Origins of Doha Project: Season 1
Archive Report
October 2013
Robert Carter and Daniel Eddisford
Contents
List of Illustrations ............................................................................................................................ 6
1. Summary ....................................................................................................................................... 9
2. Acknowledgements .................................................................................................................... 10
3. Organisations, Project and Personnel ......................................................................................... 10
4. Historical Background to Doha.................................................................................................... 11
5. Archaeological Investigations in the Heritage Quarter of the Msheireb Development ............... 17
5.1 Introduction ........................................................................................................................... 17
5.2 Aims of the Excavation ........................................................................................................... 20
5.3 Methods ................................................................................................................................ 20
5.3.1 Trench Locations ............................................................................................................. 20
5.3.2 Excavation Methods ........................................................................................................ 21
5.4 Phased Summary ................................................................................................................... 23
5.4.1 Phase 1 – Pre-architecture .............................................................................................. 23
5.4.2 Phase 2 – Early architecture ............................................................................................ 24
5.4.3 Phase 3 – Joining of two compounds ............................................................................... 26
5.4.4 Phase 4 – Later architecture ............................................................................................ 27
5.4.5 Phase 5 – Reconstruction by Private Engineering Office .................................................. 29
5.5 Technical Report: Excavations in the Radwani House ............................................................. 30
5.5.1 Room 6 ........................................................................................................................... 30
5.5.2 Room 8 ........................................................................................................................... 32
5.5.3 Room 9 ........................................................................................................................... 37
5.5.4 Room 10 ......................................................................................................................... 40
5.5.5 Room 12 ......................................................................................................................... 43
5.5.6 Trench 1 .......................................................................................................................... 46
5.5.7 Trenches 2 and 3 ............................................................................................................. 52
5.5.8 Trench 4 .......................................................................................................................... 54
5.6. Watching Brief at Company House ........................................................................................ 56
5.6.1 Room 1 ........................................................................................................................... 57
5.6.2 Room 3 ........................................................................................................................... 57
5.6.3 Room 15 ......................................................................................................................... 58
5.6.4 Room 21 ......................................................................................................................... 59
5.7. Discussion ............................................................................................................................. 59
2
5.8. Finds and Analytical Material ................................................................................................ 60
5.8.1 Miscellaneous Finds ........................................................................................................ 60
5.8.2 Toys ................................................................................................................................ 62
5.8.3 Beads and Bangles .......................................................................................................... 63
5.8.4 Textiles ........................................................................................................................... 64
5.8.5 Electrics .......................................................................................................................... 65
5.8.6 Ceramic Finds (pottery) ................................................................................................... 65
5.8.7 Glass ............................................................................................................................... 69
5.8.8 Metal Finds ..................................................................................................................... 70
5.8.9 Animal bone .................................................................................................................... 72
5.8.10 Shell .............................................................................................................................. 73
5.8.11 Company House Finds ................................................................................................... 74
6. Recording Archaeological Deposits Exposed in Section............................................................... 75
6.1 Introduction ........................................................................................................................... 75
6.2 Site ODQ 3 ............................................................................................................................. 76
6.2.1 Site location .................................................................................................................... 76
6.2.2 Section 1 ......................................................................................................................... 78
Section 2 .................................................................................................................................. 79
6.3 ODQ 4 .................................................................................................................................... 83
6.3.1 Site location .................................................................................................................... 83
6.3.2 Section 1 ......................................................................................................................... 84
6.4 Discussion .............................................................................................................................. 87
7. Archaeological Survey of an Abandoned Agricultural Area in Rayyan ........................................ 89
7.1 Introduction ........................................................................................................................... 89
7.2 Site location and scope of work.............................................................................................. 89
7.3 Archaeological and Historical Background .............................................................................. 92
7.4 Results ................................................................................................................................... 95
7.4.1 Traditional box well ......................................................................................................... 96
7.4.2 Pump Houses .................................................................................................................. 97
7.4.3 Bore holes ....................................................................................................................... 99
7.4.4 Rectangular Tanks ........................................................................................................... 99
7.4.5 Irrigation Features......................................................................................................... 101
7.4.6 Structures ..................................................................................................................... 102
7.5 Discussion ............................................................................................................................ 106
3
8. Standing Building Recording in Doha ........................................................................................ 108
8.1 Introduction ......................................................................................................................... 108
8.2 Rationale ............................................................................................................................. 108
8.3 Aims .................................................................................................................................... 110
8.4 Methodology ....................................................................................................................... 110
8.4.1 Desk-based work ........................................................................................................... 110
8.4.2 Definition of research area ............................................................................................ 110
8.4.3 Recording ...................................................................................................................... 112
8.5 Discussion ............................................................................................................................ 113
9. Conclusions ............................................................................................................................... 115
10. Bibliography ............................................................................................................................ 116
11. Appendix A – Radwani House: Specialist Report on Plaster .................................................... 118
11.1 Background, Sample Materials and Aims of Analysis ...................................................... 118
11.2 Methodology .................................................................................................................... 118
11.3 Results .............................................................................................................................. 119
11.3.1 Type 1 - Hydraulic lime binder with oolitic limestone aggregate ............................. 119
11.3.2 Type 2 - Gypsum and lime binder without aggregate .............................................. 121
11.3.3 Type 3 - Lime and gypsum with oolitic limestone aggregate ................................... 122
11.3.4 Microstratigraphy...................................................................................................... 122
11.3.5 Discussion .................................................................................................................. 123
11.3.6 References Cited ......................................................................................................... 125
11.3.7 Tables and Figures ..................................................................................................... 126
12. Appendix B – Radwani House: Preliminary Specialist Report on Bone .................................... 138
12.1 Introduction and Methodology .......................................................................................... 138
12.2 Results ............................................................................................................................... 138
12.2.1 Caprines ...................................................................................................................... 138
12.2.2 Other mammals .......................................................................................................... 139
12.2.3 Birds............................................................................................................................ 139
12.2.4 Fish ............................................................................................................................. 140
12.2.5 Other marine taxa ....................................................................................................... 140
12.2.6 Phase comparison ....................................................................................................... 140
12.3 Discussion .......................................................................................................................... 141
12.4 References ......................................................................................................................... 141
13. Appendix C – Radwani House: Preliminary Specialist Report on Botanical Remains ............... 145
14. Appendix D – Radwani House: Preliminary Specialist Report on Wood Charcoal ................... 145
4
15. Appendix E – Radwani House: Context List ............................................................................. 147
16. Appendix F: Radwani House: Finds List ................................................................................... 152
5
List of Illustrations
Figure 4.1:
Figure 4.2:
Figure 5.1:
Figure 5.2:
Figure 5.3 :
Figure 5.4:
Figure 5.5:
Figure 5.6:
Figure 5.7:
Figure 5.8:
Figure 5.9:
Figure 5.10:
Figure 5.11:
Figure 5.12:
Figure 5.13:
Figure 5.14:
Figure 5.15:
Figure 5.16:
Figure 5.17:
Figure 5.18:
Figure 5.19:
Figure 5.20:
Figure 5.21:
Figure 5.22:
Figure 5.23:
Figure 5.24:
Figure 5.25:
Figure 4.26:
Figure 5.27:
Figure 5.28:
Figure 5.29:
Figure 5.30:
Figure 5.31:
The Turkish fort with Doha in the background, 1904.
Part of the Hunting Survey of 1957 showing the Radwani House.
Location of the Msheireb redevelopment project (Google Earth image
2006).
Location of Radwani House and Company House (Google Earth image
2006).
Radwani House prior to reconstruction in 2006
Location of Excavation Areas in relation to existing walls of the Radwani
House.
The courtyard of the Radwani House looking southwest, prior to
excavation of Trenches 1 and 2.
Phase 1 features and deposits.
Phase 2 architecture.
Phase 3 architecture.
Phase 4 architecture and features.
Surfaces (174) and (191) overlain by wall (170). Cut for later soakaway
visible at far right.
Limestone wall (201) and render (239), the metal sewer pipe is a later
addition to the building.
Substantial early wall (15), with earlier postholes [101], [103] and
[105] in foreground and pit [109] in background. Looking south.
External well [126] and associated walls (86) and (15). Photographed in
plan.
Early architecture and internal courtyard floor in Room 8, looking south.
A section through the floor surfaces associated with Phase 2, looking
north.
Concert or gypsum threshold (113) and surface (87) to north of wall 15
blocking earlier alleyway.
Metal drain cover, draining into soakaway [118].
Walls (46) / (47) in foreground and wall (64) in background ,looking south.
North facing section at southern end of Room 9.
Members of the Origins of Doha Project team working in Room 9.
Phase 2 or Phase 3 limestone wall, constructed on levelling layer (123)
sealed reconstructed walls of Radwani House. Looking west.
Well (143) with bath (140) in background and truncated floors to the left
the well. Looking north.
Concrete baths (140), looking northwest.
Layer (79) overlain by several superimposed building phases. Looking
northeast.
Archaeological deposits in section in Room 12, looking east.
West facing through the occupation sequence excavated in Room 12
Early architecture (219) and (159) with later enclosure wall (172) in
background under scale. Looking south.
Postholes [215] and [218] adjacent to wall (178).
Walls (178) and (235) with alleyway and northern compound wall (172) in
background. Looking north.
North facing section through courtyard deposits in Trench 1.
Courtyard surface (227) cut by drain [185] with limestone capping stones
(181) still in place, looking west.
6
Figure 5.32:
Figure 5.33:
Figure 5.34:
Figure 5.35:
Figure 5.36:
Figure 5.37:
Figure 5.38:
Figure 5.39:
Figure 5.40:
Figure 5.41:
Figure 5.42:
Figure 5.43:
Figure 5.44:
Figure 5.64:
Courtyard surfaces in Trench 2 and circular Trench 3, looking north.
Heavily truncated wall in the western end of Trench 2, looking south.
Wall (192) and partially excavated mortar surface to the south of it.
Location of archaeological features in Company House.
Partially exposed well in Room 1, looking east.
Well and later drain in Room 3, looking west.
Large circular metal feature in Room 15.
Part of the oven in Room 21.
Limestone incense burner. Context 166, with small fragment in 171, Find 55
Fragment of probable clay prayer tablet – turba. Context 12, Find 54
Saudi Arabian 25 halala coin, probably 1970-1976 (Context 7, Find 57)
Unidentified bronze coin (Context 21, Find 60) the 1920s-1930s)
Marbles from Radwani House. Top left and top middle: Context 51, Find 1; top right:
Context 107, Find 31; bottom left: Context 34, Find 21; bottom right: Context 227,
Find 47.
Small metal allow object from the Radwani House. Perhaps a whistle.
Beads and Bangle fragments from the Radwani House.
A mother-of-pearl button. Context 41, Find 59
Fabric with metallic thread. Context 39, Find 82
Electric light fittings and bulbs. Left: Context 174, Finds 63, 61; Right: Context 151,
Finds 64, 62
Water-pipe and burner from the well in Room 10 at the Radwani House (Context
300, pot no. 147, 148)
Earthenwares from the Radwani House (Context 22), including water jar spout
(bottom right), storage jar fragment (top right), waterpipe burner (top left) and
waterpipe neck (bottom left)
Porcelains and European "Refined White Wares" from the Radwani House (Context
167)
Typical coffee cup from the 1920s assemblage (Context 24)
Small bottles from the Radwani House. Left and middle: Context 160, Find 34; Right:
Context 47, Find 27
Miniature perfume bottles from the Radwani House (left: Context 21, Find 29; right:
Context 180, Find 41)
Fragmentary glass assemblage from Context 83 (a levelling fill in Room 12)
Large corroded iron nails (Context 102, Find 118)
Padlocks (Left: Context 242, Find 147; Right: Context 83, Find 113)
Copper ring (Context 19, Find 89)
Cartridge casing (Context 22, Find 90)
Copper drain cover (Context 87, Find 115)
Animal bone from the well (Context 300, Find 65)
Shell from the Radwani House. Top left: Context 160, Find 67; top right: Context 8,
Find 68; all below: Context 232, Find 69
Chinese saucers from Company House
Figure 6.1:
Figure 6.2:
Figure 6.3:
Figure 6.4:
Figure 6.5:
Figure 6.6:
Figure 6.7:
Site locations.
Location of ODQ3 Sections 1 and 2.
ODQ3 looking south with occupied house visible.
Remains of traditional building visible in elevation, looking northeast.
ODQ3, Section 1.
ODQ3, Section 1.
ODQ3, Section 2.
Figure 5.45:
Figure 5.46:
Figure 5.47:
Figure 5.48:
Figure 5.49:
Figure 5.50:
Figure 5.51:
Figure 5.52:
Figure 5.53:
Figure 5.54:
Figure 5.55:
Figure 5.56:
Figure 5.57:
Figure 5.58:
Figure 5.59:
Figure 5.60:
Figure 5.61:
Figure 5.62:
Figure 5.63:
7
Figure 6.8:
Figure 6.9:
Figure 6.10:
Figure 6.11:
Figure 6.12:
Figure 6.13:
Figure 6.14:
In-situ upturned ceramic vessel used as an oven.
ODQ3, Section 2.
Location of ODQ4, Section 1.
ODQ4, Section 1.
ODQ4, Section 1.
Limestone wall (005).
External surfaces and make up (002), and (003).
Figure 7.1:
Figure 7.2:
Figure 7.3:
Figure 7.4:
Figure 7.5:
Figure 7.6:
Figure 7.7:
Figure 7.8:
Figure 7.9:
Figure 7.10:
Figure 7.14:
Figure 7.15:
Figure 7.16:
Figure 7.17:
Figure 7.18:
Figure 7.19:
Figure 7.20:
Figure 7.21:
Figure 7.22:
Figure 7.23:
Figure 7.24:
Site overview, looking north.
Site location.
Detailed site location.
Aerial photograph of the site in 1966.
Aerial photograph of the site in 1971.
Topographic map of al Rayyan, 1971.
Agricultural area still in use directly to the east of the site.
Plan of the site.
Bo ell
ith st u tu e
ehi d. Looki g to the southeast.
Pump house (001) looking northwest. Note borehole pipe in front of
building.
Structure (006) looking northwest, with well (005) in background.
Tank inside structure (006), draining into well (005) to west.
Bore Hole (008) in background and concrete pump base in foreground,
looking west.
Large tank (007), looking northwest.
Small tank (015), looking east.
Peripheral irrigation channel (016), looking east.
Panoramic view of fields and irrigation channels (009).
Rectangular fields and irrigation channels (009), looking south.
Building (003), looking northwest.
Wooden timbers framing the door of building (003).
Building (010), looking west.
Boundary wall (014) at the northern extent of the site, looking north.
Enclosure (017) with channel visible in background, looking north.
Window in the eastern wall of enclosure (017), looking west.
Figure 8.1:
Figure 8.2:
Figure 8.3:
Map of Old Al-Ghanim, including Buildings Survey plots
Example of building material guidance for survey participants.
Stitched and digitally rectified elevation of building GH01.
Figure 7.11:
Figure 7.12:
Figure 7.13:
8
1. Summary
This report was made possible by NPRP grant no. 5-421-6-010 from the Qatar National Research
Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of
the authors.
The Origins of Doha Project is a University College London based Qatar National Research Fund
funded research project. This project aims to explore the foundations and historic growth of Doha,
its transformation to a modern city, and the lives and experiences of its people, through a
combination of archaeological investigation, historical research and oral testimony.
The project carried out its first season of fieldwork in Doha between November 2012 and February
2013. During this time four major archaeological recording projects were undertaken. Archaeological
excavations conducted within the Heritage Quarter of the Msheireb development, located in the
historic centre of Doha, identified early 20th century occupation in this area. Apart from yielding
detailed information on the lives of the inhabitants of Doha in the 1920s and 1930s, these
excavations also revealed the time at which Doha spread westwards towards Bida', a process which
culminated in the joining of the towns into a single urban conglomeration in the 1950s. This kind of
data is not available from the historical records, and underscores the importance of archaeological
investigations.
Non-intrusive recording of exposed archaeological deposits in the Salata district of the city identified
similar early occupation horizons. An archaeological survey of an agricultural area in Rayyan
consisted of mapping the site in conjunction with a detailed description and photographic record of
the features present. Finally a detailed survey was made of the standing buildings in the Old AlGhanim area of the city. The purpose of this survey was to record the historic structures present and
to identify how many historic structures survive in this part of the city.
The work undertaken by the Origins of Doha Project represents the first systematic study of the
origins and development of the city. Archaeological excavation conducted in the centre of Doha
identified buried architectural remains and a range of occupation deposits and artefacts.
Archaeological deposits exposed in section in the Salata district of Doha also indicated several
phases of occupation. These deposits all date to the late 19th or early 20th century and are likely to
represent some of the earliest occupation associated with Doha. A survey of the standing buildings
in the Old Al-Ghanim area of the city identified a number of buildings constructed in a traditional
style. These buildings probably date to the first half of the 20th century and are an important
ele e t of Doha s ultu al he itage. I additio , the odifi atio s a d additio s to these uildi gs
show how the city was transformed in the second half of the twentieth century, as Qata s
hydrocarbon deposits began to be exploited.
This report presents the results of the four major areas of fieldwork undertaken by the Origins of
Doha Project in its first season. These consist of archaeological excavation in the Msheireb area
(Chapter 4), recording exposed archaeological deposits in Salata (Chapter 5), surveying an
abandoned agricultural area in Rayyan (Chapter 6) and recording standing buildings in Old AlGhanim (Chapter 7). A discussion of the results and recommendations for future fieldwork is
presented in Chapter 8.
9
2. Acknowledgements
This report was made possible by NPRP grant no. 5-421-6-010 from the Qatar National Research
Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of
the authors.
The authors would like to thank H.E. Sheikha Al-Mayassa bint Hamad Al-Thani and H.E. Sheikh
Hassan bin Mohammed Al-Thani of QMA. Thanks are also due to Mr Faisal Al-Naimi of the
Department of Antiquities, Professor Sultan Al-Muhesen, Professor Thomas Leisten and Alicia
Bianchi, from QMA. From Msheireb Properties and their partners we thank Ms Fatima Fawzy
Hassan, Mr Mohammed Ali Abdullah, Mr John McAslan, Mr Fanos Panayides, Mr Mohsin Allarakhia,
Ms Nur Alah Valdeolmillos, Mr Timothy Makower and Dr Scott Cooper. The authors would also like
to thank the Arabian Construction Company for all their support during our excavations in the
Msheireb Project. Finally we would like to thank the UCL Qatar students who worked with us in the
field, and all the workmen who laboured so hard alongside us.
3. Organisations, Project and Personnel
UCL Qatar is a branch campus of University College London and partner of Hamad bin Khalifa
University, a member of Qatar Foundation. Its remit is to research and teach cultural heritage and
archaeology in the Arab world and the Islamic world.
The Origins of Doha Project is a QNRF-funded research project that aims to explore the foundation
and historic growth of Doha, its transformation to a modern city, and the lives and experiences of its
people, through a combination of archaeological investigation, historical research and oral
testimony. The project will apply detailed analytical techniques to the archaeological record of Doha
and Bida', according to current best practice in archaeological research. Details about the project
and current research are available from: http://originsofdoha.wordpress.com/
Dr. Robert Carter (Director) is a Senior Lecturer at UCL Qatar and is a leading expert in the historical
archaeology of the Gulf, the archaeology of the Gulf towns and historic pearl fishing. A CV can be
provided on request, or downloaded from:
http://ucl.academia.edu/RobertCarter/CurriculumVitae
Daniel Eddisford (Field Director) has been a professional field archaeologist for the last 15 years,
working extensively in the Gulf, Anatolia and Britain. He worked at the site of Al Zubarah with a team
from Copenhagen University since 2009, undertaking excavation and archaeological survey in the
north of Qatar. A CV can be provided on request, or downloaded from:
http://independent.academia.edu/DanielEddisford/CurriculumVitae
Michal Michalski has worked as professional archaeologist in Poland, the United Kingdom and the
UAE and is currently responsible for surveying and GIS for The Origins of Doha Project. Michal
received an MA in Archaeology from the Nicolaus Copernicus University Poland in 2006 and an MSc
in Geographic Information Science from the University of Edinburgh in 2012.
All fieldwork was conducted by a team of highly experienced professional archaeologists, consisting
of Katie Campbell, Kirk Roberts, Michal Michalski and Colleen Morgan. Details of all project staff are
available from: http://originsofdoha.wordpress.com/project-staff/
10
4. Historical Background to Doha
Prior to the Origins of Doha Project s i estigatio s, there have been no other archaeological
excavations in the centre of Doha, and as a result little is known about the earliest occupation of the
capital. Despite the a dearth of archaeological investigation, there are a number of documentary
sources that provide some background information on the development of Doha.
The first known reference to settlement in the bay of Doha comes from Carsten Nie uh s ap of
the region recorded in 1765. Although lacking detail and containing obvious inaccuracies, Niebuhr
e o ds the to s of Huale Hu aila , Yusofie Yusufi a , Fa aha F eiha , Adsja probably
Al-Uqair, Saudi Arabia, misplaced on the map , as ell as Gatta
p o a l Bida' or another
unknown settlement in the bay where modern Doha is located).
Later sources state that Al-Bida' (Bidda) was founded before Doha, by members of the Sudan (AlSuwaidi) tribe who had migrated from Abu Dhabi (thus later than 1762, the foundation date of Abu
Dhabi). The two settlements co-existed throughout most of the 19th century, and although locals
made a clear distinction between the two towns, outsiders generally referred to the two together as
Bida' until the last quarter of the 19th century, when Doha tended to be used. This reflects the
shifting location of the seat of power and prosperity during the second half of the century. The first
mention of Bida' is from Captain Seton's diaries, which state that he attempted to attack it with the
Imam of Muscat in 1802. He noted fortified buildings at "Bedih" either side of a valley, possibly
indicating that both Bida' and Doha were already present, either side of the Wadi Msheireb.
One of the first descriptions of Bidda was made by David Seton, the British representative in Oman
from 1800 - 1809:
Bedih is Situated in 25.18 N. Lat. and is a large open bay full of Coral banks with
very unequal soundings from twelve to three fathoms, the land is low and sandy,
hardly to be seen at the distance of ten Miles, on a nearer approach, it seems
gradually rising from both extremes towards the centre, where it forms a ridge at
the distance of half a Mile from the Shore, under this ridge near the sea, are two
hillocks and a Valley between them, off each of the hillocks runs a Shoal with half a
fathom at high water and between them a Channel with one and a half fathoms,
and at the distance of a Mile and a half three fathoms, on the Northern hillock is a
fortified House with a Wall and Square tower, in the Valley a breast Work with two
Guns, and on the southern hillock two large huts with some kind of defence, and
half a Mile to the Southward near the ridge is another Square building with a flag
staff, under the Northern hillock is a sandy beach on which two Buglas, one Dow &
one Botella were drawn up with a breast work of Stones, the only direct landing
place is in the mouth of the Valley, but it would be attended with great loss
without ships to drive the Enemy away as it is flanked by the breast work and
boats, in which were a number of Men and ten Guns, and fronted by the two Guns
in the Valley, about two Miles to the South is a Sandy beach without cover for the
Enemy's snipers, but the Square building with the flagstaff must be stormed before
the hillocks can be got at. (Seton 1995: 27)
In 1820 Majo Cole ook s report on the Persian Gulf littoral describes the settlement of Bida' in
more detail:
Ul Budee [al Bida'], once a considerable town, is protected by two square
Ghurries near the sea shore; but containing no fresh water they are incapable of
11
defence except against sudden incursions of Bedouins, another Ghurry is situated
two miles inland and has fresh water with it. This could contain two hundred men.
There are remaining at Ul Budee about 250 men, but the original inhabitants, who
may be expected to return from Bahrein, will augment them to 900 or 1,000 men,
and if the Doasir tribe, who frequent the place as divers, again settle in it, from
600 to 800 men ‘ah a
, -4).
Colebrook observes that the population of al Bida' had recently decreased, a significant proportion
of the population having left for Bahrain following recent insecurities. Either way the movement of
relative large tribal groups between settlements in Qatar and the wider Gulf region is a recurring
theme in this period.
Subsequent to Nie u h s ap, the fi st s ste ati su e of the A a ia Gulf as a ied out
et ee
a d
the Offi es of the East I dia Co pa s Bo a Ma i e. The task as
completed in 1825 and published in 1829. Captain George Barnes Brucks describes the settlement at
Bida' at this time.
Al Bidder To is situated at the otto of a har our, for ed y the reefs before
mentioned... This place contains four hundred Arabs of the Nahune, Dooasir, and
Abookara Tribes, and is frequented by the Monasir and other wandering tribes. In
the pearl season the inhabitants are augmented to about twelve hundred, it being
convenient to the banks, and so completely sheltered. The people are mostly
fishermen; they have one or two trading boats, and, like all the other inhabitants
of this coast, take a large share of the pearl fishery. They are subject to Bahrain.
Cattle and poultry are procurable here, but they are very dear. Water, pretty good,
can also be procured. The defences are a small Ghuree in the town, two or three
towers, indifferent, and a Ghuree a mile to the north-westward of the town, on the
risi g grou d (Brucks 1829, 559).
Brucks provided a map which shows two small separate but adjacent settlements, likely to indicate
Bida' and Doha.
Instability prevailed in Qatar throughout the mid 19th century and control of Bida' and Doha changed
hands repeatedly, with several instances of destruction taking place. In 1848 the al-Thani family left
Fuwairit for al-Bida', beginning the process of unification of Qatar under the leadership of Sheikh
Mohammed bin Thani, continued by his son Jassim bin Mohammed (Rahman 2005, 41-42). Soon
after they moved a couple of kilometers to the east to settle in Doha, marking the beginning of the
eclipse of Bida'. In 1867 the Bida' and Doha were destroyed for the final time by forces from Bahrain
and Abu Dhabi. The two were once again rebuilt but remained separate towns until the mid 20th
century, referred to collectively by western sources as either Doha or Bida' ("Bidder") and
sometimes as Qatar ("Gutter"). Burchardt, in 1904, also refers to Salata ("Al-Solata").
In the 1870s Sheikh Jassim bin Mohammed Al Thani placed himself under Ottoman protection in
order to counter the combined pressures from the Bahraini tribes, the British and the Al “a ud.
Jassim demonstrated his allegiance by flying the Ottoman flag over his residence in Doha, but was
also prepared to fight then Ottomans when he felt threatened by their presence, defeating them in
battle at Wajba in 1893 (Anscombe 1997, 32; Rahman 2005: 107). An Ottoman presence in Qatar
nonetheless continued until the Turkish garrison was evacuated in 1915. The Ottomans based
themselves in a large fort in the area of today's Emiri Diwan, strategically located on high ground in
the still-open space between the towns of Bida' and Doha.
12
Continued documentation of the occupation and the architecture of the region comes from
travelogues of travellers visiting Doha in the 19th and 20th centuries. Visiting Qatar in 1904, a German
traveller Hermann Burchardt stayed as a guest of Sheikh Ahmed bin Mohammed in Doha. He wrote:
The “heikh, a portl ge tle a of dig ified ou te a e, appeared ith a large
following... After about two hours the magilis ended and a good meal brought;
then I was assigned the hut as a dwelling. I remained here as the guest of the
sheikh for si da s a d as e elle tl fed.
To supple e t his a ou t of sta i g i a hut, du i g Bu ha dt s isit he took a u e of photos of
Doha, including an image with the Turkish fort in the foreground with Doha in the background(Figure
4.1; Nippa et al 2006, 132). This image indicates that Msheireb, the area in the right of the photo,
has not yet been developed with standing architecture. The town of Doha, surrounded by a city wall,
is clearly visible in the background of the photo.
Figure 4.1: The Turkish fort with Doha in the background, 1904.
From the time of Doha's earliest occupation, the water present in the Msheireb area was an
important source of water for the settlement. The wells at Msheireb were significant enough to be
mentioned by J.G. Lorimer during his summary of Doha in 1908, and were clearly of strategic
importance; Lorimer indicates that the Turkish troops garrisoned in Doha guarded them from a
watch tower.
Dohah itself possesses o l o e ell of ra kish ater, a ed Ai Walad “a'id,
which is a half mile to the south of the Dohah quarter, but there is a group of
others called Mushairib with fairly good water, at 1 mile to the west of the Dohah
quarter. Three miles inland is Bir-al-Jadidah, a large masonry well of indifferent
water on which the town mainly depends for its supply... The Turkish troops obtain
most of their water from Mushairib, where there is a military outpost of 8 men in a
13
tower to watch the wells. The soldiers now have a vegetable garden at this place;
a d s ur , hi h as for erl o
o a o g the garriso , has disappeared
(Lorimer 1986).
In his recollections of Doha in the first half of the 20th century Nasser Othman reinforces the
importance of the Msheireb area as a water source for the city. He also mentions a natural pool that
collected rainfall, and presumably helped recharge the aquifer in this area.
Most houses had a ess to a water well providing brackish water for domestic
purposes, such as cleaning and bathing, but the water from such wells was not fit
for drinking. Sweet water had to be taken from springs at Muraikh, Nuejjah and
Musheirib, where the precious few days of winter rainfall collected in a pool known
as al-Sail. The most frequented public well was the Youssef Ahmad well near the
open-air Eid prayer-house in Al-Jassrah.
Water was carried from the spring to houses by donkey, camel or by a watercarrier [kandari]. The kandari carried two pails balanced on a wooden yoke across
his shoulders and was a familiar sight in those days. Once in the house, water was
stored in large earthenware pots. Drinking water was stored in special small
pitchers [zir] which kept the water cool and fresh even in the most extreme heat.
The taste of the ater as i pro ed
addi g a little asti to the pit her
(Othman 1984).
It appears that Doha continues to expand in the first few decades of the 20 th century. It is likely that
extension into the Msheireb area, and thus the joining of the two towns of Bida' and Doha into a
single city, began soon after the Ottomans left in 1915. The Al Thani ruler moved his seat of power
to the area of their fort and established the Emiri Diwan. Soon after this time, according to the
archaeological research presented below, settlement began in the previously-open land between
Bida' and Doha.
In 1920 a United States Hydrographical survey of the region gives a description of the now
conjoining settlements of Doha and al Bida', with "Little Doha" between them:
Al Bida or Gutteh is the a e of the o i ed three to s … Doha, Al Bida, a d
Little Doha. Doha, the eastern town, ½ mile south-westward of Ras Nessa, is
partiall alled, ith se eral to ers. The “heikh’s house is at a large rou d to er,
with a flagstaff, on the beach about the middle of town; westward of this tower is
a small bight, where boats are hauled up for repair. The reef dries ¼ mile off.
Little Doha, north-westward of and joining Doha, has a square fort on the rising
ground at its south-western corner. Al Bida joins Little Doha, and the three places
extend 1 mile along the shore. Al Bida is situated on the side of the rising ground,
a d the “heikh’s flag is flo
fro the astle. Al Bida fort, hi h is situated o
risi g grou d, has a large to er a d is o spi uous (US Hydrographical Office
1920, 115-116).
"Little Doha" refers to Jasra and new adjacent quarters, which were spreading westwards at that
time towards Msheireb and the former Ottoman fort (probably the "square fort on the rising
ground"). "Al Bida fort" probably refers to the 19th century fortification of Bida', still evident in aerial
imagery of the 1950s but since demolished, though sometimes the former Ottoman fort may have
been referred to by that name.
14
The Hunting Survey Of Doha was undertaken in 1957, and represents the first detailed survey of the
city. By the mid 20th century the Msheireb area has been developed, with a number of buildings
including the Radwani House clearly visible (Figure 4.2). The building layout recorded in the Hunting
Survey remains largely unchanged until the extensive redevelopment of Suq Waqif and the
Msheireb district began in c.2006.
Figure 4.2:Part of the Hunting Survey of 1952 showing the Radwani House.
However, the decades preceding the Hunting Survey had been marked by economic hardship in the
region. During the early 20th ce tu , u h of Qata s e o o
still depe ded o fishi g a d
pearling. At about 1900 Doha had a population of around 12,000 and around 350 pearling boats.
The introduction of Japanese cultured pearls and the global depression of the 1930s had a
disastrous effect on the whole region, and the town of Doha suffered a major downturn. There were
shortages of food, while inflation and the crippling debt plaguing most of the population
exacerbated the situation. The earliest aerial images of Doha in the late 1940s show large areas of
the city apparently abandoned and in ruins. This is reinforced by a description of the city in 1949 by
Sir John Arthur Wilton, the first British Political Agent in Qatar.
Qatar i 1949 as ot a parti ularl agreea le pla e to e. Decades of poverty
and, during the War, near famine had led to mass emigration and endemic
diseases of dirt and malnutrition. The appearance of the capital, Doha, suggested
the aftermath of an air raid as unoccupied and even occupied houses crumbled
into decay. In the countryside, at its best a harsh and barren moonscape of a land,
the desert advanced pitilessly against the few patches of cultivation. Walled
gardens and date groves bore signs of drought and neglect as the laborious
business of irrigation from deep wells by donkey-po er as ot sustai ed.
15
Not all districts of Doha were in decline in the 1920s and 1930s, as demonstrated by the expanding
westwards occupation of Jasra towards Msheireb, and it seems that early revenues from oil
exploration concessions were able to sustain the economy to a certain extent. Oil was discovered in
the region in the 1930s, and was to be the saviour of Doha. Judging from the date of the finds
associated with the earliest layers and buildings beneath the Radwani House (see below), its
surroundings were first occupied in the 1920s or 1930s. This extension of the town may have been
stimulated by oil revenues, as well as the moving of the Amiri residence to the area following the
departure of the Turkish garrison. However, the exploration of oil resources and the extraction
process took time and was halted for a number of years due to the Second World War. It was not
until 1949 that oil started to flow in commercial amounts, and in December of the same year the
first ship left the shores of Qatar bearing a consignment of crude oil. Today the nation as a whole
produced o e
,
a els of oil dail , aki g it o e of the o ld s ajo oil p odu e s. Not
everyone shared the the gloomy impression of Doha given by Wilton at this time, and it is clear that
the town had already started to recover. Charles Spencer Denham, later Lord Denham, was then a
British merchant who arrived in 1949 and shipped in building materials to construct the first modern
buildings in the town:
I don't remember it as falling down, I remember it as a typical mud building Arab
town. There we no fine buildings or anything like that, it was like a rabbit warren
with the little narrow streets going through. Our job was to bring in all the building
materials for developing the place once they had some oil money. They were
building the hospital, and we were largely responsible for that" (interview by
Frances Gillespie, 12 February 2010).
Significant social changes occurred as oil wealth began to filter down through the population of
Qatar. Wealthier Qatari families began to leave the historic core of the city, preferring to live in
newly built, modern villas rather than antiquated 19th to mid 20th century traditional limestone
houses. An initial building boom in the 1950s and 1960s saw a rapid expansion of Doha through the
widespread construction of largely traditional buildings, along with new concrete architecture. Piped
water, electricity, air conditioning and a host of other modern conveniences were added to
traditional buldings old and new, which must have drastically altered the day to day experience of
city living. Excavations at the Radwani House, built some time between the mid 1920s and the mid
1930s, shows that bit underwent a number of ad-hoc alterations along these lines in the mid to late
20th century. In addition ever increasing oil and gas revenues combined with rapid development and
expansion in Doha resulted in the arrival of large numbers of migrant workers from Asia, Africa,
Western Europe, and beyond. As time went by and Qatari families moved out of the old buildings,
this migrant labour force occupied many of the now less desirable traditional buildings in the centre
of Doha. The new inhabitants often enacted a range of alterations to the buildings, transforming
these building to suit their own needs. The inhabitants sub-divided courtyards in order to create a
number of smaller dwellings, they added second stories constructed of modern concrete blocks to
the main structure, added walls and rooms and cut openings for air conditioning units into the walls.
16
5. Archaeological Investigations in the Heritage Quarter of the Msheireb
Development
Daniel Eddisford
With contributions from Katie Campbell and Kirk Roberts
5.1 Introduction
In December 2012 the Origins of Doha team was asked by Msheireb Properties to undertake an
assessment of archaeological deposits encountered during ground works in the Heritage Quarter of
the Msheireb development. Archaeological excavation then ensued as part of the Origins project's
remit for fieldwork in Doha. The Msheireb redevelopment project covers an area of 31 hectares in
the historic core of Doha. The development is located directly to the west of Souq Waqif, and is
bounded by Al Rayyan Road to the north, Al Diwan Street to the west and Wadi Msheireb Street to
the south (Figure 5.1). The mixed-use development will comprise more than 100 buildings, with a
combination of commercial and residential properties, retail, cultural and entertainment areas.
Located in the northeast area of the development, the Heritage Quarter consists of four buildings;
Jalmoud House, Company House, Mohamed Bin Jassim House and al Radwani House (Figures 5.1
and 5.2). Extensively rebuilt by the Private Engineering Office in the mid 2000s, these historic
structures are currently being turned into cultural centres, four new museums and exhibition
buildings.
Figure 5.1: Location of the Msheireb redevelopment project (Google Earth image 2006).
Following the initial assessment archaeological excavation was undertaken between the 9th and the
19th of December 2012 inside four rooms of the Radwani House that had be identified as having high
archaeological potential. In addition several archaeological features were recorded in Company
House. Between the 3rd and the 21st of February 2013 further excavations took place, focusing on
the colonnade and courtyard of the Radwani House.
17
Figure 5.2: Location of Radwani House and Company House (Google Earth image 2006).
Prior to its extensive reconstruction by the Private Engineering Office in 2006 the building looked
significantly different than it does today (Figure 5.3). The reconstruction work undertaken involved
the demolition of most of the historic fabric of the Radwani House, and the extensive rebuilding of
the structure on a similar same floor plan. Original architectural motifs were recreated using a
mixture of traditional and modern building techniques.
Figure 5.3 : Radwani House prior to reconstruction in 2006
The excavations revealed an archaeological sequence comprising of an initial occupation of the site
that predates any significant architectural remains. This was followed by several phases of buried
walls as well as associated floors and occupation deposits. In its earliest phases—probably dating to
the early 20th, century—the buildings on the site appear to consist of two houses, separated by an
18
alleyway. These were later combined into a single complex with the closure of the alley. The
combined structures were then remodelled to make a single courtyard building. Significant features
associated with this building include a unique bath and well complex in one of the rooms. This later
phase of building was extensively reconstructed in 2006 to create the current Radwani House.
The archaeological deposits excavated were rich in artefacts, including pottery and glass. Future
analyses of other finds such as animal bones and plant remains will allow a clearer picture of
traditional life in Doha to be established. Traces of a much older occupation, probably of the
Sasanian Period, were identified in the ceramic assemblage, suggesting much earlier occupations in
the Doha area.
This archive report provides a technical summary of the results of these archaeological
investigations. The stratigraphy of each excavation area is described in detail, followed by a phased
summary of the development of the site. Analysis of the artefacts and soil samples from the
excavation is ongoing, but where preliminary data is available it has been integrated into the report.
Excavations at several contemporary archaeological sites in Qatar are briefly discussed, along with
similar urban sites elsewhere in the Arabian Gulf. These comparative sites are used to give context
to the excavations in order to place the work in a wider context and highlight the potential
importance of this and continued archaeological work in Doha.
19
5.2 Aims of the Excavation
The initial aim of the excavation was to assess the potential for buried archaeological deposits in the
Radwani House. To this end, two documents were prepared, detailing the nature and extent of
visible archaeological remains on the site, and assessing the anticipated impact of the current
development these remains (Carter and Eddisford 2012a; 2012b).
Following the initial assessment four areas of high archaeological potential were identified in Room
8, Room 9, Room 10 and Room 12. The archaeological deposits in these four areas were threatened
with imminent destruction. In order to mitigate the impact of the development of these areas the
first phase of excavation aimed to fully excavate archaeological deposits in order to ensure their
preservation by record.
A second phase of excavation focussed on the exteriors of the rooms in the colonnade and courtyard
in order to assess the character, date, location and level of preservation of any archaeological
remains in this area. This phase of work aimed to provide further information on the development of
the Radwani House and allow a better informed decision to be made on how best to mitigate for
further intrusive groundworks in the Radwani House.
In addition an intermitte t at hi g ief as ai tai ed i Co pa House th oughout the tea s
presence on site. No archaeological excavation was undertaken, but several archaeological features,
including two wells, were recorded.
Beyond the need to mitigate the impact of the current development of the Radwani and Company
Houses on the buried archaeological remains, the data from the excavation will be incorporated into
the ongoing research of the Origins of Doha Project. These data will be used to better understand
the foundation and historic growth of Doha, and its transformation to a modern city.
5.3 Methods
5.3.1 Trench Locations
The location of all the excavation areas in respect to the standing architecture of the Radwani House
is presented below in Figure 5.4. The Site of the Radwani House is centred on Qatar National Grid
231598 382715.
Room 8, Room 9 and Room 10 were ranged along the western side of the courtyard of the Radwani
House. The archaeological deposits in Room 9 were fully excavated. The remains in Room 8 were
excavated to the earliest phase of architecture, which was left in situ. In Room 10 the architectural
remains, consisting of a well and a bath, were recorded and a small trench was excavated through
early deposits in the southeast corner of the room. The archaeological deposits at the eastern end of
Room 12 were fully excavated, the rest of the deposits having been truncated by the current
restoration of the house.
Trench 1 was L-shaped in plan and located in the southwest corner of the colonnade. The trench was
located in an attempt to locate a continuation of an east-west orientated wall identified by
excavations in Room 8.
Trench 2 measured 3.10m by 1.50m and was located on the western side of the courtyard. It was
located to assess the survival of deposits in the courtyard area of the house, and relate them to the
deposits identified in Trench 1.
20
Trench 3 was located in a modern circular planter in the centre of the courtyard. It was excavated to
establish whether there was an early feature in the centre of the courtyard, such as a well or tree
planter.
Trench 4 measured 3.00m by 1.80m and was located on the eastern side of the colonnade. It was
placed here to assess the archaeological survival in this area of the site, specifically to see whether
any architectural features survived on this side of the site.
Figure 5.4: Location of excavation areas in relation to existing walls of the Radwani House
5.3.2 Excavation Methods
Where necessary the concrete sealing the excavation trenches was cut and broken out with hand
tools, with the assistance of the Arabian Construction Company.
21
All archaeological excavation and recording was carried out in accordance with the Museum of
London Archaeology Service archaeological site manual (MOLAS 1994) and where relevant the Code
of Approved Practice for fieldwork as set out by the Institute for Archaeologists (IFA revised 2008).
All identified finds, artefacts, and faunal remains were collected and retained.
Environmental sampling during the evaluation targeted a representative range of contexts from
each phase and focused contexts with high potential for botanical survival.
The site code ODQ22 has been allocated to all work in the Heritage Quarter of the Msheireb
development. This code will be used to label all recording sheets, plans and finds. The work
undertaken has been added to the Qatar National Heritage Register, under the Event Number 119.
All archaeological finds from the excavations remain the property of the Qatari state. As agreed at
the beginning of the project some of the artefacts and samples recovered are temporarily retained
for recording and analysis by UCL Qatar. All archaeological finds and materials currently held within
UCL Qatar for this purpose, will be being returned to the appropriate authority after the analysis is
complete.
Figure 5.5: The courtyard of the Radwani House looking southwest, prior to excavation of Trenches 1 and 2.
22
5.4 Phased Summary
Several major architectural phases, relating to the development of the Radwani House, were
identified during the archaeological excavations. A summary of these phases is described below. A
more detailed description of the deposits associated with each phase is given in the following
section (Section 6). The location of the excavation areas is detailed above in Figure 5.4.
5.4.1 Phase 1 – Pre-architecture
The earliest occupation of the site pre-dates any architectural remains, and is represented by several
postholes, spreads of ashy charcoal rich material, and small pits. These deposits are likely to
represent 19th and early 20th century activity on the outskirts of al Bida' and the new settlement of
Doha. The deposits associated with Phase 1 activity were relatively extensive across the site.
However, this early activity on the site was only exposed in small areas due to a combination of
extensive truncation by later features and limited exposure in the lowest levels of our relatively
small excavation trenches.
Although no stone architecture is present in this first phase of occupation three postholes [101],
[103] and [105] may represent posts associated with an early wooden building or barasti hut. These
postholes could also be associated with scaffolding for the construction of the Phase 2 architecture.
In Room 6 and in Trenches 1 and 3 trampled floor deposits (194), (171), (238) and (116) / (106) are
likely to represent relatively intensive, although possibly seasonal, activity on the site. Several pits
cut into the natural limestone. Pit [109] in Room 8 was relatively large, measuring 1.30m in diameter
and 0.60m deep. In Room 9 a series of intercutting pits contained ash rich fills. The function of these
features is not entirely clear, however some may be small quarry pits used to excavate limestone
and gypsum, then filled in with the ash of raked-out fires.
Across the site early occupation deposits appear to attest to a range of activities including
occupation and dumping of domestic waste. These charcoal rich deposits often survived only in
shallow depressions in the natural limestone. However at the southern end of the site these deposits
were more extensive and possibly better preserved as the natural limestone dips down towards the
southern end of the site.
23
Figure 5.6: Phase 1 features and deposits.
5.4.2 Phase 2 – Early architecture
The earliest building on the site was a heavily robbed out structure on the western side of the site. A
single east-west aligned wall (46) / (47) in Room 9 was rendered with a hard light pinkish grey plaster
(96). The full extent of this building was not clear, as this single wall appeared to be the only
surviving element of this structure. It may well have been a relatively small structure, without an
associated enclosed compound wall. This earliest building was associated with a thin, lime-flecked,
compact, sandy clay surface (55) / (54) and overlaying loose ashy occupation deposit (45).
The eastern extent of the building may have been truncated by a raised structure, (219) / (159), built
over the earlier building. The building was accessed via the narrow plastered stairway, (159). The
structure had been vertically truncated by later construction, and as a result any associated floor
surfaces have been destroyed. Initially building (219) / (159) was not enclosed by a courtyard wall.
However, possibly as a result of increasingly dense surrounding occupation, a courtyard wall, (172) /
24
(64), was built. This created an enclosed compound, presumably with raised building (219) / (159) in
the southwest corner. Phase 2 wall (69), located directly beneath the northern wall of the current
Radwani House, appears to represent the northern extent of this courtyard. A slightly later addition
to this courtyard wall (97), which abutted the northern compound wall, was associated with the
subdivision of the courtyard in order to create at least one more room on the northern edge of the
courtyard.
At the same time as the northern building is enclosed with a courtyard wall a second compound, (15)
/ (178) / (201), is constructed directly to the south. A small alleyway created between the two
compounds may have provided access to both dwellings. North-south aligned wall (235) was heavily
robbed out, but was associated with wall (178), forming an internal division within the southern
compound. Wall (235) probably represents the eastern wall of an internal space, the western extent
being defined by wall (15) in Room 8. To the east of wall (235) a large open area would probably
have been an external courtyard. A sequence of floors and levelling layers in this space represent an
extended period of occupation associated with the southern compound (178) / (235).
Both the northern and southern Phase 2 compounds have wells associated with them, highlighting
the availability of water in the Msheireb area. Abutting the western side of the southern compound
a circular stone lined well, [126], was set in a raised plinth. Although apparently associated with the
southern compound this well appears to be located in an external street area, directly to the west of
the compound. In the northern compound a circular cement or gypsum lined well, (143), was
surrounded by a similar sub-rectangular plinth.
On the eastern side of the site an east-west orientated wall (192) may be associated with the
southern compound, or may represent a separate structure located to the east of both compounds.
25
Figure 5.7: Phase 2 architecture.
5.4.3 Phase 3 – Joining of two compounds
The two separate compounds that occupied the site in the previous phase were joined together in
Phase 3, and the alleyway between them blocked off. Several drainage features and soakaways were
also added to the buildings in this phase of use.
A concrete or gypsum threshold (113) was constructed at the western end of the alleyway, closing
access from the street to the west. Directly to the east of this feature a concrete or gypsum surface
(87) drained through a copper alloy drain cover soakaway [118]. This was one of several stone-lined
oval soakaways constructed in this phase.
This joining of the houses may indicate that a new owner had bought both houses, or the owner of
one had bought the other, and decided to join them into a single structure. This somewhat ad hoc
26
fusion of the buildings was soon superseded by the construction of a single unified building (Phase 4,
below).
Figure 5.8: Phase 3 architecture.
5.4.4 Phase 4 – Later architecture
Phase 4 is associated with an extensive rebuilding of the Radwani House. The two separate
compounds associated with the previous architectural phase of use were combined into a single
unified building. The southern building was demolished in that process, but the walls of the northern
building were incorporated into the new built, or perhaps just their foundations. The alleyway
between them was incorporated into the domestic space. This major rebuilding of the Radwani
House is represented in a number of the excavation areas by the presence of loose rocky dump
layers that seal the earlier architectural phase.
The most striking feature associated with Phase 4 occupation were two concrete baths, (140), This
impressive and unique structure consisted of two curved basins divided by a narrow partition. Both
ends of the bath had a rounded niche and the western bath was accessed two low steps. The baths
27
were constructed of an extremely hard, light grey, plaster render applied directly over roughly hewn
limestone. A significant quantity of large moulded fragments associated with the baths were
recovered from the infill of the feature, indicating that the bath could originally have stood 0.50m
higher. The baths extend underneath the internal walls of the current Radwani House at their
eastern extent. This wall represents part of the 2006 reconstruction of the Radwani House, and the
original room that the baths occupied was larger than the current space.
The layout of the rebuilt Phase 4 Radwani House would probably have been similar to the present
day structure. A number of earlier limestone walls visible were below the walls of the current
structure, often used as foundations for the walls visible today. East-west wall, (25), was built of
unfaced limestone in a linear construction cut. This represents an internal division of the Phase 4
building, that was not replicated in the later Phase 5 restoration work. A heavily truncated northsouth aligned wall, (226), probably represents the eastern extend of a raised colonnade associated
with the Phase 4 house. This colonnade was entirely rebuilt by the Private Engineering Office in
Phase 5 using a ring beam of steel reinforced modern concrete.
A compact surface (155) was located to the east of wall (226) and represents an associated Phase 4
occupation level. A sequence of compact cement or gypsum floors and associated bedding layers
abutted the southern side of the baths (140). An earlier well, (143), located directly to the south of
the baths appears to have been turned into a soakaway. An earlier soakaway [53] was unfilled and
sealed by a compact light grey cement floor. Sub-rectangular stone lined soakaway [169] also well
out of use in Phase 4. The introduction of modern drainage, presumably replacing the earlier wells
and soakaways, was visible in the form of metal drainage pipe (13), located at the eastern edge of
the Room 9.
Other water management additions can been seen in a drainage channel which was constructed in
linear a cut, truncating the courtyard surface at the southern end of Trench 1. A cement of gypsum
drain (184) would have carried waste water or sewerage. The drain was capped with limestone slabs
(181) and covered with sandy silt and limestone rubble (179).
28
Figure 5.9: Phase 4 architecture and features.
5.4.5 Phase 5 – Reconstruction by Private Engineering Office
In 2006 the Radwani House underwent extensive reconstruction work, under the direction of the
Private Engineering Office. This reconstruction involved the extensive demolition of the historic
fabric of the building, followed by rebuilding many of the walls on the same ground plan. Earlier
walls were used as foundations, with concrete foundations added when necessary. The materials
used in the reconstruction were similar to the traditional building fabric, but the limestone is roughly
dressed into blocks as opposed to undressed blocks in original house. The reconstructed House
represents the Radwani House seen today. A detailed survey of this structure was undertaken prior
to the present building works.
29
5.5 Technical Report: Excavations in the Radwani House
Details of all archaeological deposits encountered during the excavations are presented below by
excavation area. All areas are described stratigraphically, stating with the earliest deposits.
5.5.1 Room 6
Recent construction work in Room 6 had truncated the archaeological deposits to 7.40mOD. In
addition the foundations of the 2006 rebuild of the Radwani House had also truncated the
archaeology, particularly in the northeast corner of the room. However, the southern extent of the
current phase of Radwani House is built directly on older walls; these earlier walls survive to a height
of c. 1.2m in the southern wall of Room 6 (Figure 5.11). In addition the lowest courses of several
internal walls survived, along with several associated surfaces and a stone lined soakaway.
Phase 1
The natural limestone in Room 6 was sealed by a dirty, compact, trampled surface (194). This was
overlain by a second packed earth floor (171). These represent activity predating the construction of
the first architecture on the site (Figure 5.10).
Phase 2
Constructed directly on top of the latest surface was an east-west aligned traditional mud packed
limestone wall (201). This wall was directly below, providing a foundation for the southern wall of
the current Radwani House. Abutting this wall, although presumably constructed at the same time,
was a similar north-south aligned limestone wall (170). These walls were covered with render (239)
and (241) respectively (Figure 5.11). The earliest phase of rendering consisted of a mud render,
which was later covered with hard gypsum or cement-based pale grey render. Wall (201) represents
the external wall of a house or compound, and wall (170) represents an internal division within the
building.
A mixed dump or levelling layer of loose dark grey ashy material (167) was rich in artefacts and
appeared to be re-deposited midden-like material. This deposit represents internal levelling within
the earliest building (201) / (170) in this area of the site. On top of this deposit another traditional
mud packed limestone wall (220) was covered with a hard gypsum or cement-based pale grey
render (240). The wall was heavily truncated by recent construction work, surviving only as a small
stub in the northern side of the room. The full original extent of the wall was therefore unclear. This
wall represents an addition or alteration to the original building, possibly subdividing a room or
courtyard space.
30
Figure 5.9: Surfaces (174) and (191) overlain by wall (170). Cut for later soakaway visible at far right.
Figure 5.10:Limestone wall (201) and render (239),The metal sewer pipe is a later addition to the building.
Phase 4
A soakaway was constructed within a sub-rectangular cut [169], lined with stones (182) and
surrounded with a concrete or gypsum apron (175) (Figure 5.10). The primary fill of the soakaway
31
consisted of an organic, rich deposit (176). The top of the soakaway had been truncated by recent
construction work, and its stratigraphic relationship to the other deposits in Room 6 destroyed. It is
included in Phase 4 as it is similar in appearance and construction to soakaways found in Room 8 and
Room 9. However it is possible it is of a slightly earlier date, and may be associated with wall (220).
Phase 5
The upper fills of soakaway [169] consisted of loose limestone rubble (168) and (174) which
contained modern artefacts. This infilling appears to be associated with a relatively recent
modification to the Radwani House. Linear construction cuts [203] contained a crudely cemented
limestone foundation (202). These trenches built for foundations support the existing walls of the
Radwani House, defining the eastern, northern and western extent of Room 6. These walls contain
ceramic blocks and appear to be of late 20th century date.
5.5.2 Room 8
A substantial wall (15) in Room 8 did not respect the current ground plan of the building, and clearly
represents an earlier phase of the building. A series of postholes directly to the west of the wall may
represent scaffolding associated with the construction of the building, or a wooden element to the
building which does not survive archaeologically. In the southeast corner of Room 8 a circular stone
lined well was uncovered, this feature would have been external to the building in this phase of use,
probably in a small street or alley directly to the west of the building.
The building exposed below Room 8 would have been accessed through a small alleyway, located at
the northern end of wall (15). Later in its life this alleyway appears to have been blocked off. The
two earlier buildings were joined together, creating a single larger house. A concrete or gypsum
threshold is built, possibly with an associated gate, and a drain excavated in the former alleyway. A
stone lined soakaway was observed directly under the walls of the current Radwani House, into
which the drain would have emptied.
Phase 1
Natural bedrock, (110), was cut by a small pit or posthole [125] and by a larger sub-circular pit, [109],
measuring 1.55m in diameter and 0.6m deep. This feature may represent a quarry pit for gypsum or
limestone. These features were sealed by compacted silty trample layer (116) / (106). This was cut
by three postholes [101], [103] and [105], which may represent posts or scaffolding during
construction of the earliest architectural phase (Figure 5.12).
Phase 2
The earliest architecture in Room 8 consisted of the northeast corner of a building constructed of
unfaced limestone and bonded with coarse mud mortar. The exposed area of wall (15) was L-shape
in plan, and extended beyond the excavation area to the east and south (Figure 5.12). The internal
area of the structure exposed in Room 8 measured at least 6.2m by 2m.This space continued in
Trench 1 to the east, and probably represents a courtyard in this original phase of use.
32
Figure 5.12:Substantial early wall (15), with earlier postholes [101], [103] and [105] in foreground and pit
[109]in background. Looking south.
To the west of wall (15) a less substantial wall of similar construction [84] is keyed into [15] and
encloses well [126], with a diameter of 1.4m and measuring at least 2.0m deep (Figure 5.13). The
well extended under the walls of the current Radwani House, and was not excavated for safety
reasons. The feature represents an external well, constructed at the same time as wall (15), and
surrounded by a low wall or raised platform.
Figure 5.13:External well [126] and associated walls (86) and (15). Photographed in plan.
33
A soft ashy silt layer (107) measuring only 20mm thick is the first of a series of internal surfaces and
occupation deposits in the courtyard area to the east of wall (15) (Figures 5.14 and 5.15). This is
overlain by loose, silty sand (42). Wall 15 is re-plastered at this point with a very pale grey, soft
gypsum plaster (121). Next a crushed lime floor (41) is laid, and then and then wall (15) is again replastered with a light greyish white smooth lime plaster (32). Thick silty occupation deposit (30)
accumulated against the re-plastered wall, and is sealed by 80mm thick sandy gravel floor (29).
Figure 5.14: Early architecture and internal courtyard floor in Room 8, looking south.
34
Figure 5.15: A section through the floor surfaces associated with Phase 2, looking north.
To the west of wall (15), compact sandy silty clay deposit (99)represents an external surface
associated with an alleyway adjacent to the building. It was sealed by a thin, loose silty sand layer
(49), which represents an accumulation of occupation material in the alleyway, or a brief period of
disuse or renovation. A second compact sandy silty clay deposit (37) sealed this deposit.
Linear cut [36] runs parallel to wall [15], truncating surface (37), and is associated with the replastering of the external face of the wall. The wall was covered with a light grey hard plaster (128).
Phase 3
At this point the Phase 2 building undergoes a major alteration. The alleyway directly to the north of
wall (15) is blocked, suggesting the structure to the north is incorporated into a single larger house.
A low step (113) appears to represent a threshold associated with the new entrance into the
combined building. To the east of the threshold, and inside the new building, a hard gypsum or
concrete surface (87) drains through a copper alloy drain cover soakaway [118] (Figures 5.16 and
4.16).
35
Figure 5.16:Concert or gypsum threshold (113) and surface (87) to north of wall 15 and blocking and earlier
alleyway.
Figure 5.17: Metal drain cover, draining into soakaway [118].
36
Phase 4
The major rebuilding of the Radwani House in this phase is represented in Room 8 by a 0.2m thick,
loose, rocky, silty sand dumps (11), (12) and (17) and are also associated with the demolition of the
building associated with wall (15). The walls of the Phase 4 structure were hard to define in Room 8
as they had been entirely truncated by Phase 5 activity. However, it seems likely that the extent of
the phase 4 building was similar to that of the existing Radwani House.
Phase 5
A foundation of roughly faced limestone blocks, bonded with soft mud mortar, and measuring at
least 5.2 m long was constructed in a linear foundation cut. This feature cut the latest demolition
deposit (17) and also truncated the Phase 4 walls in this area of the site. The foundation supports
the current western wall of the Radwani House and is likely to be associated with the 2006
rebuilding works undertaken by the Private Engineering Office.
5.5.3 Room 9
A relatively complex sequence of archaeological deposits was excavated in Room 9, with several
east-west oriented walls representing different phases of building and occupation. In addition a
stone lined soakaway was located in the south part of the room.
Two parallel, early phase walls are likely to be contemporary with the earliest buildings identified in
Room 8. The southern wall may be an external courtyard wall, dividing the house from the alleyway
recorded in Room 8. The northern wall was partially truncated by later activity, and represents an
internal division within the building.
At the southern end of the Room 9 a series of crushed limestone and gypsum surfaces are associated
with an alleyway between two separate buildings in this early phase of occupation. Later concrete or
gypsum surfaces are probably associated with the blocking of this alleyway in a later phase. A stone
lined soakaway was recorded in the southwest corner of Room 9.
Phase 1
Natural limestone bedrock, (81), was sealed by a patchy, thick levelling deposit of sandy, charcoal
rich occupation debris (76), measuring up to 0.15m thick. A heavily truncated pit, [75], measuring
0.50m in diameter and 0.20m deep was filled loose, ashy, silty sand (74). These deposits were sealed
by a sandy, charcoal rich levelling layer (65) /(66). This was in turn cut by pit [73], which contained an
ashy fill (72). The original size and extent of this feature was unclear as it was heavily truncated by
the west wall of the Radwani House. This sequence of occupation activity predates the architecture
in this area of the site, and represents the earliest activity recorded in Room 9
Phase 2
The earliest architecture in Room 9 consisted of an east-west running wall (46) / (47), which survived
to a maximum of two courses high (Figure 5.18). The southern face of the wall was rendered with a
hard, light pinkish grey plaster (96). The wall was associated with a thin, lime flecked compact sandy
clay surface (55) / (54) and was overlaying loose ashy occupation deposit (45). This feature is difficult
to relate to any other features, and wall (46) / (47) does not appear to continue to the east in Trench
1. It seems likely that this early structure was rebuilt, and largely destroyed by feature (159), visible
directly to the east in Trench 1.
37
Figure 5.18:Walls (46) / (47) in foreground and wall (64) in background ,looking south.
Wall (46) / (47) was heavily truncated by linear robber cut [44]. Following this phase of robbing,
there was a second early building phase consisting of an east-west aligned wall (64). This feature
measured 0.50m wide and survived to a length of 2.56m. The western extent of the wall was
truncated by a modern intrusion. Wall (64) was constructed of unfaced limestone, with a hard, dark
grey cement render on its south face. Wall (64) was contemporary with similarly constructed walls
(60) and (61), which were visible only in elevation below the present eastern wall of Room 9.
Phase 3
A series of deposits and features to the south of wall (64) are associated with closing of an earlier
alleyway and the addition of a number of drainage features and compact concrete or gypsum
surfaces. These deposits are likely a continuation of threshold (113), surface (87) and soakaway
[118] recorded directly to the south in Room8.
On the southern side of Room 9,a series of laminated occupation deposits (58) survive. These
deposits were cut by a soakaway [53] and a small drain [94]. Construction cut [53] was lined with
seven courses of unfaced limestone. (52), to form a sub-circular soakaway measuring 1.20m by 0.6m
and 0.8m deep. The primary fill of the feature, (51), consisted of a coarse sandy silt associated with
the use of the feature.
To the south of the soakaway, visible only in section, a grey concrete or gypsum drain (93) was built
within construction cut [94] and measured 0.48m wide and 0.46m deep. A similar small rendered
channel (59), located slightly to the south, is probably also associated with drainage. A similar small
drain (62) was inserted into cut [63], truncating the southern face of wall (64) (Figure 5.19).
38
Figure 5.19: North facing section at southern end of Room 9.
A crushed mortar bedding layer (57) was sealed by truncated hard grey cement or gypsum surface
(56), measuring 0.50m by 0.35m.A 0.35m thick levelling dump (92) was in turn sealed by a hard grey
cement floor (91); these two deposits are the same as (114) and (87) in directly to the south in Room
8.
Phase 4
The major reconstruction of the Radwani House associated with Phase 4 is represented in Room 9 by
an east-west wall, (25), built of unfaced limestone. The wall was built in a linear construction cut
[27], which truncated the earlier occupation deposits in this part of the room.
As in Room 8, most of the Phase 4 architecture was hard to identify as it was either directly below,
or truncated by the walls of the existing Radwani House. It seems likely that the Phase 4 structure
had a similar floor plan to the existing Building. Wall (25) represents an internal room division, which
was presumably removed prior to the 2006 reconstruction of the building.
Soakaway [53] was unfilled by limestone and sand rubble (34) and sealed by a compact light grey
cement floor (31), which is probably an external surface. This suggests the feature fell out of use in
this phase, possibly no longer required after the introduction of piped water and sewage. The
introduction of modern drainage, presumably replacing the earlier wells and soakaways, is visible in
the form of metal drainage pipe (13), within cut [14] located at the eastern edge of the Room 9.
Phase 5
Dump layers of poorly sorted loose building rubble (4) and (9) at the northern end of Room 9, and
dump (16) in the southern end of the space, represent demolition activity associated with the 2006
restoration of the building.
A levelling deposit (90) is also associated with the modern restoration works. Construction cut [89],
contained the foundations for the existing walls of the Radwani House, which were entirely rebuilt in
this area as part of the 2006 reconstruction. The foundations, (91), consisted of roughly faced
limestone, bonded with sandy light brown mortar.
39
Figure 5.20: Members of the Origins of Doha Project team working in Room 9
5.5.4 Room 10
A well and bath were exposed in Room 10, directly below a modern concrete slab associated with
the 2006 restoration of Radwani House by the Private Engineering Office. These architectural
features were recorded, and a small trench was excavated in the southeast corner of Room 10 to
examine the earlier deposits in this area of the site. The architecture in Room 10 was left in-situ and
unexcavated, so that the remains could be preserved for presentation to the public.
In the centre of Room 10 a rectangular concrete plinth surrounded a circular stone lined well. This is
similar in form to the well recorded in Room 8. The deposits overlying the well were removed prior
to our arrival on site; however it is likely that a series of partially truncated concrete or gypsum
surfaces to the north of the well were associated with its use.
The most striking feature in Room 10 was a sub-circular bath, with two internal divisions,
constructed of a hard grey material. The construction material is currently being analysed to
understand its composition. This type of bath has not been found in Qatar before, and further
research is required in order to identify any comparable feature in the region.
Phase 1
Natural bedrock (132) was sealed by a sequence of ashy levelling and occupation deposits. Deposits
(131) and (130) filled undulations in the natural, and consisted of compact sandy gravel and thin
laminated trampled layers. Compact light brown sand and crushed mortar crush, (123), represents a
more deliberate levelling layer, possibly used to prepare the area prior to building on the site.
Phase 2
An unfaced limestone wall constructed directly on top of levelling dump (123) was visible in section
at the eastern side of the room (Figure 5.21). This wall formed the foundation for the existing
Radwani House. It seems likely that the Phase 4 activity in Room 10 described below mostly
40
truncated the earlier activity, or unexcavated architecture seals these earlier deposits in this part of
the site.
Located the southern side of Room 10 layers (122) and (199) represent make-up and levelling
deposits and are sealed by a beaten earth floor surface (120). Modern disturbance make it
impossible to relate these deposits to the rest of the space, however they probably represent
occupation and floor deposits associated with the earlier architectural phases of the site.
Figure 5.21:Phase 2 or Phase 3 limestone wall, constructed on levelling layer (123) and sealed reconstructed
walls of Radwani House. Looking west.
A circular cement or gypsum lined well, (143), had an internal diameter of 2.00m, and measured
over 4.00m deep. The well was surrounded by a sub-rectangular plinth, (141), with evidence for
multiple re-plastering and repairs. East-west aligned wall (142) formed the southern extent of plinth
(141) and appears to have reinforced or stabilised the upper part of the well (Figure 5.22). This
feature is similar to the well and associated plinth [126] in Room 8.
The absence of floor surfaces directly associated with the well make it difficult to ascertain whether
it was located in an external space or inside a room. The central location of the well in Room 10,
directly inside the current entrance and approximately 30cm lower than the present colonnade,
along with its stratigraphic relationship with the bath structure immediately to the north, indicates
that it certainly belongs to an architectural phase which preceded the current layout of the Radwani
House.
41
Figure 5.22: Well (143) with bath (140) in background and truncated floors to the left of the well. Looking
north.
Phase 4
Directly to the north of well (143) two concrete baths, (140), cut the superstructure of the earlier
well. This impressive unique structure consisted of two curved basins divided by a narrow partition.
Both ends of the bath had a rounded niche and the western bath was accessed two low steps (Figure
5.23). The baths measured 3.20m east-west, 2.10m north-south and 0.90m deep. The baths were
constructed of an extremely hard, light grey, concrete or gypsum render applied directly over
roughly hewn dolomitic limestone. A significant quantity of large moulded fragments associated with
the baths were recovered from the infill of the feature, indicating that the bath would have originally
stood at least 0.50m higher. Originally the sides of the baths would have risen above the floor level
of the room. The baths both drain through separate outlet holes to the north, presumably into a
soak-away or drain originally located under the adjacent Room 11. Extensive modern disturbance in
Room 11 mean that no archaeological remains were preserved in this area. The baths extend
underneath the internal walls of the current Radwani House at their eastern extent. This wall
represents part of the 2006 reconstruction of the Radwani House, and the original room that the
baths occupied was larger than the current space.
42
Figure 5.23: Concrete baths (140), looking north west.
A sequence of compact cement or gypsum floors and associated bedding layers abutt the southern
side of baths (140). These surfaces are presumably contemporary with the baths. It seems likely that
the floors would have sealed well (143), which appears to have fallen out of use or been turned into
a soakaway by this point. The well recorded in Room 3 of Company House was reused as a soakaway
in a similar fashion (see section 4.6 of this report). However, the floors had been heavily truncated
making it impossible to establish their stratigraphic relationship to the rest of the room. The earliest
floor surface, (139),was, sealed by gravel bedding layer (138). Above this bedding deposit was a
second cement or gypsum floor surface, (136) / (137). Cement drain setting (135) was a later
addition to the floor surfaces, and appears to drain through the western wall of Room 10. The floor
surfaces described above also slope slightly towards the west, allowing the entire room to drain
efficiently.
5.5.5 Room 12
A sequence of archaeological deposits survived to a depth of 1.5m in Room 12 (Figures 5.24 and
4.25). These deposits represent floor surfaces associated with the current Radwani House,
occupation associated with earlier architectural phases of the building, and deposits associated with
activity in this location prior to construction on the site. Several phases of architecture were present,
the earliest phases associated with a mud rendered limestone wall and later phases with a gypsum
rendered wall directly above it.
Phase 1
Natural limestone (80) was sealed by a dark grey ashy silt layer (71). This context consisted of several
laminated ash rich occupation deposits, possibly associated with more ephemeral wooden
structures or more nomadic occupation pre-dating the construction of the Radwani House (Figure
5.24).
43
Figure 5.24:Layer (79) overlain by several superimposed building phases. Looking northeast.
Phase 2
Built on top of this deposited was a traditional limestone wall (69), consisting of unworked limestone
and mud mortar. The internal face was covered with a mud render (77). A layer of crushed limestone
(70) measured up to 0.50m thick, and appears to represent a levelling dump associated with the
initial construction of the Radwani house. This was overlain by charcoal rich mixed trample and
occupation deposits (67), associated with the earliest occupation of the Radwani House.
These deposits were sealed by a mud rendered limestone wall (97) which abutted wall (69). This
later phase of building appears to be associated with the subdivision of the courtyard in order to
create at least one more room on the northern edge of the courtyard. The floor of newly created
room is initially levelled with brown sandy silt (83). The occupation of the room is represented by a
highly laminated deposit (50), consisting of at least four thin floor layers each measuring c. 5mm
thick. A second dump of brown sandy silt (39) appears to be another levelling event, and is sealed by
laminated ashy grey occupation deposits (38).
Phase 4
The House undergoes a major remodelling, or possibly rebuilding, phase at this point. The external
wall in Room 12 is replaced by (78), which uses the earlier wall as a foundation. The new wall is
covered with a pale yellow grey render (40), possibly made of gypsum. A thick dump of silty sand
(19) was used to level the room and was sealed by ashy occupation deposits (8). A shallow cut [82] /
[28] running along the base of the wall of Room 12 is associated with the re-plastering of the room;
and a thin pale yellow render is applied to wall. I series of occupation deposits, some very charcoal
rich (6), (3) and (5) are all associated with the later occupation of room 12.
Phase 5
A second rebuilding of the external wall now occurs; a new wall (79) is constructed of roughly faced
limestone and a hard pale grey mortar. This wall is probably associated with the reconstruction of
the Radwani House by the Private Engineering Office in 2006. Layers of loose silty sand and stones
(2) and (1) are levelling deposits associated with a modern concrete slab that formed the latest floor
44
surface in Room 12. This slab would also have been laid as part of the recent reconstruction of the
building.
Figure 5.25: Archaeological deposits in section in Room 12, looking east.
Figure 4.25: West facing through the occupation sequence excavated in Room 12
45
5.5.6 Trench 1
Located in the southeast corner of the colonnade, Trench 1 contained a series of well-preserved
walls associated with the earliest architectural phase of the site. Two separate compounds were
identified, separated by an alleyway. A series of surfaces and associated levelling deposits were
excavated in the southern area of Trench 1, associated with the southern compound.
Late i the uildi g s life histo these two compounds were combined, before being extensively
rebuilt into the current form of the Radwani House. Earlier surfaces were truncated by a limestone
capped drain, part of several drainage features associated with this slightly later occupation.
Phase 1
Natural limestone in Trench 1 (198) was cut by a series of irregular channels, possibly the product of
natural hydrological action. At the southern end of the trench the natural limestone was sealed by
loose dark grey sand (233). This deposit contained a small quantity of charcoal and bone, and
appears to represent deliberate levelling prior to the first phase of construction in this area of the
site. Loose, sterile, silty sand (208) filled a fairly deep depression in the natural and may be the result
of natural depositional possess.
At the northern end of the trench natural limestone was sealed by an un-excavated compact beaten
earth surface (238). This represents an occupation deposit predating the earliest architecture in this
area, this is likely be associated with an earlier phase of more ephemeral occupation of the site.
Phase 2
At the northern end of the trench, constructed on top of surface (238), wall (219) and steps (159)
formed the base of a raised building platform that was filled by rubble infill (207). The building was
accessed via the narrow plastered stairway, (159). This structure measured 2.90m by 0.80m and
survived to a height of 0.60m (Figure 5.27). The building continues beyond the trench limits to north
and west. The area to the north was not excavated, to the west little evidence of the building was
recorded in Room 9. The building was constructed of unfaced limestone, with a mud render applied
to the east and south faces. The building had been vertically truncated by later construction, and as
a result any associated floor surfaces have been removed (Figure 4.29). To the east of the building
beaten earth floor surface(158) represents an external courtyard surface associated with the use of
this structure.
Initially building (219) / (159) was not enclosed by a courtyard wall. However, possibly as a result of
increasingly dense occupation, a courtyard wall, (172), was built, cutting into surface (158). This
created an enclosed compound, presumably with raised building (219) / (159) in the southwest
corner. Built in a linear construction cut, [210], wall(172) measured 0.50m wide and was again built
of unfaced limestone. Initially the external southern face of the wall does not appear to have been
rendered, however a thick, hard, light grey render was applied to the wall slightly later.
46
Figure 5.27: Early architecture (219) and (159) with later enclosure wall (172) in background under scale.
Looking south.
At the same time the northern building is enclosed with a courtyard wall a second compound is
constructed directly to the south, with a small alleyway created between the two compounds. Two
small sub-circular postholes, [215] and [218], possibly represent scaffolding associated with the
construction of this southern compound (Figure 5.28). Running parallel to (172) wall (178) measured
0.60m wide and survived to a height of 0.20m (Figure 5.28). Constructed of unfaced limestone and
soft mud mortar the northern external face was with very hard, light grey gypsum or concrete
render (237).
47
Figure 5.28: Postholes [215] and [218] adjacent to wall (178).
Between walls (178) and (172) a narrow alleyway measured 2.00m across and may have provided
access to both compounds. The earliest surface in this alleyway was a compact, silty clay
surface(190). This was overlain by a mid grey beaten earth surface (177), which represents a further
accumulation of trampled occupation deposits.
North-south aligned wall (235) was heavily robbed out, but was associated with wall (178), forming
an internal division within the southern compound (Figure 5.29). Wall (235) probably represents the
eastern wall of an internal space, the western extent being defined by wall (15) in Room 8. To the
east wall (235) a large open area would probably have been an external courtyard. A sequence of
floors and levelling layers in this space represent an extended period of occupation associated with
the southern compound (178) / (235) (Figure 5.30).
48
Figure 5.29: Walls (178) and (235) with alleyway and northern compound wall (172) in background. Looking
north
Loose sandy make up (232) was overlain by a similar light grey sandy levelling deposit (231). This
was sealed by a compact metalled gravel and sand floor surface, (230). The associated occupation
with this floor surface contained a significant quantity of pottery, glass and metal artefacts. Sealing
this surface was a further levelling deposit, (229), consisting of a 50mm thick layer of reddish brown
sand. This deposit formed a bedding to beaten earth floor surface (228). Above (228) was a further
silty beaten earth floor deposit (227) with associated occupation debris.
Deposits (232), (231), (230), (229) and (228) were not fully excavated as a trench measuring 1.20m
by 5.00m was excavated through these deposits in the south west corner of Trench 1.
49
Figure 5.30: North facing section through courtyard deposits in Trench 1.
Phase 4
A drainage channel was constructed in linear cut [185], which truncated the latest courtyard surface
in the southern end of Trench 1 (Figure 5.31). A cement of gypsum drain (184) would have carried
waste water or sewerage to a soakaway similar to those excavated in Rooms 6 and 9. The drain was
surrounded by compact sand and gravel (234) construction backfill and infilled by soft sandy silt
(183). The drain was capped with limestone slabs (181) and covered with sandy silt and limestone
rubble (179).
50
Figure 5.31:Courtyard surface (227) cut by drain [185] with limestone capping stones (181) still in place,
looking west.
51
5.5.7 Trenches 2 and 3
Located in the courtyard area, Trenches 2 and 3 exposed a series of compact floor surfaces (Figures
5.32 and 5.33). This area of the site appears to have been heavily truncated by later 20th century
reconstruction work; however the lowest course of a wall survived in the western end of Trench 2
(Figure 5.33).
Phase 1
The earliest feature in Trench 2 consisted of a sub-circular pit cut [163], which extended beyond the
trench to the west. The fill of this pit, (162), was sealed by a dirty trampled surface (161).
In Trench 3 a single posthole [200] was filled by (199). This feature may relate to a temporary
structure pre-dating the architectural phases, or to the construction of the first phase of the
Radwani House. Sealing this feature sandy deposit (196) and ashy, charcoal-rich deposits (193), (195)
and (197) all appear to be associated with activity that pre-dates the early phase architecture.
Phase 2
Beaten earth courtyard surface (157) is both at a similar level and has a similar composition to the
earliest courtyard surfaces in Trench 1. This deposit appears to be an early surface, associated with
the first phase of architecture on the site. There is no evidence of a well or planting feature in the
courtyard of the earliest phase of building on the site.
Phase 4
Built in construction cut [225] north-south aligned wall (226) survived only as a single course of
stones in the western extent of Trench 3. The construction cut was filled by (224). This wall was
heavily truncated by the reconstruction of the building undertaken by the Private Engineering Office.
Wall (226) probably represents the eastern extend of a raised colonnade associated with the Phase 4
buildings on the site. This colonnade was destroyed and rebuilt in concrete in 2006 undertaken by
the Private Engineering Office. A compact surface (155) was located to the east of wall (226) and
represents an associated occupation level.
Phase 5
Deposits (154) and (151) are mixed levelling layers with modern materials, including plastic, in them.
Over these deposits circular concrete planter (150) was built as part of the remodelling of the
Radwani House by the Private Engineering Office in 2006. Inside the planter modern deposits (149)
and (148) contain very recent material and represent the infilling of the circular planter after the
2006 reconstruction.
52
Figure 5.32:Courtyard surfaces in Trench 2 and circular Trench 3, looking north.
Figure 5.33:Heavily truncated wall in the western end of Trench 2, looking south.
53
5.5.8 Trench 4
The archaeological deposits in Trench 4 had been truncated to a depth of 1m by the extensive
reconstruction of the Radwani House undertaken by the Private Engineering Office in 2006. Below
this truncation c. 0.30m of archaeological deposits survived.
Phase 1
The earliest archaeological feature in Trench 4 was a shallow pit cut [206], measuring 0.20m deep.
The rocky fill (205) of this feature did not contain any dateable artefacts, however it is likely this
feature is of late 19th century date and is contemporary with similar pits in excavated in Rooms 7 and
8. The function of these pits is unclear; they may represent opportunistic quarrying of limestone or
gypsum.
The pit was sealed by a dark charcoal rich layer (212), which appears to be associated with an early
occupation of the site pre-dating the first building on the site. This may be associated with more
ephemeral Bedouin occupation or peripheral activities on the edge of Doha before the town spread
to include the Msheireb area.
Phase 2
Constructed on top of (212) an east-west orientated wall (192) measured 0.60m across and was
constructed of sub-angular unfaced limestone and bonded with red brown packed dirt (Figure 5.34).
At its western extent this wall appears to stop, possibly demarking the edge of a doorway. To the
south of this wall a series of compact mortar surfaces (211) represent well laid floors. To the north of
wall (192) these surfaces are notably absent, suggesting this may represent an external area.
Phase 3/ Phase 4
The wall was sealed by mixed deposits (180) and (187) containing mortar and large pieces of angular
limestone. These layers appear to be represent levelling deposits of dumped demolition material,
and may be associated with the demolition of wall (192). A charcoal rich deposit (186) was probably
dumped rake out from an oven or fire.
54
Figure 5.34: Wall (192) and partially excavated mortar surface to the south of it.
55
5.6. Watching Brief at Company House
Figure 5.35:Location of archaeological features in Company House.
56
An intermittent watching brief was conducted on groundwork within Company House. Two wells, a
possible oven and a large circular metal feature were recorded (Figure 5.35). No excavation was
undertaken; however some unstratified pottery sherds were collect from groundworks in the area.
5.6.1 Room 1
A traditional, hand dug, circular, stone-lined well was recorded in Room 1 of Company House (Figure
5.36). The well had been partially truncated by the restoration of Company House undertaken by the
Private Engineering Office.
Figure 5.36: Partially exposed well in Room 1, looking east.
5.6.2 Room 3
A similar, circular stone-lined well was recorded in Room 3 (Figure 5.37). A drain running in two the
well indicated that this feature had been reused as a soakaway later in the history of the building.
Figure 5.37: Well and later drain in Room 3, looking west.
57
5.6.3 Room 15
A large metal feature was visible in the exposed section of one of the trenches excavated in Room
15. The feature consisted of a large metal ring measuring 2.00m in diameter and 0.20m high, and
was probably part of a cover for a well or drainage feature (Figure 5.38). It appeared to be a
relatively recent feature, and is unlikely to be older than mid 20th century in date. It was on top of
an associated concrete surface and was sealed by a levelling dump of modern material.
Figure 5.38: Large circular metal feature in Room 15.
58
5.6.4 Room 21
A large possible oven was recorded in Room 21. The feature was built of a hard grey material,
probably a gypsum based concrete (Figure 5.39). The feature had been badly damaged by recent
groundworks.
Figure 5.39: Part of the oven in Room 21.
5.7. Discussion
The archaeological deposits recorded, specifically those in the Radwani House, are of high
significance. The study of the buried heritage in the Radwani House is important in both
understanding the history of the house and the wider 19th and early 20th century development of
Doha. No other detailed archaeological investigations have been conducted in Doha, and the results
of this work are therefore of national importance.
These data have the potential to reveal the development of the Radwani House from the time of its
initial construction through to the present day. Moreover, the buried remains of earlier buildings
and underlying levels will reveal the history of the Msheireb area from the time that it was first
settled.
The excavations are providing tangible output, in the form of architecture and finds, which illustrate
the lives, economic activities, material culture and diet of the inhabitants of Doha before the coming
of oil. Apart from their research value, these finds can be utilized in the museum which will be
situated at the Radwani House. The latest archaeological levels also illustrate the changing lives of
the inhabitants of Doha in the mid 20th century, as piped water and electricity were introduced to
the building.
The excavations reveal that this part of Doha was first occupied after the start of the 20th century,
probably some time in the 1920s, judging from the likely date of the pottery at the bottom of the
sequence. The first buildings beneath the Radwani House were also likely built some time in the
1920s or 1930s, with the single unified Radwani House likely to have been built some time between
the mid 1920s and mid 1930s. The whole area was urbanized by 1937, according to aerial
photographs, and the Radwani House itself can be seen in aerial photographs of 1947.
59
The historical context of these building activities can not easily be tied to the growth of Doha during
the pearling boom of the late 19th and early 20th century, as the industry was contracting after
1912. The extension of Doha westwards during those decades therefore occurred despite the
economic hardships of the time. There was nonetheless a brief recovery of pearl prices in the late
1910s, which might have fuelled some of the expansion in this area. Equally or more signficant, oil
exploration concession money began flowing into the Gulf in the 1920s and 1930s, which sustained
the ruling families of the region and allowed the towns to continue to function. In the case of the
area around the Radwani House (western Jasra and Fareej Mohammed bin Jassim) we might also
suggest that the focus of Doha was reorienting around the new seat of the ruler (now on the site of
the old Ottoman fort), which encouraged building in an area which had previously been kept empty
by the Ottomans.
The full a haeologi al se ue e of Doha ill e u h lo ge i the e t e of to a d i the Bida
area, going back at least to the start of the 19th century. The presence of probable pre-Islamic
pottery (Sasanian Period) in the assemblage of the Radwani House implies an even older occupation.
Fu the e a atio s a ou d the old to of Doha a d Bida ould e eal the it s o plete histo .
5.8. Finds and Analytical Material
Archaeological finds from the excavation include:
Ceramic sherds
Animal and Fish Bone
Glass vessel fragments
Metal objects
Textile fragments
Building material samples (wall plaster, concrete and gypsum surfaces)
Environmental samples
These materials will contribute to a better understanding of the development of the Radwani House
site in the later 19th and 20th centuries. The data produced will also contribute to our knowledge of
traditional life in Doha and Qatar, including diet, trading relations and everyday life. Preliminary
results of specialist analyses are presented below in the Appendices (i.e. animal bones, botanical
remains, wood charcoal, plaster composition). A brief report on the other finds is presented here.
In most cases the artefacts recovered from the Radwani House consist of discarded rubbish left
during slightly earlier occupation in the area. This rubbish was spread out to level the ground during
initial construction and the laying and relaying of floors, and was thus incorporated into the two
buildings which underlie the Radwani House.
5.8.1 Miscellaneous Finds
Part of an incised conical incense burner in fine white limestone was recovered from Room 6. The
largest part was from an unstratified context, but a smaller fragment was found in a lower level,
implying that it originated in the 1920s-1930s layers. It still contains ashes from its last usage. An
impressed clay tablet was found, almost certainly a prayer tablet used by some Muslims to press
their foreheads against during prayer (known as a turba). It was from a late demolition level. Also
found were two coins, one being a Saudi 25 halala coin from the 1970s, the other being
unidentifiable. The former suggests that someone was still living in the house as late as the 1970s,
though no other clear evidence of this period has been found, perhaps because it was removed
during the 2006 reconstruction of the house.
60
Figure 5.40: Limestone incense burner. Context 166, with small fragment in 171, Find 55
Figure 5.41: Fragment of probable clay prayer tablet – turba. Context 12, Find 54
61
Figure 5.42: Saudi Arabian 25 halala coin, probably 1970-1976 (Context 7, Find 57)
Figure 5.43: Unidentified bronze coin (Context 21, Find 60) the 1920s-1930s)
5.8.2 Toys
A small numbers of toys, namely glass marbles, were found all or mainly in the later levels of the site
(1950s onwards). In addition a small bouble-barrelled metal object resembling a tiny pair of
62
binoculars was found. It is not clear what this. Piercings on one side imply that it might be a whistle,
rather than toy binoculars.
Figure 5.44: Marbles from Radwani House. Top left and top middle: Context 51, Find 1; top right: Context 107,
Find 31; bottom left: Context 34, Find 21; bottom right: Context 227, Find 47.
Figure 5.45: Small metal allow object from the Radwani House. Perhaps a whistle.
5.8.3 Beads and Bangles
Several small beads and glass bangles were recovered, mostly of glass. All were from late contexts
(1950s or later) or unstratified (i.e. not found in a layer which could be assigned to the sequence).
Also found was a small mother-of-pearl button, likewise from a late context.
63
Figure 5.46: Beads and Bangle fragments from the Radwani House.
Figure 5.47: A mother-of-pearl button. Context 41, Find 59
5.8.4 Textiles
Seven finds of fabric were recovered. One consists of fragments of a tasselled hem or sleeve with a
band of woven metallic thread which has corroded to green. All the fabic finds require further
conservation work, during and after which closer identifications of the cloth might be possible.
64
Figure 5.48: Fabric with metallic thread. Context 39, Find 82
5.8.5 Electrics
Electricity is considered to have been introduced to the Msheireb area in the early 1950s. Two
contexts yielded early electrical fittings, both relatively high in the sequence, being located in a soakaway drain in Room 6 (Context 174) and a late levelling deposit in the courtyard (Trench 3, Context
151). Together these mark a significant transformation in the lives of the inhabitants of the house.
Figure 5.49: Electric light fittings and bulbs. Left: Context 174, Finds 63, 61; Right: Context 151, Finds 64, 62
5.8.6 Ceramic Finds (pottery)
The majority of the pottery from the Radwani House and older underlying buildings was extremely
fragmentary, because most of it was redeposited midden material (i.e. rubbish) which had been
incorporated into the flooring. One exception is a waterpipe and burner from the well in Room 10
(see fig. below). This is likely to have been used by inhabitants of the house and was presumably
65
thrown into the well when both it and the well had falled out of use. The Qatari term for this kind of
waterpipe is gidu.
Most of the pottery dates to the 1920s and perhaps 1930s, and compares well to material in the
collection of the National Museum of Qatar, and to excavated material from Muharraq (Bahrain).
This material pre-dates the construction of the buildings, and indicates that the two houses
underlying the Radwani House can not have been built before the 1920s. The 1920s to 1930s
assemblage gives a broad indication of the time at which Doha expanded to this area of Msheireb.
Some of the layers, particularly the courtyard fills, contained later ceramics, dating to the 1930s to
1960s. Further research is required to narrow the date range of this material.
Some of the ceramics may be suitable for display in the museum, as it is indicative of the lifestyle
and aesthetics of the earliest inhabitants of Msheireb. The commonest forms were coffee cups
(comprising more than a quarter of the assemblage), small water jars and saucers of different types,
followed by waterpipes. The 1920s-1930s assemblage was characterised by a very distinctive kind of
porcelain fluted coffee cup decorated with sprays of gold and green, red or blue, with an Arabic
motto written inside in gold. For some reason tea-cups were not clearly identifiable in the ceramic
or the glass collections, though saucers were common, and teapots were also found. Moreover,
most or all of the cups occurred late in the sequence, perhaps in the 1950s or later. We therefore
speculate that tea was taken in coffee-cups during the 1920s and 1930s.
The place of manufacture of the porcelain coffee cups has not been established. Of the other
ceramics, most of the earthenwares are of local origin in the Gulf, including Iran, Bahrain and Iraq.
Some of the porcelains (especially saucers and teapots) were from Japan. Colourful painted and
transfer-printed bowls and dishes ("Refined White Wares") mainly originated from the Netherlands.
Storage…
Flowerp…
Bowl
Lid
Cup
Large…
Dish
Teapot
Waterpipe
Waterpi…
Saucer
Coffee cup
40
30
20
10
0
Small…
Percentage of Different Ceramic
Forms at the Radwani House
66
Figure 5.50: Water-pipe and burner from the well in Room 10 at the Radwani House (Context 300, pot no. 147,
148)
Figure 5.51: Earthenwares from the Radwani House (Context 22), including water jar spout (bottom right),
storage jar fragment (top right), waterpipe burner (top left) and waterpipe neck (bottom left)
67
Figure 5.52: Porcelains and European "Refined White Wares" from the Radwani House (Context 167)
Figure 5.53: Typical coffee cup from the 1920s assemblage (Context 24)
68
5.8.7 Glass
Like most of the ceramics, the glass is highly fragmentary. The considerable variety of colours, forms
and techniques implies a very wide range of origins across Europe, the Middle East and Asia. Apart
from bottles, typical forms included miniature perfume bottles, a kind of small decorated bottle
(also for perfume?) and square bottles (one marked as being from Germany). One fragment is likely
to be from an oil lamp. A Japanese bottle was also found in the early levels (Dai Nippon company,
which existed between 1906 and 1949).
Figure 5.54: Small bottles from the Radwani House. Left and middle: Context 160, Find 34; Right: Context 47,
Find 27
Figure 5.55: Miniature perfume bottles from the Radwani House (left: Context 21, Find 29; right: Context 180,
Find 41)
69
Figure 5.56: Fragmentary glass assemblage from Context 83 (a levelling fill in Room 12)
5.8.8 Metal Finds
Many layers contained numerous fragments of corroded iron, and less commonly copper or copper
alloy. Nails of various sizes were frequently identifable, usually very badly corroded. Some were very
large and can be assumed to have originated from ships timbers. Other finds include padlocks, a
copper ring of unknown function (perhaps a handle), three bullet casings (perhaps .303 rounds) and
a heavy copper drain cover.
Figure 5.57: Large corroded iron nails (Context 102, Find 118)
70
Figure 5.58: Padlocks (Left: Context 242, Find 147; Right: Context 83, Find 113)
Figure 5.59: Copper ring (Context 19, Find 89)
Figure 5.60: Cartridge casing (Context 22, Find 90)
71
Figure 5.61: Copper drain cover (Context 87, Find 115)
5.8.9 Animal bone
Plentiful animal bone was recovered from the site mainly fish and medium-sized mammal (sheep
and/or goat). This is currently under study and will reveal details of the diet of the inhabitants of
Msheireb from the 1920s/30s into the coming of the oil era.
Figure 5.62: Animal bone from the well (Context 300, Find 65)
72
5.8.10 Shell
Shell was not particularly plentiful, but examples of the Black-Lipped Oyster (Pinctada margaritifera,
sadiaifi) and Pearl Oyster (Pinctada radiata, mahhar) were found. The former, a large species, was
mainly gathered for mother-of-pearl, as pearls occurred only rarely in them, while the latter were
the main pearl-bearing species. Although pearl oysters were normally opened and dumped at sea by
the divers, the small examples found at the Radwani House testify to the practice of shore-gathering,
when people would gather examples from the intertidal zone to open on shore.
Figure 5.63: Shell from the Radwani House. Top left: Context 160, Find 67; top right: Context 8, Find 68; all
below: Context 232, Find 69
73
5.8.11 Company House Finds
There was no systematic recovery of finds at Company House. A collection of unstratified ceramics
was made, which indicates that material comparable to that of the Radwani House was present in
the area. In addition workmen collected five Chinese-made saucers with decal transfer decoration,
of relatively late date (1950s or later).
Figure 5.64: Chinese saucer from Company House
74
6. Recording Archaeological Deposits Exposed in Section
By Daniel Eddisford
6.1 Introduction
During our first season of work the Origins of Doha Project undertook a detailed walkover survey of
the centre of Doha. This survey identified a number of potential excavation areas around the centre
of the city, with the potential for buried archaeological remains. In addition several locations were
identified with exposed archaeological deposits visible in section.
Three sections were recorded; two at site ODQ 3 and one at site ODQ 4 (Figure 6.1). The sections
were cleaned by hand and recorded at a scale of 1:10 in accordance with archaeological best
practice (MOLAS 1991).
No archaeological excavation was undertaken at ODQ 3 and ODQ 4. However after cleaning one of
the sections at ODQ 3, a ceramic vessel was found to be protruding from feature [47]. As the vessel
was in immediate threat of eroding out of the section and being destroyed it was careful removed,
and its contents sampled. The ceramic sherds are currently held at UCL Qatar for further analysis,
prior to their return to the relevant department of the Antiquities Service.
Although not providing as much detailed data as archaeological excavation, the sections recorded
provide important information about the depth and nature of the buried archaeological remains at
these locations. In addition the data recorded can be used to make generalisations about the
potential for intact buried archaeology nearby.
Figure 6.1: Site locations
75
6.2 Site ODQ 3
6.2.1 Site location
Located in the Salata district of the city, and to the east of Jabr bin Mohd Street site ODQ3 is centred
on QNG 232651E 392829N. The site consists of a substantial area of upstanding archaeology
measures c.50m north-south by 30m east-west (Figures 6.1 and 6.2).
Figure 6.2:Location of ODQ3 Sections 1 and 2
Ground reduction in the surrounding area has truncated the archaeological deposits surrounding the
site. On the site archaeological deposits survive to 2.00m above the current ground level, visible in
all the exposed sections of the site (Figure 6.3). A house on the northern side site is constructed of
concrete blocks and appears to be of later 20th century date, although elements of the building may
be earlier. A second house stood on the southern side of the site until relatively recently. This
structure was constructed in a more traditional style using unfaced limestone. The remains of the
limestone built walls are visible in elevation against the surviving building (Figure 6.4). The one
remaining house on the site is still occupied, and appears to be the reason that this area has, as yet,
not been developed.
76
Figure 6.3:ODQ3 looking south with occupied house visible
Figure 6.4:Remains of traditional building visible in elevation, looking north east
Two sections were recorded at ODQ3, a shorter section on the eastern side of the site (Section 1)
and a longer section on the western side of the site (section 2) (Figure 6.2).
In the east facing section of the site (Section 1, Figure 6.5) at least two archaeological horizons are
visible in section, possibly relating to late 19th and mid 20th century occupation. The west facing
section of the site (Section 2, Figure 6.7) contains a more complicated sequence with several
possible phases of occupation, a possible wall, and a sunken ta ūr type oven.
This area of the city has undergone radical changes in the last century. An extensive area has been
reclaimed from the sea to the north of the site, and in the 19th century the site of ODQ 3 would have
77
been located directly on the coast (Figure 6.1). More recently ground reduction close to the site has
removed most archaeological deposits; development in the form high-rise buildings continues to
truncate the limited surviving areas of archaeological deposits.
6.2.2 Section 1
The eastern side of the site was truncated by a road cut. The exposed section contained two distinct
phases of occupation, separated by thick, sterile, poorly sorted dumps of limestone and sand. Two
poorly defined walls, running obliquely to the section, may relate to these two phases of activity.
Figure 6.5:ODQ3, Section 1.
Context
No.
1
2
3
4
5
6
7
8
9
10
11
12
Description
Modern levelling
Loose gravel layer
Ashy occupation material
Loose gravel layer
Ashy occupation material
Natural gravel accumulation, possibly beach
environment
Natural
limestone
Possible wall
Ashy later against wall (8)
Natural gravel accumulation
Natural gravel accumulation, possibly beach
environment
Possible
wall
Summary of the deposits in Section 1
The natural limestone (7) was overlain by naturally deposited layers of gravel (11), (10) and (6),
probably representing a beach or wadi environment. A poorly defined wall (8) was constructed of
78
unfaced limestone on, and possibly cutting into, these deposits. The wall appeared to run northwestsoutheast, obliquely to the section, making it hard to define.
Two phases of occupation associated with wall (8) were visible in section. The earliest occupation
consists of ashy layers (5) and (9); these deposits possibly represent domestic cooking activities. A
small assemblage of ceramics recovered from layer (5) is of late 19th or early 20th century date. This
occupation phase is sealed by a layer of poorly sorted small angular limestone (4). This may
represent naturally deposited material, associated with a period of abandonment of the site. As
second phase of occupation consists of another layer of charcoal rich ash, (3).
A second possible limestone wall (12) was constructed on top of layer (3). This wall was on a similar
northwest—southeast alignment, running obliquely to the section. Another layer of loose angular
limestone, (2), was deposited against wall (12). This was sealed by modern levelling dumps (1)
Figure 6.6:ODQ3, Section 1
Section 2
The western section at ODQ 3 has a more complicated sequence of deposits; an infilled gully was
sealed by at least two phases of occupation and a substantial traditional limestone wall. A modern
manhole and drain [25] truncates the archaeological deposits in the centre of the trench, making it
hard to associated deposits at northern end of the trench to those at the south (Figure 6.7).
79
Figure 6.7:ODQ3, section 2.
Context
Description
No.
13
Modern levelling
14
Dump layer with large stones
15
Dump layer with gravel
16
Ashy layer or possible fire pit
17
Thick dump layer
18
Red dump layer
19
Pit or gully with charcoal. Fill of [22]
20
Laminated layer with charcoal, Fill of [23]
21
Natural
22
Small pit or gully
23
Large pit or gully
24
Modern drain / manhole with concrete blocks, fill of
[25]
25
Modern
drain cut
26
Possible modern layer
27
Dump layer
28
Ashy layer
29
Possible shell rich floor layer
30
Laminated dump or wind blown material
31
Compact dump or naturally deposited layer
32
Ashy layer
33
Poorly sorted sand and limestone fill of [23]
34
Ashy layer, fill of [23]
35
Sand and shell fill of [23]
36
Wall
37
Dump layer with trampled upper surface
38
Rubble layer
39
Ashy layer
40
Thick dump or levelling deposit.
41
Ashy layer
42
Sandy silt with limestone, possibly naturally deposited
43
Ashy layer
44
Homogenous dump of levelling layer
45
Charcoal rich layer
46
Fill of [47], soft ash and upturned pot
47
Cut for oven / ta ūr
48
Fill of [49]
49
Small pit cut
50
Fill of [51]
51
Small pit cut
52
Fill of [53]
53
Large pit cut
54
Laminated ashy layers
Summary of the deposits in Section 2
A large cut feature, possibly a gully or small wadi channel [23] cuts the natural limestone. This
feature is filled by a layer of sand and shell (35) that may have been deposited by alluvial action. A
second fill, (34), consists of a dark grey ashy deposit that may represent material being dumped into
the gully. This is sealed by poorly sorted sand and limestone (33), indicating further alluvial
deposition occurring. The upper fills of the pit consist of possibly naturally deposited sandy silts with
limestone inclusions, (31), (30) and (42); again possibly the result of alluvial activity.
The upper fills of gully [23] were truncated by the construction cut for a traditional wall (36),
constructed of unfaced limestone and bonded with packed dirt. To the north of this wall a shell
surface (29) was sealed by a dark charcoal rich ashy layer, representing occupation associated with
wall (36). To the south of the wall an ashy occupation layer (41) may represent an external surface.
At the southern end of the section the natural limestone (21) was overlain by a series of thin
laminated ashy deposits (54), representing trampled surfaces or oven rake out deposits. Further to
the north, thin lenses of charcoal rich material (45) represents similar early activity on the site. Layer
(54) was truncated by a relatively large pit [53], possibly a limestone quarry pit.
Notably, a large, broken ceramic storage jug was placed in a pit [47] cutting the natural limestone to
create a ta ūr type oven (Figure 6.8). The jug and surrounding cut was filled with a solt, silty ash
(46), associated with the use of the oven. The vessel that was reused in this feature had small
handles on the shoulder, a flat, squared rim and a gritty fabric. It is of 19th or 20th century date,
which suggests that the earliest activity visible on this site is of a similar date. In close proximity to
the ta ūr was another small pit cut [49] that contained stony fill (48). This feature is also
contemporary with early activity on the site, though its function is unclear. These features may be
contemporary with wall (36), or may represent an earlier phase of activity—the connection has been
obscured by the large modern cut bisecting the stratigraphy of the section.
At the northern end of the section, early activity on the site is represented by a layer of red brown
sandy silt with shell lenses, (18). This may be a deliberate levelling deposit or naturally deposited
beach sediments. A small pit or gully [19] may also be a natural feature. This is sealed by a thick,
homogonous dump of sandy silt (17). This appears to be naturally deposited, and may be the same
as (40) at the southern end of the trench. These deposits may represent a period of reduced activity
on the site.
Figure 6.8:In-situ upturned ceramic vessel used as an oven.
At the northern end of the section a second phase of activity is represented by a charcoal rich, ashy
occupation layer, (39). This may be associated with a possible ta ūr [51], located directly to the
82
north and filled with an ashy burnt fill (50). A similar charcoal rich dump or fire instillation (16) is
likely to be contemporary with this second phase of activity.
These occupation deposits are sealed by another phase of ground raising activity. Dump layers (15)
and (44) may be associated with the deliberate levelling of this area. A third phase of activity is
represented by an ashy occupation layer (43). This is sealed by dump layers (38) and (37), which are
possibly associated with the destruction of building (36) and the levelling of the area for the
construction of the existing building on the site. At the northern end of the trench levelling dump
(14) is probably also associated with this activity.
A modern manhole lined with hollow concrete blocks and an associated drain [25] truncate the
archaeological deposits in the centre of the section, and are associated with modern activity on the
site. A layer of modern material (13), includes concrete surfaces, and represents the latest phase of
modern activity on the site.
Figure 6.9:ODQ3, Section 2.
6.3 ODQ 4
6.3.1 Site location
Located 200m to the east of ODQ 3 a large rectangular area has been excavated to provide
foundations and a basement for a modern high-rise development. Site ODQ 4 is centred on QNG
232909E 392771N. Archaeological remains were visible in the exposed northern section of the site
(Figure 6.10), including two masonry walls and a number of external surfaces. The open area to the
north of the site was a graveyard in the early 20th century.
83
Figure 6.10:Location of ODQ4, Section 1.
6.3.2 Section 1
The northern section of the site contained two rendered limestone walls, with a series of external
surfaces to the west. A concrete built addition was added to the eastern side of the earlier limestone
building, before the entire structure was demolished and the area levelled.
84
Figure 6.11: ODQ4, Section 1
Context
No.
1
2
3
4
5
6
7
8
Description
Hard cement or gypsum surfaces
Series of laminated external surfaces
Limestone levelling layer
Thick dump layer / demolition layer
Traditional limestone wall with brown render
Traditional limestone wall with brown render
Concrete block wall abutting wall (5)
Natural
Summary of the deposits in Section 1
At the eastern end of the section two traditionally built limestone walls (5) and (6) were constructed
on top of natural limestone (8). These walls represent the opposite sides of a single building and are
covered in a hard dark brown render. To the east, wall (7) abuts the earlier structure, and is built of a
transitional fabric of shelly concrete blocks. No floor surfaces associated with this building were
visible in section (Figure 6.11).
At the western end of the section the natural was overlain by a levelling layer (3) consisting of
crushed limestone. This was sealed by a series of external courtyard like surfaces of crushed shell and
later lard gypsum or concrete surfaces (2) (Figure 6.14). These surfaces are also visible in the western
and northern sections of the site, and appear to represent a large external space. The relationship of
these surfaces to wall (6) has been truncated by a modern drain, however it is likely they are
contemporary with this building.
The inside of the structure, in-between walls (5) and (6), is infilled with compact sandy silt (4), with
frequent limestone inclusions (Figure 6.13). This represents the demolition of the building, and the
levelling of the site for further construction.
A number of modern drains and intrusions truncate the archaeological deposits. Modern levelling and
a tarmac surface (1) seal the archaeological deposits.
Figure 6.12:ODQ4, Section 1.
86
Figure 6.13:Limestone wall (005).
Figure 6.14:External surfaces and make up (002), and (003).
6.4 Discussion
Site ODQ3 is located near the historic core of Doha, and in the 19th century would have been very
close to the coastline. Site ODQ4 Is located slightly back from the coastline, adjacent to a historic
cemetery.
87
At least two distinct phases of occupation were identified at ODQ3, associated with several traditional
limestone walls and charcoal rich ashy surfaces. The first occupation phase may represent an
expansion of Doha in late 19th and early 20th century, associated with a period of relative prosperity.
This is followed by a possible hiatus, corresponding to the collapse of the pearling economy. The site
is reoccupied, possibly in the mid 20th century as natural energy resources are increasingly exploited
in Qatar and the city begins to flourish once more. However, it is important to treat this scenario as
hypothetical. Higher resolution data from archaeological excavations would be required to support
the details suggested by the exposed archaeological sections.
The archaeological remains recorded in the exposed section at ODQ4 are possibly later in date. The
limestone walls visible may represent the expansion of the Salata district in the mid 20 th century. A
building, probably a domestic structure, was built of limestone in a traditional style. A second phase
of building consists of a concrete block built addition to the earlier structure, and is likely to date to
the later 1950s or 1960s.
At ODQ 3 archaeological deposits were recorded from a level of c.5.70m OD; at ODQ 4 archaeological
deposits were recorded from a level of c.5.50mOD. The archaeological deposits visible in section at
these two sites suggest that where archaeological deposits remain undisturbed it is likely that
significant archaeological deposits are likely to be present. Although extensive ground reduction has
occurred in the Salata area there is the potential for archaeological deposits to survive from c.
5.50mOD to c. 0.20m below the existing ground surface.
All the archaeology within ODQ4 was recently destroyed by the excavations undertaken ahead of the
development of the site. Similar high-rise developments continue have a similar destructive effect on
the archaeology of the area. A more considered approach to development is required, that assesses
and mitigates for effect of development on buried archaeology that is likely to be present on most
new development sites.
88
7. Archaeological Survey of an Abandoned Agricultural Area in Rayyan
By Daniel Eddisford
7.1 Introduction
As part of the Origins of Doha Project a team of archaeologists from UCL Qatar undertook an
archaeological survey of an abandoned agricultural area located to the west of education city in the al
Rayyan district of Doha.
The redevelopment of the area meant the site was threatened with imminent destruction, and the
aim of the project was to record the site in as much detail as possible. The Origins of Doha Project
aims to explore the foundation and historic growth of Doha, and its transformation to a modern city.
Agricultural sites such as the one recorded would have played an important economic and social role
in the development of the city.
7.2 Site location and scope of work
The site is centred at Qatar National Grid 221383N 394809E, and is located directly to the west of
Education city (Figures 7.2 and 7.3). The site measures c.420m NE-SW by 230m NW-SE and is
bounded by a limestone wall. The eastern extent of the site has been truncated by modern activity.
Several phases of disturbance by mechanical excavators and the dumping of material on the site have
occurred.
The site is threatened by imminent destruction; a development plan has been approved to transform
the area into a golf course. Before the site was completely destroyed, a detailed archaeological
survey of the area was undertaken to ensure the preservation by record of these early agricultural
remains. An initial walkover survey identified a number of features on the site and an initial sketch
plan of the site was made. A Leica total station and a Leica differential GPS were used to make a
detailed plan of the features identified (Figure 7.9). The features were given unique numbers and a
detailed description of each was recorded on pro-forma context sheets. Finally a detailed
photographic record of the site was made.
89
Figure 7.1: Site overview, looking north.
90
Figure 7.2: Site Location.
Figure 7.3: Detailed Site Location.
91
7.3 Archaeological and Historical Background
There is a limited amount of historical data about the development of Doha, and even less describing
everyday activities such as agriculture. In 1908 J.G. Lorimer visited Doha and mentions several water
sources close to the town in his survey of the Gulf:
Dohah itself possesses o l o e ell of ra kish ater, a ed Ai Walad “a'id,
which is a half mile to the south of the Dohah quarter, but there is a group of others
called Mushairib with fairly good water, at 1 mile to the west of the Dohah quarter.
Three miles inland is Bir-al-Jadidah, a large masonry well of indifferent water on
which the to
ai l depe ds for its suppl . (Lorimer 1986)
The 1947 aerial photograph of Doha shows an agricultural area directly to the west of Doha. This
agricultural area, now known as Msheireb West, was probably originally the one cultivated by the
Ottoman garrison, later taken by the British consulate, which made use of the fresh water associated
with the main Msheireb wells. The same aerial image also shows an agricultural area to the south
east of Doha, which is likely to be the area Lorimer refers to as Bir-al-Jididah. This location is now the
site of the Al Muntazah Park.
There are no known early historic references to the Rayyan area, but that does not necessarily mean
the area was uninhabited until the modern period. The name Rayyan means 'the source of irrigation',
and it is clearly a relatively well watered area. The earliest documentary evidence of the site is an
aerial photograph from 1966 (Figure 7.4), which shows the Rayyan area to be heavily irrigated, with a
large number of well watered farms visible. By this time diesel pumps are likely to be in use, pumping
water from a relatively shallow aquifer.
Five years later, in 1971, a similar situation is visible on a second aerial photograph (Figure 7.5). A
topographic map of the same date, and likely based on the 1971 aerial photograph clearly shows the
e te t of the site, although ot a ked as a g ee ag i ultu al a ea it see s likel the site is still i
use at this time (Figure 7.6). The earliest images available from Google Earth date to 2004, and show
the site to be in a similar abandoned state as today.
92
Figure 7.4: Aerial photograph of the site in 1966.
Figure 7.5: Aerial photograph of the site in 1971.
93
Figure 7.6: Topographic map of al Rayyan, 1971.
Until the late 1950s the groundwater system of Qatar maintained a state of equilibrium in which
outflow, including extractions related to agriculture, were balanced by inflow and water levels and
salinity remained static (Pike et al., 1976). Research into hydrological changes in Qatar show that by
1979 the level of the fresh water in the water table had fallen as a consequence of increased
extraction. This drop in the water table and the associated intrusion of saline water could be
observed in the salinity of wells across the country. (Lloyd et al., 1987). While saline intrusion is an
obvious source of salinity in the aquifer, another source of salinity occurs from over pumping, which
auses the u de l i g sali e ate e eath a f esh ate le s to up o e into the bore and the
surrounding freshwater aquifer (Macumber 2009, 93). A recent assessment of the impact of
groundwater extractions by Amer et al., (2008), shows that there was a five-fold increase in
producing wells between 1975/76 (660 wells) and 2000/2001 (3400 wells). Amer et al observe the
impact of increased salinity and lower groundwater levels causing the abandonment of farms and soil
degradation, especially in coastal areas, and the salinization of water sources for coastal towns.
Today the agriculture in the al Rayyan area is almost all abandoned, and it is likely that the processes
of over extraction of groundwater resulting in salination are largely responsible. A large part of the al
Rayyan area has been re-developed as Education City, which has resulted in the destruction of all
earlier evidence of occupation. However, at least one farm is still in operation in the al Rayyan area.
Located 1.3km to the east of the site, this agricultural area uses electric pumps to access water,
presumably from a deep aquifer. The fields on the farm are enclosed by low earthen berms and
irrigated with shallow canals or water (Figure 7.7).
94
Figure 7.7: Agricultural area still in use directly to the east of the site.
7.4 Results
The results of the survey of the site are presented below.
Figure 7.8: Plan of the site.
95
Context
Number
Description
1
Cement block pump house structure
2
Concrete block water tank
3
Concrete block building
4
Concrete block water tank
5
Traditional well
6
7
Concert block structure, possibly a pump house. Addition
to
(005) block water tank
Concrete
8
Borehole and pump base
9
Fieldsystem
10
Concrete block building and borehole
11
Borehole
12
Cement block pump house structure
13
Concrete block water tank
14
Limestone wall
15
Concrete block water tank and bore hole
16
Irrigation channel
17
Limestone structure, possibly animal enclosure
Summary of archaeological features
7.4.1 Traditional box well
One of the features on the site was a substantial stone lined square well, (005), measured c.8.50m
across and at least 10m deep. The well contained water that was at least 3m deep, measured by
dropping a weighted line into the well. The upper portion of the well was lined with unfaced
limestone. The feature was enclosed by a low wall also constructed of unfaced limestone and covered
with a hard render.
Well (005) has clear parallels with abandoned wells at a number of other sites in Qatar, such as
Murayr and Shwail in northern Qatar (Eddisford 2012). The box wells at these sites provided water to
the site of al Zubarah and its surrounding settlements from the Dammam limestone aquifer. These
wells are also associated with evidence of agricultural activity. The wells at al Rayyan and the other
examples in northern Qatar are hand dug and date to the late 19th century or earlier. The shape and
size of this type of well suggests that a relatively thin layer of fresh water was skimmed off the
underlying denser saline water (Macumber 2009). In this situation a wide well giving a large surface
area is required; a deeper well would just encounter increasingly salinated water.
96
Figure 7.9: Bo
ell
ith st u tu e
ehi d. Looki g to the southeast.
7.4.2 Pump Houses
Several rectangular structures constructed of concrete blocks with shell inclusions were recorded.
Structures (001) and (012) measured c3.30m long and 3.10m wide and both would have housed
diesel water pumps. Both buildings had a similar plan; an internal concrete block with bolts provided
a base for the pump. A bore hole in front of each building has an aluminium pipe in the top of it.
Structure (001) also had a concrete trough and the base for a smaller pump in front of it (Figure 7.10).
The structures would have been roofed with corrugated tin or asbestos, but this has been removed.
Crude holes in the walls of the structures appear to be later alterations, possibly associated with
changes in the method of pumping water.
Figure 7.10: Pump house (001) looking northwest. Note borehole pipe in front of building.
97
Figure 7.11: Structure (006) looking northwest, with well (005) in background.
Figure 7.12: Tank inside structure (006), draining into well (005) to west.
Structure (006) was also built of concrete blocks with shell inclusions and was a later addition, built
on the eastern extent of well (005) (Figures 7.10 - 7.12). The northern part of the structure had an
external wall similar to (001) and (012), but contained a rendered tank with steps into it, which
drained into well (005). The southern part of the structure consisted of a rectangular low wall. The
function of this feature is not immediately apparent, it might represent another pump house that
incorporates elements of the earlier well (005).
98
7.4.3 Bore holes
Several isolated boreholes (008) and (011) consisted of concrete casings surrounding a metal pipe.
Two concrete pump bases adjacent to (008) measured 1.40m x 0.70m and 0.60m x 1.00m
respectively (Figure 7.13). Bore hole (011) was at the base of a large machine dug depression and
appeared to be a later addition to the site, possibly after some of the irrigation works had fallen out
of use.
Figure 7.13: Bore Hole (008) in background and concrete pump base in foreground, looking west
7.4.4 Rectangular Tanks
Several large rectangular tanks were built of concrete blocks with shell inclusions and lined with a
hard waterproof grey render. Several large tanks appeared to act as reservoirs, and would have fed
into a series of small canals to irrigate the surrounding fields. Tank (002) measured 6.70m by 6.60m
and 1.30m high and drained into two smaller tanks to south. This feature is possibly associated with
the pump house (001). Tank (004) measured 6.80m by 6.60m and 1.40m high gain. Again a smaller
tank to north allowed water to drain into a drainage channel to the east. A very large tank (007)
measured 11.00m by 11.80m and was associated with smaller tanks to east and west (Figure 7.13).
99
Figure 7.14: Large tank (007), looking northwest
Two smaller tanks were associated with bore holes and pumps. These features appeared to have had
a similar function to the larger tanks, acting as temporary storage before distributing water into the
irrigation system. Tank (013) measures 5.90m by 8.90m and was associated with a bore hole and
concrete pump base. The pump base measured 2.45m by 1.00m and was similar in appearance to
feature (008).
A smaller tank (015) measured 3.20m by 3.30m and was associated with a borehole and two pump
bases (Figure 7.15). Tank (005) drained directly into an irrigation channel (016), which ran around the
perimeter of the site. Directly to the south of (015) a small circular trough also fed directly into
channel (016), and was identical in appearance to the trough collecting water on a nearby farm, seen
in Figure 7.7.
100
Figure 7.15: Small tank (015), looking east.
7.4.5 Irrigation Features
Irrigation channel (016) runs around the edge of the site. Most of the channel was lined with
limestone; however in places concert blocks are also used (Figure 7.16). The addition of concrete
could represent a later repair to the channel. Tank (015) drained directly into this channel; the larger
tanks may have also fed into this channel indirectly through small canals or pipes.
Figure 7.16: Peripheral irrigation channel (016), looking east.
A number of rectangular fields (009) survive in the south-western area of the site. The fields are
defined by low mounds and measure 4.00m – 6.00m across and 13.00 – 15.00m long. A number of
irrigation channels are also visible (Figures 7.17 and 7.18). A concentration of broken brown ceramic
101
water pipes in this area of the site, and several in-situ examples indicate that these channels and the
external irrigation channel (016) would have been linked to the fields by a network of pipes. These
pipes could be blocked off, allowing water to be diverted to specific fields or areas of the site.
Figure 7.17: Panoramic view of fields and irrigation channels (009).
Figure 7.18: Rectangular fields and irrigation channels (009), looking south
7.4.6 Structures
A rectangular building (003) measured 9.40m by 3.20m (Figure 7.19). This building had three rooms
and was built of concrete blocks with shell inclusions. The external faces of the building were covered
with a hard, grey render. No roof or roof timbers survived. Round timbers wrapped in rope were used
to support the top of the door and window frames (Figure 7.20), a traditional construction technique
that pre-dates the concrete building material used. Raised windows opened to the east and south of
102
the largest eastern room. A tank directly to the north west of the building was lined with waterproof
grey render and measured 1,90m by 3.20m. The function of the building was not obvious, but it could
represent a small, low status or possibly seasonal, domestic structure.
Figure 7.19:Building (003), looking northwest.
Figure 7.20:Wooden timbers framing the door of building (003).
Another rectangular building (010), constructed of concrete blocks with shell inclusions, measured
8.00m by 4.00m (Figure 7.21). The northern extent of the building had been truncated by a
mechanical excavator, and the area around the building was heavily disturbed. The original asbestos
tile roof had collapsed, and as a result a limited examination of the building was undertaken, for
health and safety reasons. Inside the building was a borehole and concrete pump base. A table or
103
work bench was built into the wall in the southeast corner of the building. A large doorway opened to
the east. Substantially larger than the other pump houses (001) and (012) on the site this building
appears to have been a multi-purpose structure; acting as a pump house, as well as a possible
workshop and storage area.
Figure 7.21:Building (010), looking west.
A substantial wall, (014), was constructed of unfaced limestone and survived to a height of 1.70m
(Figure 7.22). This feature had been completely truncated by modern activity at the eastern extent of
the site, and had be partiality destroyed along the sites southern extent. This wall formed the
external boundary of the site.
Figure 7.22:Boundary wall (014) at the northern extent of the site, looking north.
104
Built against this external wall, an enclosure (017) consisted of two rectangular spaces. The eastern
extent of this enclosure is entirely truncated by modern activity (Figure 7.23). A small rectangular
opening in the western wall of the enclosure provided a window into the feature (Figure 7.24). An
open water channel or trough crossed the northern side of the enclosure, and probably originally
connected to the peripheral irrigation channel (016). The function of the enclosure was not entirely
clear; however it might have acted as an animal enclosure with a water trough at its northern side
Figure 7.23:Enclosure (017) with channel visible in background, looking north.
Figure 7.24:Window in the eastern wall of enclosure (017), looking west.
105
7.5 Discussion
Despite a lack of historical references to al Rayyan, the number of farms in the area in the mid 20 th
century attests to the availability of water in this location. The development of agriculture proximate
to Doha in the 1950s and 60s is clearly linked to the expansion of the city as wealth related to
exploitation of oil resources starts to flow into the region. The rapid growth of farming in this period
also involves new technological developments, specifically the introduction of diesel powered water
pumps.
The presence of a large, traditional box well (005) on the site suggests the area may have been a
source of water, and possibly even an agricultural area, before the mid 20th century. This large well
would not have been initially been used in conjunction with mechanical water pumps, although may
have been adapted with the addition of a later concrete block structure (006). No other evidence of
earlier occupation was identified, possibly a result of the extensive alterations made to the site in
from the mid 20th century to the present day.
The limestone external wall (014) that defined the extent of the site, a limestone enclosure (017) and
elements of the peripheral irrigation channel (016) could all predate the mid 20 th century. However,
all of these features continued in use until the latest use of the farm in the later 20th century.
Most of the site appears to be of a mid 20th century date. Cement block pump houses (001) and (012)
would have contained large diesel powered pumps, lifting water from a relatively shallow aquifer. The
water would have been pumped into large rectangular tanks such as (002), (004) and (007). These
tanks would have acted as storage tanks, allowed sediment to settle out and would have permitted
gravity fed irrigation of surrounding fields. These large tanks drained through smaller adjacent tanks
into a network of drainage channels, including a channel running around the periphery of the site,
(016). Smaller tanks such as (015) would have been drained directly into this peripheral drainage
channel. A series of pipes allowed the water to irrigate specific fields.
The uildi gs o the site utilise t a sitio al a hite tu al st le. Co ete lo ks a e used, ut the
contain large quantities of shell and are likely to date to the 1950s. More traditional construction
techniques, such as rope wrapped beams over doors and windows also suggest an initial construction
in the mid 20th century. Documentary evidence, including the earliest aerial imagery of the site,
suggests agriculture was expanded in the al Rayyan area in this period.
The introduction of mechanical water pumps in the mid 20th century had a rapid and detrimental
impact on the fresh water aquifers in Qatar (Macumber 2009. In northern Qatar the salination of the
aquifer led to the abandonment of a large number of villages and newly established agricultural
areas. The continuing rapid population growth, and expansion of Doha has led to increased pressure
on freshwater and food production; problems that are currently addressed by a reliance on
importation and de-salination. In 1971 the farm was still in use; however it seems likely problems
associated with the salination of the fresh water aquifer rapidly had an impact on farming on this site
and in the country in general.
A second phase of irrigation on the site may have employed much deeper bore holes and electric
pumps, similar to that still in use nearby. This second phase of mechanical irrigation would be
associated with bore hole (011), located at the bottom of a large mechanically excavated hole. The
alterations to pump houses (001) and (012) would also be associated with this later phase of
irrigation. However, attempts to maintain the farm appeared to have failed. It is seems likely the site
was largely abandoned at some point the later in the 1970s or in the 1980s.
106
Documentary research may be able shed further light on the individuals who owned and ran this
farm. It is possible that structure (003) represents a domestic building. If this is the case it is likely to
be of low status, and not the primary dwelling of the owner of the farm – lots of investment required
to set up an agricultural area on this scale.
This farm is interesting because it shows changing use of water over time, with different technologies
used to procure scarce ground water. This site provided insights to a regional pattern on an intimate
scale. Investigation of this site allows us to gain a multiscalar perspective on a time of intense change
in Qatar and in the Gulf in general. The lived experience of attempting to grow produce in this harsh
environment, temporarily mitigated by technology, and then ultimately foiled by this same
technology allows archaeologists to gather insights regarding past experiences of living in Qatar that
are not part of the nomadic narrative that emphasizes pastoral herding. Finally, recording this farm
sheds light onto the historical record of this region and calls into question the primacy of the British
Naval record. Archaeological investigation is a valuable tool for decentering colonizing narratives of
the past, and further research allows a counter-narrative regarding the regional importance of Doha
to emerge.
107
8. Standing Building Recording in Doha
By Kirk Roberts & Katie Campbell
8.1 Introduction
This document details the methodology and preliminary results of a survey and mapping
project of the historic buildings of old Doha. The work was conducted as part of the UCL
Qatar Origins of Doha Project with the support of the Qatar Foundation, during November
and December 2012.
The primary aims of the project were to document and record the remaining buildings within
the old Al Ghanem district, identify those structures which are most at risk from
development or structural decay, and provide a visual record which can be used in order to
keep track of changes to the buildings in question. Additionally, the project aims to broadly
categorize building types in terms of architectural style, function and construction materials,
using these data as a basis for a chronology of construction methods.
At a larger scale, the project proposes further comparative work with buildings in other
quarters of Old Doha, as a means of characterising discrete districts in terms of their
architectural and socio economic development.
8.2 Rationale
The historic buildings of the districts outlined in this document represent the last remaining urban
fabric of the pre-oil era in Doha. These buildings constitute the only architectural evidence for a way
of life which is rapidly disappearing, and without work to document and record them, their place in
understanding the development of this fast changing city may soon be lost.
The accelerating pace of development in central Doha, combined with the pressure of increasing
population, t affi a d loosel e fo ed pla i g egulatio
ea that a of the it s histo i
buildings are at risk of structural decay and unchecked demolition. Of particular concern is the
increasing practice of demolishing buildings in order to make way for increased vehicle parking space.
This often involves the unscheduled destruction of entire blocks comprising multiple buildings, with
o p io e o d o assess e t of those uildi gs sig ifi a e, a hite tu al fo
o st u tu al
integrity.
At present little formal work has been undertaken to record and document the architectural heritage
of pre-oil era Doha. The formative period represented by these remaining buildings is the focus of
current social and academic study, and represents a key phase in the development of modern Qatar.
108
Figure 8.1: Map of Old Al-Ghanim, including Buildings Survey plots
109
8.3 Aims
The current project aims to illuminate the early architectural history of Doha, assess the
significance of the surviving buildings, and offer recommendations on the appropriateness of
potential conservation strategies. A preliminary written and visual record of at risk buildings
will be generated, which is intended to be integrated with the information gathered by the
UCL Qata Mappi g Li i g He itage ou se. The p oje t ai s a e desig ed to e oadl
consistent with both the research aims of this course, and professional standards in buildings
recording. A teaching component will form a core aim of the project, involving postgraduate
students from UCL Qatar in the process of buildings recording and archive research.
More generally, the current project is eventually intended to form part of a comparative
study of the historic buildings in other districts of Doha, with the aim of illuminating our
understanding of the socio-economic fabric of pre-oil era Qatar. In so doing, the project will
form a valuable component in the current research trend towards understanding the growth
of modern Doha, and of the Gulf states in general.
8.4 Methodology
The current project is borne out of a broader research context, and is intended to
complement research and teaching currently underway at UCL Qatar, as well as conform to
p ofessio al sta da ds i uildi gs e o di g. The p oje t ai s to o ple e t the Mapping
Living Heritage sho t ou se, o-ordinated by UCLQ in conjunction with Msheireb Properties,
and to be broadly consistent with the principles set out in the 2006 English Heritage
Understanding Historic Buildings guideli es. The st u tu e a d ethodology are intended to
be flexible and subject to modification as research questions are refined in future seasons.
8.4.1 Desk-based work
Preliminary desk-based work consisted of gathering and cataloguing 20th century aerial
photographs from a range of sources. Digital copies of these images were then manually
scaled and rectified using fixed points of reference (e.g mosques & Doha Fort) to fit the most
recent Google Earth image for the same area, from May of 2011. Image regressions,
including comparison of satellite imagery from 2008, provided a broad visual indication of
the ea l ha a te a d sp ead of Doha s u a fa i , as ell as the i pa t of o e e e t
development and demolition. These images formed the basis for defining the survey area
and each of the plots within it. Digital map regression using aerial photography and Google
Earth indicate that the majority of buildings in the Al Ghanem district were not present in
1947, but that by the early 1960s the current street plan was well established.
8.4.2 Definition of research area
Three key geographical research areas (Ghanem, Asmakh and Najada) were defined,
identified as historic districts containing a significant amount of surviving traditional style
architecture. The primary district selected for standing buildings survey was Al Ghanem,
110
which is bounded by Ali bin Abdullah street to the North, Jabr bin Mohammed street to the
East, Al Mahar street to the South and Ibn Malik street to the West. Work involving
volunteers and students had already been undertaken in Asmakh, under the guidance of Tim
Ma o e as pa t of the UCL Qata Mapping Living Heritage Cou se. The u e t p oje t
aims to generate a corpus of data capable of being integrated with this work. A map of Al
Ghanem based on the most recent Google Earth satellite images was produced, and split
according to approximate building plots, for use in the field (Fig 7.1).
Figure 8.2: Example of building material guidance for survey participants.
In addition to the detailed work carried out in Al Ghanem, a preliminary walkover survey was
completed in the districts of Asmakh and Najada, with the aim of eventually extending the building
recording into these areas. Najada lies to the East of Al Ghanem, and is bounded by Ali bin Abdullah
Street to the North, Banks Street to the East, Ahmed bin Mohammed al Thani Street to the South and
Al Asmakh Street to the West. This area has been heavily redeveloped in recent years and very few
historic buildings have survived. The walkover survey showed that the area to the North of Al
Maymoun Street and West of Al Aasim Street had the greatest density of historic buildings, while the
area south of Al Maymoun Street contained few buildings constructed from traditional materials, and
was probably developed in the 1960s and later.
The district of Asmakh is bounded by Wadi Musheireb Street to the North, Al Asmakh Street to the
East, Ahmed bin Mohammed bin Thani Street to the South and Abdullah bin Thani Street to the West.
Initial survey showed that a number of historic buildings survive throughout the district, although
111
many are abandoned or in a poor state of repair. The historic buildings in Asmakh are in general
larger and more elaborate than those in Al Ghanem and Najada, hinting at demographic and
economic differences between the districts.
A potential area for further investigation is situated behind the Emiri Diwan, North of Al Rayyan Road
and East of Al Diwan Street, which contains a number of abandoned historic buildings, however
se u it o e s due to the a ea s p o i it to the Di a
a pote tiall ha pe e o di g o k.
Further to the West of Ghanem, the district of Al Salata may also contain historic buildings which
have survived the new National Museum development; the area east of Sheraouh Street between Al
Loulou Street and Yafa Street may be appropriate areas for further investigation.
8.4.3 Recording
Each plot in the survey area was assigned a unique code, comprising a two letter prefix denoting the
district (GH for Al Ghanem) and consecutive numbers within the district. The coded plots were
sketched on to a Google earth image of the area, and printed for use in the field.
The written record comprises a pre printed building recording sheet for each plot, including prompts
for the recording of architectural attributes, building materials, GPS co-ordinates and structural
modifications. The sheets are aimed at generating a standardised data set, independent of the person
responsible for recording, and are accompanied by a handout providing guidance and examples for
participants. Since access to the majority of the buildings included in the survey was limited or not
possible, the recording methodology is aimed primarily at gathering information on the visible,
exterior walls and street frontages. An inherent bias is acknowledged in this approach, however it was
felt that the issues involved in obtaining access to properties were beyond the scope of the current
project. The written archive was entered into spreadsheet form, in order to be compatible with the
range of GIS software employed. These data formed the basis for the map in figure 8.1.
A photographic procedure involving multiple overlapping digital images was adopted,
enabling panorama style photographs to be digitally stitched together (Figure 8.3). These
photographs currently form the most detailed and systematic visual record of the street
frontages of Al Ghanem. Among the suggestions for further work is the possibility of
integrating these images into Google Earth, enabling users to navigate detailed street views.
112
Figure 8.3: Stitched and digitally rectified elevation of building GH01.
8.5 Discussion
Building materials were split into three broad categories, with reference to photographs and
descriptions in the Notes on Buildings Recording handout sheet, in an effort to standardise recording.
These categories; Historic, Transitional and Modern, were based on the presence of particular
characteristic building materials such as Danshal roofing beams, roughly hewn limestone, compactedshell cinder blocks, concrete, and the range of modern materials currently in use, such as steel reenforced cement and corrugated asbestos roofing. Although no specific date range is assigned to
each of these categories, they are assumed to represent an approximately chronologically linear
development in terms of construction styles and materials. Further documentary research is required
in order to be able to assign dates for the introduction of particular materials, and thus refine the
interpretations based on these categories. However, some preliminary observations can be made
based on the available data.
The majority of modern development has taken place along the major roads, with most of the intact
historic buildings located in the central part of Al Ghanem. The exception to this is along Al Ghanim Al
Qadeem, where historic building fabric was observed along the majority of the West side, and a
portion of the East side of the street. It is worth noting that this implies a bias towards the survival of
domestic architecture, with street-fronting commercial properties either lost, or poorly represented
in this district. All of the historic and transitional buildings surveyed had been subjected to some form
of modification, usually to accommodate a service such as electricity, or for air-conditioning. In
several properties, blocked doors and windows hinted at a modified internal layout.
In general, the corner properties located at the intersection of minor streets were the most
susceptible to damage, collapse and demolition, and were assigned the highest conservation
priority scores. This is likely a result of the tendency for expanding vehicle parking space
where access is available from two sides. One of the recommendations resulting from the
project is therefore that where appropriate, any future buildings conservation should
prioritise those structures which are most threatened by the trend to demolish corner
properties. This observation will enable a more focused and predictive strategy for the
p ote tio of Doha s histo i uildi gs.
113
The current street layout of Al Ghanem seems little altered from historic aerial photographs of the
district, in spite of the extensive level of redevelopment. This implies an adherence to historically
significant plots, and the absence of a centralised effort to redevelop large parts of the urban fabric
by, for example, modifying roads. Several of the houses comprise a traditional style courtyard house,
with later transitional and modern extensions added within the courtyard. These likely originally
represented extended family dwellings. In particular, the layout of buildings around Al Ghanim Al
Qadeem and Khalid Bin Al Walid street seems largely unchanged.
Due to the ethical and practical issues involved in gaining access to dwellings and in
interviewing their inhabitants, no consistent, rigorous assessment was made of the modern
demographic of Al Ghanem. However, casual observation suggests that the vast majority of
residents are non-Qatari, primarily male migrant workers. Living conditions are typically
cramped, with heavy pressure on infrastructure and services such as water and electricity
provision. A more in depth sociological study of the historic districts of Doha would have
merit, and Al Ghanem exemplifies the late 20th Century pattern of Qatari migration from the
city centre. A study of the demographic history of Al Ghanem would have implications for our
understanding of the changing social and economic development of Qatar, and of the Gulf
states in general.
Further buildings recording work is recommended in the Asmakh and Najada districts. It is
hoped that a comparative study, bringing together observations and data from each of these
districts and from the current study, will enhance our understanding of the early character
and later demographic changes of historic Doha. In addition, these data should form a basis
for any conservation strategies aimed at the historic architecture of the city, and inform
planning decisions relating to the redevelopment of Old Doha.
114
9. Conclusions
The concept of urban archaeology is new to the Gulf, and few of the inhabitants of todays cities
realise that an unwritten, unexplored record of the cities' development and its people lies beneath
their feet. Even in countries where the level of archaeological activity is comparatively high, such as
Qatar, no archaeological excavation work takes place in urban contexts, even when the opportunity is
presented as areas are redeveloped. For example, a range of recent archaeological excavation and
survey projects in Qatar have focused on the later Islamic (17th to 19th century) occupation in
Northern Qatar, involving excavation at Zubarah (Richter et al 2011), Furayha (Rees at al 2011), al
Ruwaydah, (Peterson and Grey 2012) , Ras Ushayriq (Peterson 2011) and Fuwairit (Eddisford 2011).
Numerous other sites have been excavated in Qatar in previous decades by teams from the
Department of Antiquities, Denmark, UK, France, Germany, Japan and elswhere. This project is the
first to explore the archaeology, history and heritage of Doha, one of the major cities of the region.
In future seasons we hope to excavate in the historic town centres of Doha and Doha, and continue
to seek permission to work towards those aims.
http://originsofdoha.wordpress.com/
115
10. Bibliography
Amer, K.M, Al-Muraiki, A.A. and Rashid, R. (2008). Management of coastal aquifers – The case of a
peninsular – State of Qatar. 20th Salt Water Intrusion Meeting, Naples, June 2008
Anscombe, F.F. (1997) The Ottoman Gulf: The Creation of Kuwait, Saudi Arabia and Qatar. New York:
Colombia University Press.
Brucks, G.B. (1829) Memoir descriptive of the Navigation of the Gulf of Persia in R.H. Thomas (ed)
Selections from the records of the Bombay Government No XXIV (1985) New York: Oleander press.
de Cardi, B. (1978) Qatar Archaeological Report: Excavations 1973. University Oxford: Oxford Press.
Carter, R. (2012) Sea of Pearls. Seven Thousand Years of the Industry that Shaped the Gulf. London:
Arabian Publishing.
Carter, R. and Eddisford, E. (2012a) Radwani House - Assessment of Archaeological Potential.
Unpublished Report
Carter, R. and Eddisford, E. (2012b) Company House - Assessment of Archaeological Potential.
Unpublished Report
Carter, R., and Naranjo-Santana, J. (2011) Muharraq Excavations 2010. Unpublished Report
Eddisford, D. (2011) Archaeological Survey at Fuwairit and Zarqa. Report produced for the Qatar
Islamic Archaeology and Heritage Project, University of Copenhagen.
Eddisford, D. (2012). An Archaeological Survey of Northern Qatar (2011-2012). Report produced for
the Qatar Islamic Archaeology and Heritage Project, University of Copenhagen.
Institute for Archaeologists (1995) Standards and Guidance for Archaeological Excavation. Revised
2008.
Availble
from:
http://www.archaeologists.net/sites/default/files/nodefiles/ifa_standards_excavation.pdf
Jaidah, I. and Bourennane, M. (2010) The History of Qatari Architecture 1800-1950. Skira
Lloyd, J.W., Pike, J.G., Eccleston, B.L and Chidley, T.R.E. (1987). The hydrology of complex lens
conditions in Qatar Journal of Hydrology, 89 (1987); 239-258
Lorimer , J.G. (1986) Gazetteer of the Persian Gulf, Oman and Central Arabia. Cambridge: Cambridge
University Press. (Originally published 1915)
Macumber, P.G. (2009). Preliminary report on the Geomorphology and hydrology of the al Zubarah
region Northern Qatar. Unpublished report.
MOLAS (1994) Archaeological Site Manual (3rd edition). London. Available
http://www.museumoflondonarchaeology.org.uk/NR/rdonlyres/056B4AFD-AB5F-45AF-90975A53FFDC1F94/0/MoLASManual94.pdf
from:
116
Montigny, A. (1980) Etude Anthrologique au Qatar. In J. Tixier (ed.) Mission Archeologique Francaise a
Qatar, Tome 1. CNRS Recherches Anthropologiques, France
Nippa, A., Herbstreuth,P. & Burchardt, H. (2006) Unterwegs am Golf: von Basra nach Maskat. Berlin:
Schiler
Othman , N. (1984) With Their Bare Hands: The Story of the Oil Industry in Qatar. London: Longman
Group
Peterson, A. (2011) Research on an Islamic Period Settlement at Ras Ushayriq in Northern Qatar and
some observations on the occurrences of date presses. Proceedings of the Seminar for Arabian
Studies 41: 245–256.
Peterson, A. and Grey, T. (2012) Palace, mosque, and tomb at al-Ruwaydah, Qatar. Proceedings of the
Seminar for Arabian Studies 42: 277–290
Pike, J.G, Harhash, I and Gehnell, B.A.P (1976). Rainfall and groundwater recharge over Qatar.
Technical note 24. State of Qatar Ministry of Industry and Agriculture Integrated Water and Land Use
Project
Rahman, H. (2005) The Emergence of Qatar: The Turbulent Years 1627-1916 . London: Thames &
Hudson
Rees, G., Richter, T., & Walmsley, A. (2011) Investigations in al-)u ā ah hi te la d at Mu a
Furayah, north-west Qatar. Proceedings of the Seminar for Arabian Studies41: 309–316.
a d al-
Richter, T., Wordsworth, P. & Walmsley , A. (2011): Pearl fishers, townsfolk, Bedouin, and shaykhs:
economic and social relations in Islamic al-)u ā ah. Proceedings of the Seminar for Arabian Studies
41: 1–16
Seton, D. (1995) The Journals of David Seton in the Gulf 1800-1809 (edited by Sultan Muhammad AlQasimi). University of Exeter Press
US Hydrographical Office (1920) Persian Gulf Pilot: Comprising the Persian Gulf the Gulf of Oman and
the Makran coast. – Hydrographical Office publication no. 158. Government Printing Office,
Washington.
117
11. Appendix A – Radwani House: Specialist Report on Plaster
Petrographic and SEM-EDS Analysis of Historic Cementitious Building Materials from Radwani
House, Doha, Qatar
Dr. Patrick Sean Quinn
11.1 Background, Sample Materials and Aims of Analysis
Thin section petrographic and SEM-EDS analysis has been undertaken on 15 samples of cementitious building materials from UCLQ excavations at Radwani House (Site ODQ22) in Doha, Qatar.
The term cementitious building materials is used here in a general sense to refer to carbonate-rich
substances composed of lime. portland cement, gypsum or ground natural carbonate material,
which has been used with or without aggregate and other modifiers to construct, coat or adjoin
built structures (Quinn, 2013, p. 224).
Excavation of this early 20th century house has revealed architectural remains that may cover a
transitional period from the use of traditional to modern building materials. In order to investigate
this hypothesis and answer specific questions about the technology and use of cementitious building
materials at the site, such as whether portland cement or other types of hydraulic substances were
used in its construction, selected samples from several phases in four rooms of the house were
submitted for compositional analysis. Petrographic and SEM-EDS analysis was conducted in order
to characterize the composition, raw materials and technology of the cementitious building
materials and reveal compositional patterns that might be related to function or the period of
construction. Details of the samples are given in Table 1.
In this report "plaster" refers to an internal wall covering, while "render" refers to an external wall covering
(Ingham 2011, p. 137)
11.2 Methodology
Small pieces of the 15 cementitious building samples were vacuum impregnated in epoxy resin
blocks and prepared as standard 30 µm petrographic thin sections at the Institute of Archaeology,
University College London. Samples 22 and 24 were consisted of two separate layers that have
been numbered 22A, 22B and 24A, 24B respectively. These were thin sectioned individually. Samples
20 and 23 were composed of two adhered layers. Thin sections were prepared across the layers in order
to examine their microstratigraphy. The prepared thin sections were compared to one another under
the polarizing light microscope and grouped according to their compositional characteristics. The
grouped thin sections were then characterized petrographically and interpreted in terms of their
constituent raw materials and manufacturing technology. Photomicrographs of the 15 samples are
presented in Figures 1-6.
Representative samples of the identified petrographic groups were prepared for SEM analysis. The
epoxy resin blocks of samples 20, 24B and 25, remaining after thin section preparation were polished
with fine silicon carbide paper and diamond paste down to a grade of 1 µm. They were then
coated with a thin film of carbon and studied with a Philips XL30 SEM with an Oxford Instruments INCA
Wave EDS system at the Institute of Archaeology, University College London. Samples were examined
in secondary electron and backscattered electron modes and selected features were analyzed chemically
via SEM-EDS. SEM analysis was conducted with an operating voltage of 20 Kv and a working distance of
10 mm. Calibration of the EDS system was made using a cobalt standard. Elements were calculated as
weight percentage oxides using stoichiometry (Table 1). SEM images of the analyzed samples are
118
presented in Figures 7-11. Using the SEM-EDS elemental characterization of the selected cementitious
building samples, the degree of hydraulicity of the binder matrix was calculated using several alternative
formulae.
11.3 Results
In thin section, the 15 analysed samples can be divided into three types based on their composition and microstructure. Sample 23 is composed of two layers that correspond to two of the
three compositional types. The three types of cementitious building materials are described and
interpreted separately below and illustrated in Figures 1-5. For the definition of specialist terminology
and a general introduction to cementitious archaeological materials see Ingham (2011, p. 137-162),
Leslie and Eden (2008), Piovesan et al. (2009) and Quinn (2013, p. 224-231).
11.3.1 Type 1 - Hydraulic lime binder with oolitic limestone aggregate
Samples 15, 16, 17, 20, 23, 25, 28 and 29
The thin sections prepared from samples 15, 16, 17, 20, 25, 28 and 29 are composed of a lime
binder with abundant oolitic sand aggregate (Figures 1A,B, 6-8). The conspicuous oolites are
rounded, spherical to prolate calcareous particles, composed of fine micrite, that have concentric
layering and in some cases a visible nucleus of quartz or a small shell particle. Many resemble oolites,
though others have less visible layering and may be better classified as peloids o oated pa ti les .
These may have originated from the crushing of an oolitic limestone. However, no intact oolitic
limestone rock fragments were observed in any of the thin sections, suggesting that the aggregate
may have derived from a loose sandy oolitic deposit. The oolitic material was well sorted, with a
modal grain size of 0.25 mm. Rounded mono- and polycrystalline quartz grains, which reach a
diameter of 1 mm, but have a similar modal size to the oolites are also present in smaller amounts
(Figure 1B,D). Rare iron-stained chert also occurs in samples 17 and 26. These non-carbonate
inclusions may have also been part of the same loose sandy deposit. Other aggregate particles include
curved shell fragments of brachiopods (e.g. sample 16) (Figure 5C,D) and cephalopods (e.g. sample 28)
(Figure 5E,F), benthic foraminifera (e.g. sample 17) and planktonic foraminifera (e.g. sample 29).
These contain well preserved wall structure are are therefore likely to have been added as aggregate.
It is possible that they were also part of the well-sorted oolitic sand aggregate, though some of the
shell fragments can reach up to 2.5 mm in size. Some shell fragments have micritic coatings (e.g.
sample 28). The shell structure and preservation of the foraminifera suggests that they are not
fossil forms. Fragments of coral also occur in some samples (e.g. sample 28). Samples 28 and 29
both contain single relatively large crystalline limestone fragments that are associated with
gypsum. These may be related to the possible lime lump in sample 25, which is discussed below. All
samples contain a relatively high proportion (40-50%) of aggregate.
The samples contain c. 30-40% of binder, based on the prepared thin sections. This is composed of
fine micritic calcite, as well as clumps of clay minerals. An area of sample 15 which has no aggregate
mixed into it reveals the composition of the binder in detail, confirming its clayey nature. This might
indicate the mixing of calcareous material with clay. SEM-EDS analysis of the binder matrix of samples
20 and 25 reveal, in addition to abundant Ca, a significant proportion of Si, Al and Fe (Table 2),
confirming its impure nature. A distinct difference exists in several of the thin sections (e.g. samples
15 and 20) between areas of lighter and darker binder. The lighter areas have the typical
appearance of micritic calcite and as a result can be difficult to distinguish from the oolites in places.
The darker areas are lighter in PPL and have an isotropic appearance in XP. SEM-EDS analysis of
these two areas in sample 20 (Table 2) (Figure 8) reveals a higher proportion of Ca in the lighter
areas and a relatively higher percentage of Si, Fe and Al in the darker areas. If the binder in this this
group of samples was made by the mixture of calcareous material and clay, then the darker regions
119
could therefore represent areas in which the two were not well mixed parts of the binder that did
not receive as much calcareous component. An alternative explanation is that these represent regions
of uncarbonated lime. Unreacted lime can have an isotropic appearance in XP, suggesting that the
darker areas in XP could be lime that was not wetted and therefore did not recarbonate.
However, this is not borne out by the SEM-EDS data. The rare occurrence of li e lu ps i sa ple
25 seems to indicate that the calcareous material used in the binder of this group of samples was true
lime. These lumps are lime particles that were not sufficiently hydrated during slaking. The scarcity of
lime lumps in most of the samples would suggest that the the lime was finely ground before slaking
and mixing with the clay component.
The samples made of this type of cementitious sample are relatively porous (c. 10-20%). They
contain pores that were entrapped and entrained during the production of the cementitious building
material, as well as cracks formed during drying or as a result of stress. The latter are easy to
identify due to their thin, parallel-sided shape. However, entrapped and entrained voids are likely
to be difficult to distinguish from voids produced during thin section preparation. The loose nature of
the samples meant that despite vacuum impregnation, some material was plucked from the thin
section as it was ground and polished to 30 µm. Ring voids exist around the oolite particles in
some samples (e.g. sample 28). In some cases rounded voids have been left from the removal of
ooliths. Some voids contain micritic and dog-tooth calcite fringes that were formed after the
setting of the material. Other deterioration and alteration products include gypsum in sample 28
and a calcareous crust on the outside of sample 25. The gypsum may have derived from the
cementitious material described below, as it does not appear to have been present in the binder or
aggregate of the first group of samples.
Detailed SEM-EDS analysis of the binder in samples 20 and 25 revealed the presence of significant Na,
Mg, Al, Si, Fe, K and S (Table 2). Using the relative proportions of these elements relative to the
abundant Ca, it is possible to assess the hydraulicity of the lime. Hydraulicity is the ability of a
cementitious substance to set in moist or wet conditions by the chemical reac¬tions with water. It is
related to the presence of calcium silicate and calcium aluminate hydrates. Different formulae are
used by different authors to calculate the Hydraulic Index (HI) (Vicat 18125; Piovesan et al. 2009, p. 69)
or Cementation Index (C.I.) (Boynton 1980, p. 312). These are given below and applied to the data
in Table 2.
Vicat (1812):
HI =
SiO 2 + Al 2 O 3
CaO
Piovesan et al. (2009) (after Boynton 1966)6:
HI =
Al2O3 + FeO + SiO2
CaO + MgO
120
Boynton (1980):
HI =
2.8 x SiO2 + 1.1 x Al2O3 + 0.7 x FeO
CaO + 1.4 x MgO
Based on the formulae above, samples 20 and 25 have a HI senso Vicat (1812) of 0.20-0.90, a HI
sensuo Piovesan et al. (2009)/Boynton (1966) of 0.22-0.96 and a C.I. sensuo Boynton )1980) of
0.49-2.17. It is not at present clear how to interpret the first HI values. Using only the relative
proportion of Al and Si in a Ca-rich binder, as in the Vicat equation, St John et al. (1998) state that
hydraulic limes vary range in Si and Al from as low as 1-2% (fat limes) to up to 50% (eminently
hydraulic limes). The two analysed samples fall somewhere between these extremes and should
therefore be considered to have been hydraulic. The calculated HI sensuo Piovesan et al.
(2009)/Boynton (1966) of samples 20 and 25 is always <1.2, which Piovesan et al. (2009) state is the
upper limit for quick setting limes. They suggest that HI values higher than this are a consequence of the
unintentional analysis of fine aggregate particles during SEM-EDS analysis. This seems to suggest that the
measurements recorded in this study analysed only the binder matrix of samples 20 and 25 as required.
Unfortunately, Piovesan et al. (2009) do not indicate a lower HI value for hydraulic lime, though they
consider samples with a HI of >0.22 in their analysis to be hydraulic. With this in mind, all SEM-EDS
analyses of the lime binder in samples 20 and 25 indicate that the cementitious material used in this
group of samples was hydraulic.
4
Based on the Cementation Index (C.I.), Boynton (1980, p. 313) classifies lime into three types:
Feebly Hydraulic (C.I. 0.30-0.50), Moderately Hydraulic (C.I. 0.50-0.70) and Eminently Hydraulic (C.I.
0.70-1.10). In this classification scheme, sample 20 varies from Moderately Hydraulic to Eminently
Hydraulic and sample 25 is Eminently Hydraulic. It is not clear how to interpret a C.I of greater than
. i Bo to s s he e. Ne e theless, the values calculated for the two samples suggest they were
made of a hydraulic material.
The variation in the proportion of Si, Al, Fe and Mg relative to Ca within sample 20 implies a variable
hydraulicity for this sample. The darker areas of binder referred to above have a considerably higher HI
and CI based on all three formulae compared to the lighter areas. This is clearly related to the
proportion of non carbonate material in these two different regions of the binder in this and other
samples of the same compositional group. It is suggested that the darker areas have a higher
concentration of argillaceous clay-rich material, which it is suspected to have been mixed with lime in
the production of the samples of this group. The function of the clay is likely to have been to act as a
pozzolanic additive (a substance that when mixed with slaked lime reacts to produce hydraulic compounds).
Common pozzolanic additives in traditional cementitious materials include silicate rock and crushed pottery.
Its incomplete mixture with the lime has therefore led to variation in the hydraulicity of the binder.
11.3.2 Type 2 - Gypsum and lime binder without aggregate
Samples 18, 22A, 22B, 23, 24A, 24B
The thin sections of samples 18, 22A, 22B, 24A and 24B are characterized by a heterogeneous
gypsum-rich composition without appreciable aggregate. The five samples have a messier, more
heterogeneous composition in thin section than the seven samples described above (Figures 3A,B,
4A,B). They contain rare rounded quartz clasts of up to 1.25 mm in diameter that may or may not
have been intentionally added as aggregate. Rare ooliths that resemble those in the other group of
samples occur in sample 22A. Limestone fragments are present in small amounts. These can be
composed of small calcite spars (sample 24A) or finer micritic material (sample 22A). It is not clear
whether these were added as aggregate or whether they represent unburnt limestone fragments
121
from the lime burning process. They can be distinguished from lime lumps that occur in some
samples. Large relatively well formed gypsum crystals are a component of all samples (Figure 2A,B).
SEM-EDS analysis of examples of these crystals in sample 24B confirmed that they are gypsum
(CaSO4.2H2O) (Table 2). The crystals vary in size but can reach 2.25 mm in sample 24A. In some cases
they form agglomerations of several subhedral crystals (e.g. sample 22A) (Figure 3E,F), which have a
similar composition to the larger crystals (Table 2). It is suspected that they were incorporated by the
insufficient crushing of gypsum during preparation rather than being added as aggregate. Obvious lime
lumps with an isotropic appearance and cracked internal microstructure occur in places (e.g.
samples 22B and 24B). These suggest the use of lime in this type of cementitious building material.
The binder appears to be a mixture of lime and gypsum. Micritic calcite crystals are visible in thin
section, however, it is not possible to make out clearly crystals of the gypsum under the polarizing light
microscope. Nevertheless, the presence of larger gypsum crystals and agglomerations as well as
the SEM-EDS analysis of the binder of sample 24B confirms its presence in the binder (Table 2). The
binder has a very muddy appearance in places, which may suggest the addition of clay to the mix.
SEM-EDS of the binder of sample 24B reveals the presence of significant Mg and Si, though no Al
or Fe (Table 1). As in the first type of cementitious building material, the binder contains
compositionally different regions that might indicate the incomplete mixing of two or more
components, or perhaps insufficient re-carbonation of lime in places.
All samples in this group contain a patchy pore pattern. Radiating crystals of gypsum have formed
within many of the pores (Figure 3E,F) (Table 2), perhaps during the setting of the material Though
some of this could have also formed after setting due to deterioration of the cementitious structures
after setting.
11.3.3 Type 3 - Lime and gypsum with oolitic limestone aggregate
Samples 26 and 27
These two samples share some similarities to the other types of cementitious material above in
that they contain oolitic and quartz aggregate, as well as some gypsum (Figures 4E,F, 5A,B).
Oolites are much rarer in samples 26 and 27 compared to samples 15, 16, 17, 20, 25, 28 and 29,
though there is a concentration of these rounded particles in one area of sample 27 (Figure 5A,B). The
rounded quartz sand is of a similar size and composition of that which was added as aggregate to
samples 15, 16, 17, 20, 25, 28 and 29. Prominent lime lumps occur in both samples, which indicate that
lime was used for the binder of the cementitious building material. This may have been mixed with an
argillaceous component, giving the matrix a muddy appearance. The binder seems to be well
carbonated, with the exception of the burnt lime lumps. Sample 26 contains an area of fine
micritic calcite several millimeters in size with little argillaceous material and no aggregate. This may
be representative of the lime used to prepare the binder in this sample.
11.3.4 Microstratigraphy
Sample 23 is composed of two adhered layers (Figure 3C,D). In thin section the inner layer is of
the oolitic lime composition and the outer layer is of the gypsum-rich type described above.
Samples 22 and 24 are also composed of two layers, which were presented separately for
analysis (22A and 22B and 24A and 24B). In both samples, the two layers are made of similar
gypsum-rich cementitious material, without obvious compositional differences. These
122
Petrographic Analysis of Cementitious Building Material from Radwani House, Doha, Qatar
may have been constructed at a similar time, with the second, outer layer added after the first
layer had set.
Sample 25 features a c. 0.75 mm thick, undulose, calcareous layer composed of successive layers of
calcite crystals, oriented perpendicular to its surface (Figure 4C,D). This is interpreted as a crust built
up by successive cycles of calcite precipitation. It might have been produced by water percolating
through the bath structure from which it originated, dissolving and depositing calcium carbonate.
The morphology of this caliche-type crust suggests that it is not a finishing layer. SEM-EDS analysis of
this layer (Table 1) (Figure 9) reveals that it is rich in Ca, but also contains significant Mg and S, that may
have been leached from the cementitious building material and deposited as salt.
Sample 20 has a 1.25 mm thick outer layer that appears to be some sort of finishing material (Figure
2C,D). This is composed of dense, muddy, micritic calcite with rare oolites. It is well bonded to the
material below, but contrasts strongly from it due to its more dense and perhaps muddier composition
and a much smaller proportion of oolitic aggregate. SEM-EDS analysis of the binder of this layer
reveals that it contains a high proportion of Si as well as significant Al and Fe. This would have given it a
moderately hydraulic behaviour. Sample 26 contains an area of fine micritic calcite several millimeters in
size with little argillaceous material and no aggregate (Figure 4E,F).
11.3.5 Discussion
Relationship between different types of cementitious building material
The two main types of cementitious building material detected in this study differ in terms of both
their binder and aggregate composition. The first type was made with a lime binder, whereas the
second type is gypsum-rich. The lime-rich samples contain as much as 50% oolite sand aggregate,
whereas the gypsum-rich samples do not appear to have had any aggregate added to them. This
difference in binder and aggregate technology sets them apart.
Despite the key differences between the two main types of cementitious building material
detected in the Radwani House samples analyzed, it is worth noting that the binder of both
cementitious building material types contains significant argillaceous material, which may have
added in the form of mud or clay. It is not possible to be sure how this material made its way into the
binder of both types of samples, though the presence in sample 20 of darker and lighter areas of binder
with varying proportions of Ca relative to Si, Al and Fe, suggests the mixing of clay and mud. No such
clear distinction was found in the gypsum-rich cementitious building material samples, though these
are more heterogeneous in general. Binder made from natural gypsum can contain clay impurities
(Ingham 2011, p. 139), which might explain the argillaceous material in the gypsum-rich samples. The
latter also appear to contain calcite lime in their binder. This can also be natural occurrence, but
can be added intentionally in modern finishing gypsum plaster to improve its working properties
(Ingham 2011, p. 142).
Samples 26 and 27 appear to have been made using materials from from the other two recipes.
They are gypsum based, but contain some calcite as well as lime lumps, suggesting that as in the
second group of samples, a mix of gypsum and lime was used. The practice of adding small
amounts of gypsum to lime is know 'gauging' (Ratcliffe 1997). This speeds up the set of the lime
aggregate mix. Unlike the other gypsum rich material, samples 26 and 27 contain a small, but
significant proportion of aggregate that was likely to have been intentionally added. This is composed of
oolites and rounded quartzose grains which are identical to the aggregate that was added in much
greater amounts in the hydraulic lime material. Samples 26 and 27 provide another link between the
two main types of cementitious building material detected in this study.
The two layers in sample 23 are composed of hydraulic lime binder with oolite aggregate and gypsum-rich
binder without aggregate respectively. Though these layers were administered separately, their
association on a single architectural feature implies a further link between the two cementitious
123
building material types.
The relationships between the three compositional groups discussed above suggest the use of several
common elements (oolite sand, lime, gypsum, clay) in different proportions, perhaps for different
purposes.
Relationship between cementitious building material type and function
Based on the analysed samples, the different cementitious recipes detected in the present study
may be broadly correlated with the construction of specific architectural features at Radwani House.
Hydraulic lime binder with oolitic aggregate is used as a render8 and for the construction of floors, a
drainage feature and baths at the site. Its hydraulic characteristics would have been well suited to its
use in the later two contexts and may also have been beneficial for the construction of floors. Its use as
a render might indicate that it was favoured as a coating for surfaces that did not need such a smooth
finish. The aggregate might have had a functional advantage by helping the render withstand the
heating-cooling cycle of the Qatari climate.
The gypsum-rich material without aggregate was used mainly as a plaster for internal walls, but
also as a external layer over the first material in the construction of a drain feature.. The fine nature
of this material would have provided a smooth finish on interior walls. That it was not used as a render
in the analysed samples may be explained by the fact that gypsum is adversely affected by water and
damp. Though Qatar has an extremely arid climate, rain does fall in winter months. This hypothesis
does not explain the use of this material as surface coating on the drain feature, which would be
expected to come in direct contact with water.
The third type of cementitious material, which was detected in samples 26 and 27 was used to construct a
floor and render a well in Room 10.
In sample 23, the coarser oolite-rich lime-based material seems to have been used for a base coat,
onto which the finer gypsum-rich material was added. Other possible finishing layers were recorded in
samples 20 and 26. These indicate the use of a finer material with less aggregate, one with significant
argillaceous material and the other composed of lime only. The latter could be a lime wash.
Chronological patterns in cementitious building material type
Taking the cemenrtitious building material samples analysed from the four rooms together and
viewing the distribution of the detected recipes by excavated phase, a broad chronological
pattern is visible. The hydraulic lime samples with oolitic aggregate come from Phases 4 and 3 of
Radwani House and the gypsum-rich samples with no aggregate come from Phase 2 and the lower
part of Phase 3. In the analysed samples, the hydraulic lime
material with oolitic aggregate does not occur in Phase 2 and the gypsum-rich material without
aggregate is not present in Phase 4.
This pattern could perhaps be taken to signify a change in cementitious building material technology
over time at the site, from the use of gypsum-based cementitious material without aggregate in the
earlier phases of Radwani House to the use of lime-based material with aggregate in the later phases.
Sample 23. which contains separate layers of both types comes from the middle of Phase 3 and
might therefore be seen to correlate roughly to a possible change in building material technology.
However, a proportion of lime was used in the binder of the gypsum-rich samples in Phase 2 and
samples 26 and 27, from phase 4. This suggests that lime was being used for the possible switch
and gypsum was being used after it had taken place. With this in mind and given the broad correlation
between cementitious building material recipe and function described above, it is perhaps unwise to
view this chronological pattern strictly in terms of a change in cementitious building material
technology at the site.
A small gypsum factory opened near to Radwani House some time in the mid 20th century. This does
not appear to be correlated with the use of gypsum for construction purposes at the site, as this is
124
more prevalent in the earlier phases based on the sample analysed in this study.
Use of cement at the Radwani House
One question posed of the analysis, was whether cement/concrete was ever used in the construction
of Radwani House. Portland cement was invented in the nineteenth century and has since the
1930s been the favored binder type for a variety of construction purposes in many parts of the
world. However, it is thought that this technology did not arrive in the Gulf area until the 1950s.
Portland cement can be identified by the presence of specific unreacted residuals (Walsh 2007, p.
4) composed of three main calcium silicate minerals, alite, belite and ferrite. Alite occurs as
colourless, high relief crystals with a six sided shape that have first order grey interference colours
(Ingham 2011, Table 19, p. 81). Belite occurs in clusters of crystals, surrounded by brown ferrite
that a e des i ed as ha i g a u h of of g apes o pholog (Walsh 2007, p. 4-5; Ingham 2011,
p. 82-83, 155-157). Belite crystals have a colourless, yellow or green colour in PPL and second
order birefingence colours in XP.
The Radwani House samples were examined at 200x magnification in order to determine whether
these phases are present. Several samples of the first and third types of cementitious building material
contained small structures that resemble the bunch of grapes, illustrated by Walsh (2007, p. 5)
and Ingham (2011, p. 82-83, 155-157) (Figure 6A-D). In some cases it was possible to make out
crystals within these mainly dark brown to black features (Figure 6C, D). However, these did not
have second order birefingence colours. Some five or six sided higher relief crystals were seen in
some samples from the first cementitious building material type, but these appeared isotropic or highly
birefringent (Figure 6E, F), so cannot be positively identified as alite. Walsh (2007, p. 5) states that
s all luste s a e dete ted i s all ua tities i h d auli li es a d What lea l
distinguishes portland cement from other historical binders is the presence of alite . With this i
mind, it is not possible in this study to confirm the use of Portland cement in the manufacture of the
cementitious building materials used at Radwani House.
11.3.6 References Cited
Boyton, R.S. 1980. Chemistry and technology of lime and limestone. 2nd edition. John Wiley & Sons
Inc., New York.
Ingham, J.P. 2011 Geomaterials Under the Microscope: A Colour Guide. Manson, London.
Leslie, A. B. and Eden, M. 2008. A Code of Practice for the Petrographic Examination of Mortars,
Plasters, Renders and Related Materials. Applied Petrography Group.
Piovesan, R., Curti, E., Grifa, C., Maritan, L. and Mazzoli, C. 2009. Petrographic and microstratigraphic
analysis of mortar-based building materials from the Temple of Venus, Pompeii, In: Quinn, P. S. (Ed.)
Interpreting Silent Artefacts: Petrographic Approaches to Archaeological Ceramics. Archaeopress: 65-79.
St John, D. A., Poole, A. B. and Sims, I. 1998. Concrete Petrography: A Handbook of Investigative
Techniques. Arnold, London.
Quinn, P. S. 2013. Ceramic Petrography: The Interpretation of Archaeological Pottery & Related
Artefacts in Thin Section. Archaeopress, Oxford.
Walsh, J. J. 1997. Petrography: Distinguishing Natural Cement from Other Binders in Historical
Masonry Construction Using Forensic Microscopy Techniques. Journal of American Society for Testing
and Materials International, Vol. 4, No. 11-12.
125
11.3.7 Tables and Figures
Sample
Room
Phas
15
9
3
e
16
9
3
17
9
3
18
9
2
20
9
3
22
8
2
23
8
3
24
12
3
25
10
4
26
10
4
27
10
4
28
8
3
29
10
4
Contex
91
t
93
95
96
89
121
113
68,40
140
141
143
128
137
Description and Comments
Render
Render
Render
Render
Mortar Sample. Two adhered layers.
Mortar Sample. Two separate layers (22A and
Mortar
22B) Sample. Two adhered layers.
Plaster. Two separate layers (24A and 24B)
Wall + Render from Bath
Floor
Well + Plinth
Render on external wall [15]
Upper grey floor on well side of bathroom
Table 1. Cementitious building material samples from Radwani House analysed in this report, with
accompanying information.
Elemen %wt oxide
t
Na
Mg Al Si
Sample 20 dark binder
1.91 5.39 29.99
Sample 20 dark binder 0.80 0.83 4.93 33.50
Sample 20 light binder
0.96 3.02 15.23
Sample 20 light binder
0.78 2.97 11.59
nder
Sample 20 finishing
1.25 1.74 4.16 20.96
Sample
1.16 1.33 3.38 22.33
layer 25 binder
Sample 25 binder
0.70 1.59 3.57 24.39
Sample 25 outer crust
1.13
Sample 24B binder
1.11 0.83
0.812
Sample 24B binder
8.09
5.28
Sample 24B gypsum crystal
Sample 24B gypsum agglomeration
Sample 24B gypsum agglomeration
HI1
Description
Ca
40.53
42.76
71.22
74.41
41.19
40.31
42.61
60.25
42.44
61.01
34.46
35.61
34.74
Fe
K
S
5.40
15.71 0.87
2.19
13.83 0.90
1.27
8.44 0.26
1.87
8.14 0.20
3.16 0.77
0.61
2.45
1.82 0.64
1.95 3.20 0.66
3.72
54.14
25.62
50.89
52.74
51.06
HI2
CI
0.96
0.93
0.27
0.22
0.66
0.68
0.63
2.17
2.29
0.65
0.49
1.50
1.61
1.61
Table 2. Detailed SEM-EDS analysis of selected features within cementitious building material samples
20, 24B and 25 from Radwani House with hydraulicity index (HI) calculated as per 1. Vicat (1812) and
2. Piovesan et al. (2009), and cementation index calculated as per Boynton (1980). Elements are
expressed as %wt oxides. Cl omitted.
126
Figure 1. Thin section photomicrographs of cementitious building material samples from Radwani
House, Doha. Qatar analyzed in this report. Image width = 2.9 mm. XP = Crossed polars, PPL = Plane
polarized light.
127
Figure 2. Thin section photomicrographs of cementitious building material samples from Radwani
House, Doha. Qatar analyzed in this report. Image width = 2.9 mm. XP = Crossed polars, PPL = Plane
polarized light.
128
Figure 3. Thin section photomicrographs of cementitious building material samples from Radwani
House, Doha. Qatar analyzed in this report. Image width = 2.9 mm. XP = Crossed polars, PPL = Plane
polarized light.
129
Figure 4. Thin section photomicrographs of cementitious building material samples from Radwani
House, Doha. Qatar analyzed in this report. Image width = 2.9 mm. XP = Crossed polars, PPL = Plane
polarized light.
130
Figure 5. Thin section photomicrographs of cementitious building material samples from Radwani
House, Doha. Qatar analyzed in this report. Image width = 2.9 mm. XP = Crossed polars, PPL = Plane
polarized light.
131
Figure 6. Thin section photomicrographs of cementitious building material samples from Radwani
House, Doha. Qatar analyzed in this report. Image width = 0.75, except E, F = 1.45 mm. XP = Crossed
polars, PPL = Plane polarized light.
132
Figure 7. Scanning electron micrograph of cementitious building material sample 20 (Type 1) from
Radwani House, Doha. Qatar analyzed in this report. This image illustrates the thin surface layer
applied to the sample, as well as the oolite-rich material beneath it. Image captured with
backscattered electron detector that reveals compositionally different regions within the polished
sample.
133
Figure 8. Scanning electron micrograph of cementitious building material sample 20 (Type 1) from
Radwani House, Doha. Qatar analyzed in this report. This image illustrates the darker (right) and
lighter areas (left) within the binder. Image captured with backscattered electron detector that
reveals compositionally different regions within the polished sample.
134
Figure 9. Scanning electron micrograph of cementitious building material sample 25 (Type 1) from
Radwani House, Doha. Qatar analyzed in this report. This image illustrates the thin calcareous crust that
has formed on the sample, as well as the oolite-rich material beneath it. Image captured with
backscattered electron detector that reveals compositionally different regions within the polished
sample.
135
Figure 10. Scanning electron micrograph of cementitious building material sample 24B (Type 2) from
Radwani House, Doha. Qatar analyzed in this report. This image illustrates the heterogeneous, gypsum
rich composition of this material. A well-formed crystal of gypsum can be seen in the center of the
image. Image captured with backscattered electron detector that reveals compositionally different
regions within the polished sample.
136
Figure 11. Scanning electron micrograph of cementitious building material sample 24B (Type
2) from Radwani House, Doha. Qatar analyzed in this report. This image illustrates the heterogeneous, gypsum rich composition of this material. In the center of the image is an area of
gypsum-rich binder. Image captured with backscattered electron detector that reveals
compositionally different regions within the polished sample.
137
12. Appendix B – Radwani House: Preliminary Specialist Report on
Bone
S. Hamilton-Dyer
12.1 Introduction and Methodology
Excavations at Radwani House in the Heritage Quarter of Doha recovered a small
assemblage of animal bones from a variety of deposits. Most of the material is likely to be
occupation deposits of early 20th century date.
Taxonomic identifications were made mainly using the author's modern comparative collections. All fragments were identified to taxon and element with the following exceptions:
ribs and vertebrae of the ungulates (other than axis, atlas, and sacrum) were identified only
to the level of cattle/horse-sized and sheep-sized. Undiagnostic shaft and other fragments
were similarly divided. Any fragments that could not be assigned even to this level have
been recorded to class only. Where possible sheep and goat were separated using the
methods of Boessneck (1969), Payne (1985) and Halstead & Collins (2002). Recently broken
fragments were joined where possible and have been counted as single bones. Tooth
eruption and wear stages of sheep and cattle mandibles were recorded following Grant
(1982). Measurements mainly follow von den Driesch (1976) for mammals and birds and
Morales & Rosenlund (1979) for fish and are in millimetres unless otherwise stated. Withers
height calculations of the domestic ungulates are based on factors recommended by von
den Driesch and Boessneck (1974). The archive includes details of metrical and other data
not presented in the text. The individual records give further details of the condition of each
specimen. These include erosion, gnawing, burning and other aspects of appearance such as
flaking and staining.
12.2 Results
An overall total of 2628 specimens was recorded, over half being of fish. These are quite
small bones and the bulk of the bone by weight is, however, of mammals. The large
mammals are of at least six taxa with caprines the most frequent. Bird bones are relatively
common but of only three taxa. The taxa and number of specimens are listed in Table 1 and
a summary by context in Table 2. The material is generally in good condition in the limey
deposits and few bones are eroded.
12.2.1 Caprines
Both sheep and goat are present with 18% of the 210 caprine bones positively identified as
goat and 7% as sheep; the majority could not be distinguished. Most of the anatomical
elements are represented, with a typical bias in favour of the largest and the sturdiest bones
(Table 3a,b). There are relatively few mandibles and pieces of skull or horncores. These are
resistant and easily recognisable elements, therefore their relative rarity indicates that the
heads were probably used or discarded elsewhere. The most well-represented element is
the tibia at 22 specimens in Phase 1 and a further 11 in Phase 2. These represent at least
five different individuals from each phase. Aging data from teeth is limited to just three
mandibles; a goat kid from context 29 in room 1, a young adult sheep/goat from context 76
in room 9 and a similar one from context 24 in room 7. There are more bones with
epiphyseal fusion data, although still not a large group (Table 4). There are a few bones in
the Phase 1 group from animals that died before about 18 months but most bones are from
animals that were older than this. Fused and unfused elements are almost equal in the 3048 month group. Data for the late fusing elements is restricted to just one fused humerus.
138
The phase 2 data are insufficient for reliable interpretation but do appear to follow a similar
pattern. A female caprine pelvis from context 39 exhibits eburnation of the acetabulum; this
pathology may be age related. The aging data taken together thus indicates a spread of ages
but with most bones being from young adults/sub adults. Measurements were taken on the
fused bones and are given in archive but only one, a sheep metacarpus from 194 in room 6,
was sufficiently complete for an estimation of withers height. This was calculated at 78.2
cm, quite a large animal. Butchery marks were observed on several of the bones. Chopping
through vertebrae indicates axial (or sometimes sub-axial) division of the carcase and the
ribs were often chopped into small sections. Several of the major limb bones were chopped
and/or broken mid-shaft. Most marks were made by a heavy blade but finer knife cuts are
also present on a few of the bones, for example around the proximal humerus where it was
removed from the shoulder.
12.2.2 Other mammals
One bone from a small goat-sized animal, from sample 7 in room 12, is the distal part of a
gazelle humerus. Bones of larger mammals are few; equid remains are restricted to a
deciduous premolar from 194 in room 6, part of an atlas vertebra from 232 in trench 1 and
another deciduous premolar from 162 in trench 2. All three are probably of donkey and in
the case of the one from 162, a naturally shed tooth. Cattle bones are better represented at
ten specimens, although none is a complete bone. These include the rear part of three
mandibles, the one from 162 in Trench 2 being from a neonate calf. Another neonate bone,
in this case a humerus, is present in trench 1 context 222. This bone has been gnawed at
both ends. A gnawed proximal fragment of a femur from context 55 in room six is from a
rather small animal with a caput depth of only 34.2 mm. The distal part of a metacarpus
from 233 in trench 1 is also quite small and these two bones are probably from females. A
proximal fragment of an unfused ulna from context 193 in trench 3 has a rather narrow
profile for cattle but would be extremely large for a young sheep or goat, a small and slim
cow is probably more likely. There are also fragments of a metatarsus, a humerus, and a
chopped astragalus. Several cattle-sized limb shaft, rib and vertebral fragments were also
recorded. The metatarsus fragment from 167 in room 6 is one of the few bones in poor
condition, being weathered and cracked, and was probably exposed on the surface before
being redeposited.
Camel is represented by an upper molar from 223 in trench 1. A piece of rib and an
indeterminate limb shaft fragment are also probably of camel, based on their large size.
No bones of dog or similar carnivores were found but some of the bones do bear gnawing
marks, giving indirect evidence of their presence on site.
A few bones of very small mammals were found, one can be identified as rat but the exact
species is not determinable.
12.2.3 Birds
Bird bones are more frequent than those of the largest mammals; most can be identified as
domestic fowl. These include several bones containing medullary deposits, indicative of
hens (Driver 1982). A few eggshell fragments were also recovered and it is assumed that
these are also from domestic fowl, although they have been recorded only as indeterminate
bird. Other unidentified bird elements, such as phalanges and limb bone fragments, are also
probably of fowl (Table 5). Two fowl ulnae show knife marks, one where cut from the
humerus and the other indicating removal of the lower part of the wing. There are just two
bones of other birds; a pigeon (cf. domestic) tibiotarsus from context 194 in room 6 and a
139
wing phalanx of a bustard from 45 in room 9. Measurements taken on the bird bones are
listed in archive.
12.2.4 Fish
Fish remains are numerous, 1923 specimens. The majority of these are indeterminate fin
rays and other small fragments, as is typical of material that includes sieved samples, but
457 bones were identified to taxon and include at least 16 species. Several of the fish
represented in the material come from families with many similar species, some of the
anatomical elements are diagnostic but others are less easily distinguished to species and in
some cases the bones have been recorded only to family level. Numerically, the most
common remains are of Clupeid fish (herrings, sardinellas and shads) but these small bones
are mainly from sieved samples of a deposit in room 8 and another deposit in room 6 (Table
6). There are about 18 Clupeids in the area (Carpenter et al 1997) and it was not possible to
identify the exact species. The next most frequent remains are of seabreams and are more
evenly distributed across the site in small numbers. At least two different fish are present;
Rhabdosargus sp. and Argyrops sp. Rabbitfish, Siganus sp., are also quite well distributed
while the more numerous bones of mullets came mostly from a deposit in room 9, and these
are probably from a single fish. Other fish that were found in several loci are groupers,
mackerels/tunas and emperors. The groupers and emperors were not identified beyond the
family level, the majority of the mackerel bones match Scomberomorus sp. A further five
Scombrid bones can be identified as being of the related tunas. Several other fish occur as a
few bones only; shark, sea catfish, needlefish, flathead, jack, silverbiddies and parrotfish.
The bone elements found of the more frequent taxa indicate that most arrived as whole fish
(Table 7). Butchery marks are very rare, probably in part because several of the fish species
are small enough to be prepared and eaten with no or minimal butchery. Of the larger fish,
two Scomberomorus vertebrae had been chopped across and a catfish head had been
chopped diagonally across the prefrontal bone. Traces of burning are restricted to one
charred fin ray and a small indeterminate vertebra that has been calcined. The fish have all
been identified from other sites in Qatar and the Gulf area (Beech 2001, 2002, 2003, Russ &
Petersen 2013); seabreams, emperors, groupers and the scombrids are often frequent while
the smaller taxa such as the clupeids and silverbiddies are dependant on sieving for their
recovery.
12.2.5 Other marine taxa
In addition to the fish bones a few remains of crabs were found in room 8 and trench 3, and
a fragment of turtle carapace was recovered from room 12.
12.2.6 Phase comparison
There are some differences in the assemblage between the two phases, although these are
difficult to interpret as the two samples are so different in size. Butchery marks are more
frequent in the later phase but other traces, such as gnawing and burning, are so few in
either phase that any differences may be artefacts of sample size. The general condition of
the bone is better in the later phase and there are fewer fragments that were not identified.
Sheep and goat bones, and also the indeterminate material of this size, are more than twice
as frequent in the later phase. Bird bones (almost all domestic fowl) are also more common.
The larger animals such as cattle are too few to see any true difference but they are also
more common in the later phase. These apparent increases are related to the amounts of
fish and indeterminate mammal fragments, which are fewer in the later group. The better
condition of the later material should result in higher frequencies of the smaller bones,
which is true for the caprines and birds but fish are actually less frequent. The sample size is
much smaller for the later material and, therefore, presence or absence of taxa in individual
small deposits can bias the overall results.
140
12.3 Discussion
This is a relatively small assemblage but important as it gives the first faunal dataset from
the early 20th century development of this part of Doha.
Most archaeozoological assemblages reported from Qatar have been from prehistoric sites
(see Beech 2002). Late Islamic material is currently being analysed from recent and ongoing
excavations at Rubayqa (Russ & Petersen 2013) and Ruwayda (Russ pers. comm.) in northern
Qatar. Results are preliminary but similar taxa have been identified from these sites, though
in varying proportions. At Rubayqa for example the bird bones are mainly of cormorant
rather than domestic fowl. The fish assemblage there is numerically dominated by
seabreams followed by emperors and then groupers, needlefish and sea catfish. There are a
few species that are present there and not in the Radwani assemblage and vice-versa, but
both assemblages are quite small and larger assemblages are needed to confirm these
differences. The most frequent catches reported in recent data include emperors, groupers,
jacks, mullets and mackerels (Sivasubramaniam & Ibrahim 1982, 1983, 1984). These data
are from the commercial fleet and do not include small-scale and personal fishing (Al-Ansi &
Priede 1996), which may explain some differences; rabbitfish for example are mainly caught
using intertidal traps.
The majority of the meat supply, as represented by this small sample of bones, was from
goat and sheep. A few bones are from very young animals but the majority are from full size
but relatively young prime adults, some quite large. The few cattle bones include those of
calves and adults, probably including small females. The calves could represent veal or
mortalities from cows kept for milk; as the calf bones were not butchered and the cattle
bones are few in number their usage is unclear. Some of the cattle-sized rib and vertebral
fragments were chopped, indicating that at least some beef was consumed. None of the
rarely found camel and equid remains has butchery marks and so it is not possible to tell
whether these were eaten. Some of the chicken bones were cut and the medullary deposits
and eggshell fragments indicate that hens were present. The exploitation of wild resources,
other than fish, is restricted to one bone of gazelle and one of bustard. Although neither has
cut marks it is assumed that these found their way to the site as a result of hunting. Marine
resources include a turtle and some crab remains as well as a good variety of fish, all still
common in the Arabian Gulf today. The larger fish were probably prepared as portions after
removal of the head while the small ones were presumably presented entire, perhaps just
gutted. Fish and fishing continue to be an important economic activity and food resource
thoughout the area.
12.4 References
Al-Ansi, M. and Priede, I.G. (1996) Expansion of fisheries in Qatar (1980–1992): growth of an
artisanal fleet and closure of a trawling company, Fisheries Research, 26, Issues 1–2, pp 101111
Beech, M. (2001) In the Land of the Ichthyophagi: Modelling fish exploitation in the Arabian
Gulf and Gulf of Oman from the 5th millennium BC to the Late Islamic Period. DPhil thesis Departments of Archaeology and Biology, University of York, U.K.
141
Bee h, M.
Fishi g i the U aid: a ‘e ie of Fish-bone Assemblages from Early
Prehistoric Coastal Settlements in the Arabian Gulf. Journal of Oman Studies 12: pp 25-40
Beech, M. (2003) The development of fishing in the United Arab Emirates: a
zooarchaeological perspective. In (D.T. Potts, H. Naboodah and P. Hellyer), Archaeology of
the United Arab Emirates: Proceedings of the First International Conference on the
Archaeology of the UAE, Trident Press Ltd., London, pp 289-308.
Boessneck, J. (1969) Osteological Differences between Sheep (Ovis aries Linné) and Goat
(Capra hircus Linné) in (D. Brothwell and E.S. Higgs) Science in Archaeology, London, Thames
and Hudson, pp 331-358
Carpenter, K. E., Krupp, F., Jones, D.A., Zajonz, U. (1997) The living marine resources of
Kuwait, Eastern Saudi Arabia, Bahrain, Qatar, and the United Arab Emirates. FAO species
identification guide for fishery purposes. Rome, FAO
Driesch, A. von den (1976) A guide to the measurement of animal bones from archaeological
sites, Peabody Museum Bulletin 1, Peabody Museum of Archaeology and Ethnology,
Harvard University, Cambridge Massachusetts
Driesch, A. von den and Boessneck, J. (1974) Kritische Anmerkungen zur
Widerristhöhenberechnung aus Längenmaßen vor- und frühgeschichtlicher Tierknochen,
Säugetierkundliche Mitteilungen 22, München, pp 325-348
Driver, J. C. (1982) Medullary bone as an indicator of sex in bird remains from archaeological
sites. In (B. Wilson, C. Grigson and S. Payne) Ageing and Sexing Animal Bones from
Archaeological Sites, British Archaeological Reports (British series), Oxford, 109, pp 251-254
Grant, A. (1982) The use of tooth wear as a guide to the age of domestic ungulates. In (B.
Wilson, C. Grigson and S. Payne) Ageing and Sexing Animal Bones from Archaeological Sites,
British Archaeological Reports (British series), 109, Oxford, p 251-254
Halstead, P. and Collins, P. (2002) Sorting the Sheep from the Goats: Morphological
Distinctions between the Mandibles and Mandibular Teeth of Adult Ovis and Capra, Journal
of Archaeological Science, 29, 545-553
Morales, A. & Rosenlund, K. (1979) Fish Bone Measurements, Steenstrupia, Copenhagen
Mo a N.C. a d O Co o T.P.
Age att i ution in domestic sheep by skeletal and
dental maturation: a study of available sources. International Journal of Osteoarchaeology
4, pp 267-285
Payne S. (1985) Morphological distinctions between the mandibular teeth of young sheep,
Ovis, and goats, Capra. Journal of Archaeological Science, 12, London, p 139-147
Russ H. and Petersen A. D. (2013) Fish and fishing during the late Islamic period at Rubayqa,
northern Qatar: preliminary results. Proceedings of the Seminar for Arabian Studies 43, pp 18
Sivasubramaniam K. and Ibrahim, M.A. (1982a) Common Fishes of Qatar. Scientific Atlas of
Qatar 1, Doha Modern Printing, Doha, Qatar
142
Sivasubramaniam K. and Ibrahim, M.A. (1982b) Demersal fish resources around Qatar, Qatar
Univ. Sci. Bull., 2, pp. 305–352
Sivasubramaniam K. and Ibrahim, M.A. (1983) Pelagic fish resources and their fishery around
Qatar, Qatar Univ. Sci. Bull., 3, pp. 297–327
Sivasubramaniam K. and Ibrahim, M.A. (1984) Fisheries in Qatar—Past, Present and Future
Development Possibilities, Al-ahlia P. Press, Doha, Qatar
Zeder M.A. (2002) Reconciling Rates of Long Bone Fusion and Tooth Eruption and Wear in
Sheep (Ovis) and Goat (Capra). In (Ruscillo D.) Recent Advances in Ageing and Sexing Animal
Bones, BAR Oxbow, pp 87-118
143
Taxon
equid, cf. donkey, Equus cf. asinus
equid, indeterminate, Equus sp.
camel, Camelus sp.
cattle, Bos taurus
sheep/goat, Ovis/Capra
sheep, Ovis aries
goat, Capra hircus
gazelle, Gazella sp.
large mammal, cattle-sized
large mammal, camel-sized
large mammal, sheep-sized
mammal, indeterminate
rat, Rattus sp.
small mammal, indeterminate rat-sized
domestic fowl, Gallus gallus
bustard, Chlamydotis undulata
pigeon, domestic/rock, Columba livia
bird, indeterminate
NISP
1
2
1
10
158
14
38
1
26
2
202
184
1
7
34
1
1
18
turtle, Chelonia
1
shark/ray, Chondrichthyes
herring family, Clupeidae
sea catfish, Ariidae, Arias gigas
needlefish, Belonidae, Tylosaurus sp.
flathead, Platycephalidae, Platycephalus sp.
groupers, Serranidae
jacks, Carangidae, cf. Scomberoides sp.
emperors, Lethrinidae
seabream, Sparidae, cf. Rhabdosargus sp. and Argyrops
sp.
grey mullets, Mugilidae
silverbiddies, Gerreidae, Gerres sp.
parrotfish, Scaridae
tunas, Thunninae, Katsuwonus/Sarda/Thunnus sp.
king mackerels, Scomberomorinae
rabbitfish, Siganus sp.
total identified fish
fish, indeterminate
crab, indeterminate
1
220
2
4
2
27
1
28
61
50
2
4
5
9
41
457
1466
3
Total
2628
144
13. Appendix C – Radwani House: Preliminary Specialist Report on
Botanical Remains
Dr Mary Anne Murray
Botanical samples taken during the Radwani House excavation have yielded evidence of diet,
agriculture, imported foods and other uses of organic material in Old Doha. A three-barreled
flotation machine was used to separate the plant remains from soil samples collected from
the excavation trenches. The plant remains were then sorted and examined under a
microscope.
The botanical study is ongoing but initial results have already yielded dietary data, including
evidence for wheat, barley, date, grape and melon, as well as the presence of possible
medicinal plants (Cordia myxa – the Egyptian plum).
14. Appendix D – Radwani House: Preliminary Specialist Report on
Wood Charcoal
Dr Rainer Gerisch
In May 2013, microscopic examinations were begun on samples of wood charcoal, which
were obtained from soil samples collected during the excavations at the site of ODQ22 in the
city of Doha. They supplement the identifications of plant remains made by M.A. Murray,
University College London. Previous works by the author were done at sites in Egypt, on
material from Syria, and on woods used for traditional ships and boats along the African and
Arabian coast.
The objectives of the study are to determine the taxonomic composition of the samples in
order to contribute to the reconstruction of the past woody vegetation, to provide information on the use and selection of wood-based fuels for domestic fires, assess the human
impact on the shrub and tree cover, supplement the history of locally cultivated plants, and
to find indications for trade relations in cases of imported wood through re-used ship and
building timbers. With regard to the contextual data, possible differences in the charcoal
assemblages of the two occupation phases and between context classes should be observed.
The examinations are complemented by the study of old photographs of Doha and written
sources.
The charcoal remains were recovered by the use of a flotation machine carried out by M.A.
Murray, and in a few cases through hand selection/dry sieving. They were embedded in
occupation deposits and fills of the two domestic buildings and comprise 33 samples, which
also include a small amount of uncharred wood. The size of the charcoal pieces ranges from
medium large to very small and tiny. The stratigraphical assignment shows that two thirds of
the samples were obtained from soil units of Phase 1, one third of those from Phase 2.
Samples with larger contents are: No. 7, Context 39, Room 12; No. 10, Context 83, Room 8;
No. 12, Context 67, Room 8; No. 39, Context 197, Trench 3 from Phase 1 and No. 14, Context
16, Room 9 from Phase 2.
The material is analysed with a Euromex binocular microscope at magnifications of 40 and
80 and with a high power Euromex reflected light microscope (60-500). Pieces are
fractured by hand in the transverse, tangential and radial planes, or already existing breaks
are used to observe the wood anatomical features. The charcoal is compared with
145
anatomical atlases, databases, and a reference collection. For documentation, the pieces of
each taxon in the sample are counted and its volume and weight measured.
Only a few species, which belong to the native flora of Qatar have an arboreal growth form,
most of them also occur as shrubs: Acacia ehrenbergiana, A. tortilis subsp. tortilis, Avicennia
marina, Calotropis procera, Haloxylon persicum, Leptadenia pyrotechnica, Prosopis cineraria,
and Ziziphus nummularia (Lycium shawii, Salvadora persica, Tamarix passerinoides and T.
ramosissima in other countries also as small trees) (Batanouny, K.H., Ecology and flora of
Qatar, 1981; Norton, J. et al., An illustrated checklist of the flora of Qatar, 2009), many other
were introduced in the course of the urban and rural development: fruits trees, Acacia
nilotica, Albizia lebbeck, Bauhinia variegata, Casuarina equisetifolia, Conocarpus lancifolius,
Delonix regia, Dodonaea viscosa, Nerium oleander, Prosopis juliflora, Sesbania sesban etc.
They are distributed together with other woody species and herbs in littoral salt marshes, in
wadis and depressions as well as along roadsides, in gardens, parks and on agricultural land.
Among charcoal analyses on material from the Arabian Peninsula are those by M. Tengberg
(several sites / Oman, United Arab Emirates, Bahrain), R. Gale (Saar / Bahrain), and R. Neef
(Tayma oasis / Saudi Arabia).
Preliminary examinations of the charcoal assemblages revealed 20 taxa, among which are:
Acacia nilotica, Acacia sp., Avicennia marina, Lycium shawii, Palmae, Pinus sp., Prunus sp.,
Rhamnus/Phillyrea sp., Rhizophora/Bruguiera sp., Tamarix sp., and Ziziphus sp. The trees
and shrubs provide fuel, timber, fruits, shade, and sand protection and served for decorative
purposes; well represented is charcoal from acacia wood. Vitis vinifera, which could be
identified from seed remains has not been found yet.
Acacia ehrenbergiana and A. tortilis subsp. tortilis are common trees and shrubs which grow
in depressions and wadis. Tamarix aphylla, T. passerinoides, and T. ramosissima occur on
saline grounds, T. aphylla also around cultivated depressions. Ziziphus nummularia grows
wild in depressions, and Z. mauritania and Z. spina-christi are cultivated. Avicennia marina is
the only mangrove species found on the eastern coast of Qatar; Rhizophora mucronata
occurs along the shores of East Africa, Saudi Arabia, Yemen, Oman, and India, Bruguiera
gymnorrhiza of East Africa and India. Pine species with window-like tracheid to ray pits are
P. nigra and P. sylvestris, which are distributed in Europe and Asia.
146
15. Appendix E – Radwani House: Context List
Context
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
Description
levelling dump
floor
floor
demolition layer
occupation deposits
occupation deposits
fill of [82]
occupation deposits
demolition layer
unstratified finds from Room 8
demolition layer
demolition layer
fill of [14]
cut - probably for modern pipe
traditional wall
wall collapse
rubble layer
fill of [28]
dump / levelling deposit
unstratified finds from Room 9
unstratified finds from Room 10
unstratified finds from Room 12
unstratified finds from Room 14
unstratified finds from Room 7
traditional wall
construction backfill associated with wall 26 in cut [27]
construction cut for wall 27
cut - possible re-plastering event
internal surface
occupation deposits
cement floor
wall render
void - same as (41)
fill of soakaway [53]
fill of [36]
cut - possible re-plastering event
surface
occupation deposits
make up / levelling dump
wall render
crushed limestone floor
floor
fill of 44
robber cut
floors / occupation deposits
traditional wall
traditional wall
fill of 127
Location
R12
R12
R12
RM9
R12
R12
R12
R12
RM9
RM8
RM8
RM8
RM9
RM9
RM8
RM9
RM8
RM12
RM12
RM9
RM10
RM12
RM14
RM7
RM9
RM9
RM9
RM12
RM8
RM8
RM9
RM8
RM8
RM9
RM8
RM8
RM8
RM12
RM12
RM12
RM8
RM8
RM9
RM9
RM9
RM9
RM9
RM8
147
Context
number
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
Description
loose deposit of silty sand
series of floors
primary fill of soakaway [53]
stone lining of soakaway
cut for soakaway
floor
floor - same as 54
cement floor
bedding layer for surface 56
laminated floors
cement drainage feature
traditional wall
wall
cement drainage feature
cut for drain
traditional wall
levelling dump
levelling dump
mixed occupation deposit
wall render
wall
limestone dump
ashy layer
fill of 73
Cut
fill of [75]
cut
occupation deposits
wall render
wall
wall
natural
natural
possible re-plastering cut
floor layer
wall
wall
linear cut
concrete surface
rubble footing
cut
make up / levelling dump
cement floor
rubble make up layer
cement render - drain. Fill of [94]
cut for drain
wall render
wall render
wall
wall render
Location
RM8
RM12
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM12
RM12
RM12
RM12
RM12
RM9
RM9
RM9
RM9
RM9
RM12
RM12
RM12
RM12
RM9
RM12
RM12
RM8
RM8
RM8
RM8
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM9
RM12
RM12
148
Context
number
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
Description
laminated floors
fill of [101]
posthole
fill of [103]
posthole
fill of [104]
small pit cut
sandy layer
ashy layer
fill of [109]
pit cut
natural
threshold?
construction packing
Concrete step
construction packing
void
levelling layer
stone lining of soakaway
construction cut for soakaway (117)
levelling for surface
surface / floor
wall render
silty sand later
make up / levelling dump
fill of [125
posthole or small pit
well
construction cut for well (126)
wall render
unstratified finds form company house
charcoal rich layer
levelling above natural
natural
fill of [134]
modern cut
concrete lined setting for drain
concrete floor
concrete floor
rubble make up layer
concrete floor
bath
concrete floor
wall
well and plinth
drain
well
well
modern gypsum surface
modern made ground
Location
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM8
RM10
RM10
RM8
RM10
RM10
RM8
RM8
RM8
RM8
RM8
CH
RM10
RM10
RM10
RM10
RM10
RM10
RM10
RM10
RM10
RM10
RM10
RM10
RM10
RM10
CH
CH
CH
TR1
TR3
149
Context
number
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
Description
modern made ground
circular planter
made ground - probably modern
modern concrete
modern concrete
made ground - probably modern
compact surface
demolition layer - same as 160
courtyard surface
floor surface
wall
demolition layer - same as 156
dirty surface over natural
fill of [163]
pit cut
modern levelling
modern levelling
Unstratified layer in room 6
loose midden like layer - disturbed
stony fill of soakaway [169]
cut for soakaway
wall
compact surface
wall
deposit between walls 172 and 159
fill of soakaway [169]
concrete base of soakaway [169]
fill of soakaway [169]
compact surface
wall
fill of drain cut [185]
demolition rich levelling layer
cap of drain [185]
stone lining of soakaway [169]
fill of drain in cut [185]
cement drain in cut [185]
Cut for drain
charcoal rich layer
demolition / levelling
fill of [189]
probable re-plastering cut
compact surface
VOID
wall
charcoal rich layer
compact surface
ashy layer
coarse make up layer
ashy layer
natural
Location
TR3
TR3
TR3
TR1
TR2
TR3
TR2
TR1
TR3
TR1
TR1
TR1
TR2
TR2
TR2
TR4
TR4
RM6
RM6
RM6
RM6
RM6
RM6
TR1
TR1
RM6
RM6
RM6
TR1
TR1
TR1
TR4
TR1
RM6
TR1
TR1
TR1
TR4
TR4
TR1
TR1
TR1
TR4
TR4
TR3
RM6
TR3
TR3
TR3
TR3
150
Context
number
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
Description
fill of [200]
posthole
wall
fill of [203]
Large modern construction cut for existing walls
modern disturbance
fill of [206]
shallow pit cut
rubble make up layer
sterile layer over natural
fill of [210]
construction cut for wall 172
mortar surface
black layer
fill of [215]
fill of [215]
posthole
fill of [218]
fill of [218]
posthole
wall
wall
natural
fill of [223]
robber cut
fill of [225]
construction cut
truncated wall
surface
surface
surface
surface
surface
make up / levelling dump
make up / levelling dump
fill of drain cut for [185]
wall
wall render
wall render
floor surface
wall render
wall render
wall render
Location
TR3
TR3
RM6
RM6
RM6
TR4
TR4
TR4
TR1
TR1
TR1
TR1
TR4
TR4
TR1
TR1
TR1
TR1
TR1
TR1
TR1
RM6
TR4
TR1
TR1
TR2
TR2
TR2
TR1
TR1
TR1
TR1
TR1
TR1
TR1
TR1
TR1
TR1
TR1
TR1
RM6
RM6
RM6
151
16. Appendix F: Radwani House: Finds List
Pottery, bone and botanical remains are not included below but are subjected to specialist
study.
left
Find or
for
Context
Conservation
bag
Cons
number
number
number
July
2013
Photo
number
Material
Object
Type
Quantity Weight
Description
1 51
No
IMG_0876 glass
marble
2
12.8 two marbles, one green,
one milky blue
2 51
No
IMG_0877 plastic
other
1
red plastic rawl plug coated
with white plaster
3 51
No
IMG_0878 bone
other
1
chicken bone
4 71
No
IMG_0879 glass
bulk glass
4
3 fragments thin flat clear
glass with whitish patina,
one fragment green glass
5 76
No 62
IMG_0932 glass
bulk glass
7
6 50
No
IMG_0881 glass
bulk glass
3
3 1 fragment of bottle body
or shoulder clear glass with
whitish patina, 1 small
fragment clear glass with
patina, 1 fragment green
glass
7 83
No 54
IMG_0908 glass
bulk glass
16
135 notable pieces include
green bottle top and
bottom, a square pale
green box frag, a clear
turquoise moulded piece in
art deco style
8 51
No
IMG_0883 glass
bead
2
tiny donut shaped bead of
pale opaque blue glass;
plus small fragments of red
tubular bead
9 29
No
IMG_0884 ceramic bead
1
small broken bead,
originally barrel-shaped, of
ceramic or fine brown
stone, glazed or painted in
black
10 3
No 67
IMG_0909 glass
bulk glass
2
21.1 2 clear glass sherds
11 8
No 45
IMG_0910 glass
bulk glass
3
10.7 1 small brown bottle, 1
clear turquoise moulded, 1
white
12 9
No 46
IMG_0912 glass
bulk glass
1
10.4 1 clear flat glass
13 10
No 47
IMG_0913 glass
bulk glass
1
0.7 blue sherd
14 10
No 47
IMG_0915 glass
bangle
1
0.9 blue moulded(?) bangle
frag
15 12
No 37
IMG_0916 glass
bead
2
0.3 1 red glass tubular section,
one opaque orange-red
subcircular bead
16 17
No 48
IMG_0917 glass
bulk glass
1
17 20
No 49-50
IMG_0918 glass
bulk glass
2
11.3 1 clear sherd one moulded
turquoise
18 22
No 59
IMG_0919 glass
bangle
2
1.4 bangle frag of clear glass
with blue backing; plus fine
8.7 2 small fragments of thin
clear glass with whitish
patina
6.3 clear sherd
152
left
Find or
for
Context
Conservation
bag
Cons
number
number
number
July
2013
Photo
number
Material
Object
Type
Quantity Weight
Description
blue bangle frag or poss
segment of curved tubular
bead
19 22
No 59
IMG_0920 glass
bulk glass
5
23.6 mix of clear and green
glass; clear glass bottle
shoulder with gold
20 24
No 58
IMG_0921 glass
bulk glass
1
44.1 green bottle top and neck,
quite uneven
21 34
No 63
IMG_0923 glass
marble
1
5.6 opaque white and green
swirls
22 38
No 66, 69
IMG_0924 glass
bulk glass
3
46.8 2 green, 1 clear with
moulded shape
23 39
No 70
IMG_0925 glass
bulk glass
5
16.1 3 clear, one is top of an oil
lamp?
24 41
No 64
IMG_0927 glass
bulk glass
4
7.7 2 green, one blue, one
brown
25 42
No 65
IMG_0928 glass
bulk glass
2
9.2 one clear flat greenish, one
thin curved "old looking"
26 45
No 51
IMG_0933 glass
bulk glass
29
74.6 mainly thin curved "old
looking" but also small
green base, clear bottle
frags and moulded
turquoise lid(?) fragment
27 47
No 60
IMG_0929 glass
bulk glass
2
16.3 one clear white moulded,
other clear small bottle
with gold
28 66
No 61
IMG_0930 glass
bulk glass
2
48.3 one small clear frag, one
brown bottle with ...PPON
BE... (Nippon Beer)
29 21
No 53
IMG_0934 glass
perfume
bottle
1
16.4 one small clear perfume
bottle
30 242
No
IMG_0935 glass
bulk glass
4
17.2 various enamelled pieces,
and a narrow tubular
screwtop lid
31 107
No 56
IMG_0936 glass
marble
1
32 108
No 44
IMG_0937 glass
bulk glass
7
33 122
No 57
IMG_0938 glass
bulk glass
1
34 160
No
IMG_0939 glass
bulk glass
9 164.5 2 small bottle bases, one
with indented pattern,
turquoise moulded piece,
and other bits
35 160
No
IMG_0940 glass
bangle
1
0.4 fine green glass bangle
36 162
No
IMG_0941 glass
bulk glass
5
6.8 clear thin frags
37 166
No
IMG_0942 glass
bulk glass
5
72.9 large brown bottle frag,
opaque white and green
and clear
54 12
Yes 31
IMG_1025 ceramic stamped
tablet
1
14.8 Broken soft brown clay
tablet, probably originally
arched shape, impressed
on one side with delicate
5.9 clear green marble, a chunk
knocked off
11.3 opaque white plus various
small pieces
7.8 dark olive green
153
left
Find or
for
Context
Conservation
bag
Cons
number
number
number
July
2013
Photo
number
Material
Object
Type
Quantity Weight
Description
design and script. Probably
a Shia prayer tablet from
somewhere like Karbala
55 166
Yes
IMG_1029 stone
56 83
No
57 7
incense
burner
5
260 incised incense burner of
soft fine white limestone.
One large piece and 5
smaller. Main part is
broken at top and bottom,
top part is conical and
incised. Ashes still adhering
to the bowl. Also a sherd of
same from context 171,
included in this finds
record. Recorded weight
refers to main piece; total
with small pieces is 360
IMG_1032 bitumen impressed
bitumen
1
10.2 thin piece of hard bitumen,
impressed on one side with
some kind of fabric or
woven surface
Yes 38
IMG_1035 metal
coin
1
4.2 bronze coin, Saudi 25
halala, with near-legible
date, almost certainly
139..., cf. Faisal bin Abdul
design not Khalid bin Abdul
Aziz so some time between
1970-1976. Date might be
legible after conservation.
58 119
No 41
IMG_1040 metal
object
1
2.5 circular metal alloy disc
with several component
parts. Perhaps the end of a
large-calibre bullet
cartidge, less likely the end
of a battery
59 41
No 36
IMG_1042 shell
button
1
0.4 small mother-of-pearl
button with double
piercing. Nacre visible on
one side and white and
brown ridging not
completely polished off on
the other
60 21
Yes 39
IMG_1044 metal
coin
1
1.9 small corroded bronze coin
61 174
No
IMG_1046 electrics light bulb
1
20.8 broken light bulb with thick
glass stem inside
62 151
No
IMG_1047 electrics light bulb
2
9.1 thick glass stem of broken
bulb, plus the broken
bottom part
63 174
No
IMG_1048 electrics bakelite
casing
1
6.9 brown disc-shaped casing,
probably for light fitting
64 151
No
IMG_1049 electrics bakelite
casing
2
4.8 2 parts of black disc-shaped
casing, probably for light
fitting. Early plastic?
Bakelite? Degraded.
65 300
No 29
IMG_1050 bone
3
bulk bone
52.4 sheep/goat jaw plus loose
tooth and another bone
154
left
Find or
for
Context
Conservation
bag
Cons
number
number
number
July
2013
Photo
number
Material
Object
Type
botanical mixed
botanicals
Quantity Weight
66 30
No 32
67 160
No
IMG_1051 shell
bulk shell
2
99.4 large P. margaritifera shell
plus small bivavle shell too
68 8
No
IMG_1053 shell
bulk shell
1
22.9 P. radiata shell - pearl
oyster
69 232
No
IMG_1054 shell
bulk shell
9
64.1 7 small P. radiata shells pearl oyster, plus another
bivalve and sea-worm cast
70 83
No
shell
bulk shell
4
73.4 mixed shell
71 162
No
shell
bulk shell
9
35.4 mixed shell
72 108
No
shell
bulk shell
3
62.1 mixed shell
73 46
No
shell
bulk shell
22
74 222
No
shell
bulk shell
3
75 194
No
shell
bulk shell
3
15 mixed shell
76 171
No
shell
bulk shell
1
3.6 small P. radiata
77 42
No
shell
bulk shell
1
1.2 shell
78 233
No
shell
bulk shell
1
26.5 shell
79 39
No
shell
bulk shell
26
8 tiny mixed shell from
flotation residue
80 183
No
shell
bulk shell
31
1.8 tiny mixed shell from
flotation residue
81 108
Yes 43
IMG_1055 textile
textile
brown textile, many pieces
82 39
Yes 35
IMG_1056 textile
textile
fine textile with green
metal thread, partly
conserved
83 8
Yes 33
IMG_1058 textile
textile
fine dark brown textile, red
also visible, also green
(from metal thread?)
84 83
Yes 34
IMG_1059 textile
textile
brown textile
85 188
Yes
IMG_1060 textile
textile
yellow and green textile
86 194
Yes
IMG_1061 textile
textile
red and green textile,
woven into a tumb-shaped
lump
87 233
Yes
IMG_1062 textile
textile
88 1
No 73
metal
89 19
Yes 40
IMG_1063 metal
copper
object
1
10.5 curved copper object with
polygonal terminal - like a
bracelet but also has
attachment rings on one
side
90 22
Yes 100
IMG_1064 metal
gun
cartridge
1
10.4 most of a gun cartridge,
perhaps .303
94 30
No 76
metal
bulk metal
95 37
No 77
metal
bulk metal
with nail
96 38
No
metal
bulk metal
bulk metal
9
Description
2.3 mix of wood charcoal,
wood, husks from date
stones(?)
4 tiny mixed shell from
flotation residue
9.1 small P. radiata
belt-shaped strip of textile
1000
guessed weight. Large
corroded iron stake and
pieces
28.4 corroded iron
2
16 corroded iron - 2 parts of a
nail
4.4 corroded iron
155
left
Find or
for
Context
Conservation
bag
Cons
number
number
number
July
2013
Photo
number
Material
Object
Type
Quantity Weight
Description
97 38
No
metal
bulk metal
64.1 corroded iron
98 39
No 78
metal
bulk metal
with nail
99 41
No 79
metal
bulk metal
with nail
77.4 corroded iron, incl nails
100 42
No 80
metal
bulk metal
with nail
53.1 corroded iron, incl nails
101 42
No 99
IMG_1065 metal
102 45
No 81
103 46
1
21 corroded iron - part of a
nail
handle
1
metal
bulk metal
2
No
metal
bulk metal
with nail
2
7.4 corroded iron nail
104 50
No
metal
bulk metal
2
1.4 corroded iron
105 51
No
metal
bulk metal
106 58
No
metal
bulk metal
107 66
No 82
metal
bulk metal
15 corroded iron
108 67
No
metal
bulk metal
7.8 corroded iron
109 70
No 83
metal
bulk metal
19.7 corroded iron
110 71
No
metal
bulk metal
20.2 corroded iron
111 74
Yes 85
IMG_1066 metal
bulk metal
metal
bulk metal
112 76
No
113 83
Yes 103
IMG_1067 metal
padlock
41.6 copper alloy handle for a
drawer or similar
8 corroded iron, perhaps part
of blade plus small nail end
1 corroded iron and copper
6.2 corroded iron
3.7 corroded copper fragments
- apparently a mesh or
textile
37.9 corroded iron
1
114 83
No 86
metal
115 87
Yes 71
IMG_1070 metal
drain
cover
1 2000 copper alloy drain cover,
dia 19, pierced. Weight is
amn estimate
116 35
Yes 42
IMG_1072 metal
nail
6
117 99
No 87
metal
bulk metal
118 102
No 88
IMG_1073 metal
bulk metal
with nail
119 8
No
metal
bulk metal
10.6 corroded iron
120 107
No 89
metal
bulk metal
24.1 corroded iron
121 108
No 90
metal
bulk metal
36.4 corroded iron
122 174
No
metal
bulk metal
with nail
123 151
No
metal
bulk metal
with nail
2
124 156
Yes
IMG_1068 metal
copper
rod
1
125 156
Yes
IMG_1069 metal
bulk metal
with nail
1 169.5 massive iron nail head
126 161
No
??
bulk metal
127 162
Yes
IMG_1074 metal
metal
bulk metal
31.1 copper alloy padlock
gun
cartridge
106.1 corroded iron
33.6 corroded iron nail,
conserved
11 corroded iron
5
58 corroded iron nails
269.5 corroded iron, including at
least 2 tin cans, iron ring,
nail
24.4 corroded iron - nails
2.2 thin copper rod
6.2 corroded iron
1
10.5 heavily encrusted but
whole gun cartridge,
resembles .303
156
left
Find or
for
Context
Conservation
bag
Cons
number
number
number
July
2013
Photo
number
Material
Object
Type
Quantity Weight
Description
128 192
No
metal
bulk metal
213.2 corroded iron
129 167
No
metal
bulk metal
with nail
130 171
No
metal
bulk metal
131 176
No
metal
bulk metal
with nail
132 177
No
metal
bulk metal
16.5 corroded iron
133 183
No
metal
bulk metal
4.2 corroded iron
134 186
No
metal
bulk metal
18.2 corroded iron
135 193
No
metal
bulk metal
7.4 corroded iron
136 194
No
metal
bulk metal
72.1 corroded iron
137 197
No
metal
bulk metal
6.2 corroded iron
138 209
No
metal
bulk metal
0.3 scraps of copper
139 212
No
metal
bulk metal
8.7 corroded iron
140 222
No
metal
bulk metal
141 230
No
metal
bulk metal
with nail
142 232
No
metal
bulk metal
143 233
No
metal
bulk metal
144 119
No
metal
bulk metal
with nail
145 232
Yes
IMG_1077 metal
object
1
23 2 joined cones of metal
alloy, resembling
binoculars. Probably a
whistle
146 232
Yes
IMG_1076 metal
gun
cartridge
1
9.8 most of a gun cartridge cf
.303
147 242
Yes
IMG_1075 metal
padlock
1 117.1 corroded copper padlock
with iron loop
148 242
No
metal
bulk metal
with nail
178 unstratified material - nail,
copper wire, part of an iron
pipe
91 22
No 74
metal
bulk metal
with nail
81.5 corroded iron pieces, incl
some nails
92 29
No
metal
nail
1
0.5 small copper tack embeded
in plaster
93 29
No 75
metal
bulk metal
1
12 corroded iron
38 167
No
IMG_0943 glass
bulk glass
5
39 171
No
IMG_0944 glass
bulk glass
2
0.3 2 tiny frags
40 177
No
IMG_0945 glass
bulk glass
2
0.4 2 tiny frags
41 180
No 113
IMG_0946 glass
perfume
bottle
1
13 one small clear perfume
bottle
42 177
No
IMG_0947 glass
bulk glass
4
23.6 incl cut glass or perhaps
moulded
43 207
No
IMG_0948 glass
bulk glass
1
74 clear rectangular bottle,
"Germany" stamped on
bottom, plus other letters
("DE..E")
44 106
No
IMG_0949 glass
bulk glass
1
0.4 1 greenish frag
15.7 corroded iron - nail
4.3 corroded iron
3
15.5 corroded iron nails
11.6 corroded iron and copper
42 corroded iron - nail
60.7 corroded iron plus copper
loop
29.6 corroded iron
337.7 corroded iron - incl large
flat pieces, nail head
14.3 misc bits, 3 green
157
left
Find or
for
Context
Conservation
bag
Cons
number
number
number
July
2013
Photo
number
Material
Object
Type
Quantity Weight
Description
45 222
No
IMG_0950 glass
bulk glass
1
16.8 moulded or cut, greenish
46 222
No
IMG_0951 glass
bangle
1
47 227
No
IMG_0953 glass
marble
1
48 227
No
IMG_0954 glass
bead
4
0.9 4 small black glass beads
49 227
No
IMG_0955 glass
bulk glass
2
1.8 1 opaque green handle (?)
fragment, one orangebrown sherd
50 229
No
IMG_0956 glass
bulk glass
2
3.4 2 flat, clear
51 230
No
IMG_0957 glass
bulk glass
2
3.5 1 green, one turquoise
52 231
No
IMG_0958 glass
bulk glass
2
53 232
No
IMG_0959 glass
bulk glass
7
1.2 thin moulded glass bangle
with gold-ish coating
7 corroded clear glass marble
with blue swirl inside
1.6 1 clear, one turquoise
18.9 misc bits of different
colours
158
Robert Carter