian Mitannian empire (Klengel 1992). Indeed, it was this new presence of Hurrian Mitannians in the southern Levant that Thutmosis III claimed to have provoked his first military campaigns, which ultimately brought much of Canaan (including western Jordan) under Egyptian control for the first time (Redford 1992).

South Levantine Late Bronze Age temple architecture changes from the simple “empty-box” form of the Middle Bronze Age temples at Shechem, Megiddo and Hazor (Area A), to more architecturally complex internally subdivided structures such as those at Hazor (Area H), Lachish (Acropolis Temple) and Beth Shan (Mekal). Architectural change need not reflect change in cult practice, but when these architectural changes occur across the south Levantine landscape at the same time as new North Syrian Hurrian cultural traditions appear, there is more strength to arguments that seek to link changed architectural forms with changing religious beliefs (Hess 1989). We view the early Late Bronze Age changes to the original Middle Bronze Age temple form at Pella in this context of widespread change in cultural and religious beliefs, largely attributable to the influence of North Syrian Hurrian religious forms.

Second Constructional Phase: Late Bronze Age II (ca 1350 – 1150 BC)

from Bourke (2004:8-9)

A major change to the re-modelled temple design occurred around 1350 BC, probably as a result of severe earthquake damage. Similar earthquake-related damage is found throughout the city and in buildings on nearby Tell Husn (Bourke et al. 1999). Two key alterations occurred thereafter. The entire temple structure was narrowed and the cella was provided with a colonnade.

The entire structure was levelled down to the stone foundations and new (much less massive) stone and mudbrick walls were built along the outer edge of the original east, south and western wall lines. However, a new north wall line was created five metres to the south of the Middle Bronze Age original, resulting in a significant narrowing of the entire structure. This was probably brought about by the sharp warping of the underlying foundations in the north temple area, still clear today from aerial photographs. At this time the original wide entrance to the Holy of Holies was narrowed and re-centred, and rebuilt using roughly dressed limestone and more carefully dressed (and drilled) basalt orthostat blocks; the latter were probably reused from earlier structures. Two small basalt column bases now flanked the re-configured entrance to the Holy of Holies. The floor of the Holy of Holies was re-laid, with new small stone foundational layers sealed by a thick, yellow, plaster floor surface. A number of distinct Egyptian-style foundation deposits were placed in shallow pits below this re-laid floor.

The new (much narrower) rectangular cella to the east of the rebuilt Holy of Holies was provided with a central colonnade at this time, indicated by the presence of three pillar bases. The western and eastern column bases had relatively small sub-structural foundations, but the central column base was provided with a massive limestone sub-structure, implying that it was designed to be the major weight-bearing support. All three column base foundations were cut into the mud brick paving of the original cella floor. Traces of burnt wooden columns were found in direct association with both of the smaller column bases. Thin, off-white plaster floors were laid across the narrowed cella area and the lower regions of the interior wall surfaces were sealed with a thick, monochrome, pale brown mud plaster.

The eastern facade of the temple was also remodelled, although subsequent Iron Age re-use in this area has made the exact form of the Late Bronze Age structure difficult to reconstruct with any confidence. However, it seems probable that the two hollow-square towers flanking the early Late Bronze Age temple entrance collapsed in the earthquake and were not rebuilt. If this interpretation is correct, then the area to the east of the reconstructed east wall would have been an open pebble-paved plaza.

These alterations to the early Late Bronze Age temple form could be interpreted as a simple structural response to severe earthquake damage, in that virtually all changes could be seen as a necessary strengthening of the original structurally unsound “hollow-box” design. However, the construction of a pillared hall, the addition of flanking columns at the entrance to the Holy of Holies, and the presence of “Egyptianising” foundation deposits may all reflect a new cultural influence at work.

While we have no reason to posit major change in local religious beliefs in the Late Bronze Age II, the architectural remodelling of the Pella temple coincides closely with the first presence of the Egyptian Nineteenth Dynasty pharaohs in the region (Redford 1992). This Late New Kingdom dynasty profoundly changed the ways in which the Canaanite empire had been administered previously, being far more inclined to interfere directly in the running of vassal states (Weinstein 1981). From this time (ca 1300 BC) an accelerated “Egyptianisation” of local elite culture can be observed, as can direct Egyptian influence on local Canaanite architectural modes (Wimmer 1990; Higginbottom 1996). With this in mind, it may be that the Egyptianising foundation deposits and the pillared hall at Pella provide evidence for an increasingly pervasive Egyptianisation of local elite culture east of the Jordan during the later New Kingdom.

The remodelled temple remained in use until the end of the Bronze Age (ca 1150 BC), when the entire site of Pella suffered a major destruction. This may also have been due to earthquake activity, although human agency remains possible, as this is the time of the enigmatic Sea People descent on Egypt, generally (if not always reliably) associated with a widespread destruction horizon throughout the region at this time (Sandars 1978).

Iron Age I Temple Use (ca 1150 – 950 BC)

from Bourke (2004:9)

The post-destruction Iron Age I (ca 1150– 950 BC) temple deposits were badly disturbed by later building activities. There is some meagre archaeological evidence for activity in and about the Holy of Holies during the two hundred years of the Iron I period, but the pillared hall and eastern facade seems to have collapsed into ruin. All areas surrounding the Holy of Holies were given over to domestic use, and much disturbed by numerous rubbish pits (Bourke et al. 2003). The immediate area in and about the Holy of Holies seems to have retained some measure of cultic function, although the mixture of cultic and domestic practice renders the precise nature of cult practice obscure

Third Constructional Phase: Iron Age II (ca 950 – 800 BC)

from Bourke (2004:10-11)

The third constructional phase marked a profound change to the form and (arguably) the function of the Pella temple. After the two hundred years of decline represented by the Iron Age I (ca 1150–950 BC), the area of the Late Bronze Age Holy of Holies was completely rebuilt as two separate storage and cultic rooms, with access to the cult room via an indirect entranceway in the southeast corner. The cult room was provided with benches around its west and north sides, and what appears to be a stepped mud brick podium was built against the eastern wall, perhaps for the display of cult paraphernalia. The northern room was filled with baskets of lentils and bags of grain, all burnt in the final destruction. Most cultic items, favissae and offering debris were located in the open courtyard area immediately to the east of the cult room. The roughly square courtyard area was dominated by a massive stone altar, positioned roughly in the centre of the courtyard. The major cult items, which included the ceramic “Cow Box”, and associated incense cups and a chalice, were found in destruction debris beside the stone altar.

Direct architectural parallels for the Iron Age II temple form are elusive. At Shechem (Stager 1999) and Tel Kittan (Eisenberg 1977), Iron Age II structures were reconstructed directly on top of Late Bronze Age originals, although the architectural forms are not particularly close to those at Pella. However, reasonably close parallels are found with the Iron II temples from Tel Qasile (Mazar 1980) on the Palestinian coast, and some individual design elements are paralleled in Iron Age temples at nearby Beth Shan (Rowe 1940). The contemporary material culture (and cult practice) at these two sites display an eclectic mixture of local Canaanite and “Aegean-Cypriot” influences, which many researchers equate (rather shakily) with the “Sea Peoples” (Tubb 2001), or more specifically with the better-known Biblical Philistines (Dothan 1982).

Architectural parallels are consistent with a significant change in cult practice, and offering vessels and figurines display relatively unambiguous links with the Palestinian coast for the first time during these Iron I-II horizons. Whilst it is probably unwise to equate specific politico-historical events with changing archaeological circumstances, the sharp change in cult practice at Pella does seem to indicate the presence of a major new influence in the region, with all archaeological indicators favouring a source on the Palestinian coastal plain (Singer 1994). It is difficult not to view these purely archaeological circumstances as consistent with Biblical testimony relating to the penetration of the originally coastal Philistine peoples into the eastern Jezreel Valley, which many regard as occurring at precisely this time (Raban 1991; Singer 1994).

The remodelled Iron Age II temple precinct at Pella was in use for perhaps 150 years (ca 950–800 BC) before the temple and the entire settlement was destroyed in an extensive conflagration (Bourke et al. 2003). While earthquake activity has been suggested as the likely cause for similarly dated destruction horizons at Deir ’Alla (Franken 1992), the same horizon of destruction at nearby Tell Hammeh (Cahill et al. 1987) and Tel Rehov (Mazar 1999) has been attributed to the military activities of either Egyptian (or just possibly) Aramaean invaders. At Pella, while earthquake destruction is still considered the most probable cause, significant militaria (specifically iron arrowheads and scale armour) are consistently associated with this destruction horizon. Whatever the ultimate cause, this destruction proved to be catastrophic to the long-term well-being of the city of Pella, as settlement ceased across the site for the next 500 years, only reviving with the Seleucid occupation of the region after 200 BC (Bourke 1997)

ARTEFACTUAL DISCOVERIES AND CULTIC FUNCTION

The Late Bronze Age II Temple: Foundation, Offering and Destruction Deposits

from Bourke (2004:13-14, 20)

... Finally, there are materials found in situ within the destruction deposits which mark the end of the Late Bronze Age temple (ca 1150 BC). These materials are normally found upon the floor surfaces, and therefore may well instruct on the spatial patterning of cult practices, although much material is found within a thick destruction deposit that sealed the entire area. Materials found within this deposit could theoretically derive from upper storey and rooftop locations, as well as lower storey floor levels.

... Destruction Deposits; Ceramic Objects

A variety of offering utensils were recovered, sometimes in pieces, sometimes more or less intact, either contained within or sealed below a thick layer of ash and brick debris.

...

The Iron Age II Temple: Offering and Destruction Deposits

from Bourke (2004:13-14, 20)

... Destruction Deposits

The destruction of the Iron II temple (ca 800 BC) ended significant occupation in the area for more than five hundred years. Thick deposits of ash and brick debris sealed the temple proper and most nearby areas. Interpretation of the final Iron II destruction horizons is complicated by the large and intrusive Late Antique (ca 550 AD) foundation trenches that cut through much of the area, largely frustrating attempts to study the spatial patterning of objects found in situ below destruction horizons. The Iron II temple proper suffered quite severely from later constructional activity. However, the area of the eastern courtyard surrounding the central altar was largely undisturbed, and it was here that many cult objects were identified. These include the ceramic model shrine (the “Cow-Box”), perforated ceramic cups, used for incense offerings, and the painted ceramic chalice, used in libation offerings.

...

Raphael and Agnon (2018)

from Raphael and Agnon (2018:776)

Iron IIB (900-700 BCE)

entire settlement destroyed (Phase 5) c. 950-850/800 BCE, probably by an earthquake. Arrowheads and scale armor were found (Bourke 2012: 190-191).

Walmsley et. al. (1993)

from Walmsley et. al. (1993:189)

AREA XXXII EXCAVATIONS

... Iron Age (Area XXXII Phases C-E)²⁹

Excavation of the well-constructed architecture of Phase D was completed this season. In plot XXXIIA this consisted of three small square rooms which abut each other from north to south, a partially covered courtyard to their east, and a small alleyway, to the north. In plot XXXIIB, to the west, the continuation of these three north south rooms was exposed, whilst a small north-south passageway was found to separate these kitchen structures from a large open courtyard further to the west.

There is nothing about these structures to suggest anything other than a well constructed domestic housing unit. Extensive ceramic material has been recovered off the floors (Fig. 15: 1-3). The complex underwent a fiery destruction, probably at some stage during the tenth century BC.

Although ceramic analysis is still at a stage, considerably more cataloguing has reinforced the impression that there is but a small interval between the destructions of Phase D (late tenth century BC?) and Phase C (early ninth century BC?).

Canadian excavations at Tell al Hammeh continue to produce material close to our own in circumstance and kind, and the extensive Iron Age destruction at Tell as-Sa'idiyeh would seem to parallel our Phase D horizon.³⁰ Together they suggest some form of linked circumstance in the destructions attested to in the region. Whilst the latter destruction (Phase C) may well have a natural cause-and recent controversy over the extent and impact of Shishak's razzia notwithstanding-the case for the earlier (Phase D) destruction being attributed to Shishak should be considered. Whilst Ussishkin is right to point out that there has never been any compelling evidence to attribute the extensive early Iron Age II destruction horizon in Palestine to Shishak, Dever is equally correct to observe that the extensive destruction horizons do exist, Shishak or no, and some, most probably linked, explanation for the extensive disruption must be attempted.³¹

Whether it be through earthquake or human action, it seems certain now that Pella shares in the misfortunes of its western neighbours in the tenth century BC.

Footnotes

29. Supervisors: Erin Crumlin (XXXIIA); Jodie Benton (XXXIIB); Graham Phillip (XXXIIC).

30. J. Cahill, G. Lipton and D. Tarler, 'Tell el Hammah, 1988', /El 38 (1988), pp.191-194; J.N. Tubb, 'Preliminary Report on the Fourth Season of Excavations at Tell es-Sa'idiyeh in the Jordan Valley', Levant 22 (1 990) pp. 21-38.

31. D. Ussishkin, 'Notes on Megiddo, Gezer, Ashdod, and Tel Balash in the Tenth to Ninth Centuries b.c.', BASOR 277-8 (1990), pp. 71-91.

Hellenistic Destructions - c. 220 - c. 200 BCE and c. 100(?) - 80/79 BCE

Tidmarsh (2024) identified two Hellenistic destruction levels. They were named the "Jannaeus Destruction" and the the "Antiochus Destruction level". The "Jannaeus Destruction" was identified in numerous areas, dated to early in the 1st century BCE (Hellenistic 3C - c. 100(?) - 80/79 BCE), and attributed to military activity during the reign of Hasmonean King Alexander Jannaeus (r. 103-76 BCE). Tidmarsh (2024:64) dated the "Antiochus Destruction level" to Hellenistic 2B (c. 220 - c. 200 BCE). The "Antiochus Destruction level" was identified in Area XXXIVB east on Tel Husn.

Polybius in The Histories (Book 5 Section 70) noted that Seleucid Emperor Antiochus III (r. 223-187 BCE) conquered Pella. Smith (1973:37-39) quotes and analyzes excerpts from Josephus and Georgius Syncellus which indicate that Alexander Jannaeus (r. 103-76 BCE) conquered Pella.

551 CE earthquake - doubtful

Discussion

Ambraseys (2009) states:

Russell also argues that Pella and Ramat Rahel were damaged in this event [551 CE Beirut Quake]. Pella is 100 km southeast of Tyre, but Ramat Rahel is just south of Jerusalem, thus it is impossible that this earthquake damaged the latter. Ambraseys et al. (1994, 24-25) wrongly place the epicentral region of this event in the Jordan Rift Valley. This was due to the bias of information from the debatable archaeological evidence in Russell (1985).

7th century CE earthquake

Figures

Fig. 1 - Map of archaeological areas

Figure 1

Map of archaeological areas at Pella

(Pella Project, modiﬁed Walmsley).

Walmsley (2007)

at Pella and environs from Walmsley (2007)
Fig. 4.7 - Map of pre- and post-749
Figure 4.7

Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level).

Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m

courtyard house, damaged in the 659 earthquake, -43 m

central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m

East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m

eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m

extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m

buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m

west residential quarter (excavation back-filled),destroyed in 749, -51 m

West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m

policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt

-2 m (Base image: Google Earth, 20 August 2021).

Blanke and Walmsley (2022)
Islamic Fiḥl from Blanke and Walmsley (2022)

Discussion

Blanke and Walmsley (2022:86) report that Urban Fiḥl was badly impacted by a 7th century CE earthquake resulting in permanent reconfigurations to the town’s layout. Dating was established by secure ceramic comparisons with the corpus from excavations of the eastern residential area on the main mound [i.e. Pella]. Reported archaeoseismic evidence, some of which is rebuilding evidence, is summarized in the table below. Smith (1973:165) attributed damage to the West Church Complex (Area I) to the 659/660 CE Jordan Valley Quake(s) while noting that this event toppled most of the upper part of the walls of the sanctuary and Atrium and damaged the north dependency. Smith et al. (1989:92) suggested that this earthquake also damaged the Area IX Church.


Location	Image(s)	Description(s)	Source(s)
‘policing and administrative garrison’ at Tall al-Husn - Fig. 4.7 [10]	Fig. 4.7 [10] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	excavations by Watson on the summit of Tall al-Ḥuṣn (Fig. 4.7 [10]), which commenced in 1988, identified a multi-phase ‘policing and administrative garrison’ similarly destroyed in 659, a date established by secure ceramic comparisons with the corpus from excavations of the eastern residential area on the main mound. The site, regardless of its commanding view over the Jordan Valley, was not reoccupied. the barracks was shattered by the force of the tremor and never rebuilt	Blanke and Walmsley (2022:86) Walmsley (2007)
The eastern end of the Pella mound (aka the main mound) - Fig. 4.7 [5]	Fig. 4.7 [5] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	a residential area suffered a partial collapse of buildings following which an east–west access street was closed off and multi-roomed houses on two levels repaired and equipped with large private courtyards. houses required substantial rebuilding and remodelling after the quake	Blanke and Walmsley (2022:86) Walmsley (2007)
The eastern end of the Pella mound (aka the main mound) - Fig. 4.7 [2]	Fig. 4.7 [2] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	a derelict courtyard house that was not rebuilt after an apparent mid-seventh-century collapse, although evidence of animal butchering reveals continued use of the space	Blanke and Walmsley (2022:86)
The valley below the tall - Fig. 4.7 [3]	Fig. 4.7 [3] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	the central church suffered significant damage that resulted in the permanent blocking of the Western portal only decades old and the reinstatement of the main entrance on the northern side (Fig. 4.7 [3]). The colonnade at the top of the church staircase seemingly collapsed at this time. An unwillingness to restore the portico and staircase suggests that the earthquake caused havoc in the valley, with broad evidence for land movement causing architectural slumping and infilling with coarse alluvium and boulders, the remediation of which would have been Sisyphean. The churches also seem to have been adversely effected by this seismic event, although there is argument about the extent to which they were still functioning in the mid-7th c.	Blanke and Walmsley (2022:86) Walmsley (2007)

References

Walmsley (2007)

Walmsley (2007) reports the following archeoseismic evidence at Pella (Fihl)

The [659/660] earthquake is particularly evident on Tall al-Husn [in Pella], where the barracks was shattered by the force of the tremor and never rebuilt, and on the main mound, where houses required substantial rebuilding and remodelling after the quake. The churches also seem to have been adversely effected by this seismic event, although there is argument about the extent to which they were still functioning in the mid-7th c.¹³

Footnotes

13 Cathedral church: Smith and Day (1989) 90–94; West Church: Smith (1973) 164–67. In the reports on both of these churches, Smith argues for an intermediary earthquake in A.D. 717, but at Pella like elsewhere the evidence for this event is most inconclusive. Quite possibly the structural damage attributed by Smith to A.D. 717 was the result of the irrefutable earthquake of 749, while the post-A.D. 717 robbing and stripping of the churches occurred after A.D. 749. If correct, this would mean both churches were still in use until destroyed in the A.D. 749 cataclysm.

Walmsley (2007) attributes some archeoseismic destruction at Pella due to the Jordan Valley Quake although this date assignment seems tentative.

Excavations in the early 1980s identified six house units destroyed in the earthquake of 749. These houses represented the last phase in a long urban development that commenced with the complete redevelopment of living quarters on Pella's main mound in the first half of the sixth century (Watson 1992). The original arrangement consisted of four-metre wide gravelled streets set out on a formal grid, each street flanked by stone and mudbrick terrace-style houses two storeys high, prefaced in some places by shops. These streets, intended to serve local needs, were not equipped with colonnades or sidewalks. Although modified, the layout remained the same until an earthquake in 659-60 required a rebuilding of the quarter, in which the linear terrace houses were replaced by independent, self-contained units centred on one or more sizeable courtyards.

Blanke and Walmsley (2022)

from Blanke and Walmsley (2022:86)

Figures

Fig. 1 - Map of archaeological areas

Figure 1

Map of archaeological areas at Pella

(Pella Project, modiﬁed Walmsley).

Walmsley (2007)

at Pella and environs from Walmsley (2007)
Fig. 4.7 - Map of pre- and post-749
Figure 4.7

Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level).

Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m

courtyard house, damaged in the 659 earthquake, -43 m

central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m

East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m

eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m

extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m

buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m

west residential quarter (excavation back-filled),destroyed in 749, -51 m

West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m

policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt

-2 m (Base image: Google Earth, 20 August 2021).

Blanke and Walmsley (2022)
Islamic Fiḥl from Blanke and Walmsley (2022)

Excerpt

Urban Fiḥl was badly impacted, resulting in further and permanent reconfigurations to the town’s layout. Early evidence for this earthquake was identified at the eastern end of the main mound in the 1979 to 1982 seasons (Fig. 4.7 [5]), where a residential area suffered a partial collapse of buildings following which an east–west access street was closed off and multi-roomed houses on two levels repaired and equipped with large private courtyards.⁵¹ In the valley below the tall, the central church suffered significant damage that resulted in the permanent blocking of the Western portal only decades old and the reinstatement of the main entrance on the northern side (Fig. 4.7 [3]).⁵² The colonnade at the top of the church staircase seemingly collapsed at this time. An unwillingness to restore the portico and staircase suggests that the earthquake caused havoc in the valley, with broad evidence for land movement causing architectural slumping and infilling with coarse alluvium and boulders, the remediation of which would have been Sisyphean. Later excavations by Watson on the summit of Tall al-Ḥuṣn (Fig. 4.7 [10]), which commenced in 1988, identified a multi-phase ‘policing and administrative garrison’ similarly destroyed in 659, a date established by secure ceramic comparisons with the corpus from excavations of the eastern residential area on the main mound.⁵³ The site, regardless of its commanding view over the Jordan Valley, was not reoccupied. Likewise, work by da Costa in a third location east of the main mound exposed a derelict courtyard house that was not rebuilt after an apparent mid-seventh-century collapse, although evidence of animal butchering reveals continued use of the space (Fig. 4.7 [2]).⁵⁴

Footnotes

51 Walmsley, ‘The social and economic regime at Fihl’. 254–255; Walmsley, ‘Households at Pella’. 250–251.

52 Smith and Day, Pella 2, 79–82, 90–94; earthquake damage was likely exacerbated by increased wādī wash, commonly allied with a variety of environmental and human causes, Walmsley, ‘The village ascendant’, 519–520.

53 Watson and Tidmarsh, ‘Pella/Tall al-Husn excavations 1993’, 303.

54 da Costa et al., ‘New light on late antique Pella’, 518, 525–527.

Walmsley in McNicoll et al (1982)

Walmsley in McNicoll et al (1982) noted the following:

only in one trench (IVE) has the Sydney team excavated much below the A.D. 746/7 surface, producing evidence for at least three Byzantine and Umayyad architectural phases. Since an attempt to establish a detailed chronology for the whole Umayyad period on the basis of this one trench would be premature, the following account concentrates on the final phase in the life of urban Pella.

...

We turn now to a consideration of the layout and use of the buildings in Areas III and IV (figs 28-29 and end-plates 2-3). A dominant feature of Pella in the Byzantine and early Umayyad periods appears to have been streets with packed mud and pebble surfaces. One such street, 5 m wide, ran east-west through Area IV. From it, north and south, doorways gave access to dwellings, hence referred to as the North and South Buildings. But at some stage during the Umayyad period the street was cut by a wall which continued south to form the west wall of the South Building. Before this event it appears that this building had covered a considerably greater area; now to the west of the north-south Umayyad wall the earlier walls were razed level with the new and final occupation surface of a courtyard. Into this surface were dug lightly fired clay tabuns. Although the date of the demolition of the western sector of the South Building and of the construction of the north-south wall is uncertain, the slight build-up of detritus on this surface points to a time not far removed from the final destruction of A.D. 746/7. Tentatively we ascribe these alterations to the period following the earthquake of A.D. 717.

The earthquake date of 717 CE was revised to 659/660 CE [Jordan Valley Quake(s)] in Walmsley (2007). The 717 CE Earthquake struck Syria and Jazira (Upper Mesopotamia) on Christmas Eve (~9 pm) 24 Dec. 717 CE. Reports from Upper Mesopotamia suggests an epicenter far from Pella indicating that another closer earthquake [e.g. Jordan Valley Quake(s)] was likely responsible for this archeoseismic evidence.

Smith (1973)

Provisional History of the West Church Complex

Introduction

from Smith (1973:164)

On the basis of the preceding descriptions of excavational results in Area I, it is possible to offer a provisional reconstruction of the history of the West Church complex. The evidence suggests that five major phases can be distinguished.

Phase 1: ca. A.D. 530-610 - phase ends with damage attributed to military activity (Byzantine-Sassanian War of 602-628 CE)

from Smith (1973:164)

The history of the West Church complex began with the construction of the sanctuary in the late 5th or 6th century, perhaps above a venerated late 1st or early 2nd century Christian grave at the eastern edge of a large cemetery which had been used as early as the Middle Bronze Age. It is possible that the church was commenced during the early years of the reign of Justinian I, when many large new ecclesiastical edifices were constructed in Syria and Palestine. The construction may have required many years for completion, and the decoration of the interior of the sanctuary — which eventually included a dado and glass mosaics on portions of the walls and ceiling, a colorful tessellated pavement for the central aisle, and other appointments — probably continued longer. Not long after the completion of the sanctuary an Atrium was begun on the west, and still later a dependency was constructed on the north, reached by a doorway cut into the church wall. This phase saw the greatest wealth and power which the worshippers were ever to enjoy. Violence seems to have terminated this prosperity. The end of the phase does not seem to have come from earthquake, although the fabric of the church was considerably damaged. The cist grave was also perhaps looted at this time. It is impossible for the damage to be attributed to the uprising of the Samaritans in 529 unless the construction of the church can be shown to have taken place earlier than the reign of Justinian I. An acceptable provisional hypothesis to account for the abuse which the church underwent at the end of this phase is the incursion of marauding Persians into Transjordan and Palestine in 610 [JW: part of the Byzantine-Sassanian War of 602-628 CE]. This date accounts reasonably well for the fact that in phase 2 (see below) the church was not restored to its former elegance, as it might well have been if the damage had occurred as early as 529, when the Byzantine empire was still rich and powerful.

Phase 2: ca. A.D. 610-660 - phase ends with damage attributed to the Jordan Valley Quake(s) of 659/660 CE

from Smith (1973:164-165)

Only a few years after the Persians invaded Transjordan and Palestine, the emperor Heraclius drove them back and brought temporary peace to the region. Sometime after 614 the church was repaired. The workmanship shows a sharp decline in quality. The dado of the sanctuary, which had been damaged or removed at the end of phase 1 was replaced by plaster. In the north Apse a low row of flat stone slabs from the former dado seem to have been reused to repair a part of the flooring of the sanctuary which presumably had been damaged. The great entrance on the west side of the Atrium was blocked off, and perhaps other measures were taken for increased security. A new interment was probably made in the cist grave during this phase. The floor of the north Apse was probably also raised 40 cm. during this phase, and the doorway which had been cut between the sanctuary and the north dependency in phase 1 was perhaps scaled. This phase apparently extends past the coming of the Islamic conquerors in 635. There is no evidence of any immediate abuse to the church by the new rulers, but the phase is generally one of much more modest operations than had been the case during phase 1. Throughout this time, however, the West Church complex was kept free of debris. The phase ends with damage to the church which can most plausibly be attributed, on the basis of present information, to the extensive and severe earthquakes of A.D. 658-660 (Willis 1928:80). One of these shocks, perhaps that of 658, toppled most of the upper part of the walls of the sanctuary and Atrium and damaged the north dependency.

JW: This earthquake or earthquakes struck in 659 and possibly 660 CE and are known as the Jordan Valley Quake(s). Smith (1973:165) is referencing the older somewhat error-prone catalog of Willis (1928).

Phase 3: ca. A.D. 660-717 - phase ends with damage attributed to an earthquake in 713 or 717 CE (both of which are unlikely quake candidates)

from Smith (1973:165)

A new phase began with repairs made to the West Church complex following the earthquake which ended the preceding phase. The damaged upper part of the walls of the sanctuary and Atrium was not repaired: rather, the walls were levelled off at a height of three to five meters and the unwanted stones were removed from the vicinity. The original massive piers of the sanctuary were likewise removed from the church, and a thin layer of soil was smoothed over the floor to fill in places where the collapse of the roof, piers and upper portions of the walls had gouged pits into the paved floor. Relatively small dressed stones were set up as replacements for the original pier bases, and upon these some sort of columns or timbers were set up to support a light-weight roof across the truncated budding. At least one column with base and capital was set up against the south wall as a support for the roof or a small gallery. The cast wall, perhaps slightly less damaged than the other walls, required at least one buttress to keep it from collapsing. The walls were replastered, less qualitatively than in phase 2. Probably connected with these repairs was a flimsy wall constructed amid several of the roof supports on the south side of the church, and a flight of steps which may have led to a pulpit or a low gallery. Whether or not by this time the southern part of the sanctuary had been appropriated by the Muslims as a mosque is not known, but it is a possibility. The Atrium continued to be used in some fashion. but may have undergone considerable remodelling, including the repaving of sections of it. As the phase progressed. the church complex grew less adequately tended, with the result that the walls deteriorated further, and possibly portions of the sanctuary, Atrium and dependencies began to be abandoned and left to silt up. This phase seems to have come to an end with the earthquake of 713 or that of 717, which, though not so extensive or severe as some, were among the strongest in the region between 660 and 746 (Willis 1928: 78,80, as corrected by Ambraseys 1962). The walls of the church, already greatly reduced in height, may have suffered relatively little fresh damage, but the flimsy repairs and constructions made during the phase, such as the stairway in the nave and the column set up next to the south wall, toppled. It may be that the columns of the Atrium had largely stood until this time, but if so they fell during the shocks. The roofs of all the parts of the complex still in use likewise collapsed. Some pottery lying on the thin soil layer in the sanctuary was shattered and buried in the debris, a large section of which fell near the south Apse, pushing through the layer of accumulated soil and indenting the limestone paving.

JW: The epicenters of both the 713 CE Syria Quake and the 717 CE Earthquake were too far away to have caused such damage in Pella.

Phase 4: ca. A.D. 717-746 - phase ends with damage attributed to the Jan. 749 CE Holy Desert Quake of the Sabbatical Year sequence

from Smith (1973:165-166)

The earthquake damage which ended phase 3 was not repaired at all during this phase, unless a few spots in the church were cleared to permit squatters to live there or some of the stones were removed for use elsewhere. Lacking a roof, the sanctuary began to silt up, particularly on the east where the wind from the northwest filled in the crevices amid the debris more rapidly than elsewhere. The ruins of the Atrium and the dependencies also began to accumulate aeolian soil, though in some unsheltered spots perhaps as much soil blew away as was deposited. Some sort of simple domestic life was sporadically carried on amid the ruins. Animals were perhaps sheltered there, and in the north dependency women baked bread in an oven, leaving most of the ashes on the floor. Discovering that usable paving stones lay beneath the debris of the church complex, robbers groveled amid the debris, pulling up as many pavers as they could conveniently move. During this time of sporadic use, a certain amount of broken pottery found its way into the debris. Since only an insignificant amount of the potsherds within this debris is of Abbasid or later date, this phase may be said to have ended with the earthquake of 746, which according to Muslim historians was exceptionally severe, devastating much of Syria and Palestine, including the Jordan Valley (Willis 1928:80: cf. Abel 1933-1938, 53-34). Some further damage may have been done by earthquakes in 756 and the years following (Willis 1928:80), although these were less severe than that of 746 (Amiran 1950-1952:50-51). Just prior to 746, all that was left of the walls of the church was five to six courses, except at the eastern end where the walls were somewhat higher, and debris stood as deep as 2 m. in some spots. When the earthquake struck, virtually all of the courses of the walls not held in place by debris collapsed, generally falling westward. Buried in this final collapse were a few vessels in domestic use in the complex at the time.

The debris of phase 4, along with the smaller quantity from phase 3, constitutes a single Umayyad stratum. This stratum predominates in the excavated plots north of the sanctuary wall, and is the sole stratum inside the church and Atrium: west of the Atrium wall in Plot 5 it is less extensive than the Byzantine stratum (fig. 49). Because of the uneven way in which debris accumulated in phases 3 and 4, as well as a considerable amount of later turning over of the debris which seems to have taken place, it is impractical to attempt to distinguish substrata. The numismatic evidence for this Umayyad stratum is corroborative; the stratum contained, in addition to stray coins of earlier date, five post-reform Umayyad coins dating from ca. 700-750.

JW: The causitive earthquake would have been the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence.

Phase 5: ca. A.D. 746 to present

from Smith (1973:166)

After the earthquake which ended phase 4, almost all that was left of the West Church complex was the stones of the fallen half-dozen courses of the walls. Part of these stones silted up, but there was so little left to catch the dust that for the next thousand years the surface of the West Church complex remained almost static. Despite the opinions of some explorers during the last century, there is little evidence that many of the remaining stones from the church were taken away, perhaps because there were generally fewer people living at Pella and in the Jordan Valley to molest the ruins. With the passing centuries the stones weathered increasingly. Visitors to the spot left virtually no trace of their presence, save an occasional horseshoe or nail, or rarely a few sherds of a broken vessel. Goatherds probably occasionally threw potsherds from the central mound at their goats in the church's ruins, thereby contributing to a certain amount of extraneous pottery in the upper layers of debris. Such was the condition of the West Church complex when excavation began there in 1967.

Caveat

from Smith (1973:166-167)

One who reads this historical reconstruction may wonder whether or not it is warranted to link the evidence of damage in the West Church complex so precisely with earthquakes. Even though the earthquake damage in Area I is clear, there is admittedly the possibility of error in such identifications. One cannot always know how accurate the Islamic records of earthquakes are, as to either geographical spread or intensity — or, for that matter, dating. Most experts who have studied ancient earthquake records for the Middle East have, however. generally accepted the accuracy of the recorded tremors as to date, location and intensity. In associating the phases of the West Church complex with known quakes, I have selected only those shocks of such magnitude (approximately 7-10 on the Richter scale, so far as can be estimated: see Amiran 1950-1952) in Syria-Palestine that it is unlikely that they did not produce a distinct effect upon Pella, located as it is in the earthquake-prone Jordan Valley. The association of specific earthquakes with phases in the history of the West Church complex has, furthermore, been done with all other kinds of evidence — ceramic, numismatic and historical — in mind.

If any revision of the dating proposed here should later be required, it might be in the direction of a somewhat earlier date for the beginning of phase 1 and the end of phase 4. Many of the churches at Gerasa were constructed during the last third of the 5th century (Kraeling 1938:171), a date which cannot be entirely excluded as a possibility for the West Church. The date of the end of phase 4, here concluded to have occurred at the time of the earthquake of 746 [JW: actually 749], could conceivably have taken place with the earthquake of 717 [JW: an unlikely earthquake candidate]; if so, a corresponding adjustment to the dates of the other phases would be required. A revision in the direction of later dating of the phases would have far less warrant. In any case, none of these revisions would involve more than half a century.

Early 8th century CE earthquake ?

Discussion

Smith (1973:165) interpreted a seismic destruction in Phase 3 of the Western Church Complex in Area I. Although Smith (1973:165) suggested the 713 CE Syria Quake and the 717 CE Earthquake as candidates, the epicenters of both events were too far away to have caused such damage. This may, therefore, suggest an earthquake unreported in the Earthquake Catalogs. Smith (1973:165)'s description of the archaeoseismic evidence is repeated below:

The walls of the church, already greatly reduced in height, may have suffered relatively little fresh damage, but the flimsy repairs and constructions made during the phase, such as the stairway in the nave and the column set up next to the south wall, toppled. It may be that the columns of the Atrium had largely stood until this time, but if so they fell during the shocks. The roofs of all the parts of the complex still in use likewise collapsed. Some pottery lying on the thin soil layer in the sanctuary was shattered and buried in the debris, a large section of which fell near the south Apse, pushing through the layer of accumulated soil and indenting the limestone paving.

Smith et al. (1989:94) also interpreted a 717 CE seismic destruction in Phase 4 of the Area I Church Complex which he described as follows:

This phase ended in A.D. 717, when a massive earthquake damaged the Church and many other structures at Pella. Much of the superstructure of the sanctuary of the Church probably collapsed at that time, although some of the columns in the Atrium may have remained standing, as certainly did the two large columns of the porch on the north side of the Atrium. Falling masonry crushed portions of the floors and shattered, among other things, the marble screen beneath the altar.

References

Smith (1973)

Provisional History of the West Church Complex

Introduction

from Smith (1973:164)

On the basis of the preceding descriptions of excavational results in Area I, it is possible to offer a provisional reconstruction of the history of the West Church complex. The evidence suggests that five major phases can be distinguished.

Phase 1: ca. A.D. 530-610 - phase ends with damage attributed to military activity (Byzantine-Sassanian War of 602-628 CE)

from Smith (1973:164)

The history of the West Church complex began with the construction of the sanctuary in the late 5th or 6th century, perhaps above a venerated late 1st or early 2nd century Christian grave at the eastern edge of a large cemetery which had been used as early as the Middle Bronze Age. It is possible that the church was commenced during the early years of the reign of Justinian I, when many large new ecclesiastical edifices were constructed in Syria and Palestine. The construction may have required many years for completion, and the decoration of the interior of the sanctuary — which eventually included a dado and glass mosaics on portions of the walls and ceiling, a colorful tessellated pavement for the central aisle, and other appointments — probably continued longer. Not long after the completion of the sanctuary an Atrium was begun on the west, and still later a dependency was constructed on the north, reached by a doorway cut into the church wall. This phase saw the greatest wealth and power which the worshippers were ever to enjoy. Violence seems to have terminated this prosperity. The end of the phase does not seem to have come from earthquake, although the fabric of the church was considerably damaged. The cist grave was also perhaps looted at this time. It is impossible for the damage to be attributed to the uprising of the Samaritans in 529 unless the construction of the church can be shown to have taken place earlier than the reign of Justinian I. An acceptable provisional hypothesis to account for the abuse which the church underwent at the end of this phase is the incursion of marauding Persians into Transjordan and Palestine in 610 [JW: part of the Byzantine-Sassanian War of 602-628 CE]. This date accounts reasonably well for the fact that in phase 2 (see below) the church was not restored to its former elegance, as it might well have been if the damage had occurred as early as 529, when the Byzantine empire was still rich and powerful.

Phase 2: ca. A.D. 610-660 - phase ends with damage attributed to the Jordan Valley Quake(s) of 659/660 CE

from Smith (1973:164-165)

Only a few years after the Persians invaded Transjordan and Palestine, the emperor Heraclius drove them back and brought temporary peace to the region. Sometime after 614 the church was repaired. The workmanship shows a sharp decline in quality. The dado of the sanctuary, which had been damaged or removed at the end of phase 1 was replaced by plaster. In the north Apse a low row of flat stone slabs from the former dado seem to have been reused to repair a part of the flooring of the sanctuary which presumably had been damaged. The great entrance on the west side of the Atrium was blocked off, and perhaps other measures were taken for increased security. A new interment was probably made in the cist grave during this phase. The floor of the north Apse was probably also raised 40 cm. during this phase, and the doorway which had been cut between the sanctuary and the north dependency in phase 1 was perhaps scaled. This phase apparently extends past the coming of the Islamic conquerors in 635. There is no evidence of any immediate abuse to the church by the new rulers, but the phase is generally one of much more modest operations than had been the case during phase 1. Throughout this time, however, the West Church complex was kept free of debris. The phase ends with damage to the church which can most plausibly be attributed, on the basis of present information, to the extensive and severe earthquakes of A.D. 658-660 (Willis 1928:80). One of these shocks, perhaps that of 658, toppled most of the upper part of the walls of the sanctuary and Atrium and damaged the north dependency.

JW: This earthquake or earthquakes struck in 659 and possibly 660 CE and are known as the Jordan Valley Quake(s). Smith (1973:165) is referencing the older somewhat error-prone catalog of Willis (1928).

Phase 3: ca. A.D. 660-717 - phase ends with damage attributed to an earthquake in 713 or 717 CE (both of which are unlikely quake candidates)

from Smith (1973:165)

A new phase began with repairs made to the West Church complex following the earthquake which ended the preceding phase. The damaged upper part of the walls of the sanctuary and Atrium was not repaired: rather, the walls were levelled off at a height of three to five meters and the unwanted stones were removed from the vicinity. The original massive piers of the sanctuary were likewise removed from the church, and a thin layer of soil was smoothed over the floor to fill in places where the collapse of the roof, piers and upper portions of the walls had gouged pits into the paved floor. Relatively small dressed stones were set up as replacements for the original pier bases, and upon these some sort of columns or timbers were set up to support a light-weight roof across the truncated budding. At least one column with base and capital was set up against the south wall as a support for the roof or a small gallery. The cast wall, perhaps slightly less damaged than the other walls, required at least one buttress to keep it from collapsing. The walls were replastered, less qualitatively than in phase 2. Probably connected with these repairs was a flimsy wall constructed amid several of the roof supports on the south side of the church, and a flight of steps which may have led to a pulpit or a low gallery. Whether or not by this time the southern part of the sanctuary had been appropriated by the Muslims as a mosque is not known, but it is a possibility. The Atrium continued to be used in some fashion. but may have undergone considerable remodelling, including the repaving of sections of it. As the phase progressed. the church complex grew less adequately tended, with the result that the walls deteriorated further, and possibly portions of the sanctuary, Atrium and dependencies began to be abandoned and left to silt up. This phase seems to have come to an end with the earthquake of 713 or that of 717, which, though not so extensive or severe as some, were among the strongest in the region between 660 and 746 (Willis 1928: 78,80, as corrected by Ambraseys 1962). The walls of the church, already greatly reduced in height, may have suffered relatively little fresh damage, but the flimsy repairs and constructions made during the phase, such as the stairway in the nave and the column set up next to the south wall, toppled. It may be that the columns of the Atrium had largely stood until this time, but if so they fell during the shocks. The roofs of all the parts of the complex still in use likewise collapsed. Some pottery lying on the thin soil layer in the sanctuary was shattered and buried in the debris, a large section of which fell near the south Apse, pushing through the layer of accumulated soil and indenting the limestone paving.

JW: The epicenters of both the 713 CE Syria Quake and the 717 CE Earthquake were too far away to have caused such damage in Pella.

Phase 4: ca. A.D. 717-746 - phase ends with damage attributed to the Jan. 749 CE Holy Desert Quake of the Sabbatical Year sequence

from Smith (1973:165-166)

The earthquake damage which ended phase 3 was not repaired at all during this phase, unless a few spots in the church were cleared to permit squatters to live there or some of the stones were removed for use elsewhere. Lacking a roof, the sanctuary began to silt up, particularly on the east where the wind from the northwest filled in the crevices amid the debris more rapidly than elsewhere. The ruins of the Atrium and the dependencies also began to accumulate aeolian soil, though in some unsheltered spots perhaps as much soil blew away as was deposited. Some sort of simple domestic life was sporadically carried on amid the ruins. Animals were perhaps sheltered there, and in the north dependency women baked bread in an oven, leaving most of the ashes on the floor. Discovering that usable paving stones lay beneath the debris of the church complex, robbers groveled amid the debris, pulling up as many pavers as they could conveniently move. During this time of sporadic use, a certain amount of broken pottery found its way into the debris. Since only an insignificant amount of the potsherds within this debris is of Abbasid or later date, this phase may be said to have ended with the earthquake of 746, which according to Muslim historians was exceptionally severe, devastating much of Syria and Palestine, including the Jordan Valley (Willis 1928:80: cf. Abel 1933-1938, 53-34). Some further damage may have been done by earthquakes in 756 and the years following (Willis 1928:80), although these were less severe than that of 746 (Amiran 1950-1952:50-51). Just prior to 746, all that was left of the walls of the church was five to six courses, except at the eastern end where the walls were somewhat higher, and debris stood as deep as 2 m. in some spots. When the earthquake struck, virtually all of the courses of the walls not held in place by debris collapsed, generally falling westward. Buried in this final collapse were a few vessels in domestic use in the complex at the time.

The debris of phase 4, along with the smaller quantity from phase 3, constitutes a single Umayyad stratum. This stratum predominates in the excavated plots north of the sanctuary wall, and is the sole stratum inside the church and Atrium: west of the Atrium wall in Plot 5 it is less extensive than the Byzantine stratum (fig. 49). Because of the uneven way in which debris accumulated in phases 3 and 4, as well as a considerable amount of later turning over of the debris which seems to have taken place, it is impractical to attempt to distinguish substrata. The numismatic evidence for this Umayyad stratum is corroborative; the stratum contained, in addition to stray coins of earlier date, five post-reform Umayyad coins dating from ca. 700-750.

JW: The causitive earthquake would have been the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence.

Phase 5: ca. A.D. 746 to present

from Smith (1973:166)

After the earthquake which ended phase 4, almost all that was left of the West Church complex was the stones of the fallen half-dozen courses of the walls. Part of these stones silted up, but there was so little left to catch the dust that for the next thousand years the surface of the West Church complex remained almost static. Despite the opinions of some explorers during the last century, there is little evidence that many of the remaining stones from the church were taken away, perhaps because there were generally fewer people living at Pella and in the Jordan Valley to molest the ruins. With the passing centuries the stones weathered increasingly. Visitors to the spot left virtually no trace of their presence, save an occasional horseshoe or nail, or rarely a few sherds of a broken vessel. Goatherds probably occasionally threw potsherds from the central mound at their goats in the church's ruins, thereby contributing to a certain amount of extraneous pottery in the upper layers of debris. Such was the condition of the West Church complex when excavation began there in 1967.

Caveat

from Smith (1973:166-167)

One who reads this historical reconstruction may wonder whether or not it is warranted to link the evidence of damage in the West Church complex so precisely with earthquakes. Even though the earthquake damage in Area I is clear, there is admittedly the possibility of error in such identifications. One cannot always know how accurate the Islamic records of earthquakes are, as to either geographical spread or intensity — or, for that matter, dating. Most experts who have studied ancient earthquake records for the Middle East have, however. generally accepted the accuracy of the recorded tremors as to date, location and intensity. In associating the phases of the West Church complex with known quakes, I have selected only those shocks of such magnitude (approximately 7-10 on the Richter scale, so far as can be estimated: see Amiran 1950-1952) in Syria-Palestine that it is unlikely that they did not produce a distinct effect upon Pella, located as it is in the earthquake-prone Jordan Valley. The association of specific earthquakes with phases in the history of the West Church complex has, furthermore, been done with all other kinds of evidence — ceramic, numismatic and historical — in mind.

If any revision of the dating proposed here should later be required, it might be in the direction of a somewhat earlier date for the beginning of phase 1 and the end of phase 4. Many of the churches at Gerasa were constructed during the last third of the 5th century (Kraeling 1938:171), a date which cannot be entirely excluded as a possibility for the West Church. The date of the end of phase 4, here concluded to have occurred at the time of the earthquake of 746 [JW: actually 749], could conceivably have taken place with the earthquake of 717 [JW: an unlikely earthquake candidate]; if so, a corresponding adjustment to the dates of the other phases would be required. A revision in the direction of later dating of the phases would have far less warrant. In any case, none of these revisions would involve more than half a century.

mid 8th century CE earthquake

Figures

Fig. 1 - Map of archaeological areas

Figure 1

Map of archaeological areas at Pella

(Pella Project, modiﬁed Walmsley).

Walmsley (2007)

at Pella and environs from Walmsley (2007)
Fig. 3 - General plan of

Figure 3

General plan of the Umayyad housing in Area IV

(Walmsley).

Walmsley (2007)

the Umayyad housing in Area IV from Walmsley (2007)
Plan of House G from

Endplate - Area IV - Plan of House - House G

Walmsley and Smith in McNicoll et al (1992)

Walmsley and Smith in McNicoll et al (1992)
Plan of House G (magnified)

Endplate - Area IV - Plan of House - House G

Walmsley and Smith in McNicoll et al (1992)

from Walmsley and Smith in McNicoll et al (1992)
Fig. 4 - View of Pella

Figure 4

View of the Pella Houses A–B in Area IV, after excavation. Part of the section shown in figure 5 remains on the left; note the fallen column (Walmsley).

Walmsley (2007)

Houses A–B in Area IV from Walmsley (2007)
Umayyad Collapse in Area IV

Area IV Umayyad Collapse

Pella Project - University of Sydney

from the Pella Project - University of Sydney
Marble paving slabs shattered

Plate LXXVIII.1

Marble paving slabs in the sanctuary of the Civic Complex Church shattered in the earthquake of A.D. 746/47. The fragments were preserved by a column which collapsed on the slabs.

JW: Actually that earthquake struck in 749

Smith in ADAJ Plates 1983

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Smith in ADAJ Plates 1983
Skeletons of camels killed

Plate LXXXI.2

Skeletons found in the courtyard of the "hall of camels" just north of the Civic Complex Church. The animals are lying as they were killed in the earthquake of A.D. 746/7

JW: Actually that earthquake struck in 749

Smith in ADAJ Plates 1983

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Smith in ADAJ Plates 1983
Skeletons of humans killed

Plate LXXXI.2

Area IV Plot P (extension). Skeletons of two charred adult human beings with covering textiles, found in the AD 746/7 destruction deposit.

JW: Actually that earthquake struck in 749

Walmsley and Smith in McNicoll et al (1992)

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Walmsley and Smith in McNicoll et al (1992)

Discussion

Archeoseismic evidence

Archeoseismic evidence for a mid 8th century CE earthquake at Tabaqat Fahl (Pella) showed up in the form of collapsed structures, human and animal skeletons, and items of value found in the rubble including coins and other items. Perhaps the most dramatic evidence was found in an early Islamic domestic occupation level in Area IV. Rooms 13, 14, 15, and 16 of House G attested to tragic circumstances stemming from apparently rapid structural collapse. Five fallen columns and a pier originally arranged in two rows on an east-west axis, with three columns to the south and a combination of two columns and a pier in the northern row were discovered in the debris of Room 15 . Human and animal skeletons were found throughout. Pottery and other finds dated the destruction level to mid 8th century CE while the numismatic evidence provided a terminus post quem of A.H. 126 (25 October 743 - 12 October 744 CE).

Mid 8th century CE archaeoseismic evidence was also found in other excavations in Pella such as the church complex in the central valley (Area IX) and the West and East churches (Areas I and V) ( Walmsley, 2007).

Smith et al. (1989:94) noted that the Church Complex of Area IX had been abandoned and deconsecrated before the 749 CE earthquake struck with perhaps some domestic occupation and animal husbandry taking place in what remained of the structures. Walmsley (2007) reports the discovery of two human skeletons and several animals, including 7 camels (one in advanced pregnancy), a horse and foal, an ass, and 4 cows amongst the architectural debris of the Area IX Church Complex. Smith et al. (1989:94) noted that Umayyad coins of the first half of the 8th century found on the floor of the Chamber of the Camels and coins in the possession of one of the victims confirm the date of the final destruction.

In the Western Church Complex in Area I, Smith (1973:166) noted that during this event, virtually all of the courses of the walls that were not buttressed by debris collapsed, generally falling westward burying a few vessels in domestic use. Smith (1973:166) characterized Phase 4 of the Western Church Complex (Area I) as a single Umayyad stratum based on debris (e.g. pottery) and coins - the 5 latest of which were post-reform Umayyad coins dating from ca. 700-750. Phase 4 lay immediately below the presumed 749 CE collapse.

Winter Earthquake

Walmsley (2013) suggested that the earthquake struck in the winter.

The animals on the ground floor were chiefly cows (Rooms 8 and 9, totaling three) and small equids (mules or donkeys; inner courtyard and Rooms 6 and 7) – more costly animals than sheep and goats, hence their owners’ wish to shelter them properly during winter, the season in which the earthquake struck.

Nighttime Earthquake

The earthquake killed apparently sleeping humans and domiciled animals suggesting that the causative earthquake struck at night. Walmsley in McNicoll et al (1982:127) noted that one of the human skeletons in Area IX was found lying, as if sleeping and that a skeleton on the ground levels in Area IV was wearing a cloak or was wrapped in a blanket (Walmsley in McNicoll et al, 1982:138). Walmsley in McNicoll et al (1982:185) also reported on the discovery of two human skeletons (male and female) that had apparently fallen through the house from the second story and were covered in textiles which turned out to be a fine-weaved silk suggestive of bed clothes. Walmsley in McNicoll et al (1982:185) noted the following:

Apart from room 16, the main living area of the household was located upstairs. Although doubt surrounds the precise layout of the rooms of the upper storey, some at least were well fitted out with plain mosaic floors, plastered and painted walls, as well as reused marble features (PJI: 140-1). Most likely the owners, including the couple found in room 15 and the individual in 13, occupied the upper floor, while the apparently well-to-do `below stairs' stable-hand, occupied the ground floor room 16.

References

Walmsley (2007)

from Walmsley (2007)

Figures

Fig. 1 - Map of archaeological areas

Figure 1

Map of archaeological areas at Pella

(Pella Project, modiﬁed Walmsley).

Walmsley (2007)

at Pella and environs from Walmsley (2007)
Fig. 3 - General plan of

Figure 3

General plan of the Umayyad housing in Area IV

(Walmsley).

Walmsley (2007)

the Umayyad housing in Area IV from Walmsley (2007)
Plan of House G from

Endplate - Area IV - Plan of House - House G

Walmsley and Smith in McNicoll et al (1992)

Walmsley and Smith in McNicoll et al (1992)
Plan of House G (magnified)

Endplate - Area IV - Plan of House - House G

Walmsley and Smith in McNicoll et al (1992)

from Walmsley and Smith in McNicoll et al (1992)
Fig. 4 - View of Pella

Figure 4

View of the Pella Houses A–B in Area IV, after excavation. Part of the section shown in figure 5 remains on the left; note the fallen column (Walmsley).

Walmsley (2007)

Houses A–B in Area IV from Walmsley (2007)
Umayyad Collapse in Area IV

Area IV Umayyad Collapse

Pella Project - University of Sydney

from the Pella Project - University of Sydney
Marble paving slabs shattered

Plate LXXVIII.1

Marble paving slabs in the sanctuary of the Civic Complex Church shattered in the earthquake of A.D. 746/47. The fragments were preserved by a column which collapsed on the slabs.

JW: Actually that earthquake struck in 749

Smith in ADAJ Plates 1983

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Smith in ADAJ Plates 1983
Skeletons of camels killed

Plate LXXXI.2

Skeletons found in the courtyard of the "hall of camels" just north of the Civic Complex Church. The animals are lying as they were killed in the earthquake of A.D. 746/7

JW: Actually that earthquake struck in 749

Smith in ADAJ Plates 1983

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Smith in ADAJ Plates 1983
Skeletons of humans killed

Plate LXXXI.2

Area IV Plot P (extension). Skeletons of two charred adult human beings with covering textiles, found in the AD 746/7 destruction deposit.

JW: Actually that earthquake struck in 749

Walmsley and Smith in McNicoll et al (1992)

in the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence from Walmsley and Smith in McNicoll et al (1992)

Excerpts

6 house units were excavated at the eastern end of the main mound, each buried in a thick destruction deposit from an earthquake dating to the mid-8th c. A.D. (fig.. 1, Area IV

Figure 1

Map of archaeological areas at Pella

(Pella Project, modiﬁed Walmsley).

Walmsley (2007)

).

... As has already been noted, graphic archaeological evidence for the severity of the A.D. 749 earthquake has been found in different locations at Pella, notably the church complex in the central valley (Area IX), the West and East churches (Areas I and V), and the domestic quarters on the main mounds (Areas IV and VIII).

... The Pella evidence when revealed in the early 1980s presented a conclusive picture of a simultaneous urban destruction in A.D. 749. First proposed by Smith for the West Church (Area I) in the mid 1970s,¹⁸ new work at the central valley church (Area IX) and on the main mound produced much more definitive and illustrative information on urban life at Pella at the time of the mid-8th c. earthquake.

... The cathedral church and related structures excavated by Smith present an intriguing case of urban change after the Islamic expansion. Significant modifications occurred north of the church building, where a paved porticoed court, which coincided with the construction of the church, was modified by the construction of two-storeyed rooms faced with porches and galleries. Standing some 8.5 m high, this structure was clearly of major importance in the life of the town as it flanked the main entrance into the church, and was probably commercial in purpose (fig. 2).¹⁹ At least, this would seem to have been its final function on the day the A.D. 749 earthquake struck Pella with all of its intensity. Caught within the collapsing structure were two people and several animals, including 7 camels (one in advanced pregnancy), a horse and foal, an ass, and 4 cows. The age and condition of the camels - juvenile, young adult or in calf - suggests they were being sheltered on a particularly harsh winter’s day [JW: night].²⁰

... With these fatalities of nature were recovered several Umayyad coins, including a dinar (94 A.H./A.D. 712/13) and 14 dirhams (between 81 A.H./A.D. 700/701 and 112 A.H./A.D. 730/31), numerous ceramic vessels and lamps, iron implements including a dagger, a bronze jug and three glass bottles.²¹ A mass of iron tools was found in a smith’s workshop in the south-west corner of the building, including a pick head, shovel, shears and sledgehammer head. What was happening in the church at this time is disputed (see note ¹³). In the court preceding the church (Atrium), three iron torch holders were found along with pottery vessels, suggesting a not insignificant level of activity. Within the church building, in the south aisle, the skeletons of two humans were found, but, apart from a few ceramic vessels (including a group of arcane pierced pottery cones), little else, except architectural debris from the collapsed church. Subsequently, from a level above the collapse that contained later Abbasid-Fatimid material, stone was robbed from the destroyed building.²² This pattern of an A.D. 749 collapse and subsequent robbing matches that recognised in Area IV, and further suggests that the looting of the church fixtures did not occur until after the earthquake of A.D. 749.

On the main mound, due to the more complete nature of the deposits, the archaeological sequence is less obscured and the finds more representative, for it was houses - with all their human and material contents - that were lost to the unexpected ferocity of the A.D. 749 quake. In Area IV
(figs. 1

Figure 1

Map of archaeological areas at Pella (Pella Project, modiﬁed Walmsley).

Walmsley (2007)

, 3,

Figure 3

General plan of the Umayyad housing in Area IV (Walmsley).

Walmsley (2007)

, and 4

Figure 4

View of the Pella Houses A–B in Area IV, after excavation. Part of the section shown in ﬁ gure 5 remains on the left; note the fallen column (Walmsley).

Walmsley (2007)

), excavations between 1979 and 1983 revealed 6 house units destroyed at the time of the earthquake.

... House G, as with Houses A and B, has exposed the utter destructiveness inflicted on Pella by the A.D. 749 earthquake. The upper level(s) of brick collapsed instantaneously into the ground floor rooms, trapping people, animals and domestic objects within them; even the chickens pecking about in the central courtyard of House G had no time to flee the tumbling building.

... Entombed on the ground floor level were more domestic items along with many victims of the tragedy, human and animal. Based on the disposition of the artefacts in the collapse and the complete entrapped skeletons on the ground floor, it can be generalised that the downstairs area of the house was devoted to the care of valuable domesticated animals and light workshop activities, whereas the primary living quarters were located upstairs.

The human victims recovered on the ground floor consisted of three adults and a child. A single male was found crouched against the west wall of room 2, with one arm raised above his head in a desperate attempt to protect himself from heavy falling material (fig. 12

Figure 12

Detailed plan of House G

(Winikoff, modiﬁed Walmsley)

Walmsley (2007)

). Across the inner courtyard in its north-east corner, an adult male and female with a young child were found huddled together under the staircase.³⁰ Very possibly the couple and the child fell from the upper level (seemingly the balcony or the top of the staircase) before being entombed in the crumbling building. Four gold dinars were found with the solitary victim in room 2, and 6 others near the couple in the inner courtyard, perhaps dropped when they fell from the upper storey.³¹ In total, 10 dinars is not an insignificant amount of money, especially considering that the annual head tax for an adult amounted to some 4 dinars. The trapped animals on the ground floor were chiefly cows (rooms 8 and 9, totalling three), small equids (mules or donkeys; inner courtyard, rooms 6 and 7), sheep, goats, chickens and a cat. Apart from the chickens and cat, which roam at will, these animals represented wealth. This is especially the case for the cows and mules (the sheep and goats were few, and may represent special cases such as sickly animals); hence the attempt by their owners to provide sufficient shelter in winter.

... The victim found in room 2, separated from the inner court by a roughly built dividing wall, may have been living in this downstairs space, perhaps intermittently. A robust adult male aged between 25 and 30 years, both his physique and location near the animals would suggest that he was their carer (a job for which, the 4 dinars would indicate, he was well rewarded).

... Just outside the doorway into room 2, evidence for a small fire was found on the bench against the west wall of the inner courtyard. Probably being used for cooking rather than heating, the fire ignited the collapsed building debris in the inner courtyard, baking the fallen bricks and carbonising organic matter such as wood.

Gradually spreading through the tumbled collapse, the burning eventually reached the trapped couple and child by the staircase (hopefully all dead by this time), unusually preserving their clothing through the process of slow but complete combustion. Although the resultant carbonization of the material destroyed evidence of its original colouring, a study of the burnt cloth, which was manufactured from silk, was able to identify the different styles of weaving used in its manufacture.³⁴ Their prestige clothing, the likelihood they fell from the upper level during the earthquake and the discovery of 6 dinars close by would indicate the greater wealth and status of the couple by the staircase. Perhaps they were the owners of the house and animals, and for whom the male in room 2 worked as a herder.

Footnotes

13 Cathedral church: Smith and Day (1989) 90–94; West Church: Smith (1973) 164–67. In the reports on both of these churches, Smith argues for an intermediary earthquake in A.D. 717, but at Pella like elsewhere the evidence for this event is most inconclusive. Quite possibly the structural damage attributed by Smith to A.D. 717 was the result of the irrefutable earthquake of 749, while the post-A.D. 717 robbing and stripping of the churches occurred after A.D. 749. If correct, this would mean both churches were still in use until destroyed in the A.D. 749 cataclysm.

18 Smith (1973) 165–66.

19 This interpretation differs in details from Smith, who dates the conversion to the early 6th c. and proposes the purpose of the building was ecclesiastical, suggesting it was intended as a ‘hospice’ or served an ‘administrative’ function. This author has, in addition, emphasised a commercial purpose for this construction, proposing they served as the early Islamic market and caravanserai of Pella. All three functions—accommodation, administrative and commercial - could have (and probably were) intended. Equally likely, Pella’s clergy-administrators were the persons responsible for this enhancement to the urban environment, acting as the local representatives within the province of al-Urdunn.

20 On camels—the ‘second pillar’ of Bedouin life—see the definitive work by Jabbur (1995) 191–237.

21 Smith and Day (1989) 67–71.

22 Smith and Day (1989) 50–52, note especially ﬁ g. 13 and the accompanying explanation.

30 Bourke (1992) 220–21.

31 Walmsley (2001). As Early Islamic dinars always carry a mint date, they provide a useful chronological ﬁ x, with dates of 96 A.H./A.D. 714/15, 97 A.H./A.D. 715/16, 106 A.H./A.D. 724/25 and 117 A.H./A.D. 735/36. In the courtyard, six further dinars were recovered at the end of the 1982 season, dating to 91 A.H./A.D. 709/10, two of 94 A.H./A.D. 712/13, 110 A.H./A.D. 728/29, 112 A.H./A.D. 730/31 and 122 A.H./A.D. 739/40. Chronologically, the best dating came from a small copper fals minted in Damascus in 126 A.H./A.D. 743/44, struck just a few years before the A.D. 749 earthquake.

34 Eastwood (1992).

Blanke and Walmsley (2022)

from Blanke and Walmsley (2022:86)

Figures

Fig. 1 - Map of archaeological areas

Figure 1

Map of archaeological areas at Pella

(Pella Project, modiﬁed Walmsley).

Walmsley (2007)

at Pella and environs from Walmsley (2007)
Fig. 4.7 - Map of pre- and post-749
Figure 4.7

Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level).

Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m

courtyard house, damaged in the 659 earthquake, -43 m

central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m

East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m

eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m

extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m

buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m

west residential quarter (excavation back-filled),destroyed in 749, -51 m

West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m

policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt

-2 m (Base image: Google Earth, 20 August 2021).

Blanke and Walmsley (2022)
Islamic Fiḥl from Blanke and Walmsley (2022)

Excerpt

The horror of entrapment and gruesome deaths inflicted on victims of the 749 earthquake is graphically documented by the excavation of six multi-storied dwellings at the east end of the main mound at Fiḥl (Fig. 4.7 [5]

). Heavy mudbricks from the walls of the upper floor plunged inwards filling the ground floor rooms, in the process trapping animals (from bovines to equids, chickens and cats) and at least five, up to ten, humans, all caught in the panic of nature’s dreadful paroxysm.⁶⁶ A large section of one house caught fire, carbonising two victims. Later, no appreciable attempt was made to recover the bodies of dead, nor were valuable objects retrieved such as coins in precious metals,⁶⁷ and no evidence for digging into the collapse was identified. The lingering reminder of the dead, left entombed in the ruins, would have evoked painful memories, and prolonged feelings of helplessness, loss and an absence of closure among survivors. For some time, maybe decades, these debilitating emotions inhibited resettlement on the Fiḥl tall, but settlement did return. In the 1987 excavation season, a deep cesspit was encountered, replete with finds of ceramic, metal, bone, glass and eggshell (Fig. 4.7 [6]), while recently extensive ninth to eleventh century occupational levels have been verified on the tall (Fig. 4.7 [7]); these discoveries only touch the edges.⁶⁸

Footnotes

66 Walmsley, ‘Households at Pella’. On pp. 247–250, the paper touches on other locations with striking evidence of the 749 earthquake at Fiḥl, notably the central church complex and market with victims (Area 9) mentioned earlier, also the west and east churches (Areas 1 and 5), and a second domestic quarter on the main mound (Area 8). On page 250 the words ‘robbing’ and ‘looting’ are used; these words are no longer part of Walmsley’s archaeological vocabulary.

67 Walmsley, ‘Islamic Coins’, 59–60, 147–149.

68 Edwards et al., ‘Preliminary report’, 81–86 (pit), O’Hea in da Costa et al., ‘New light on late antique Pella’, 523–525 on lustre-painted Islamic glass; McPhillips, pers. comm.

Walmsley and Smith in McNicoll et al (1982)

from Walmsley and Smith in McNicoll et al (1982)

Disinterred from amongst the debris that filled rooms 13-15 were numerous skeletons, both human and animal, as well as finds of pottery, stone and metals.

... In the north-east corner of room 15, two adult humans (a male and a female) were found in conjunction with a large mass of textile fragments (see Appendices 3 and 8). A number of equid (probably donkeys) and chicken skeletons were also uncovered at floor level, along with a severely crushed cat. Underneath a drum from one of the fallen columns a further six Umayyad dinars were recovered, with the latest dating to AH 122/AD 739-40. Chronologically more important, however, is the bronze coin of AH 126/AD 743-4 from room 16, minted just three years before the AD 747 earthquake. A list of these and other Umayyad coins will be found in Appendix 9.

Other finds from room 15 included more examples of mid-8th-century domestic pottery, especially cooking bowls (cf. PJ1: pl. 143: 2) as well as a hoe, harnessing rings, door hinges, and an iron lock.

In the northern part of room 15, the effects of a fire after the collapse of the building were clearly discernible. Column drums were cracked and blackened and the yellow clay bricks baked red from the heat. This fire also engulfed the human couple (pl. 120) trapped in the north-east corner of the room, although their tragedy is our blessing, as the fire carbonized and preserved organic remains usually lost by decay at Pella. Of note are the textiles (see Appendix 8), oak beams, straw from mats, date stones, and olive pips.

While removing the deposit in room 15, a number of interesting observations were made on the nature of materials used in the construction of the upper storey of the house. Numerous yellow clay and pebble bricks had fallen into this and the surrounding rooms from the upstairs walls along with segments of wooden beams used to support the floor and roof. A considerable number of large white tesserae were also found in room 15, in some cases still adhering to a pebble and mortar base. These originated, it would seem, from the floor of the room located above 15. The incinerated couple would have also fallen from this upper room when its floor collapsed during the first shocks of the AD 747 earthquake.

In the adjacent room 13, more equid skeletons were uncovered, as well as three hens and another human, the latter also from the upper storey. Objects from the deposit in this room included lamps, two of pottery and one of bronze, and a glass vase.

Unlike rooms 13 and 15, no skeletal material was excavated from the Umayyad levels in room 14. However, copious quantities of sherds from storage vessels were found at floor level; at least three large jars were crushed in situ according to the area supervisor (Edwards 1982).

Walmsley (2001)

from Walmsley (2001:57)

All three churches and the market were destroyed and, on the main mound, the large domestic quarters of two or three stories were completely flattened, entrapping the occupants and their possessions.

Smith (1973)

Provisional History of the West Church Complex

Introduction

from Smith (1973:164)

On the basis of the preceding descriptions of excavational results in Area I, it is possible to offer a provisional reconstruction of the history of the West Church complex. The evidence suggests that five major phases can be distinguished.

Phase 1: ca. A.D. 530-610 - phase ends with damage attributed to military activity (Byzantine-Sassanian War of 602-628 CE)

from Smith (1973:164)

The history of the West Church complex began with the construction of the sanctuary in the late 5th or 6th century, perhaps above a venerated late 1st or early 2nd century Christian grave at the eastern edge of a large cemetery which had been used as early as the Middle Bronze Age. It is possible that the church was commenced during the early years of the reign of Justinian I, when many large new ecclesiastical edifices were constructed in Syria and Palestine. The construction may have required many years for completion, and the decoration of the interior of the sanctuary — which eventually included a dado and glass mosaics on portions of the walls and ceiling, a colorful tessellated pavement for the central aisle, and other appointments — probably continued longer. Not long after the completion of the sanctuary an Atrium was begun on the west, and still later a dependency was constructed on the north, reached by a doorway cut into the church wall. This phase saw the greatest wealth and power which the worshippers were ever to enjoy. Violence seems to have terminated this prosperity. The end of the phase does not seem to have come from earthquake, although the fabric of the church was considerably damaged. The cist grave was also perhaps looted at this time. It is impossible for the damage to be attributed to the uprising of the Samaritans in 529 unless the construction of the church can be shown to have taken place earlier than the reign of Justinian I. An acceptable provisional hypothesis to account for the abuse which the church underwent at the end of this phase is the incursion of marauding Persians into Transjordan and Palestine in 610 [JW: part of the Byzantine-Sassanian War of 602-628 CE]. This date accounts reasonably well for the fact that in phase 2 (see below) the church was not restored to its former elegance, as it might well have been if the damage had occurred as early as 529, when the Byzantine empire was still rich and powerful.

Phase 2: ca. A.D. 610-660 - phase ends with damage attributed to the Jordan Valley Quake(s) of 659/660 CE

from Smith (1973:164-165)

Only a few years after the Persians invaded Transjordan and Palestine, the emperor Heraclius drove them back and brought temporary peace to the region. Sometime after 614 the church was repaired. The workmanship shows a sharp decline in quality. The dado of the sanctuary, which had been damaged or removed at the end of phase 1 was replaced by plaster. In the north Apse a low row of flat stone slabs from the former dado seem to have been reused to repair a part of the flooring of the sanctuary which presumably had been damaged. The great entrance on the west side of the Atrium was blocked off, and perhaps other measures were taken for increased security. A new interment was probably made in the cist grave during this phase. The floor of the north Apse was probably also raised 40 cm. during this phase, and the doorway which had been cut between the sanctuary and the north dependency in phase 1 was perhaps scaled. This phase apparently extends past the coming of the Islamic conquerors in 635. There is no evidence of any immediate abuse to the church by the new rulers, but the phase is generally one of much more modest operations than had been the case during phase 1. Throughout this time, however, the West Church complex was kept free of debris. The phase ends with damage to the church which can most plausibly be attributed, on the basis of present information, to the extensive and severe earthquakes of A.D. 658-660 (Willis 1928:80). One of these shocks, perhaps that of 658, toppled most of the upper part of the walls of the sanctuary and Atrium and damaged the north dependency.

JW: This earthquake or earthquakes struck in 659 and possibly 660 CE and are known as the Jordan Valley Quake(s). Smith (1973:165) is referencing the older somewhat error-prone catalog of Willis (1928).

Phase 3: ca. A.D. 660-717 - phase ends with damage attributed to an earthquake in 713 or 717 CE (both of which are unlikely quake candidates)

from Smith (1973:165)

A new phase began with repairs made to the West Church complex following the earthquake which ended the preceding phase. The damaged upper part of the walls of the sanctuary and Atrium was not repaired: rather, the walls were levelled off at a height of three to five meters and the unwanted stones were removed from the vicinity. The original massive piers of the sanctuary were likewise removed from the church, and a thin layer of soil was smoothed over the floor to fill in places where the collapse of the roof, piers and upper portions of the walls had gouged pits into the paved floor. Relatively small dressed stones were set up as replacements for the original pier bases, and upon these some sort of columns or timbers were set up to support a light-weight roof across the truncated budding. At least one column with base and capital was set up against the south wall as a support for the roof or a small gallery. The cast wall, perhaps slightly less damaged than the other walls, required at least one buttress to keep it from collapsing. The walls were replastered, less qualitatively than in phase 2. Probably connected with these repairs was a flimsy wall constructed amid several of the roof supports on the south side of the church, and a flight of steps which may have led to a pulpit or a low gallery. Whether or not by this time the southern part of the sanctuary had been appropriated by the Muslims as a mosque is not known, but it is a possibility. The Atrium continued to be used in some fashion. but may have undergone considerable remodelling, including the repaving of sections of it. As the phase progressed. the church complex grew less adequately tended, with the result that the walls deteriorated further, and possibly portions of the sanctuary, Atrium and dependencies began to be abandoned and left to silt up. This phase seems to have come to an end with the earthquake of 713 or that of 717, which, though not so extensive or severe as some, were among the strongest in the region between 660 and 746 (Willis 1928: 78,80, as corrected by Ambraseys 1962). The walls of the church, already greatly reduced in height, may have suffered relatively little fresh damage, but the flimsy repairs and constructions made during the phase, such as the stairway in the nave and the column set up next to the south wall, toppled. It may be that the columns of the Atrium had largely stood until this time, but if so they fell during the shocks. The roofs of all the parts of the complex still in use likewise collapsed. Some pottery lying on the thin soil layer in the sanctuary was shattered and buried in the debris, a large section of which fell near the south Apse, pushing through the layer of accumulated soil and indenting the limestone paving.

JW: The epicenters of both the 713 CE Syria Quake and the 717 CE Earthquake were too far away to have caused such damage in Pella.

Phase 4: ca. A.D. 717-746 - phase ends with damage attributed to the Jan. 749 CE Holy Desert Quake of the Sabbatical Year sequence

from Smith (1973:165-166)

The earthquake damage which ended phase 3 was not repaired at all during this phase, unless a few spots in the church were cleared to permit squatters to live there or some of the stones were removed for use elsewhere. Lacking a roof, the sanctuary began to silt up, particularly on the east where the wind from the northwest filled in the crevices amid the debris more rapidly than elsewhere. The ruins of the Atrium and the dependencies also began to accumulate aeolian soil, though in some unsheltered spots perhaps as much soil blew away as was deposited. Some sort of simple domestic life was sporadically carried on amid the ruins. Animals were perhaps sheltered there, and in the north dependency women baked bread in an oven, leaving most of the ashes on the floor. Discovering that usable paving stones lay beneath the debris of the church complex, robbers groveled amid the debris, pulling up as many pavers as they could conveniently move. During this time of sporadic use, a certain amount of broken pottery found its way into the debris. Since only an insignificant amount of the potsherds within this debris is of Abbasid or later date, this phase may be said to have ended with the earthquake of 746, which according to Muslim historians was exceptionally severe, devastating much of Syria and Palestine, including the Jordan Valley (Willis 1928:80: cf. Abel 1933-1938, 53-34). Some further damage may have been done by earthquakes in 756 and the years following (Willis 1928:80), although these were less severe than that of 746 (Amiran 1950-1952:50-51). Just prior to 746, all that was left of the walls of the church was five to six courses, except at the eastern end where the walls were somewhat higher, and debris stood as deep as 2 m. in some spots. When the earthquake struck, virtually all of the courses of the walls not held in place by debris collapsed, generally falling westward. Buried in this final collapse were a few vessels in domestic use in the complex at the time.

The debris of phase 4, along with the smaller quantity from phase 3, constitutes a single Umayyad stratum. This stratum predominates in the excavated plots north of the sanctuary wall, and is the sole stratum inside the church and Atrium: west of the Atrium wall in Plot 5 it is less extensive than the Byzantine stratum (fig. 49). Because of the uneven way in which debris accumulated in phases 3 and 4, as well as a considerable amount of later turning over of the debris which seems to have taken place, it is impractical to attempt to distinguish substrata. The numismatic evidence for this Umayyad stratum is corroborative; the stratum contained, in addition to stray coins of earlier date, five post-reform Umayyad coins dating from ca. 700-750.

JW: The causitive earthquake would have been the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence.

Phase 5: ca. A.D. 746 to present

from Smith (1973:166)

After the earthquake which ended phase 4, almost all that was left of the West Church complex was the stones of the fallen half-dozen courses of the walls. Part of these stones silted up, but there was so little left to catch the dust that for the next thousand years the surface of the West Church complex remained almost static. Despite the opinions of some explorers during the last century, there is little evidence that many of the remaining stones from the church were taken away, perhaps because there were generally fewer people living at Pella and in the Jordan Valley to molest the ruins. With the passing centuries the stones weathered increasingly. Visitors to the spot left virtually no trace of their presence, save an occasional horseshoe or nail, or rarely a few sherds of a broken vessel. Goatherds probably occasionally threw potsherds from the central mound at their goats in the church's ruins, thereby contributing to a certain amount of extraneous pottery in the upper layers of debris. Such was the condition of the West Church complex when excavation began there in 1967.

Caveat

from Smith (1973:166-167)

One who reads this historical reconstruction may wonder whether or not it is warranted to link the evidence of damage in the West Church complex so precisely with earthquakes. Even though the earthquake damage in Area I is clear, there is admittedly the possibility of error in such identifications. One cannot always know how accurate the Islamic records of earthquakes are, as to either geographical spread or intensity — or, for that matter, dating. Most experts who have studied ancient earthquake records for the Middle East have, however. generally accepted the accuracy of the recorded tremors as to date, location and intensity. In associating the phases of the West Church complex with known quakes, I have selected only those shocks of such magnitude (approximately 7-10 on the Richter scale, so far as can be estimated: see Amiran 1950-1952) in Syria-Palestine that it is unlikely that they did not produce a distinct effect upon Pella, located as it is in the earthquake-prone Jordan Valley. The association of specific earthquakes with phases in the history of the West Church complex has, furthermore, been done with all other kinds of evidence — ceramic, numismatic and historical — in mind.

If any revision of the dating proposed here should later be required, it might be in the direction of a somewhat earlier date for the beginning of phase 1 and the end of phase 4. Many of the churches at Gerasa were constructed during the last third of the 5th century (Kraeling 1938:171), a date which cannot be entirely excluded as a possibility for the West Church. The date of the end of phase 4, here concluded to have occurred at the time of the earthquake of 746 [JW: actually 749], could conceivably have taken place with the earthquake of 717 [JW: an unlikely earthquake candidate]; if so, a corresponding adjustment to the dates of the other phases would be required. A revision in the direction of later dating of the phases would have far less warrant. In any case, none of these revisions would involve more than half a century.

Smith et al. (1989)

Provisional History of the Civic Church Complex in Area IX

Introduction

from Smith et al. (1989:82)

The history of the Civic Complex Church falls into six major phases, the first four of which are reconstructed in plans (Figs. 23, 25, 26 and 28), in elevations (Figs. II and 19), and in perspective drawings (Figs. 24 and 27). In the final two phases there were no significant architectural changes other than gradual deterioration of the Phase 4 features.

Phase 1: ca. A.D. 400 to 525-550 - phase ends with renovation in Phase 2

from Smith et al. (1989:82-86)
Figs. 23 and 24

The Church was very likely Pella's Cathedral, perhaps constructed as a response to the city's having been granted a bishopric — a distinction which probably had come during the second half of the 4th century. The name of the saint(s) to whom the edifice was dedicated is not known. Historical texts record the names of three of Pella's bishops: Zebennos in the mid-5th century and Paulos and Zacharias in the first half of the 6th century (see texts 67-68, 70, 73-74 in Smith 1973:60-66). Although frugally incorporating a large quantity of architectural elements taken from buildings of the Roman period, the structure must have been erected at very considerable expense. The cost of construction was doubtless borne largely by Pella's own citizens, for the time of Constantinian benefactions was past and the era of Justinian's largesse to churches was still well in the future (see Avi-Yonah 1958:43-45). It is unlikely that the Church received any significant economic benefit from the magnanimity of pilgrims that contributed to ecclesiastical prosperity in Palestine during the Byzantine period.

The structure was built on what was probably the only spacious public area in the city other than the Forum, namely the large Parvis which had been constructed on a low rise above the Wadi Jirm in the Early Roman period. Foregoing the customary orientation due west in order to be able to utilize existing Roman constructions, the architects positioned the Church at the southwestern corner of the Parvis, with its south wall resting on the south retaining wall of the Parvis.

The edifice does not display many architectural idiosyncrasies. The rectangular plan was standard not only for churches in Palestine and Transjordan (e.g.. churches at Madaba; see Manfredi 1899), but also throughout the Byzantine world at the time. Southern Levantine churches were often more influenced by the ecclesiastical conventions of Constantinople than by Syrian church architecture, but some regional practices are evident. Typical of Levantine architecture was the use of stone rather than brick, undoubtedly because of the abundance of suitable material (Ovadiah 1970:212).

The construction of the sanctuary can be dated to around A.D. 400. In the absence of inscriptional or historical records this date must be regarded as approximate, based on both archaeological evidence and stylistic criteria. The coin of Constans or Constantine II (A.D. 337-350) found in the bedding of the floor of the Atrium suggests a terminus post quem for the building. The relatively high proportion of coins of 4th or 4th-5th century types, particularly ones dating from the reign of Arcadius (A.D. 395-408) or shortly before that time, that were found in various loci in the Church, should probably be given some weight as reflecting the time of the construction of the Church. Stratigraphic evidence was provided by several soundings below the Church. Potsherds found in the fill for the Bema and immediately beneath the floor of the Church in several soundings (Pl. 43D) were not sufficient in quantity or diagnostic features to provide extremely close dating, but none of them appeared to be later than the middle of the 5th century, if that late.

Architectural criteria tend to confirm this proposed due. The closest parallels for the original plan of the structure are in churches dated to the 4th-5th centuries. Church EA at Sardis, dated to the second half of the 4th century, with an Atrium constructed by the early 5th century, has its main entrance on the north side of its Atrium (Hanfmann 1983:199-200 and Fig. 289), just as the Pella edifice does. The design of a central Apse flanked by side chambers is found in churches of similar date in northern Syria (Lassus 1947:63), Israel (Negev 1982:15-16), and Italy (Mathews 1971:106). Furthermore. the arrangement of the freestanding lunate Bema has parallels chiefly in churches of the early 5th century, notably at Tiberias (Ovadiah 1970:180) and Et-Tabgha in Israel, although there is some uncertainty about the date of the latter (the beginning of the 5th century according to Ovadiah and de Silva 1982:130 but about A.D. 500 according to Krautheimer 1979:168, 504). The Atrium on the west supports a date anywhere from the 4th to the mid-5th century (cf. Ovadiah 1970:200; Crowfoot 1941:54).

The sanctuary has a length-width proportion of 3:2. but that fact itself is not narrowly diagnostic of date. The dimensions of the side aisles are somewhat more helpful: in 5th-century churches, the north and south aisles tend to be long and narrow, later becoming proportionately wider (Kraeling 1938:189). The width-length ratios of the north and south aisles of the Church are 1:7.21 and 1:7.48, respectively. These proportions are similar to those of early 5th-century churches at Shavei Zion and Falfirtin (both 1:7), and the Cathedral at Gerasa (1:8), ca. A.D. 365; in churches from the mid-6th century onward the ratio is closer to 1:5 (Prausnitz 1967:20). The width of the Church's nave in proportion to the aisles is 2.4:1 (north aisle) and 2.5:1 (south aisle), a proportion which more than fulfills Crowfoot's (1941:61) specification of a 2:1 ratio for early churches in Palestine, and is in contrast to the wider aisles of the somewhat later West Church and East Church at Pella (Smith 1973; Plan I: McNicoll, Smith and Hennessy 1982:119).

The earliest mosaic floors at the east end of the north and south aisles can also be used as a confirmation of dating, although without great precision. The presence of crosses in the border of the mosaic in the south aisle suggests that the mosaic was laid before A.D. 427, when an edict of Theodosius II forbade crosses as decorative elements on floors; this rule, however, was not always strictly followed in Palestine (Kitzinger 1965:8). Comparisons for the south mosaic belong to the late 4th and early 5th centuries, e.g., in churches at Zahrani in Lebanon (Chehab 1957:Pl. 48) and Shavei Zion (Prausnitz 32-36).

Other features of the Church provide little evidence for its date. The architectural details such as mouldings, capitals and bases largely, and perhaps entirely, represent spolia from Roman buildings of various dates and styles; these constitute, however, an approximate terminus post quem, since none of them dates from later than the 4th century.

The position of the Church on the Parvis suggests that even if the Atrium was not built at the time the sanctuary was constructed, the original plan for the ecclesiastical complex provided for such a structure, and left a square space in front of the sanctuary on the west for that purpose. It is possible that initially a single portico was built against the west front, perhaps serving some of the functions that a Narthex later did in churches. When the Atrium was constructed that portico became the east side of the quadrangle. This hypothesis finds support in the fact that the east portico differs from the other porticos of the Atrium in elevation, width and paving. The remainder of the Atrium very likely was built soon afterward, still within Phase I. From that time until the 7th century the main access to the Church was through a large portal in the north wall of the Atrium, which was embellished with an elaborate door surround that may have been taken from a Roman-period building.

Inside the sanctuary two symmetrical east-west colonnades, composed mostly or entirely of reused architectural elements of the Roman period, divided the sanctuary into a wide nave and two relatively narrow side aisles. The walls were decorated with Ajlun limestone veneer, decorative inlays and mosaics. A freestanding Bema of lunate shape, raised some 40 cm above the Church's floor and doubtless bounded by some sort of screening, was constructed at the east end of the nave. The presence of an inscribed Apse is a Palestinian tradition (Crowfoot 1941:65) and has parallels in Transjordan (e.g., Manfredi 1899:160-161), but the freestanding lunate Bema in this Church is unusual.

From the 4th century on, various Christian writers expressed the opinion that the architecture of a church had symbolical meaning, although not all of them agreed on the precise nature of that meaning. An overarching concept was that as one advanced into a Byzantine church one encountered increasing sanctity, each stage of the approach to God being demarcated in some way. The sanctuary was a symbol, indeed even a microcosm, of God's heavenly realm. Eusebius clearly expressed that viewpoint, and in the Testamentum Domini Nostri Jesu Christi the three west portals of a church are said not only to correspond to the trinity but also to mark the transition between the secular and the divine (Kraeling 1938:175). The raised platform at the east end of the nave, with its altar, represented the throne of God. This view of reality presupposed the validity of analogy, and held that the Old Testament reflected divine truths through prefiguration; hence the church was seen as a perfect, cosmic Temple which supplanted the earthly temple of Jerusalem (Eusebius. Hist. Eccl., x.iv.46). The Church was probably designed with at least some of these concepts in mind. The pair of distinctive columns that flanked the central portal of the west front were perhaps intended as an analogy to the columns which flanked the door of Solomon's temple, and the curtain that hung between than may have been interpreted as a symbolical veil of the Temple.

In Byzantine churches some special liturgical functions were carried out at the east end of the north and south aisles, where special chambers often existed. In the sanctuary of the Church these pastophoria are clearly identified as places of liturgical importance by the presence of polychrome mosaic pavements: the only question is what purposes each of these served. Major sacerdotal functions in Byzantine churches of the period were performed in the prothesis, a place used for the preparation of the sacrament, and the diaconicon, used as a sacristy and vestry (see Cyril of Scythopolis. Vita Euthymii 39; Lassus 1947:163; Krautheimer 1979:503, n. 21, and Ovadiah 1970:195). Textual and inscriptional evidence for early churches shows that the prothesis and diaconicon had no fixed positions (Kraeling 1938:175-178;Orlando, 1952-54:151-52); hence the function of these areas in the Church cannot be identified conclusively, even though, as mosaic designs in the door show, the east end of each side aisle clearly had special importance. The west end of each side aisle also had a special function as a place where persons not yet admitted to full chuck membership could stand during services.

Adjoining structures to the north and south also belong to this phase. One of the most important of these was a long portico of Ionic columns that was constructed parallel to the north wall of the sanctuary and Atrium, facing the Parvis. The length of this stoa is not known, but the row of columns possibly extended past at least the central door in the north wall of the Church. A large dependency, possibly a clergy house, was constructed or reused on the south. The baptistry has not yet been located. A row of shops, whether owned by the Church or by others, was built — or rebuilt on Roman foundations — immediately to the west of the Atrium and one storey lower.

This phase was unquestionably one of relative stability for the Church at Pella, as paganism rapidly succumbed to Christianity while external threats were minimal. Christians were relatively secure under the Byzantine emperor, and burgeoning trade kept the citizens of Pella prosperous. The Church probably underwent some modifications during this time, but there are few discernible traces of repair except to the mosaic floors. It is possible that the mosaics had suffered damage in one of several major earthquakes that occurred during the period (Willis 1928, corrected by Ambrascys 1962). Although the end of this phase appears to have been in the second quarter of the 6th century, there is no evidence of any damage attributable specifically to the Samaritan uprising of A.D. 529.

Phase 2: ca. A.D. 525-50 to ca. 614 - phase ends with damage attributed to military activity (Byzantine-Sassanian War of 602-628 CE)

from Smith et al. (1989:86-90)
Fig. 25

Around the second quarter of the 6th century the Church underwent significant renovation. The alterations took place within a favorable economic. political and religious climate at Pella. Christianity enjoyed virtually a religious monopoly in the city, as elsewhere in the empire. The Emperor Justinian was encouraging the building and refurbishing of churches throughout the East. It was probably not by chance that the large West Church was constructed (Smith 1973:164) about the time that the Civic Complex Church was undergoing renovation.

The major alteration to the Church was the removal of the east wall of the sanctuary and the installation of a new wall having three semicircular apses. The apses were probably roofed with hemidomes. An inscription in mosaic (text 102 p. 139) possibly referring to the remodeling of the Church, may have extended around the base of the hemidome in the north Apse. Decorative columns and colonnettes enriched the interior. The chancel, enlarged by the construction of the central Apse and outfitted with low stone benches on both the north and the south sides, could accommodate a growing number of clergy. The floors of the north and south apses were raised approximately 40 cm, to the height of the central Apse, and all three apses were paved with marble flagstones. The areas immediately in front of the north and south apses were repaved in opus sectile. A mosaic pavement in a popular floret pattern was installed in the south aisle. In the southwest corner a large cable was placed against the wall, its two massive pier legs resting on this mosaic.

The remodelled interior doubtless retained much of the decoration of the preceding phase, but added new richness, particularly at the new east end. The walls had richly variegated veneers, inlaid stone designs and mosaics, the floors were covered with vivid patterns of stone tiling and polychrome mosaics, and at least portions of the ceilings of the apses had glass mosaics. Locations of special sanctity within the sanctuary were set apart by meter-high imported marble screens, decorated with geometric or openwork patterns or symbolical bas-reliefs. Glass windows and liturgical vessels added sumptuousness to the interior. Cloth hangings may have served to separate other specialized areas within the Church, such as aisles set aside for women. The effect upon the worshipper would have been not entirely unlike that produced by the churches at Gaza as described by Choricius (Laudatio Marciani II:28ff.) and Mark the Deacon (Vita Porphyrii.,71ff).

Accompanying the expansion of the Church in this phase, the north entrance into the Atrium was made more impressive by the addition of a monumental porch. Two Corinthian columns, reused from the defunct Temple of the Roman period that probably had been situated near the Church, were installed in front of the portal. On the west the shops were filled in and new chambers were erected one storey higher, probably to serve some ecclesiastical purpose. Chamber I was provided with a mosaic floor made of large white tesserae, while the other chambers were paved with mudstone. A pavement also was installed west of the chambers to provide access to them. The south dependency continued to be used, as did the triangular chamber. The latter was provided with a decorative mosaic floor in the same pattern as that in the south aisle of the sanctuary.

This phase of the Church is difficult to date closely, since the only stratigraphic evidence came from deposits in the three apses, all of which had been disturbed. The potsherds excavated below the enlarged Bema were largely from the 4th or 5th century, with a few Umayyad fragments intermingled in a robber's trench. The numismatic evidence from soundings in the north and south apses contained no specimens later than the 6th century, but some of the potsherds from the fill below the raised floors of these apses belonged typologically to the second half of the 7th century A.D. The floors showed, however, signs of disturbance, and may have undergone repairs in the 7th century, when makeshift steps were installed in the south Apse.

Architectural style is informative for the date of this phase. Triapsidal churches are rare before A.D. 500. but begin to appear widely in the 6th century (see Krautheimer 1979:168 and Negev 1982:15-16). The plans of the sanctuaries of both the Procopius Church (dated A.D. 526-27) and the Church of Saints Peter and Paul (dated to about 540) at Gerasa (Kraeling 1938:Pl. 39) are similar to that of the Pella Church. The preference for semicircular apses, if not primarily aesthetic (Colt 1962:9), may have been faddish, influenced by some famous building such as the Church of St. Simeon Stylites (Delougaz and Haines 1960:24). Liturgical developments also may have influenced this architectural development. Freshfield (1873:391) long ago suggested that the introduction of the triapsidal plan was related to changes in the liturgy in the reign of Justin II (A.D. 563-578), and many scholars have used this criterion as a fixed point, dating churches of this form to the third quarter of the 6th century or later (e.g., Ovadiah 1970:195). More recently, however, those studying the liturgy, in particular the rite of the Great Entrance (e.g.. Mathews 1971; Taft 1975) have found no evidence for Freshfield's claim. The beginning of this phase of the Church can therefore be dated to about the second quarter of the 6th century. Margalit (1987:116) dates the remodeling of the monoapsidal church at Shivta in the Negev into a triapsidal church to the beginning of the 6th century A.D., and agrees with Negev that the triapsidal churches may be linked with the development of the cult of saints and martyrs.

The interior appointments attributable to this phase tend to confirm the date of A.D. 525-550 for the remodelling of the sanctuary. The mosaic in the south aisle, with its floret pattern, is similar to ones found in buildings at Antioch dating from A.D. 500-525 (Levi 1947:Pl. 137a, b), the Church of Saints Peter and Paul (A.D. 540) and the Synagogue Church (A.D. 530/1) at Gerasa (Kraeling 1938:Pls. 65 and 76a), the Shumata Church of A.D. 555 (Avi-Yonah 1934:Pl. 30.2), the church at Khalde in Lebanon (Chehab 1957:Pl. 38), the church at `Ain es-Samake dating from the beginning of the 6th century (Chehab 1957:Pl. 113.1) and the Church of St. Menas in Rihab, constructed in A.D. 529 (Piccirillo 1980:Pl. 18). The style of the latticework floral screens is consistent with a mid-6th century date. Similar screens have been dated to the 6th century at Mytika (Bokotopoulou 1979:Pl. 89) and Nea Anchialos (Sotiriou 1929:86) in Greece, as well as at Ravenna (Orlandos 1952-56, 2:523). A geographically closer parallel can be seen in the inscribed screen from the synagogue at Khirbet Susiya (Yevin 1974: Pls. 42-43).

The end of this phase is marked by some sort of damage that the Church suffered around the time of, and possibly as a result of, Persian incursions into the southern Levant in A.D. 610-614 [JW: part of the Byzantine-Sassanian War of 602-628 CE], a situation perhaps exacerbated by the aging of the building and prior earthquake shocks. This dating is broadly confirmed by the artifacts of the 6th to early 7th centuries found in the fill that was thrown into the shops and portions of the Baths during renovations in Phase 3.

Phase 3: ca. A.D. 614 to 658-60 - phase ends with damage attributed to the Jordan Valley Quake(s) of 659/660 CE

from Smith et al. (1989:90-92)
Figs. 26, 27, and 19

Figures

Normal Size

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 19 Reconstruction

Fig. 19

Reconstruction of the gallery that was constructed in the north dependency during Phase 3 (excavated as Loci 43 and 50), viewed schematically from the north. On the right, behind the upper and lower Ionic colonnade, is the large, reused column of Ajlun limestone that formed the porch of the north portal of the atrium.

(See also the perspective drawing in Fig. 27.)

Smith et al. (1989)

of Phase 3 gallery in the north dependency of the Area IX church from Smith et al. (1989)
Fig. 26 Reconstructed

Fig. 26

Reconstructed plan of the church in Phase 3, around A.D. 614 until 658-660. The bema has been extended further west. The reconstruction of the west portico of the atrium is purely conjectural, particularly the walls shown at the top of the small staircase. The precise positions of the columns in the west collonade is not known, and the wall at the north end of the monumental staircase is inferred.

Smith et al. (1989)

plan of the Area IX church in Phase 3 from Smith et al. (1989)
Fig. 27 Reconstructed

Fig. 27

Reconstructed perspective drawing of the church in Phase 3. Details of the way in which the arch was installed in the west wall of the atrium are conjectural. The bases and shafts of the collonade at the top of the monumental staircase are shown as being more uniform than they probably were.

Smith et al. (1989)

perspective drawing of the Area IX church in Phase 3 from Smith et al. (1989)

Magnified

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 26 Reconstructed

Fig. 26

Reconstructed plan of the church in Phase 3, around A.D. 614 until 658-660. The bema has been extended further west. The reconstruction of the west portico of the atrium is purely conjectural, particularly the walls shown at the top of the small staircase. The precise positions of the columns in the west collonade is not known, and the wall at the north end of the monumental staircase is inferred.

Smith et al. (1989)

plan of the Area IX church in Phase 3 from Smith et al. (1989)

Discussion

This phase of the Church's history was inaugurated by major constructions which may have been started soon after the departure of the Persians in A.D. 614, but in any case probably prior to the Islamic conquest of A.D. 635. The most ambitious undertaking was a monumental western approach to the Church from the west. Ever since the Roman Period the Wadi Jirm probably had served as the main approach to Pella from the west, with the result that the Civic Complex and the Church stood directly in view of persons approaching the city from the Jordan Valley. The rising water table in the Wadi Jirm and the need to elevate the occupational level in the little valley gave occasion for a bold new architectural venture. All structures from the shops adjoining the Atrium to the exedra of the Baths were razed and paved over. A well-planned approach to the Church was built, consisting of a wide staircase and colonnade, a terrace, and a new portal cut through the west wall of the Atrium. Impressive though the results must have been, much of the workmanship was hasty, carried out, as usual, largely with spolia. Several factors may have prompted this ambitious undertaking, one of which may have been the clergy's desire that the Church have an approach commensurate with Christianity's growing importance in the city. There may also have been a liturgical interest, since the stairway was appropriate for public processions. A practical factor was the city's abandonment of the Baths, which previously would have constituted a physical barrier to the construction of such an approach.

In this phase a rather poorly constructed synthronon was added to the central Apse. The 1.40-m high bench could not have provided seating for anyone except a single ecclesiastical leader, the bishop, sitting at the center; hence the construction did not increase the available space for clergy in the chancel, but actually decreased it. Perhaps in part to compensate for the loss of space in the Apse, the Bema was extended westward by 2.70 m. The extension was paved with marble flagstones laid on a brick-red clay bedding, and was made continuous with the existing marble paving. Below the altar, which was moved slightly farther west, a marble chancel screen — for some reason displaced from its original context — was installed horizontally into the floor, level with the pavers, its obverse facing upward. There was also some intermittent repair to the flooring of the sanctuary, with an increased use of coarse white tesserae, and to the walls, where plaster increasingly replaced segments of the polished stone dado.

The building of the monumental western approach and the shift of the main entrance from north to west was accompanied by major changes in the north portico. Some of the columns of the stoa were moved about to form a two-storey, L-shaped structure with chambers and porches, to be used as quarters for an expanded clergy, as a hospice, as administrative offices, or for some other ecclesiastical purpose. Thus the north dependency came into existence.

The triangular chamber on the south underwent modifications as well. The mosaic floor of Phase a was cut through by a plastered drainage channel installed considerably above the former floor level. At the same time a new wall and a small arch were constructed in the chamber. Somewhat later the level of this chamber was raised and a new paved floor was laid, with a small cylindrical basin or bath built into the floor.

By this time the Byzantine Empire was already faltering: nevertheless, the ecclesiastical leaders could hardly have anticipated that only two decades after the construction of the western approach the city would fall to Muslim conquerors. Although that event in A.D. 635 did not oblige Christians to renounce their religion, life unquestionably became increasingly difficult for them. This third phase may be said to end with one or more earthquakes that occurred in 658-60, which doubtless damaged the Church.

JW: This earthquake or earthquakes struck in 659 and possibly 660 CE and are known as the Jordan Valley Quake(s).

Phase 4: ca. A.D. 658-60 to 717 - phase ends with damage attributed to an earthquake in 717 CE (which is an unlikely quake candidates)

from Smith et al. (1989:92-94)
Fig. 28

Figures

Normal Size

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 28 Reconstructed

Fig. 28

Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis.

Smith et al. (1989)

plan of the Area IX church in Phase 4 from Smith et al. (1989)

Magnified

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 28 Reconstructed

Fig. 28

Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis.

Smith et al. (1989)

plan of the Area IX church in Phase 4 from Smith et al. (1989)

Discussion

This phase, which falls entirely within the Umayyad period, may be said to begin with the repairing of the Church after the earthquake of A.D. 658-60. There can be little question but that the Church no longer enjoyed its former prestige and prosperity. The north dependency was probably sold, lost by civic expropriation, or abandoned by the Church. Non-ecclesiastical buildings may have encroached upon the Church's property; indeed, all of the Church's property west of the Atrium may have changed hands; in any case, the western approach ceased to be used, and the portal in the west wall of the Atrium was inelegantly sealed off.

The closing of the western approach meant that the main entrance to the Church was once again the portal in the north wall of the Atrium. Crude walls were erected among the columns of the north dependency, dividing it into numerous small chambers. Some of these rooms were provided with simple stone benches built against the walls, perhaps a reflection of changing functions in parts of the north dependency. The fact that two of the three doors in the north wall of the sanctuary were not closed off suggests that part of the north dependency, east of Walls 67 and 77 (Fig. 10), may have continued to be used by the Church. The northern approach to the Atrium was altered physically in various ways, all of which display extremely poor workmanship that utilized all kinds of disused building stones, among which were several bases from the Temple of the Roman period that had probably once stood nearby. The low state to which the Church had fallen is seen not only in the manner in which the walls of the passageway were formed, but also in the fact that the ecclesiastical authorities felt it necessary to place a stile across the passageway to prevent animals from entering the Church. The stile was scarcely better constructed than the passageway, although veneering may have made it less crude in appearance. Access to the Church was further restricted by the blocking off of the doorway from the portico into the east vestibule.

Between the portals of the west front of the sanctuary, two crude stone benches were constructed for some unknown purpose. Inside the sanctuary stone chips were used to repair the damaged floor mosaics, and mudstone frequently replaced marble paving. Plaster was liberally substituted for damaged veneer. The south dependency may have been abandoned by the Church during this phase, though the building continued to be used for some purpose; the staircase outside the southeast corner of the Church may have been constructed during this phase so as to create access to the upper storey of the building.

During this phase Christians at Pella became caught up in an early phase of the iconolastic controversy which was to wax and wane during the next two centuries. Artistic representations of living creatures became offensive to some Christians — whether citizens of Pella or outsiders is not known — who were able to exert sufficient influence either to have such representations removed or obliterated. The bas-reliefs of a pair of sheep on the marble screen below the altar of the Church were meticulously chiseled away by iconoclasts, leaving only their outlines.

This phase ended in A.D. 717, when a massive earthquake damaged the Church and many other structures at Pella. Much of the superstructure of the sanctuary of the Church probably collapsed at that time, although some of the columns in the Atrium may have remained standing, as certainly did the two large columns of the porch on the north side of the Atrium. Falling masonry crushed portions of the floors and shattered, among other things, the marble screen beneath the altar.

JW: The epicenter of the 717 CE Earthquake was too far away to have caused such damage in Pella.

Phase 5: ca. A.D. 717-747 - phase ends with damage attributed to the Jan. 749 CE Holy Desert Quake of the Sabbatical Year sequence

from Smith et al. (1989:94)

Figures

Normal Size

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 18 Detailed Plan

Fig. 18

Detailed plan of part of the north dependency and Parvis. Locus 50, the "Chamber of the Camels," contains the skeletons of seven camels, a camel foetus, an ass, and Human 2. The small chamber, Locus 43, in the northeast corner of Locus 50, held the skeleton of Human 1. Locus 66, also a chamber, held the skeletons of two cows; the outline of stones beneath the skeletons is somewhat uncertain. Locus 67, a part of the Parvis, had the skeleton of two cows, and Locus 22, an extension of the north dependency, had fragile remains of a horse and a foal. In Locus 23 is the stile across the entrance to the north portal of the atrium, which prevented animals from wandering into the atrium and Church. Excavated portions of drainage channels, declining in a generally westward direction beneath the paving, are indicated by bold straight lines; hypothetical continuations, based on paving patterns, are indicated by rows of dots.

Smith et al. (1989)

of Area IX part of the north dependency and Parvis with human and animal skeletons from Smith et al. (1989)
Fig. 28 Reconstructed

Fig. 28

Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis.

Smith et al. (1989)

plan of the Area IX church in Phase 4 from Smith et al. (1989)

Magnified

Fig. 2 Simplified Plan

Fig. 2

Simplified plan of Area IX, showing the major features

Smith et al. (1989)

of Area IX from Smith et al. (1989)
Fig. 18 Detailed Plan

Fig. 18

Detailed plan of part of the north dependency and Parvis. Locus 50, the "Chamber of the Camels," contains the skeletons of seven camels, a camel foetus, an ass, and Human 2. The small chamber, Locus 43, in the northeast corner of Locus 50, held the skeleton of Human 1. Locus 66, also a chamber, held the skeletons of two cows; the outline of stones beneath the skeletons is somewhat uncertain. Locus 67, a part of the Parvis, had the skeleton of two cows, and Locus 22, an extension of the north dependency, had fragile remains of a horse and a foal. In Locus 23 is the stile across the entrance to the north portal of the atrium, which prevented animals from wandering into the atrium and Church. Excavated portions of drainage channels, declining in a generally westward direction beneath the paving, are indicated by bold straight lines; hypothetical continuations, based on paving patterns, are indicated by rows of dots.

Smith et al. (1989)

of part of the Area IX north dependency and Parvis with human and animal skeletons from Smith et al. (1989)
Fig. 28 Reconstructed

Fig. 28

Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis.

Smith et al. (1989)

plan of the Area IX church in Phase 4 from Smith et al. (1989)

Discussion

The earthquake of A.D. 717 had shaken the Church so severely that much of its superstructure, as well as much of the Atrium, must have collapsed or subsequently been pulled down. Many of the fallen architectural and ordinary building stones were taken away, a fact attested all too clearly by the few architectural members of the superstructure found in the debris of the Church.

The piles of trash found on the floor may in part be the result of a halfhearted attempt at a cleanup after the earthquake of A.D. 717, but also may be the result of people's grubbing about in the ruins for usable items. Soil afterward accumulated on the robbed paving, a development indicating that the building lay open for some time before the final destruction. Some chambers in the north dependency continued to be used despite the destruction and the piles of trash, but for non-ecclesiastical purposes such as animal husbandry.

Any attempt to clean up the debris from the earthquake of 717 was soon abandoned. The Church must have been deconsecrated, stripped of its furniture and fittings, and abandoned. The Atrium was occasionally used for domestic purposes by persons who were willing to live amid the tumbled architectural remains. The result of this activity was that piles of debris were still lying on the floor of the building on January 18, 747, when another major earthquake shook down much of the remaining architecture of the Church, killing at least two persons in the sanctuary as well as two other persons and a number of animals in the north dependency. Umayyad coins of the first half of the 8th century found on the floor of the Chamber of the Camels and coins in the possession of one of the victims confirm the date of the final destruction.

JW: The causitive earthquake would have been the January 749 CE Holy Desert Quake of the Sabbatical Year Sequence.

Phase 6: ca. A.D. 747 to the present

from Smith et al. (1989:94)

There was only sporadic habitation at Pella after the earthquake of 747, although herders continued to bring their flocks to the spring. In the Abbasid and Mamluk periods, a few flimsy shelters may have been constructed amid the debris in the central and north apses, where the walls probably were standing to a greater height than elsewhere, but there was no permanent reoccupation of the area. Casual visitors left occasional potsherds amid the ruins. A thick layer of loess and colluvium washed down toward the Wadi Jirm from the saddle east of the mound. A few columns still stood amid the ruins and the accumulating soil, among them the two large columns of the Church's north portal and some columns in the north dependency. Centuries later, in an earthquake of unknown date, these, too, tumbled to the ground.

Early Bronze II Earthquake - c. 2800 BCE


Effect	Location	Image(s)	Description(s)
Collapsed Walls Fire	site-wide		EB II [Early Bronze II] occupational strata are sealed by a thick layer of destruction debris, associated with quantities of burnt mudbrick and stone - Bourke et. al. (2009)
Collapsed and Damaged Structures	site-wide ?		destruction of domestic architecture, public buildings and defense wall - Raphael and Agnon (2018:773), citing Bourke (2000:235) and Bourke et. al. (2009)

Phase VIII Earthquake - Middle Bronze IIB - ca. 1700 BCE or slightly earlier


Effect	Location	Image	Description
Collapsed Walls	Walls 41 and 47 of Plot IIIC of Area III on Tabaqat Fahl (Pella) Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008)		Walmsley et. al. (1993:178-180) noted that mudbrick and stone debris from Walls 41 and 47 suggest that Phase VIII was rendered uninhabitable through earthquake activity. East-West Wall 47 was preserved to a height of 50 cm. and most of the eastern metre of the wall had collapsed to the south in antiquity. Wall 41, the main city wall, runs north-south.

Phase VIC Earthquake (?) - Middle Bronze III - c. 1600 BCE

Effect	Location	Description
Collapsed Walls	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	Both Wall 27 and the city wall (Wall 41) suffered major structural damage. Several large pieces of the inner face of the city wall fractured and collapsed onto floor surfaces. - Bourke, Sparks, and Schroder (2006:26) Wall 41 runs N-S and Wall 27 runs E-W
Fractured Walls	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	Several large pieces of the inner face of the city wall fractured and collapsed onto floor surfaces. - Bourke, Sparks, and Schroder (2006:26)
Broken Pottery found in fallen position	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	much broken pottery and other objects sealed by a thick brick-filled debris layer - Bourke, Sparks, and Schroder (2006:26)
Fractured floors and walls	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	clear evidence of wall and floor fracturing - Bourke, Sparks, and Schroder (2006:26)
Debris	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	sealed by a thick brick-filled debris layer - Bourke, Sparks, and Schroder (2006:26)
Fire	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	The three small cubicles built against the inner face of the city wall suffered a fiery destruction - Bourke, Sparks, and Schroder (2006:26)

LB IIA Earthquake - towards the end of the 14th century BCE


Effect	Location	Image	Description
Titled Walls	Trench XXXIVF (Area XXXIV on Tell el-Husn) Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008)	Plate 8.1 - Plate 8.1 Tell Husn (Pella). Trench XXXIVF. LBA Structures canted over (to left) in earthquake. Bourke, Sparks, and Mairs (1999)	LBA Structures canted over (to left) in earthquake - Bourke, Sparks, and Mairs (1999)
Collapsed Walls	Trench XXXIVF (Area XXXIV on Tell el-Husn) Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008)		The entire complex was destroyed in a massive earthquake, probably dating towards the end of the LBIIA period, based on pottery sealed in destruction layers - Bourke, Sparks, and Mairs (1999)
Valuable Objects found in rubble	Trench XXXIVF (Area XXXIV on Tell el-Husn) Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008)		A hoard, consisting of eleven leaf-shaped arrowheads of copper alloy and three stunning bolts, all in an excellent state of preservation, (fig. 6) was recovered from the lip of one of two large stone-lined storage bins (Feature 6: fig. 4; pl. 7: 2). The uniformity of the collection suggests that they may have been from a quiver of weapons. The hoard dates to the period of the final destruction of the Husn complex by earthquake, based on ceramic evidence (fig. 5). The quiver probably fell into the storage bin during the destruction of the building - Bourke, Sparks, and Mairs (1999)
foundation damage and warping Folded and Distorted Walls	Temple Complex in Area XXII on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		Sharp warping of the underlying foundations in the north temple area [in Area XXII], still clear today from aerial photographs - Bourke (2004:8-9) western wall of the temple revealed stress-twisting and shattering - Raphael and Agnon (2018:775) citing Bourke (2012)
Widespread Damage	throughout Tabaqat Fahl (Pella) and on Tell el-Husn Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		similar earthquake-related damage is found throughout the city and in buildings on nearby Tell Husn (Bourke et al. 1999) - Bourke (2004:8-9)

Iron I Earthquake (?) - ca. 1150 BCE


Effect	Location	Image	Description
Widespread Damage - Collapsed Walls	Temple Complex in Area XXII on Tabaqat Fahl (Pella) throughout Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		The remodelled temple remained in use until the end of the Bronze Age (ca 1150 BC), when the entire site of Pella suffered a major destruction. This may also have been due to earthquake activity, although human agency remains possible, as this is the time of the enigmatic Sea People descent on Egypt, generally (if not always reliably) associated with a widespread destruction horizon throughout the region at this time (Sandars 1978). - Bourke (2004:8-9)

Iron IIB Earthquake (?) - ~800 BCE


Effect	Location	Image	Description
Collapsed Walls	Temple Complex in Area XXII on Tabaqat Fahl (Pella) and nearby areas Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		Thick deposits of ash and brick debris sealed the temple proper and most nearby areas. - Bourke (2004:13-14, 20)
Widespread Damage	throughout Tabaqat Fahl (Pella) and on Tell el-Husn Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		The remodelled Iron Age II temple precinct at Pella was in use for perhaps 150 years (ca 950–800 BC) before the temple and the entire settlement was destroyed in an extensive conflagration (Bourke et al. 2003). While earthquake activity has been suggested as the likely cause for similarly dated destruction horizons at Deir ’Alla (Franken 1992), the same horizon of destruction at nearby Tell Hammeh (Cahill et al. 1987) and Tel Rehov (Mazar 1999) has been attributed to the military activities of either Egyptian (or just possibly) Aramaean invaders. At Pella, while earthquake destruction is still considered the most probable cause, significant militaria (specifically iron arrowheads and scale armour) are consistently associated with this destruction horizon. Whatever the ultimate cause, this destruction proved to be catastrophic to the long-term well-being of the city of Pella, as settlement ceased across the site for the next 500 years, only reviving with the Seleucid occupation of the region after 200 BC (Bourke 1997) - Bourke (2004:13-14, 20)

7th century CE earthquake

Caveat - some of the effects listed by Walmsley (2007) and Blanke and Walmsley (2022) may have been based on rebuilding evidence.


Effect	Location	Description(s)
Collapsed Walls	‘policing and administrative garrison’ at Tall al-Husn Fig. 4.7 [10] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	the barracks was shattered by the force of the tremor and never rebuilt - Walmsley (2007)
Partial Collapse	The eastern end of the Pella mound (aka the main mound) Fig. 4.7 [5] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	a residential area suffered a partial collapse of buildings following which an east–west access street was closed off and multi-roomed houses on two levels repaired and equipped with large private courtyards. - Blanke and Walmsley (2022:86) houses required substantial rebuilding and remodelling after the quake - Walmsley (2007)
Collapsed Walls	The eastern end of the Pella mound (aka the main mound) Fig. 4.7 [2] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	a derelict courtyard house that was not rebuilt after an apparent mid-seventh-century collapse, although evidence of animal butchering reveals continued use of the space - Blanke and Walmsley (2022:86)
Fallen Columns land movement in the valley architectural slumping and infilling with coarse alluvium and boulders	The valley below the tall (i.e. the Civic Church Complex of Area IX) Fig. 4.7 [3] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022) Fig. 2 Simplified plan of Area IX, showing the major features Smith et al. (1989) Fig. 28 Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis. Smith et al. (1989)	the central church suffered significant damage that resulted in the permanent blocking of the Western portal only decades old and the reinstatement of the main entrance on the northern side (Fig. 4.7 [3]). The colonnade at the top of the church staircase seemingly collapsed at this time. An unwillingness to restore the portico and staircase suggests that the earthquake caused havoc in the valley, with broad evidence for land movement causing architectural slumping and infilling with coarse alluvium and boulders, the remediation of which would have been Sisyphean. - Blanke and Walmsley (2022:86) The churches also seem to have been adversely effected by this seismic event, although there is argument about the extent to which they were still functioning in the mid-7th c. - Walmsley (2007)
Collapsed Walls	Areas I and V - West and East churches Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Plan of West Church Complex in Area I Smith (1973)	Phase [4 of the Western Church Complex (AreaI)] ends with damage to the church which can most plausibly be attributed, on the basis of present information, to the extensive and severe earthquakes of A.D. 658-660 (Willis 1928:80). One of these shocks, perhaps that of 658, toppled most of the upper part of the walls of the sanctuary and Atrium and damaged the north dependency. - Smith (1973:166) JW: Dates are slightly off - should be attributed to the 659/660 CE Jordan Valley Quake(s)

Early 8th century CE earthquake ?


Effect	Location	Image(s)	Description(s)
Fallen column set up next to the south wall Fallen stairway in the nave Collapsed Roofs Shattered pottery Indented Paving	Area I West Church Complex Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Plan of West Church Complex in Area I Smith (1973)	Fig. 13 Stratigraphic sections through A-A' in the sanctuary of the Church, and B-B' in the Parvis (for locations, see plan in Fig. to). Five levels are identified. Note that Level 2 is not present in section B-B'. Level 1: Gray topsoil containing small stones and pockets of Marnluk and post-Mamluk sherds. Level 2: Hard-packed brown soil, with some gravel; largely sterile but containing some sherds of various periods washed from upslope. (This level is not present in section B-B'.) Level 3: Dense, yellowish-brown loess, largely sterile but containing a few Abbasid-Fatimid sherds. At the beginning of this level a robber-pit was cut into level 4. Level 4: Debris from the earthquake of A.D. 747, loose, grayish-brown soil with rubble and architectural stones. Level 5: Brown, clayey soil containing trash accumulation from the earthquake of A.D. 717, including rubble, roof tile fragments, bits of mosaic, chunks of cement and artifacts (see Pls. 56—57). Smith et al. (1989)	The walls of the church, already greatly reduced in height, may have suffered relatively little fresh damage, but the flimsy repairs and constructions made during the phase, such as the stairway in the nave and the column set up next to the south wall, toppled. It may be that the columns of the Atrium had largely stood until this time, but if so they fell during the shocks. The roofs of all the parts of the complex still in use likewise collapsed. Some pottery lying on the thin soil layer in the sanctuary was shattered and buried in the debris, a large section of which fell near the south Apse, pushing through the layer of accumulated soil and indenting the limestone paving. - Smith (1973:166)

mid 8th century CE earthquake


Effect	Location	Image	Description
Collapsed Walls Human and animal remains found in the rubble Objects of value found in the rubble Fallen Columns Crushed and fallen pottery Debris Fire	Areas IV and VIII - domestic quarters on main mounds Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Fig. 3 General plan of the Umayyad housing in Area IV (Walmsley) Walmsley (2008) Figure 4 View of the Pella Houses A–B in Area IV, after excavation. Part of the section shown in ﬁ gure 5 remains on the left; note the fallen column (Walmsley). Walmsley (2007)	Endplate - Area IV - Plan of House - House G Walmsley and Smith in McNicoll et al (1992) Fig. 12 Figure 12 Detailed plan of House G (Winikoff, modiﬁed Walmsley) Walmsley (2007) Area IV Umayyad Collapse Pella Project - University of Sydney Plate LXXXI.2 Area IV Plot P (extension). Skeletons of two charred adult human beings with covering textiles, found in the AD 746/7 destruction deposit. JW: Actually that earthquake struck in 749 Walmsley and Smith in McNicoll et al (1992)	Description On the main mound, due to the more complete nature of the deposits, the archaeological sequence is less obscured and the finds more representative, for it was houses - with all their human and material contents - that were lost to the unexpected ferocity of the A.D. 749 quake.In Area IV, excavations between 1979 and 1983 revealed 6 house units destroyed at the time of the earthquake. ... House G, as with Houses A and B, has exposed the utter destructiveness inflicted on Pella by the A.D. 749 earthquake. The upper level(s) of brick collapsed instantaneously into the ground floor rooms, trapping people, animals and domestic objects within them; even the chickens pecking about in the central courtyard of House G had no time to flee the tumbling building. ... Entombed on the ground floor level were more domestic items along with many victims of the tragedy, human and animal. Based on the disposition of the artefacts in the collapse and the complete entrapped skeletons on the ground floor, it can be generalised that the downstairs area of the house was devoted to the care of valuable domesticated animals and light workshop activities, whereas the primary living quarters were located upstairs. The human victims recovered on the ground floor consisted of three adults and a child. A single male was found crouched against the west wall of room 2, with one arm raised above his head in a desperate attempt to protect himself from heavy falling material (fig. 12). Across the inner courtyard in its north-east corner, an adult male and female with a young child were found huddled together under the staircase.³⁰ Very possibly the couple and the child fell from the upper level (seemingly the balcony or the top of the staircase) before being entombed in the crumbling building. Four gold dinars were found with the solitary victim in room 2, and 6 others near the couple in the inner courtyard, perhaps dropped when they fell from the upper storey.³¹ In total, 10 dinars is not an insignificant amount of money, especially considering that the annual head tax for an adult amounted to some 4 dinars. The trapped animals on the ground floor were chiefly cows (rooms 8 and 9, totalling three), small equids (mules or donkeys; inner courtyard, rooms 6 and 7), sheep, goats, chickens and a cat. Apart from the chickens and cat, which roam at will, these animals represented wealth. This is especially the case for the cows and mules (the sheep and goats were few, and may represent special cases such as sickly animals); hence the attempt by their owners to provide sufficient shelter in winter. ... The victim found in room 2, separated from the inner court by a roughly built dividing wall, may have been living in this downstairs space, perhaps intermittently. A robust adult male aged between 25 and 30 years, both his physique and location near the animals would suggest that he was their carer (a job for which, the 4 dinars would indicate, he was well rewarded). ... Just outside the doorway into room 2, evidence for a small fire was found on the bench against the west wall of the inner courtyard. Probably being used for cooking rather than heating, the fire ignited the collapsed building debris in the inner courtyard, baking the fallen bricks and carbonising organic matter such as wood. Gradually spreading through the tumbled collapse, the burning eventually reached the trapped couple and child by the staircase (hopefully all dead by this time), unusually preserving their clothing through the process of slow but complete combustion. Although the resultant carbonization of the material destroyed evidence of its original colouring, a study of the burnt cloth, which was manufactured from silk, was able to identify the different styles of weaving used in its manufacture.³⁴ Their prestige clothing, the likelihood they fell from the upper level during the earthquake and the discovery of 6 dinars close by would indicate the greater wealth and status of the couple by the staircase. Perhaps they were the owners of the house and animals, and for whom the male in room 2 worked as a herder. Footnotes 30 Bourke (1992) 220–21. 31 Walmsley (2001). As Early Islamic dinars always carry a mint date, they provide a useful chronological ﬁ x, with dates of 96 A.H./A.D. 714/15, 97 A.H./A.D. 715/16, 106 A.H./A.D. 724/25 and 117 A.H./A.D. 735/36. In the courtyard, six further dinars were recovered at the end of the 1982 season, dating to 91 A.H./A.D. 709/10, two of 94 A.H./A.D. 712/13, 110 A.H./A.D. 728/29, 112 A.H./A.D. 730/31 and 122 A.H./A.D. 739/40. Chronologically, the best dating came from a small copper fals minted in Damascus in 126 A.H./A.D. 743/44, struck just a few years before the A.D. 749 earthquake. 34 Eastwood (1992). - Walmsley (2007) The horror of entrapment and gruesome deaths inflicted on victims of the 749 earthquake is graphically documented by the excavation of six multi-storied dwellings at the east end of the main mound at Fiḥl. Heavy mudbricks from the walls of the upper floor plunged inwards filling the ground floor rooms, in the process trapping animals (from bovines to equids, chickens and cats) and at least five, up to ten, humans, all caught in the panic of nature’s dreadful paroxysm.⁶⁶ A large section of one house caught fire, carbonising two victims. Later, no appreciable attempt was made to recover the bodies of dead, nor were valuable objects retrieved such as coins in precious metals,⁶⁷ and no evidence for digging into the collapse was identified. The lingering reminder of the dead, left entombed in the ruins, would have evoked painful memories, and prolonged feelings of helplessness, loss and an absence of closure among survivors. For some time, maybe decades, these debilitating emotions inhibited resettlement on the Fiḥl tall, but settlement did return. In the 1987 excavation season, a deep cesspit was encountered, replete with finds of ceramic, metal, bone, glass and eggshell (Fig. 4.7 [6]), while recently extensive ninth to eleventh century occupational levels have been verified on the tall (Fig. 4.7 [7]); these discoveries only touch the edges.⁶⁸ Footnotes 66 Walmsley, ‘Households at Pella’. On pp. 247–250, the paper touches on other locations with striking evidence of the 749 earthquake at Fiḥl, notably the central church complex and market with victims (Area 9) mentioned earlier, also the west and east churches (Areas 1 and 5), and a second domestic quarter on the main mound (Area 8). On page 250 the words ‘robbing’ and ‘looting’ are used; these words are no longer part of Walmsley’s archaeological vocabulary. 67 Walmsley, ‘Islamic Coins’, 59–60, 147–149. 68 Edwards et al., ‘Preliminary report’, 81–86 (pit), O’Hea in da Costa et al., ‘New light on late antique Pella’, 523–525 on lustre-painted Islamic glass; McPhillips, pers. comm. - Blanke and Walmsley (2022:86) Disinterred from amongst the debris that filled rooms 13-15 were numerous skeletons, both human and animal, as well as finds of pottery, stone and metals. ... In the north-east corner of room 15, two adult humans (a male and a female) were found in conjunction with a large mass of textile fragments (see Appendices 3 and 8). A number of equid (probably donkeys) and chicken skeletons were also uncovered at floor level, along with a severely crushed cat. Underneath a drum from one of the fallen columns a further six Umayyad dinars were recovered, with the latest dating to AH 122/AD 739-40. Chronologically more important, however, is the bronze coin of AH 126/AD 743-4 from room 16, minted just three years before the AD 747 earthquake. A list of these and other Umayyad coins will be found in Appendix 9. Other finds from room 15 included more examples of mid-8th-century domestic pottery, especially cooking bowls (cf. PJ1: pl. 143: 2) as well as a hoe, harnessing rings, door hinges, and an iron lock. In the northern part of room 15, the effects of a fire after the collapse of the building were clearly discernible. Column drums were cracked and blackened and the yellow clay bricks baked red from the heat. This fire also engulfed the human couple (pl. 120) trapped in the north-east corner of the room, although their tragedy is our blessing, as the fire carbonized and preserved organic remains usually lost by decay at Pella. Of note are the textiles (see Appendix 8), oak beams, straw from mats, date stones, and olive pips. While removing the deposit in room 15, a number of interesting observations were made on the nature of materials used in the construction of the upper storey of the house. Numerous yellow clay and pebble bricks had fallen into this and the surrounding rooms from the upstairs walls along with segments of wooden beams used to support the floor and roof. A considerable number of large white tesserae were also found in room 15, in some cases still adhering to a pebble and mortar base. These originated, it would seem, from the floor of the room located above 15. The incinerated couple would have also fallen from this upper room when its floor collapsed during the first shocks of the AD 747 earthquake. In the adjacent room 13, more equid skeletons were uncovered, as well as three hens and another human, the latter also from the upper storey. Objects from the deposit in this room included lamps, two of pottery and one of bronze, and a glass vase. Unlike rooms 13 and 15, no skeletal material was excavated from the Umayyad levels in room 14. However, copious quantities of sherds from storage vessels were found at floor level; at least three large jars were crushed in situ according to the area supervisor (Edwards 1982). - Walmsley and Smith in McNicoll et al (1982) Five fallen columns and a pier originally arranged in two rows on an east-west axis, with three columns to the south and a combination of two columns and a pier in the northern row were discovered in the debris of Room 15 - JW: I forgot where I got this from
Collapsed Walls Human and animal remains found in the rubble Objects of value found in the rubble Fallen Columns Debris	Area IX - Civic Church Complex in the Central Valley Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Fig. 2 Simplified plan of Area IX, showing the major features Smith et al. (1989) Fig. 28 Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis. Smith et al. (1989)	Fig. 18 Detailed plan of part of the north dependency and Parvis. Locus 50, the "Chamber of the Camels," contains the skeletons of seven camels, a camel foetus, an ass, and Human 2. The small chamber, Locus 43, in the northeast corner of Locus 50, held the skeleton of Human 1. Locus 66, also a chamber, held the skeletons of two cows; the outline of stones beneath the skeletons is somewhat uncertain. Locus 67, a part of the Parvis, had the skeleton of two cows, and Locus 22, an extension of the north dependency, had fragile remains of a horse and a foal. In Locus 23 is the stile across the entrance to the north portal of the atrium, which prevented animals from wandering into the atrium and Church. Excavated portions of drainage channels, declining in a generally westward direction beneath the paving, are indicated by bold straight lines; hypothetical continuations, based on paving patterns, are indicated by rows of dots. Smith et al. (1989) Plate LXXXI.2 Skeletons found in the courtyard of the "hall of camels" just north of the Civic Complex Church. The animals are lying as they were killed in the earthquake of A.D. 746/7 JW: Actually that earthquake struck in 749 Smith in ADAJ Plates 1983 Plate 37A Three of the five skeletons excavated in the [Area IX] Civic Complex. A. Remains found in the southwest corner of the sanctuary Smith et al. (1989) Plate 37B Three of the five skeletons excavated in the [Area IX] Civic Complex. B. Skeleton found in the drainage channel in the triangular chamber. Smith et al. (1989) Plate 37C Three of the five skeletons excavated in the [Area IX] Civic Complex. C. Skeleton found inside the entrance to Locus 50, the Chamber of the Camels. Smith et al. (1989) Plate 39B The intertwined skeletons of Camels B and C [in Locus 50, the "Chamber of Camels" in the Area IX Civic Complex], viewed from the east. Smith et al. (1989) Plate 40C Skeleton of Camel F [in the Area IX Civic Complex] Smith et al. (1989) Plate 40D Skeleton of foetus of Camel B in Locus 50 [the "Chamber of Camels" in the Area IX Civic Complex] Smith et al. (1989) Plate 40E Ass Skeleton, with legs of Camel E at upper left and the skeleton of Camel D at right [in the Area IX Civic Complex] Smith et al. (1989) Fig. 13 Stratigraphic sections through A-A' in the sanctuary of the Church, and B-B' in the Parvis (for locations, see plan in Fig. to). Five levels are identified. Note that Level 2 is not present in section B-B'. Level 1: Gray topsoil containing small stones and pockets of Marnluk and post-Mamluk sherds. Level 2: Hard-packed brown soil, with some gravel; largely sterile but containing some sherds of various periods washed from upslope. (This level is not present in section B-B'.) Level 3: Dense, yellowish-brown loess, largely sterile but containing a few Abbasid-Fatimid sherds. At the beginning of this level a robber-pit was cut into level 4. Level 4: Debris from the earthquake of A.D. 747, loose, grayish-brown soil with rubble and architectural stones. Level 5: Brown, clayey soil containing trash accumulation from the earthquake of A.D. 717, including rubble, roof tile fragments, bits of mosaic, chunks of cement and artifacts (see Pls. 56—57). Smith et al. (1989)	Description The cathedral church and related structures excavated by Smith present an intriguing case of urban change after the Islamic expansion. Significant modifications occurred north of the church building, where a paved porticoed court, which coincided with the construction of the church, was modified by the construction of two-storeyed rooms faced with porches and galleries. Standing some 8.5 m high, this structure was clearly of major importance in the life of the town as it flanked the main entrance into the church, and was probably commercial in purpose (fig. 2).¹⁹ At least, this would seem to have been its final function on the day the A.D. 749 earthquake struck Pella with all of its intensity. Caught within the collapsing structure were two people and several animals, including 7 camels (one in advanced pregnancy), a horse and foal, an ass, and 4 cows. The age and condition of the camels - juvenile, young adult or in calf - suggests they were being sheltered on a particularly harsh winter’s day [JW: night].²⁰ ... With these fatalities of nature were recovered several Umayyad coins, including a dinar (94 A.H./A.D. 712/13) and 14 dirhams (between 81 A.H./A.D. 700/701 and 112 A.H./A.D. 730/31), numerous ceramic vessels and lamps, iron implements including a dagger, a bronze jug and three glass bottles.²¹ A mass of iron tools was found in a smith’s workshop in the south-west corner of the building, including a pick head, shovel, shears and sledgehammer head. What was happening in the church at this time is disputed (see note ¹³). In the court preceding the church (Atrium), three iron torch holders were found along with pottery vessels, suggesting a not insignificant level of activity. Within the church building, in the south aisle, the skeletons of two humans were found, but, apart from a few ceramic vessels (including a group of arcane pierced pottery cones), little else, except architectural debris from the collapsed church. Subsequently, from a level above the collapse that contained later Abbasid-Fatimid material, stone was robbed from the destroyed building.²² This pattern of an A.D. 749 collapse and subsequent robbing matches that recognised in Area IV, and further suggests that the looting of the church fixtures did not occur until after the earthquake of A.D. 749. Footnotes 19 This interpretation differs in details from Smith, who dates the conversion to the early 6th c. and proposes the purpose of the building was ecclesiastical, suggesting it was intended as a ‘hospice’ or served an ‘administrative’ function. This author has, in addition, emphasised a commercial purpose for this construction, proposing they served as the early Islamic market and caravanserai of Pella. All three functions—accommodation, administrative and commercial - could have (and probably were) intended. Equally likely, Pella’s clergy-administrators were the persons responsible for this enhancement to the urban environment, acting as the local representatives within the province of al-Urdunn. 20 On camels—the ‘second pillar’ of Bedouin life—see the definitive work by Jabbur (1995) 191–237. 21 Smith and Day (1989) 67–71. 22 Smith and Day (1989) 50–52, note especially ﬁ g. 13 and the accompanying explanation. - Walmsley (2007) The result of this activity was that piles of debris were still lying on the floor of the building on January 18, 747 [JW: Actually January 17 ± 1, 749 CE] when another major earthquake shook down much of the remaining architecture of the Church, killing at least two persons in the sanctuary as well as two other persons and a number of animals in the north dependency. Umayyad coins of the first half of the 8th century found on the floor of the Chamber of the Camels and coins in the possession of one of the victims confirm the date of the final destruction. - Smith et al. (1989:94)
Seismic Destruction Collapsed Walls Fallen Column	Areas I and V - West and East churches Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Plan of West Church Complex in Area I Smith (1973)	Plate 12 Plate 12 I U: view toward the east. The massive block of fallen masonry on the left and the toppled column drum at the rear are the results of the earthquake which ended phase 4 of the [Western] church's history. The capital in the foreground may belong to the same column. Smith (1973)	As has already been noted, graphic archaeological evidence for the severity of the A.D. 749 earthquake has been found in different locations at Pella, notably the church complex in the central valley (Area IX), the West and East churches (Areas I and V), and the domestic quarters on the main mounds (Areas IV and VIII). - Walmsley (2007) When the earthquake struck, virtually all of the courses of the walls [of the Western Church Complex in Area I] not held in place by debris collapsed, generally falling westward. Buried in this final collapse were a few vessels in domestic use in the complex at the time. - Smith (1973:166) The massive block of fallen masonry on the left and the toppled column drum at the rear are the results of the earthquake which ended phase 4 of the [Western] church's history - Smith (1973:Plate 12)

mid 8th century CE earthquake

Modified by JW from Fig. 3 of Walmsley (2008)
Collapse areas are limited to those described in the text but were likely more extensive
Smith (1973:166) noted that during this event, virtually all of the courses of the walls of the Western Church Complex (Area I) that were not buttressed by debris collapsed, generally falling westward burying a few vessels in domestic use

Deformation Map

Modified by JW from Fig. 3 of Walmsley (2008)

Early Bronze II Earthquake - c. 2800 BCE

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(s)	Intensity
Collapsed Walls Fire	site-wide		EB II [Early Bronze II] occupational strata are sealed by a thick layer of destruction debris, associated with quantities of burnt mudbrick and stone - Bourke et. al. (2009)	VIII+ ?
Collapsed and Damaged Structures (collapsed walls)	site-wide ?		destruction of domestic architecture, public buildings and defense wall - Raphael and Agnon (2018:773), citing Bourke (2000:235) and Bourke et. al. (2009)	VIII+

This 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 VIII Earthquake - Middle Bronze IIB - ca. 1700 BCE or slightly earlier

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	Description	Intensity
Collapsed Walls	Walls 41 and 47 of Plot IIIC of Area III on Tabaqat Fahl (Pella) Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008)		Walmsley et. al. (1993:178-180) noted that mudbrick and stone debris from Walls 41 and 47 suggest that Phase VIII was rendered uninhabitable through earthquake activity. East-West Wall 47 was preserved to a height of 50 cm. and most of the eastern metre of the wall had collapsed to the south in antiquity. Wall 41, the main city wall, runs north-south.	VIII+

This 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 VIC Earthquake (?) - Middle Bronze III - c. 1600 BCE

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
Collapsed Walls	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	Both Wall 27 and the city wall (Wall 41) suffered major structural damage. Several large pieces of the inner face of the city wall fractured and collapsed onto floor surfaces. - Bourke, Sparks, and Schroder (2006:26) Wall 41 runs N-S and Wall 27 runs E-W	VIII +
Fractured Walls (penetrative fractures)	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	Several large pieces of the inner face of the city wall fractured and collapsed onto floor surfaces. - Bourke, Sparks, and Schroder (2006:26)	VI +
Broken Pottery found in fallen position	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	much broken pottery and other objects sealed by a thick brick-filled debris layer - Bourke, Sparks, and Schroder (2006:26)	VII +
fravtured floors and walls (penetrative fractures)	Area III on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney	clear evidence of wall and floor fracturing - Bourke, Sparks, and Schroder (2006:26)	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).

LB IIA Earthquake - towards the end of the 14th century BCE

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	Description	Intensity
Titled Walls	Trench XXXIVF (Area XXXIV on Tell el-Husn) Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008)	Plate 8.1 - Plate 8.1 Tell Husn (Pella). Trench XXXIVF. LBA Structures canted over (to left) in earthquake. Bourke, Sparks, and Mairs (1999)	LBA Structures canted over (to left) in earthquake - Bourke, Sparks, and Mairs (1999)	VI+
Collapsed Walls	Trench XXXIVF (Area XXXIV on Tell el-Husn) Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008)		The entire complex was destroyed in a massive earthquake, probably dating towards the end of the LBIIA period, based on pottery sealed in destruction layers - Bourke, Sparks, and Mairs (1999)	VIII+
foundation damage and warping Folded and Distorted Walls	Temple Complex in Area XXII on Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		Sharp warping of the underlying foundations in the north temple area [in Area XXII], still clear today from aerial photographs - Bourke (2004:8-9) western wall of the temple revealed stress-twisting and shattering - Raphael and Agnon (2018:775) citing Bourke (2012)	IX VII+

This archeoseismic evidence requires a minimum Intensity of IX (9) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224).

Iron I Earthquake (?) - ca. 1150 BCE

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	Description	Intensity
Widespread Damage - Collapsed Walls	Temple Complex in Area XXII on Tabaqat Fahl (Pella) throughout Tabaqat Fahl (Pella) Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		The remodelled temple remained in use until the end of the Bronze Age (ca 1150 BC), when the entire site of Pella suffered a major destruction. This may also have been due to earthquake activity, although human agency remains possible, as this is the time of the enigmatic Sea People descent on Egypt, generally (if not always reliably) associated with a widespread destruction horizon throughout the region at this time (Sandars 1978). - Bourke (2004:8-9)	VIII+

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

Iron IIB Earthquake (?) - ~800 BCE

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	Description	Intensity
Collapsed Walls	Temple Complex in Area XXII on Tabaqat Fahl (Pella) and nearby areas Pella contour plan showing excavation Areas, 2019 Pella Project - University of Sydney		Thick deposits of ash and brick debris sealed the temple proper and most nearby areas. - Bourke (2004:13-14, 20)	VIII+

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

7th century CE earthquake

Caveat - some of the effects listed by Walmsley (2007) and Blanke and Walmsley (2022) may have been based on rebuilding evidence.
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(s)	Intensity
Collapsed Walls	‘policing and administrative garrison’ at Tall al-Husn Fig. 4.7 [10] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	the barracks was shattered by the force of the tremor and never rebuilt - Walmsley (2007)	VIII+
Partial Collapse (Displaced Walls?)	The eastern end of the Pella mound (aka the main mound) Fig. 4.7 [5] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	a residential area suffered a partial collapse of buildings following which an east–west access street was closed off and multi-roomed houses on two levels repaired and equipped with large private courtyards. - Blanke and Walmsley (2022:86) houses required substantial rebuilding and remodelling after the quake - Walmsley (2007)	VII+?
Collapsed Walls	The eastern end of the Pella mound (aka the main mound) Fig. 4.7 [2] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022)	a derelict courtyard house that was not rebuilt after an apparent mid-seventh-century collapse, although evidence of animal butchering reveals continued use of the space - Blanke and Walmsley (2022:86)	VIII+
Fallen Columns land movement in the valley (seismic uplift/subsidence?) architectural slumping and infilling with coarse alluvium and boulders (anthropic compacted substratum?)	The valley below the tall (i.e. the Civic Church Complex of Area IX) Fig. 4.7 [3] Figure 4.7 Features of pre- and post-749 Islamic Fiḥl. Note all height measurements are minus (below sea level). Early Islamic commercial centre [khan] composed of two compounds, eight to tenth centuries (partially excavated), -44 m courtyard house, damaged in the 659 earthquake, -43 m central church and Odeon, -54 m, west (left) of which lies the heavily silted Wādī al-Jirm’, -60 m East church, damaged in the 659 tremor and perhaps destroyed by the 749 earthquake, -20 m eastern tall residential quarter of the early sixth century, damaged in 659 tremor, destroyed in 749, with evidence for a ninth century rebuilding in the southern part, -42 m extremely abraded occupation area of surfaces, floors,stone wall stubs and deep refuse pits dating to the eighth/ninth to eleventh centuries, -38 m buildings with late tenth to eleventh century lustre-decorated glass and, north of the number, a later mosque of probable twelfth century date, -43 m west residential quarter (excavation back-filled),destroyed in 749, -51 m West church, evidence for earlier earthquake(s), probably 659, then great damage in 749, -60 m policing and administrative garrison on the summit of Tall al-Ḥuṣn, destroyed in 659 and not rebuilt -2 m (Base image: Google Earth, 20 August 2021). Blanke and Walmsley (2022) Fig. 2 Simplified plan of Area IX, showing the major features Smith et al. (1989) Fig. 28 Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis. Smith et al. (1989)	the central church suffered significant damage that resulted in the permanent blocking of the Western portal only decades old and the reinstatement of the main entrance on the northern side (Fig. 4.7 [3]). The colonnade at the top of the church staircase seemingly collapsed at this time. An unwillingness to restore the portico and staircase suggests that the earthquake caused havoc in the valley, with broad evidence for land movement causing architectural slumping and infilling with coarse alluvium and boulders, the remediation of which would have been Sisyphean. - Blanke and Walmsley (2022:86) The churches also seem to have been adversely effected by this seismic event, although there is argument about the extent to which they were still functioning in the mid-7th c. - Walmsley (2007)	V+ or VIII+ VI+? VI+?
Collapsed Walls	Areas I and V - West and East churches Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Plan of West Church Complex in Area I Smith (1973)	Phase [4 of the Western Church Complex (AreaI)] ends with damage to the church which can most plausibly be attributed, on the basis of present information, to the extensive and severe earthquakes of A.D. 658-660 (Willis 1928:80). One of these shocks, perhaps that of 658, toppled most of the upper part of the walls of the sanctuary and Atrium and damaged the north dependency. - Smith (1973:166) JW: Dates are slightly off - should be attributed to the 659/660 CE Jordan Valley Quake(s)	VIII+

This archeoseismic evidence requires a minimum Intensity of VIII (8) when using the Earthquake Archeological Effects chart of Rodríguez-Pascua et al (2013: 221-224). Caveat - some of the effects listed by Walmsley (2007) and Blanke and Walmsley (2022) may have been based on rebuilding evidence.

Early 8th century CE 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(s)	Intensity
Fallen column set up next to the south wall Fallen stairway in the nave Collapsed Roofs (suggesting displaced or folded walls) Shattered pottery Indented Paving	Area I West Church Complex Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Plan of West Church Complex in Area I Smith (1973)	Fig. 13 Stratigraphic sections through A-A' in the sanctuary of the Church, and B-B' in the Parvis (for locations, see plan in Fig. to). Five levels are identified. Note that Level 2 is not present in section B-B'. Level 1: Gray topsoil containing small stones and pockets of Marnluk and post-Mamluk sherds. Level 2: Hard-packed brown soil, with some gravel; largely sterile but containing some sherds of various periods washed from upslope. (This level is not present in section B-B'.) Level 3: Dense, yellowish-brown loess, largely sterile but containing a few Abbasid-Fatimid sherds. At the beginning of this level a robber-pit was cut into level 4. Level 4: Debris from the earthquake of A.D. 747, loose, grayish-brown soil with rubble and architectural stones. Level 5: Brown, clayey soil containing trash accumulation from the earthquake of A.D. 717, including rubble, roof tile fragments, bits of mosaic, chunks of cement and artifacts (see Pls. 56—57). Smith et al. (1989)	The walls of the church, already greatly reduced in height, may have suffered relatively little fresh damage, but the flimsy repairs and constructions made during the phase, such as the stairway in the nave and the column set up next to the south wall, toppled. It may be that the columns of the Atrium had largely stood until this time, but if so they fell during the shocks. The roofs of all the parts of the complex still in use likewise collapsed. Some pottery lying on the thin soil layer in the sanctuary was shattered and buried in the debris, a large section of which fell near the south Apse, pushing through the layer of accumulated soil and indenting the limestone paving. - Smith (1973:166)	V+ ? VI+ or VII+ VII+ ?

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

mid 8th century CE 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	Description	Intensity
Collapsed Walls Human and animal remains found in the rubble Objects of value found in the rubble Fallen Columns Crushed and fallen pottery Debris Fire	Areas IV and VIII - domestic quarters on main mounds Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Fig. 3 General plan of the Umayyad housing in Area IV (Walmsley) Walmsley (2008) Figure 4 View of the Pella Houses A–B in Area IV, after excavation. Part of the section shown in ﬁ gure 5 remains on the left; note the fallen column (Walmsley). Walmsley (2007)	Endplate - Area IV - Plan of House - House G Walmsley and Smith in McNicoll et al (1992) Fig. 12 Figure 12 Detailed plan of House G (Winikoff, modiﬁed Walmsley) Walmsley (2007) Area IV Umayyad Collapse Pella Project - University of Sydney Plate LXXXI.2 Area IV Plot P (extension). Skeletons of two charred adult human beings with covering textiles, found in the AD 746/7 destruction deposit. JW: Actually that earthquake struck in 749 Walmsley and Smith in McNicoll et al (1992)	Description On the main mound, due to the more complete nature of the deposits, the archaeological sequence is less obscured and the finds more representative, for it was houses - with all their human and material contents - that were lost to the unexpected ferocity of the A.D. 749 quake.In Area IV, excavations between 1979 and 1983 revealed 6 house units destroyed at the time of the earthquake. ... House G, as with Houses A and B, has exposed the utter destructiveness inflicted on Pella by the A.D. 749 earthquake. The upper level(s) of brick collapsed instantaneously into the ground floor rooms, trapping people, animals and domestic objects within them; even the chickens pecking about in the central courtyard of House G had no time to flee the tumbling building. ... Entombed on the ground floor level were more domestic items along with many victims of the tragedy, human and animal. Based on the disposition of the artefacts in the collapse and the complete entrapped skeletons on the ground floor, it can be generalised that the downstairs area of the house was devoted to the care of valuable domesticated animals and light workshop activities, whereas the primary living quarters were located upstairs. The human victims recovered on the ground floor consisted of three adults and a child. A single male was found crouched against the west wall of room 2, with one arm raised above his head in a desperate attempt to protect himself from heavy falling material (fig. 12). Across the inner courtyard in its north-east corner, an adult male and female with a young child were found huddled together under the staircase.³⁰ Very possibly the couple and the child fell from the upper level (seemingly the balcony or the top of the staircase) before being entombed in the crumbling building. Four gold dinars were found with the solitary victim in room 2, and 6 others near the couple in the inner courtyard, perhaps dropped when they fell from the upper storey.³¹ In total, 10 dinars is not an insignificant amount of money, especially considering that the annual head tax for an adult amounted to some 4 dinars. The trapped animals on the ground floor were chiefly cows (rooms 8 and 9, totalling three), small equids (mules or donkeys; inner courtyard, rooms 6 and 7), sheep, goats, chickens and a cat. Apart from the chickens and cat, which roam at will, these animals represented wealth. This is especially the case for the cows and mules (the sheep and goats were few, and may represent special cases such as sickly animals); hence the attempt by their owners to provide sufficient shelter in winter. ... The victim found in room 2, separated from the inner court by a roughly built dividing wall, may have been living in this downstairs space, perhaps intermittently. A robust adult male aged between 25 and 30 years, both his physique and location near the animals would suggest that he was their carer (a job for which, the 4 dinars would indicate, he was well rewarded). ... Just outside the doorway into room 2, evidence for a small fire was found on the bench against the west wall of the inner courtyard. Probably being used for cooking rather than heating, the fire ignited the collapsed building debris in the inner courtyard, baking the fallen bricks and carbonising organic matter such as wood. Gradually spreading through the tumbled collapse, the burning eventually reached the trapped couple and child by the staircase (hopefully all dead by this time), unusually preserving their clothing through the process of slow but complete combustion. Although the resultant carbonization of the material destroyed evidence of its original colouring, a study of the burnt cloth, which was manufactured from silk, was able to identify the different styles of weaving used in its manufacture.³⁴ Their prestige clothing, the likelihood they fell from the upper level during the earthquake and the discovery of 6 dinars close by would indicate the greater wealth and status of the couple by the staircase. Perhaps they were the owners of the house and animals, and for whom the male in room 2 worked as a herder. Footnotes 30 Bourke (1992) 220–21. 31 Walmsley (2001). As Early Islamic dinars always carry a mint date, they provide a useful chronological ﬁ x, with dates of 96 A.H./A.D. 714/15, 97 A.H./A.D. 715/16, 106 A.H./A.D. 724/25 and 117 A.H./A.D. 735/36. In the courtyard, six further dinars were recovered at the end of the 1982 season, dating to 91 A.H./A.D. 709/10, two of 94 A.H./A.D. 712/13, 110 A.H./A.D. 728/29, 112 A.H./A.D. 730/31 and 122 A.H./A.D. 739/40. Chronologically, the best dating came from a small copper fals minted in Damascus in 126 A.H./A.D. 743/44, struck just a few years before the A.D. 749 earthquake. 34 Eastwood (1992). - Walmsley (2007) The horror of entrapment and gruesome deaths inflicted on victims of the 749 earthquake is graphically documented by the excavation of six multi-storied dwellings at the east end of the main mound at Fiḥl. Heavy mudbricks from the walls of the upper floor plunged inwards filling the ground floor rooms, in the process trapping animals (from bovines to equids, chickens and cats) and at least five, up to ten, humans, all caught in the panic of nature’s dreadful paroxysm.⁶⁶ A large section of one house caught fire, carbonising two victims. Later, no appreciable attempt was made to recover the bodies of dead, nor were valuable objects retrieved such as coins in precious metals,⁶⁷ and no evidence for digging into the collapse was identified. The lingering reminder of the dead, left entombed in the ruins, would have evoked painful memories, and prolonged feelings of helplessness, loss and an absence of closure among survivors. For some time, maybe decades, these debilitating emotions inhibited resettlement on the Fiḥl tall, but settlement did return. In the 1987 excavation season, a deep cesspit was encountered, replete with finds of ceramic, metal, bone, glass and eggshell (Fig. 4.7 [6]), while recently extensive ninth to eleventh century occupational levels have been verified on the tall (Fig. 4.7 [7]); these discoveries only touch the edges.⁶⁸ Footnotes 66 Walmsley, ‘Households at Pella’. On pp. 247–250, the paper touches on other locations with striking evidence of the 749 earthquake at Fiḥl, notably the central church complex and market with victims (Area 9) mentioned earlier, also the west and east churches (Areas 1 and 5), and a second domestic quarter on the main mound (Area 8). On page 250 the words ‘robbing’ and ‘looting’ are used; these words are no longer part of Walmsley’s archaeological vocabulary. 67 Walmsley, ‘Islamic Coins’, 59–60, 147–149. 68 Edwards et al., ‘Preliminary report’, 81–86 (pit), O’Hea in da Costa et al., ‘New light on late antique Pella’, 523–525 on lustre-painted Islamic glass; McPhillips, pers. comm. - Blanke and Walmsley (2022:86) Disinterred from amongst the debris that filled rooms 13-15 were numerous skeletons, both human and animal, as well as finds of pottery, stone and metals. ... In the north-east corner of room 15, two adult humans (a male and a female) were found in conjunction with a large mass of textile fragments (see Appendices 3 and 8). A number of equid (probably donkeys) and chicken skeletons were also uncovered at floor level, along with a severely crushed cat. Underneath a drum from one of the fallen columns a further six Umayyad dinars were recovered, with the latest dating to AH 122/AD 739-40. Chronologically more important, however, is the bronze coin of AH 126/AD 743-4 from room 16, minted just three years before the AD 747 earthquake. A list of these and other Umayyad coins will be found in Appendix 9. Other finds from room 15 included more examples of mid-8th-century domestic pottery, especially cooking bowls (cf. PJ1: pl. 143: 2) as well as a hoe, harnessing rings, door hinges, and an iron lock. In the northern part of room 15, the effects of a fire after the collapse of the building were clearly discernible. Column drums were cracked and blackened and the yellow clay bricks baked red from the heat. This fire also engulfed the human couple (pl. 120) trapped in the north-east corner of the room, although their tragedy is our blessing, as the fire carbonized and preserved organic remains usually lost by decay at Pella. Of note are the textiles (see Appendix 8), oak beams, straw from mats, date stones, and olive pips. While removing the deposit in room 15, a number of interesting observations were made on the nature of materials used in the construction of the upper storey of the house. Numerous yellow clay and pebble bricks had fallen into this and the surrounding rooms from the upstairs walls along with segments of wooden beams used to support the floor and roof. A considerable number of large white tesserae were also found in room 15, in some cases still adhering to a pebble and mortar base. These originated, it would seem, from the floor of the room located above 15. The incinerated couple would have also fallen from this upper room when its floor collapsed during the first shocks of the AD 747 earthquake. In the adjacent room 13, more equid skeletons were uncovered, as well as three hens and another human, the latter also from the upper storey. Objects from the deposit in this room included lamps, two of pottery and one of bronze, and a glass vase. Unlike rooms 13 and 15, no skeletal material was excavated from the Umayyad levels in room 14. However, copious quantities of sherds from storage vessels were found at floor level; at least three large jars were crushed in situ according to the area supervisor (Edwards 1982). - Walmsley and Smith in McNicoll et al (1982) Five fallen columns and a pier originally arranged in two rows on an east-west axis, with three columns to the south and a combination of two columns and a pier in the northern row were discovered in the debris of Room 15 - JW: I forgot where I got this from	VIII+ ? ? V+ or VIII+ VII ? ?
Collapsed Walls Human and animal remains found in the rubble Objects of value found in the rubble Fallen Columns Debris	Area IX - Civic Church Complex in the Central Valley Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Fig. 2 Simplified plan of Area IX, showing the major features Smith et al. (1989) Fig. 28 Reconstructed plan of the church in Phase 4, from A.D. 658-660 until 717. The plan remained essentially the same, but with continued deterioration, in Phase 5 (A.D. 717-747). The position of columns in the west collonade of the atrium is somewhat uncertain. The drainage channel in the small chamber in the north dependency continued northwest into the Parvis. Smith et al. (1989)	Fig. 18 Detailed plan of part of the north dependency and Parvis. Locus 50, the "Chamber of the Camels," contains the skeletons of seven camels, a camel foetus, an ass, and Human 2. The small chamber, Locus 43, in the northeast corner of Locus 50, held the skeleton of Human 1. Locus 66, also a chamber, held the skeletons of two cows; the outline of stones beneath the skeletons is somewhat uncertain. Locus 67, a part of the Parvis, had the skeleton of two cows, and Locus 22, an extension of the north dependency, had fragile remains of a horse and a foal. In Locus 23 is the stile across the entrance to the north portal of the atrium, which prevented animals from wandering into the atrium and Church. Excavated portions of drainage channels, declining in a generally westward direction beneath the paving, are indicated by bold straight lines; hypothetical continuations, based on paving patterns, are indicated by rows of dots. Smith et al. (1989) Plate LXXXI.2 Skeletons found in the courtyard of the "hall of camels" just north of the Civic Complex Church. The animals are lying as they were killed in the earthquake of A.D. 746/7 JW: Actually that earthquake struck in 749 Smith in ADAJ Plates 1983 Plate 37A Three of the five skeletons excavated in the [Area IX] Civic Complex. A. Remains found in the southwest corner of the sanctuary Smith et al. (1989) Plate 37B Three of the five skeletons excavated in the [Area IX] Civic Complex. B. Skeleton found in the drainage channel in the triangular chamber. Smith et al. (1989) Plate 37C Three of the five skeletons excavated in the [Area IX] Civic Complex. C. Skeleton found inside the entrance to Locus 50, the Chamber of the Camels. Smith et al. (1989) Plate 39B The intertwined skeletons of Camels B and C [in Locus 50, the "Chamber of Camels" in the Area IX Civic Complex], viewed from the east. Smith et al. (1989) Plate 40C Skeleton of Camel F [in the Area IX Civic Complex] Smith et al. (1989) Plate 40D Skeleton of foetus of Camel B in Locus 50 [the "Chamber of Camels" in the Area IX Civic Complex] Smith et al. (1989) Plate 40E Ass Skeleton, with legs of Camel E at upper left and the skeleton of Camel D at right [in the Area IX Civic Complex] Smith et al. (1989) Fig. 13 Stratigraphic sections through A-A' in the sanctuary of the Church, and B-B' in the Parvis (for locations, see plan in Fig. to). Five levels are identified. Note that Level 2 is not present in section B-B'. Level 1: Gray topsoil containing small stones and pockets of Marnluk and post-Mamluk sherds. Level 2: Hard-packed brown soil, with some gravel; largely sterile but containing some sherds of various periods washed from upslope. (This level is not present in section B-B'.) Level 3: Dense, yellowish-brown loess, largely sterile but containing a few Abbasid-Fatimid sherds. At the beginning of this level a robber-pit was cut into level 4. Level 4: Debris from the earthquake of A.D. 747, loose, grayish-brown soil with rubble and architectural stones. Level 5: Brown, clayey soil containing trash accumulation from the earthquake of A.D. 717, including rubble, roof tile fragments, bits of mosaic, chunks of cement and artifacts (see Pls. 56—57). Smith et al. (1989)	Description The cathedral church and related structures excavated by Smith present an intriguing case of urban change after the Islamic expansion. Significant modifications occurred north of the church building, where a paved porticoed court, which coincided with the construction of the church, was modified by the construction of two-storeyed rooms faced with porches and galleries. Standing some 8.5 m high, this structure was clearly of major importance in the life of the town as it flanked the main entrance into the church, and was probably commercial in purpose (fig. 2).¹⁹ At least, this would seem to have been its final function on the day the A.D. 749 earthquake struck Pella with all of its intensity. Caught within the collapsing structure were two people and several animals, including 7 camels (one in advanced pregnancy), a horse and foal, an ass, and 4 cows. The age and condition of the camels - juvenile, young adult or in calf - suggests they were being sheltered on a particularly harsh winter’s day [JW: night].²⁰ ... With these fatalities of nature were recovered several Umayyad coins, including a dinar (94 A.H./A.D. 712/13) and 14 dirhams (between 81 A.H./A.D. 700/701 and 112 A.H./A.D. 730/31), numerous ceramic vessels and lamps, iron implements including a dagger, a bronze jug and three glass bottles.²¹ A mass of iron tools was found in a smith’s workshop in the south-west corner of the building, including a pick head, shovel, shears and sledgehammer head. What was happening in the church at this time is disputed (see note ¹³). In the court preceding the church (Atrium), three iron torch holders were found along with pottery vessels, suggesting a not insignificant level of activity. Within the church building, in the south aisle, the skeletons of two humans were found, but, apart from a few ceramic vessels (including a group of arcane pierced pottery cones), little else, except architectural debris from the collapsed church. Subsequently, from a level above the collapse that contained later Abbasid-Fatimid material, stone was robbed from the destroyed building.²² This pattern of an A.D. 749 collapse and subsequent robbing matches that recognised in Area IV, and further suggests that the looting of the church fixtures did not occur until after the earthquake of A.D. 749. Footnotes 19 This interpretation differs in details from Smith, who dates the conversion to the early 6th c. and proposes the purpose of the building was ecclesiastical, suggesting it was intended as a ‘hospice’ or served an ‘administrative’ function. This author has, in addition, emphasised a commercial purpose for this construction, proposing they served as the early Islamic market and caravanserai of Pella. All three functions—accommodation, administrative and commercial - could have (and probably were) intended. Equally likely, Pella’s clergy-administrators were the persons responsible for this enhancement to the urban environment, acting as the local representatives within the province of al-Urdunn. 20 On camels—the ‘second pillar’ of Bedouin life—see the definitive work by Jabbur (1995) 191–237. 21 Smith and Day (1989) 67–71. 22 Smith and Day (1989) 50–52, note especially ﬁ g. 13 and the accompanying explanation. - Walmsley (2007) The result of this activity was that piles of debris were still lying on the floor of the building on January 18, 747 [JW: Actually January 17 ± 1, 749 CE] when another major earthquake shook down much of the remaining architecture of the Church, killing at least two persons in the sanctuary as well as two other persons and a number of animals in the north dependency. Umayyad coins of the first half of the 8th century found on the floor of the Chamber of the Camels and coins in the possession of one of the victims confirm the date of the final destruction. - Smith et al. (1989:94)	VIII+ ? ? V+ or VIII+ ?
Seismic Destruction Collapsed Walls Fallen Column	Areas I and V - West and East churches Fig. 1 Map of archaeological areas at Pella Pella Project, modified Walmsley Walmsley (2008) Plan of West Church Complex in Area I Smith (1973)	Plate 12 Plate 12 I U: view toward the east. The massive block of fallen masonry on the left and the toppled column drum at the rear are the results of the earthquake which ended phase 4 of the [Western] church's history. The capital in the foreground may belong to the same column. Smith (1973)	As has already been noted, graphic archaeological evidence for the severity of the A.D. 749 earthquake has been found in different locations at Pella, notably the church complex in the central valley (Area IX), the West and East churches (Areas I and V), and the domestic quarters on the main mounds (Areas IV and VIII). - Walmsley (2007) When the earthquake struck, virtually all of the courses of the walls [of the Western Church Complex in Area I] not held in place by debris collapsed, generally falling westward. Buried in this final collapse were a few vessels in domestic use in the complex at the time. - Smith (1973:166) The massive block of fallen masonry on the left and the toppled column drum at the rear are the results of the earthquake which ended phase 4 of the [Western] church's history - Smith (1973:Plate 12)	VIII+? VIII+ 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). The interpreted second story collapse in Area IV and inability of animals and human victims to exit the premises before getting crushed by the structural collapse suggest a fairly rapid collapse of the structures, proximity to the epicenter, and an intensity well above VIII (8). The generally westward collapse of virtually all of the unbuttressed walls of the West Church Complex suggests that the epicenter was to the west.

References

Articles and Books

Bourke S. 1997. Pre-Classical Pella in Jordan: a conspectus of ten years' work (1985-1995). Palestine Exploration Quarterly 129:94-115.

Bourke, S. J. 2000 Pella in the Early Bronze Age. In G. Philip and D. Baird (eds.), Ceramics and Change in the Early Bronze Age of the Southern Levant, 233-253. Sheffield: Sheffield Academic Press.

Bourke S, Lawson E, Lovell J, Hua Q, Zoppi U, Barbetti M. 2001. The chronology of the Ghassulian Chalcolithic period in the southern Levant: new 14C determinations from Teleilat Ghassul, Jordan. Radiocarbon 43(3):1217-22.

Bourke S, Lawson E, Zoppi U, Hua Q. 2004a. The chronology of the Late Neolithic and Early Chalcolithic periods in the north Jordan Valley: new 14C determinations from Pella in Jordan. In: Higham T, Bronk Ramsey C, Owen C, editors. Radiocarbon and Archaeology. Proceedings of the 4th Symposium, Oxford 2002. Oxford: Oxford University School of Archaeology, Monograph 62. p 119-25.

Bourke S, Zoppi U, Meadows J, Hua Q, Gibbins S. 2004b. The end of the Chalcolithic period in the south Jordan Valley: new 14C determinations from Teleilat Ghassul, Jordan. Radiocarbon 46(1):315-23.

Bourke, S. J. 2004 Cult and archaeology in Pella in Jordan: excavating the Bronze and Iron Age temple precinct. Journal and Proceedings of the Royal Society of New South Wales 137: 1-31.

Bourke, S., Sparks, R. and Schroder, M. 2006 Pella in the Middle Bronze Age. In P.M. Fischer (ed.), The Chronology of the Jordan Valley during the Middle and Late Bronze Ages: Pella, Tell Abu al-Kharaz, and Tell Deir ‘Alla, 9- 58. Vienna: Velger der Ӧsterreichischen Akademie der Wissenschaften.

Bourke S. 2008 The Chalcolithic period. In: Adams R, editor. Jordan: An Archaeological Reader. London: Equinox. p 109-60.

Bourke, S., Zoppi, U., Meadows, J., Hua, Q., & Gibbins, S. (2009). The Beginning of the Early Bronze Age in the North Jordan Valley: New 14C Determinations from Pella in Jordan. Radiocarbon, 51(3), 905-913.

Bourke, S. J. 2012 The six Canaanite temples of Tabaqat Phahil: excavating Pella’s ‘fortress’ temple (1994- 2009). In Temple Building and Temple Cult: Architecture and Cultic Paraphernalia of Temples in the Levant (2-1. Mill. B.C.E.), 159-201. Wiesbaden: Harrassowitz. Abhandlungen des Deutschen Palästina-Vereins 41.

Bourke S. 2013, Pre-Classical Pella in Jordan: Pre-Classical Pella in Jordan: A Conspectus of Recent Word ACOR Newsletter Volume 25.1 Summer 2013

Blanke, L. and A. Walmsley (2022). Resilient cities: Renewal after disaster in three late antique towns of the East Mediterranean. Remembering and Forgetting the Ancient City, Oxbow Books: 69-109.

FALCONER, S. 2001 The Middle Bronze Age, 271–289, in: MACDONALD, B. ADAMS, R. and BIENKOWSKI, P. (eds.), The Archaeology of Jordan. Sheffield.

MAGNESS-GARDINER, B. 1997 Jordan in the Middle Bronze Age, 303–326. in: OREN, E. (ed.), The Hyksos: New Historical and Archaeological Perspectives. Philadelphia. - open access at archive.org - bookmarked to the article

Petit, L. P., 2009, Settlement Dynamics in the Middle Jordan Valley during the Iron Age II. British Archaeological Reports: International Series 2033. - accessible at the Getty

Shea, William H., 1989, The Inscribed Tablets from Tell Deir 'Alla Part I, Andrews University Seminary Studies, Spring 1989 Vol. 27 No.1, 21-37, Andrews University Press

Steiner, M. L. (2019). "Iron Age Cultic Sites in Transjordan." Religions 10(3): 145. - see page 3 for a one paragraph description of archaeoseismic evidence at Deir Alla

Van der Kooij, G., (2006), The Chronology of the Jordan Valley during the Middle and Late Bronze Ages: Pella, Tell Abu al-Kharaz and Tell Deir ‘Alla (Contributions to the Chronology of the Eastern Mediterranean; ed. P. Fischer), Wien (2006).

Walmsley, A. G. (1997) SETTLED LIFE IN MAMLÛK JORDAN VIEWS OF THE JORDAN VALLEY FROM FAHL (PELLA) Aram Periodical Vol. 9.1-2: 129-43

Walmsley, A. G. (2007a) ‘Households at Pella, Jordan: The domestic destruction deposits of the mid-eighth century’ in L. Lavan, E. Swift and T. Putzeys (eds.), Objects in Context, Objects in Use. Material Spatiality in Late Antiquity, Late Antique Archaeology 5 (Leiden, 2007), pp. 239–272.

Walmsley, A. (2007b). "Economic Developments and the Nature of Settlement in the Towns and Countryside of Syria-Palestine, ca. 565-800." Dumbarton Oaks Papers 61: 319-352.

Walmsley, A. (2008). "HOUSEHOLDS AT PELLA, JORDAN: DOMESTIC DESTRUCTION DEPOSITS OF THE MID-8TH C." Late Antique Archaeology 5(1): 239-272.

Walmsley, A. (2013). Early Islamic Syria, Bloomsbury Publishing.

Watson, P.M., “Change in Foreign and Regional Economic Links with Pella in the Seventh Century A.D.: The Ceramic Evidence,” in La Syrie de Byzance à l’Islam, VIIe–VIIIe siècles: Actes du colloque international Lyon, ed. P. Canivet and J.-P. Rey-Coquais (Damascus, 1992), 233–48.

Excavation Reports

Pella in Jordan 1 and 2

McNicoll, A., et al. (1982). Pella in Jordan 1: An Interim Report on the Joint University of Sydney and the College of Wooster Excavations at Pella 1979-1981, Australian National Gallery.

Walmsley, A. G., ‘Islamic Coins’ in K. A. Sheedy, R. Carson and A. Walmsley (eds.), Pella in Jordan 1979–1990: The Coins, Adapa Monograph Series 1 (Sydney, 2001), pp. 57–66, 147–153.

Chapter 7 - Umayyad Period by Walmsley and Smith in McNicoll, A., et al. (1982). Pella in Jordan 1

McNicoll, A. (1992). Pella in Jordan 2 : the second interim report of the joint University of Sydney and the College of Wooster excavations at Pella 1982-1985 : by A.W. McNicoll ... [et al.] with additional contributions by J.A. Black ... [et al.]. Sydney, Meditarch.

Chapter 9 - Islamic Period by Walmsley and Smith in McNicoll, A. (1992). Pella in Jordan 2

Walmsley, Alan G., and Robert H. Smith. 1982. The Islamic Period. In Pella in Jordan 2, ed. A. W. McNicoll et al. pp. 183-98. Sydney: Mediterranean Archaeology Supplement 2.

Eastwood G. M. (1992) “The Pella textiles”, in Pella in Jordan 2. The Second Interim Report of the Joint University of Sydney and College of Wooster Excavations at Pella 1982–1985, ed. A. W. McNicoll et al. (Sydney 1992) 257–65.

Pella of the Decapolis 1 and 2

Smith, Robert Houston ed. (1973) Pella of the Decapolis. Vol. 1 The 1967 Season of the College of Wooster Expedition to Pella, The College of Wooster - can be borrowed with a free account from archive.org

Smith et al. (1989) Pella of the Decapolis. Vol. 2. Final report on the College of Wooster excavations in area IX, the Civic Complex, 1979-1985 [Wooster, Ohio], College of Wooster. - can be borrowed with a free account from archive.org

Other Excavation Reports

Schumacher, Gottlieb (1988) Pella, Palestine Exploration Fund - open access at archive.org

Hennessy, A.W. et. al. (1983) Preliminary Report on the Fourth Season of Excavations at Pella, 1982 Annual of the Department of Antiquities of Jordan 27

Hennessy, A.W. et. al. (1983) Preliminary Report on the 1983 Season of the Sydney/Wooster Joint Expedition to Pella: Spring Session Annual of the Department of Antiquities of Jordan 27

Department of Antiquities of Jordan PLATES_1983 - contains Plates from Pella

Walmsley et. al., 1993, The Eleventh and Twelfth Seasons of Excavations at Pella (Tabaqat Fahl) 1989-1990 ADAJ XXXVII

Bourke, S.J., Sparks, R.T., and Mairs, L.D., 1999, Bronze Age Occupation on Tell Husn (Pella): Report on the University of Sydney's 1994/95 Field Seasons, Mediterranean Archaeology Vol. 12 1999

Bourke S, Sparks R, McLaren B, Sowada K, Mairs L, Meadows J, Hikade T, Reade W. 2003. Preliminary report on the University of Sydney's eighteenth and nineteenth seasons of excavations at Pella (Tabaqat Fahl) in 1996/97. Annual of the Department of Antiquities Jordan 47:335—88.

Tidmarsh, J. (2024). After Alexander: The Hellenistic and Early Roman Periods at Pella in Jordan, Sydney University Press.

Websites

Alan G. Walmsley academia.edu page

Stephen J. Bourke, Pella Publications - University of Sydney

search for "Pella in Jordan" on academia.edu

ACOR Photo Archive - a search for Pella will return many photos

Photos of Pella at PBase.com

Pella's Canaanite Temple

Pella Project - University of Sydney

Bibliography from Stern et al (1993 v. 3)

History

Abel, GP 2, 405-406 and passim

W. F. Albright, BASOR 81 (1941), 19; 83 (1941), 33, 36; 89 (1943), 9-13, 15, 17

Pritchard, ANET, 243, 253, 329, 486.

Main Publications

R. H. Smith, Pella of the Decapolis I, The 1967 Season of the College of Wooster Expedition to Pella, Wooster, 1973

id. (and L. P. Day), Pella of the Decapolis 2, Final Report on the College of Wooster Excavations in Area IX, The Civic Complex, 1979-1985, Wooster 1989

A. McNicoll et al.. Pella in Jordan l, An Interim Report on the Joint University of Sydney and the College of Wooster Excavations at Pella 1979-1981, Canberra 1982

id., Pella in Jordan 2, Second Interim Report (ibid.), 1982- 1985, Sydney (in prep.).

Other Studies

S. Merrill, East of the Jordan, New York 1881, l84f., 442-447, 463

G. Schumacher, Across the Jordan, London 1886, 271-278

id., Abila, Pella, and Northern Ajlun, London 1895, 78ff.

W. F. Albright, AASOR 6 (1924-1925), 39-42

J. Richmond, PEQ 66 (1934), 18-31

N. Glueck, BASOR 89 (1943), 3, Sf.

90 (1943), 3

91 (1943), 16

id., Explorations in Eastern Palestine 4 (AASOR 25-28, 1945- 1949), New Haven 1951, 254-257

R. W. Funk and H. N. Richardson, BA 21 (1958), 82-96

H. N. Richardson, RB 67 (1960), 242-243

R. H. Smith,JLN6711 (Mar. 16, 1968), 26-27

id., Archaeology 21 (1968), 134-137

26 (1973), 250-256

34/5 (1981), 46-53

id., RB75 (1968), 105-112

id., Wooster Alumni Magazine 82 (1968), 4-13

id., ADAJ 14 (1969), 5-10

24 (1980), 13-14

25 (1981), 311-326

26 (1982), 323-334

27 (1983), 363-373, 647

id., PEQ 101 (1969), 2-3, 55

id., Berytus 21 (1972), 39-53

id., Pella of the Decapo/is 1 (Review), ADAJ 19 (1974), 169-172

2 (Review), LA 39 (1989), 283-286

id., ASOR Newsletter(Apr. 1980), 22-26

(May 1981), 7-10

(May 1983), 13-17

id., BA 43 (1980), 61-62

id.,AJA 85 (1981), 218

87 (1983), 260

91 (1987), 276, 308-309

id. (et al.), BASOR 243 (1981), 1-30

249 (1983), 45- 78

id., LA 31 (1981), 345-347

34 (1984), 433-436

id., Syria 60 (1983), 303-306

id. (et al.), Current Anthropology 25 (1984), 234-236

id., National Geographic Research I (1985), 470-489

id. (and A. W. McNicol!), BASOR Supplement 24 (1986), 89-116

id., SHAJ 3 (1987), 35-58

id., Aram 4 (in prep.); A. Spijkerman, LA 20 (1970), 353-358

N. W. Boweret al., JFA 2 (1975), 390-398

A. McNicol! et al., ADAJ24 (1980), 14-40

25 (1981), 358-360

26 (1982), 343-363

28 (1984), 55-86

30 (1986), 155-198; id., BASOR 240 (1980), 63-84

id., MdB 22 (1982), 34-36

id., SHAJ l (1982), 339-345

id., Pella in Jordan l (Reviews),AJA 88 (1984), 426-427.-PEQ 116 (1984), 156-157

id., Pella inJordan2 (Review), LA 38 (1988), 465-467

J. B. Hennessyet al., ADAJ25 (1981), 267-309

27 (1983), 325-361

id., Ancient Middle East Ceramics and Australian Archaeology (eds. C. A. Hope and J. K. Zimmer), Melbourne 1983, 13-24

id., Palestine in the Bronze and Iron Ages (0. Tufnell Fest.), London 1985, 100-113

H.-F. Vogenbeck, Antike Welt 12/3 (1981), 13-16

F. Manns, MdB22 (1982), 37-39;A. Walmsley, RB89(1982), 245-247

M. Piccirillo, MdB 35 (1984), 8-9

T. F. Potts, LA 34 (1984), 436-437

id. (et al.), ADAJ 29 (1985), 181-210

32 (1988), 115-149

id., Antiquity 60/230 (1986), 217-219

id., SHAJ3 (1987), 59-71; J. Hanbury-Tenison, PEQ 117 (1985), 100-101

J. Balensi, MdB 46 (1986), 8-9

A. B. Knapp, BASOR 266 (1987), 1-30

id. (et al.), Journal of Mediterranean Archaeology 1/2 (1988), 57-113

id., IEJ39 (1989), 129-148

Khouri, Antiquities, 21-26

id., Pella: A Brief Guide to the Antiquities (AI KutbaJordan Guides), Amman 1988

Weippert 1988 (Ortsregister)

Akkadica Supplementum 7-8 (1989), 406-441

P. C. Edwards et al., ADAJ 34 (1990), 57-93

id., The Natufian Culture in the Levant (International Monographs in Prehistory, Archaeology Series I, eds. 0. Bar-Yosefand F. R. Valla), Ann Arbor 1991, 123-148

F. V. Richards, Scarab Seals from a Middle to Late Bronze Age Tomb at Pella in Jordan (Orbis Biblicus et Orientalis 117), Freiburg (in prep.)

T. Weber, Pella Decapolitana

Studien zur Geschichte, Architektur und figiirlichen Kunst einer hellenisierten Stadt des nOrdlichen Ostjordanlandes (Abhandlungen des Deutsch en Paliistinavereins 18), Wiesbaden (in prep.).

Bibliography from Stern et al (2008)

Main Publications

A. W. McNicoll et al., Pella in Jordan, 2: The 2nd Interim Report of the Joint University of Sydney and College of Wooster Excavations at Pella 1982–1985 (Mediterranean Archaeology Suppl. 2), Sydney 1992

ibid. (Reviews) PEQ 128 (1996), 78–79. — ZDPV 111 (1995), 195–198

F. V. Richards, Scarab Seals from a Middle to Late Bronze Age Tomb at Pella in Jordan (Orbis Biblicus et Orientalis 117), Freiburg 1992

ibid. (Reviews) JAOS 114 (1994), 97–98. — Orientalia N.S. 63 (1994), 129–132. — JEA 82 (1996), 223

A. B. Knapp, Society and Polity at Bronze Age Pella: An Annales Perspective (JSOT/ASOR Monographs 6), Sheffield 1993

ibid. (Reviews) Antiquity 68/259 (1994), 456–457. — AJA 99 (1995), 155–156. — JAOS 115 (1995), 131–132. — BASOR 310 (1998), 81–82

T. M. Weber, Pella Decapolitana: Studien zur Geschichte, Architektur und Bildenden Kunst einer hellenisierten Stadt des nördlichen Ostjordanlandes (Abhadlungen des Deutschen Palästinavereins 18), Wiesbaden 1993

K. Sheedy et al., Pella in Jordan 1979–1990: The Coins (Adapa Monograph Series 1), Sydney 2001

ibid. (Reviews) American Journal of Numismatics 13 (2001), 147–150. — The Numismatic Chronicle 162 (2002), 450–451. — BASOR 329 (2003), 97–99. — PEQ 135 (2003), 74–75. — Revue Numismatique 160 (2004), 395–398

P. B. McLaren, The Military Architecture of Jordan during the Middle Bronze Age: New Evidence from Pella and Rukeis (BAR/IS 1202), Oxford 2003

The Chronology of the Jordan Valley during the Middle and Late Bronze Ages: Pella, Tell Abu al-Kharaz and Tell Deir ‘Alla (Contributions to the Chronology of the Eastern Mediterranean; ed. P. Fischer), Wien (forthcoming)

Studies

J. M. C. Bowsher, PEQ 124 (1992), 74–75 (Review)

R. de Vries, AJA 96 (1992), 520–522, 539– 541

P. G. Macumber, Paléorient 18 (1992), 31–44

id. (et al.), SHAJ 6 (1997), 87–92

M. O’Hea, Aram 4 (1992), 253–264

id., Annales du Congrès de l’Association Internationale pour l’Histoire du Verre 14 (1998), 219–222

R. H. Smith, ABD, 5, New York 1992, 219–221

id., Aram 4 (1992), 197–214

id., The Archaeology of Jordan and Beyond, Winona Lake, IN 2000, 465–475

A. G. Walmsley, La Syrie de Byzance a l’Islam, VIIe–VIIIe siécles. Actes du Colloque International, Lyon, 11–15.9.1990 (eds. P. Canivet & J. -P. Rey-Coquais), Damas 1992, 249–261

id., SHAJ 4 (1992), 377–385

id. (et al.), ADAJ 37 (1993), 155–240; id., Mediterranean Archaeology 8 (1995), 113–114

id., Trade, Contact and the Movement of Peoples in the Eastern Mediterranean (J. B. Hennessy Fest.

Mediterranean Archaeology Suppl. 3

eds. S. J. Bourke & J. -P. Descoeudres), Sydney 1995, 321–324

id., Aram 9–10 (1997–1998), 129–143

P. M. Watson, La Syrie de Byzance a l’Islam (op. cit.), Damas 1992, 233–248

id., AJA 98 (1994), 557–558

id. (& J. Tidmarsh), ADAJ 40 (1996), 293–313

id., Levant 28 (1996), 63–76

30 (1998), 219

id., MdB 105 (1997), 52

id., SHAJ 7 (2001), 485–492

id., Gadara-Gerasa und die Dekapolis (Zaberns Bildbände zur Archäologie

Antike Welt Sonderbände

eds. A. Hoffmann & S. Kerner), Mainz am Rhein 2002, 59–71

S. J. Bourke, Mediterranean Archaeology 5–6 (1992–1993), 161–163

12 (1999), 51–66 (et al.), 147–155

id., AJA 97 (1993), 467–471; 100 (1996), 522–523

103 (1999), 494–496 (& K. Da Costa)

id. (et al.), ADAJ 38 (1994), 81–126

42 (1998), 179–211

47 (2003), 335–388

id. (& R. T. Sparks), Trade, Contact (op. cit.), Sydney 1995, 149–167; id., Orient Express 1996, 3–5

id., PEQ 129 (1997), 94–115

id., Ancient History in Modern University (eds. T. W. Hillard et al.), 1, Grand Rapids, MI 1998, 103–110

id., Occident and Orient 4 (1999), 56–57; 8/2 (2003), 5–6

id., Ceramics and Change, Sheffield 2000, 233–253

id., ASOR Annual Meeting Abstract Book, Boulder, CO 2001, 14

F. Blanchetiere & R. Pritz, Aux origins juives du christianisme (Cahiers du Centre de Recherche Français de Jérusalem, Serie Hommes et Sociétés 2

eds. F. Blanchetiere & M. D. Herr), Jerusalem 1993, 93–110

B. E. Denton, ACOR: The First 25 Years, Amman 1993, 66–68

I. Melhem & N. Khasawneh, ADAJ 38 (1994), 31*–38*

J. A. Sauer, BASOR 293 (1994), 83–84 (Review)

P. Artzi, Journal for Semitics 7 (1995), 118–124

P. C. Edwards & P. G. Macumber, Trade, Contact (op. cit.), Sydney 1995, 1–14

S. -A. Wallace, ibid., 291–302

A. G. Walmsley, SHAJ 5 (1995), 657–668

M. Signidou, Basileia: Die Paläste der hellenistischen Könige. Internationales Symposium, Berlin, 16–20.12.1992 (Schriften des Seminars für klassische Archäologie der Freien Universität, Berlin

eds. W. Höpfner & G. Brands), Mainz am Rhein 1996, 144–147

F. Baker, ADAJ 41 (1997), 69–73

id., Levant 30 (1998), 215–216; 31 (1999), 313–314

J. B. Hennessy, MdB 104 (1997), 18–19

id., OEANE, 4, New York 1997, 256–259

H. -V. Kieweler, Jounal for Semitics (University of South Africa, Pretoria) 9 (1997), 1–38

A. M. Maeir, The Material Culture of the Central Jordan Valley during the Middle Bronze II Period: Pottery and Settlement Pattern, 1–2 (Ph.D. diss.), Jerusalem 1997

id., PEQ 132 (2000), 37–58

J. Magness, JAOS 117 (1997), 481–486

Z. al-Muheisen & D. Terrier, LA 47 (1997), 493

Y. Garfinkel, Levant 30 (1998), 191–194

F. Zayadine, NEAS Bulletin 43 (1998), 39–40

F. Israel, Guerre et conquete dans le Proche-Orient Ancien. Actes de la Table Ronde, 14–11.1998 (ed. L. Nehme), Paris 1999, 63–77

E. J. Van der Steen, PEQ 131 (1999), 176–192

T. Weber, Damaszener Mitteilungen 11 (1999), 433–451

id., Gadara-Umm Qes I: Decapolitana Untersuchungen zur Topographie, Geschichte, Architektur und der Bildenden Kunst einer “Polis Hellenis” im Ostjordanland (Abhandlungen des Deutschen-Palästina-Vereins 30), Wiesbaden 2002, 483–484

P. M. Fischer (& K. Kopetzky), Synchronisation, Wien 2000, 132

id., Ägypten und Levante 13 (2003), 51–68; J. L. Lovell, Ceramics and Change, Sheffield 2000, 59–71

G. Touchais et al., Bulletin de Correspondance Hellenique 124 (2000), 919–920

The Archaeology of Jordan (Levantine Archaeology 1

eds. B. MacDonald et al.), Sheffield 2001

Australians Uncovering Ancient Jordan: 50 Years of Middle Eastern Archaeology (ed. A. Walmsley), Sydney 2001

B. Bagatti, Ancient Christian Villages of Galilee (SBF Collectio Minor 37), Jerusalem 2001, 259–263

A. Michel, Les églises d’époques byzantine et umayyade de Jordanie, Ve–VIIIe siécle: étude de typologie et inventaire, Turnhout 2001

R. T. Sparks, SHAJ 7 (2001), 259–264

K. da Costa et al., ADAJ 46 (2002), 503–533

Z. A. Kafafi, Kein Land für sich allein, Freiburg 2002, 20–30

H. Richter, Die Phönizischen Anthropoiden Sarkophage, 2: Tradition, Rezeption, Wander (Forschungen zur PhönizischPunischen und Zyprischen Plastik I/2

ed. S. Frede), Mainz am Rhein 2002, 243–271

B. Andreae, Antike Bildmosaiken, Mainz am Rhein 2003, 19–25

K. Dahmen & A. Lichtenberger, Boreas 26 (2003), 39–47; P. H. R. Houwelingen, Westminster Theological Journal 65 (2003), 181–200

A. Lichtenberger, Kulte und Kultur der Dekapolis: Untersuchungen zu numismatischen, archäologischen und epigraphischen Zeugnissen (Abhandlungen des Deutschen Palästina-Vereins 29), Wiesbaden 2003

G. Philip et al., Levant 35 (2003), 71–100

M. Donnelly, SHAJ 8 (2004), 97–108

Y. Elitzur, Ancient Place Names in the Holy Land: Preservation and History, Jerusalem 2004, 60–64

Y. Goren et al., Inscribed in Clay, Tel Aviv 2004, 260–261

J. Haser, SHAJ 8 (2004), 155–160

I. Shachar, PEQ 136 (2004), 5–33

J. Tidmarsh, SHAJ 8 (2004), 459–468; E. Villeneuve, MdB 158 (2004), 29.

Bibliography from Meyers et. al. (1997)

Hennessy, J. Basil, et al. "Pella." In Archaeology of Jordan, vol. 2, Field Reports, edited by Denys Homes-Fredericq and J. Basil Hennessy, pp. 406-441. Louvain, 1989. Lengthy survey of the Pella excavations; the bibliography should also be consulted.

McNicoll, Anthony W „ Robert Houston Smith, and J. Basil Hennessy. Pella in Jordan 1: An Interim Report on the Joint University of Sydney and the College of Wooster Excavations at Pella, 1979-1981. Canberra, 1982. Contains some material that will not be repeated in the final excavation report series, Pella of the Decapolis.

McNicoll, Anthony W., and Robert Houston Smith, et al. Pella in Jordan 2: The Second Interim Report of the Joint University of Sydney and the College of Wooster Excavations at Pella, 1982-1985. Sydney, 1992. Detailed report of excavations to the end of 1985, containing some material that will not be repeated in the final excavation report series.

Smith, Robert Houston. The 1967 Season of the College of Wooster Expedition to Pella. Pella of the Decapolis, vol. 1. Wooster, Ohio, 1973. First volume in a series of final excavation reports of the Wooster Expedition to Pella and the Sydney-Wooster Joint Expedition.

Smith, Robert Houston. "Excavations at Pella of the Decapolis, 1979- 1985." National Geographic Research 1 (1985): 470-489. Overview of the results of the first seven seasons of the Sydney-Wooster Joint Expedition to Pella.

Smith, Robert Plouston. "Trade in the Life of Pella of the Decapolis." In Studies in the History and Archaeology of Jordan, vol. 3, edited by Adnan Hadidi, pp . 53-58. Amman , 1987. Discusses the extent to which international commerc e was a factor in Pella's history.

Smith, Robert Houston, and Leslie P. Day. Final Report on the College of Wooster Excavations in Area IX, the Civic Complex, 1979-198$. Pella of the Decapolis, vol. 2. Wooster, Ohio, 1989. Second volume in a series of final excavation reports.

Smith, Robert Houston. "The Southern Levant in the Hellenistic Period." Levant 22 (1990): 123-130. Discusses the Hellenistic period in Palestine and Transjordan in light of historical and archaeological evidence, with particular reference to Pella.

Wikipedia pages

Pella

from Wikipedia

Decapolis

from Wikipedia

Flight to Pella

from Wikipedia

Battle of Fahl

from Wikipedia