Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar (Archaeological Heritage of Oman, 10) 9781803274539, 9781803274546, 1803274530

Citation preview

T H E A R C H A E O LO G I C A L H E R I TA G E O F O M A N - V O L . 1 0

PERSISTENT PASTORALISM MONUMENTS AND SETTLEMENTS IN THE ARCHAEOLOGY OF DHOFAR Joy McCorriston

M IN ISTRY OF H E RITAGE A N D TOUR I SM - SULTANAT E OF OM AN 2023

The Archaeological Heritage of Oman

PER SISTENT PASTOR A LISM

Monuments and Settlements in the Archaeology of Dhofar JOY McCORRISTON

Archaeopress Publishing Ltd Summertown Pavilion 18-24 Middle Way Summertown Oxford OX2 7LG www.archaeopress.com Ministry of Heritage and Tourism Sultanate of Oman P.O. Box 200, Postal Code 115 Thaqafah Street Muscat, Sultanate of Oman © Joy McCorriston 2023 Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar (Includes bibliographical references and index). 1. Arabia. 2. Oman 3. Dhofar. 4. Pastoralism 5. Antiquities. This edition is published by Archaeopress Publishing Ltd in association with the Ministry of Heritage and Tourism, Sultanate of Oman. This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective agreements, no reproduction of any part may take place without the written permission of the Ministry of Heritage and Tourism, Sultanate of Oman. Cover image: High Circular Tomb (D104-001) on a limestone terrace spur overlooking an eastern Wādī Dhahabūn tributary (photograph by J. McCorriston and K. Olson). First published 2023 ISBN: 978-1-80327-453-9 ISBN: 978-1-80327-454-6 (e-Pdf)

Contents

List of illustrations and tables

vii

Acknowledgments

xv

Glossary notes 1 A Story and Its Meaning

xvii 1

2 Dhofar’s Pastoral Landscape

11

3 Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

23

4 Cultural Inheritance of the Dhofar Pastoralists

37

5 Building Pastoral Communities in the Neolithic (5000–4500 BCE)

45

6 Tribes in the Bronze Age (3200–1500 BCE)

64

7 Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

96

8 Mobility and Community in the Late Iron Age (300 BCE–500 CE)

112

9 Patterns of Monuments and Settlement Shaped Dhofar

126

Bibliography

131

Index

143

v

List of illustrations and tables

FIGURES 1.1.

D001-001 with the Muḍayy springs in background left (photograph by T. SteimerHerbet).

1

1.2.

D001-001 Individual A (‘Awal). In this and all subsequent images, we masked human bone to provide illustration respecting the dead while showing the archaeological context of excavations (photograph by T. Steimer-Herbet and E. Lagan).

3

1.3.

D001-001 Commingled remains, including neonatal remains (photograph by T. Steimer-Herbet and E. Lagan).

3

1.4.

Sheikh Suha‘il Al-Rujdi (left) and an aide visit an HCT excavated by Drs. Tara Steimer and Kimberly Williams (both right) (photograph by M. Harrower).

5

1.5.

View of the springs at Muḍayy from D001-001 HCT (photograph by T. Steimer-Herbet).

6

1.6.

D001-001 HCT on plateau before excavation (photograph by J. McCorriston).

8

1.7.

Diagram of a (tribal) segmentary lineage kinship system in which contingent and contextual affinity is measured by descent from common ancestors, usually through patrilineal lines. Triangles denote male descent; females not shown (image by K. Olson).

9

2.1.

Image map of four major ecosystems in Dhofar: coastal plain, wooded escarpment, grassland plateau, and Nejd desert (image by by K. Olson).

11

2.2.

Cloud forest of Terminalia dhofarica across the seaward-facing Dhofar escarpment (photograph by J. McCorriston).

12

2.3.

Parkland of the Dhofar plateau with modern short grassland (photograph by J. McCorriston).

13

2.4.

Native palms in the sandy channel of Wādī Dhahabūn (photograph by J. Everhart).

13

2.5.

Travertine deposits, like speleothems, form with the evaporation of water that has seeped through limestone, carrying with it an oxygen proxy signal of temperature and a carbon signal of vegetation (photograph by J. McCorriston and K. Olson).

15

2.6.

Image map of enhanced Early-Middle Holocene monsoon precipitation and paleolacs in Southern Arabia (precipitation data from Brown et al. 2018; Fordham et al. 2017) (image by K. Olson).

16

2.7.

Archaeological evidence of anthropogenic fires can be seen embedded in sediment terraces in Yemen’s highland deserts, an ecosystem comparable to the backslope of Dhofar’s mountain plateau and near Nejd. No comparable sediment terraces exist in Dhofar to preserve such traces of former activity, but the tall grasslands themselves suggest a comparable antiquity to the practice of burning. Here the terrace surface has eroded to reveal a broad swath of grey ash and micro-charcoal, center (photograph by J. McCorriston).

19

2.8.

Cattle grazing short grassland on the Dhofar plateau (photograph by K. Pustovoytov and K. Olson).

20

vii

viii

2.9.

Hypothetical effect on human transhumance from reduction in forest, fog precipitation, and groundwater recharge. Note the reduced transhumance of (cattle) herders into the Nejd with the replacement of deep-rooted trees by grassland on the plateau. Adapted from Hildebrandt and El-Tahir 2006 (image by C. Hickman and E. Lagan).

21

3.1.

Archaeological survey team recording a circular tomb. Left to right: Sarah Fargo, Matthew Senn, Ali Ahmad Al-Kathīrī (photograph by J. Everhart).

23

3.2.

One of many triliths on a rock terrace, seen across a shallow tributary near Hanūn (photograph by M. Harrower and K. Olson).

24

3.3.

Muruźәd (Jibbāli), which appear very like the madhāby (Arabic) grilling hearths and sacrificial commemoratives in Ḥaḍramawt, Yemen. These examples occur in the Wādī Ḥalūf near D014-003 (see Chapter 6) (photograph by M. Harrower).

25

3.4.

High resolution images surveyed across Dhofar (Image 5 not accessed) (image by K. Olson)

29

3.5.

Development of Computational Algorithms for Automated Detection. The process examines each window of an image with a series of filters; the statistic JB measures the relative difference in intensity between the inner and outer windows; the next filter (Jv) calculates the normalized difference vegetation index (NDVI), and so forth (details in Schuetter et al. 2013) (image by J. Schuetter and K. Olson)

29

3.6.

Using GPS rover to register ground control point (GCP) for image rectification in Wādī Haylāʾ, south of Muḍayy (photograph by J. McCorriston).

30

3.7.

To develop automated detection software, the team visited many computer-generated locations that were false positives for monuments. By ground-truthing these sites, the team refined software accuracy (photograph by J. McCorriston).

31

3.8.

Matthew Senn and Joy McCorriston documenting a likely Neolithic platform near Andhur (photograph by J. Everhart).

31

3.9.

Aerial view of D113-001, a prominent, immense, flat-topped boulder beside the track through one of Wādī Dhahabūn’s southernmost major tributaries. The elaborate images on its surface can only be seen from above, and there is no nearby overlook. This was a landmark known and shared as cultural memory. It is unlikely that all markings were produced at one time (image by W. AbuAzizeh and T. Everhart).

34

4.1.

Al-‘Alīy herder with goats in Southern Yemen (photograph by J. McCorriston).

37

4.2.

Neolithic points from surface collections at Bā Mashnayq, Muḍayy (third from left) and in Wādī Ghārah, a tributary to Wādī Ghadūn (photographs by M.Senn and E. Lagan).

39

4.3.

Neolithic ring of skulls from sacrificed cattle at Shiʿb Kheshiya, Yemen. The skulls were inserted fresh into soft mud. Later, visitors erected a commemorative stone platform nearby (photograph by M. Harrower).

40

4.4.

Painted images on the rock walls of natural rock shelters in Dhofar’s coastal mountains often include images of domesticated dogs, camels, cattle, and goats. These are at the site of Mthbon, southeast of Jibjat (photograph by J. McCorriston and K. Olson).

42

4.5.

Traditional style cattle byre in contemporary Dhofar mountain village. Confined animals produce a large accumulation of dung, useful as fertilizer or fuel (photograph by J. McCorriston).

43

5.1.

Image map of Neolithic platform monuments located by the AHSD Project survey team (image by K. Olson).

46

5.2.

Bedrock terrace at D028 overlooking vegetation fed by one of Muḍayy’s springs; note the stone monuments on the terrace and accumulated sandy sediments in the drainage below (photograph by J. Everhart).

46

5.3.

Finely knapped bifacial points from the surface at D028 (image by M. Senn and E. Lagan).

47

5.4.

Neolithic platform sectioned in one quarter, with exposed fill and large perimeter slabs supported by interior blocks. A second large monument is visible in right background (photograph by T. Steimer-Herbet).

48

5.5.

Found in the lower fill of D028-001, a pierced shell of Impages hectica, a marine shell likely used as an ornament (image by C. Heyne, J. McCorriston and E. Lagan).

48

5.6.

Overview of Neolithic platform D028-001 with excavated northwest quadrant (photograph by T. Steimer-Herbet).

50

5.7.

Plan of Neolithic platform D028-001 (image by T. Steimer-Herbet and K. Olson).

50

5.8.

Hard Umm er-Radhuma limestone layers harbor rock pools at Hanūn (photograph by J. Everhart and K. Olson).

51

5.9.

D038-003 on a low terrace; a trilith in the foreground may have robbed stone from the less preserved Neolithic monument under excavation in the rear (photograph by M. Senn).

52

5.10.

Plan of Neolithic platform D038-003 (image by T. Steimer-Herbet, J. McCorriston and K. Olson).

52

5.11.

Twin coral beads from the fill of D038-003 (image by C. Heyne, J. McCorriston and E. Lagan).

53

5.12.

Deliberate stone arrangement several meters to the north of D038-003 (photographs by J. McCorriston and K. Olson).

54

5.13.

Hearth D103.002 at the east end of the terrace on which sits D103.001. Although the hearth is of indeterminate date, it is one of few signs of subsequent prehistoric visits on this terrace (photograph by J. McCorriston),

56

5.14.

D103-001 looking east before excavation note alignment of upright boulders facing east (photograph by J. McCorriston).

56

5.15.

D103.001 viewed from the east, after excavation. Note the shattered limestone betyl in the left foreground and knife sharpening scars on the central upright (photograph by J. McCorriston).

57

5.16.

D103.001 east-facing interior profile viewed from the east (photograph by J. McCorriston).

57

5.17.

D106-001 SCAB with east-facing alignment of larger uprights. Note the meter-long shaped stone fallen at southeast (orthophoto by W. AbuAzizeh and T. Everhart).

58

5.18.

D106-001 SCAB with east-facing alignment of larger uprights. Note the betyl stone with shaped, tapered end in foreground, approximately 1 meter length (photograph by J. McCorriston).

59

5.19.

Pedagenic carbonates (white, left) formed by the interaction of soil biota and limestone upright after its placement in the monument. Because the living organisms depositing carbon in soil carbonates postdate the monument construction, a radiocarbon age would secure a terminus ante quem for the construction (photograph by J. McCorriston).

59

5.20.

D112-001 viewed from the south (photograph by J. McCorriston).

60

ix

x

5.21.

D112-001 SCAB after clearing to expose rocks in-situ. Note the shallow excavation 1 meter wide in the northwest against a perimeter stone (orthophoto by T. Everhart).

61

5.22.

D027-001, D027-003, and D027-002 at Muḍayy (photograph by J. McCorriston).

62

6.1.

Matthew Senn documents the perimeter of an HCT monument using kinetic highprecision GPS (photograph by J. Everhart and K. Olson).

64

6.2.

Image map of Bronze Age tombs documented in survey and excavation (image by K. Olson).

65

6.3.

D001-001 after excavation (photograph by T. Steimer-Herbet and K. Olson).

66

6.4.

Chamber fill of D001-001 with minor collapse of corbelled roof (photograph by T. Steimer-Herbet and K. Olson).

67

6.5.

Two beads from the base of D001-001 chamber (photographs by C. Heyne and E. Lagan).

68

6.6.

D013-001 overlooking Wādī Haylāʾ (photograph by K. Williams and J. McCorriston).

69

6.7.

Chamber of D013-001 at the end of excavation. Note the dry-wall technique, here consisting of an inner facing and outwardly faced or buttressed with heaped stone (photograph by J. Williams and E. Lagan).

69

6.8.

Articulated left tibia, fibula, and foot of the male buried in D013-001 (image by K. Williams and E. Lagan).

70

6.9.

Right maxilla with erupted third molar from the adult male buried in D013-001 (image by K. Williams and E. Lagan).

70

6.10.

Mother-of-pearl ornament from D013-001 (image by C. Heyne, J. McCorriston and E. Lagan).

71

6.11.

Section D013-001 (image by K. Williams, T. Steimer-Herbet and K. Olson).

71

6.12.

D013-002 Overlooking Wādī Haylāʾ (photograph by K. Williams and J. McCorriston).

72

6.13.

Chamber of D013-002 after excavation (photograph by K. Williams).

72

6.14.

Pierced Conus shell recovered from D013-002 chamber (image by C. Heyne and E. Lagan).

73

6.15.

D013-002 Plan and section (image by K. Williams, T. Steimer-Herbet and K. Olson).

73

6.16.

Overview of D014-001 (fore and center) before excavation. Behind and left is D014-002 and D014-003 (photograph by C. Heyne and J. McCorriston).

74

6.17.

Chamber of D014-001. The original desert reg terrace surface has been cleared to bedrock at the base of the chamber (photograph by C. Heyne and J. McCorriston).

75

6.18.

D014-001 Plan and elevation (image by C. Heyne, T. Steimer-Herbet and K. Olson).

75

6.19.

D014-002 Plan and elevation (image by C. Heyne, T. Steimer-Herbet and K. Olson).

76

6.20.

Long bone shaft at the base of the chamber in D014-002 (image by T. Steimer-Herbet and E. Lagan).

77

6.21.

D033-001 Documented by survey team (photograph by J. Everhart).

77

6.22.

D033-001 Plan and elevation (image by T. Steimer-Herbet and K. Olson).

78

6.23.

D036-002 in overview with D036-001 in the background (photograph by T. SteimerHerbet).

79

6.24.

D006-001 before excavation (photograph by J. McCorriston).

80

6.25.

D104-001 viewed from southwest (photograph by J. McCorriston and K. Olson).

81

6.26.

D104-001 base of chamber with mother-of-pearl ornament in situ (photograph by J. McCorriston).

82

6.27.

View of sealed chamber (center) D022-005 after cleaning the monument, probing construction, and before excavation of the chamber (photograph by T. Steimer-Herbet and K. Olson).

84

6.28.

D022-005 Plan and elevation (image by T. Steimer-Herbet and K. Olson).

84

6.29.

Ḥalūf tomb D014-003 after excavation (photograph by J. McCorriston and K. Olson).

85

6.30.

D014-007 Unexcavated Ḥalūf monument (photograph by J. McCorriston).

86

6.31.

Excavated chamber in Ḥalūf tomb D014-003 (photograph by J. McCorriston and K. Olson).

87

6.32.

Detail of undressed stone work, Ḥalūf tomb D014-003 (photograph by J. McCorriston and K. Olson).

87

6.33.

Copper dagger recovered from inner face of Ḥalūf tomb D014-003 (and shown in situ) (image by C. Heyne, J. McCorriston and E. Lagan).

88

6.34.

D014-003 Plan with hearths aligned to the west (image by J. McCorriston, T. SteimerHerbet and K. Olson).

88

6.35.

Heat-treated agate bead from the upper fill of D014-003 (image by C. Heyne, J. McCorriston and E. Lagan).

89

6.36.

View of D014-003 (background) from excavated D014-002 (foreground). Note the alignments of cobble rings in front of Ḥalūf tomb D014-003 (photograph by J. McCorriston).

90

6.37.

D001-006 on horizon viewed from the south; black cobble heaps align between tomb and cliff edge (photograph by T. Steimer-Herbet).

91

6.38.

Ḥalūf tomb D001-006 after clearing one half of the exterior, viewed from the south (photograph by T. Steimer-Herbet and J. McCorriston).

92

6.39.

Ḥalūf tomb D001-006 Plan and elevation (image by T. Steimer-Herbet and K. Olson).

92

6.40.

View of the chamber of Ḥalūf tomb D001-006 after excavation and exterior cleaning (photograph by T. Steimer-Herbet and K. Olson).

93

6.41.

Exterior detail, Ḥalūf tomb D001-006 (photograph by T. Steimer-Herbet).

94

7.1.

Image map of Shakeel and Halqoot (image by K. Olson).

96

7.2.

Occupation site D114 viewed from the east (photograph by J. McCorriston).

97

7.3.

D114 architecture and graves. Excavated sites indicated in red (image by A. Buffington, W. AbuAzizeh and T. Everhart).

98

7.4.

D069 Map of architectural sites (image by M. Senn and B. Baaske).

98

7.5.

D114-016 Boulders form a corral perimeter, possibly once topped by brush (photograph by W. AbuAzizeh and K. Olson).

99

7.6.

D114-004 A. Interior of a room with paver floor and dry-stone facing to the interior. Steps lead from a courtyard accumulation to the interior living surface (photograph by W. AbuAzizeh and K. Olson).

100

7.7.

D114-004A South section (image by A. Buffington, W. AbuAzizeh, B. Baaske and K. Olson).

101

xi

xii

7.8.

D114-001 corral and D114-004 attached dwelling in overview; lintels-supported entrance intact and stepped-down entrance highlighted (image by W. AbuAzizeh, J. McCorriston and B. Baaske).

102

7.9.

Beads and ornaments from D114 (image by J. McCorriston, C. Heyne and E. Lagan).

103

7.10.

Radiocarbon calibrations from D069 Shakeel and D114 Halqoot (image by J. McCorriston).

104

7.11.

D069-002 Trench A northeast section (image by J. McCorriston and B. Baaske)

105

7.12.

D069-002 House plan after excavation (image by J. McCorriston, B. Baaske and K. Olson).

106

7.13.

D069-002 Overview after excavation. Note construction style matches D114 houses (photograph by L. Proctor and K. Olson).

107

7.14.

D069-001 Quad A Platform or partial re-pavement with house entrance at upper left (photograph by J. McCorriston and K. Olson).

107

7.15.

Handstone abandoned on the floor of D069-004 (image by M. Senn and E. Lagan).

108

7.16.

High circular tombs and houses or corrals on the southern slope of Wādī Kharshīt (photograph by J. McCorriston).

109

7.17.

Images of camels in a rockshelter beside a Late Iron Age site; nearby images show camels, men with swords and shields, and perhaps a ship (photograph by J. McCorriston).

111

8.1.

Trilith schema after Yule 2014:73 (image by E. Lagan).

114

8.2.

Image map of trilith locations surveyed. Details in Harrower et al. 2014 (image by K. Olson).

115

8.3.

Excavation of trilith hearth D043-004a (photograph by C. Heyne).

116

8.4.

Trilith D005-004 (photograph by Joy McCorriston).

117

8.5.

Within recent memory, bedouin used dry walls to seal niches in the cliff face, behind which multiple burials may be found (photograph by J. McCorriston).

118

8.6.

D102-003 looking north to D102-001 (photograph by J. McCorriston).

118

8.7.

D102-001 Excavation to the original ground surface on which D102-001 was constructed. Note the thin pedogenic carbonates on the underside of the stone (photograph by J. McCorriston and K. Olson).

119

8.8.

Excavation beside D102-005 reveals insufficient carbonate formations on underside of perimeter cobbles (photograph by J. McCorriston).

120

8.9.

Hearth beside D102-005 sectioned (photograph by J. McCorriston and K. Olson).

121

8.10.

Syntax of a trilith: platform (rear), 4 small boulders squared (mid), and hearths in front (foreground) (photograph by J. McCorriston).

121

8.11.

Tombs D001-005 (foreground) and D001-004 before excavation (photograph by T. Steimer-Herbet).

122

8.12.

D001-004 Plan and section (image by T. Steimer-Herbet).

123

8.13.

Flexed adult male burial in D001-004 (image by K. Williams and E. Lagan).

124

8.14.

D001-005 Adult female buried in D001-005 (image by K. Williams and E. Lagan).

124

8.15.

D001-005 Plan and sections (image by T. Steimer-Herbet).

125

9.1.

Chronology of monuments in Dhofar. Yellow bars are labelled by tomb type; the gray bars represent (left to right) Ḥalūf tombs, Wall tomb, Chamber cairns, and Islamic graves (image by J. McCorriston).

127

TABLES 3.1.

Excavated Monuments

33

5.1.

Bayesian analysis (Chronomodel software) with 210-year marine calibration per Southon et al. 2002, Saliège et al. 2005.

54

5.2.

Volume mean and variance by monument type.

63

6.1.

Survey results 2009-2018. Bronze Age tomb types, numbers per type, and chronology in Dhofar.

66

xiii

Acknowledgments

To Thuwaiba Al-Riyami and ‘Ali Aḥmad Al-Kathīrī in gratitude for your friendship. Often have I sat in the offices of the Ministry of Heritage and Tourism in Muscat, gazing at the hazy horizon where heavy tankers ply the territorial waters. Endlessly, they stream slowly by where sea meets sky, funneling petroleum flow to a thirsty planet. Oman is at the pulse of the world, extracting copper, dates, stallions, incense, cloves, and oil across a rich maritime trade that stretches far into prehistory. And all throughout, there have been the persistent pastoralists of the interior, ancestral Omanis whose mobile lives and oral traditions left few traces in the desert lands. These pastoral folk herded cattle, goats, and camels. This other Oman sometimes engaged with maritime traders and oasis farmers, and yet at other times, the pastoralists eschewed them, living lives of relative isolation. This book offers some results of ten years’ field and analytical research into the lives of pastoral Arabians. Fundamentally, it represents the work of multi-talented, multi-disciplinary, multi-cultural teams and multi-national teams. Furthermore, the results reported here reflect the collaborations and contributions of many. While this book presents some new ideas that are mine, it presents data and results previously reported under joint authorship with the colleagues and students who produced them. To them I offer both acknowledgement and profound gratitude. Thank you, Michael Harrower, Matthew Senn, Tara SteimerHerbet, Kimberly Williams, Jennifer Everhart, Catherine Heyne, Mas‘ūd Al-Hādhari, Mas‘ūd Al-Kathīrī, ‘Ali Ahmad Al-Kathīrī, Ali Salim Al-Mashaanī, Ali Mehri, Prem Goel, Jihaye Park, Jared Schuetter, Dorota Grejner-Brzezinska, Louise Martin, Kyle Olson, Lucas Procter, Anne Skidmore, Mark Moritz, Ian Hamilton, Lawrence Ball, Sarah Ivory, Abigail Buffington, Tim Everhart, Konstantin Pustovoytov, Wael AbuAzizeh, Annalee Sekulic, Anna Berlekamp, Sarah Fargo Evinsky, and Jean-François Saliège. This work has been supported logistically, financially, and with infinite patience by Oman’s Ministry of Heritage and Tourism (formerly Heritage and Culture). I honor the leadership of former Minister of Heritage and Culture, H.M. Sultan Haitham bin Tariq Al-Said, and former Undersecretary for Heritage, H.E. Sultan Al-Harthy, who first authorized this research. In Muscat, I particularly thank H.E. Salim M. Almahruqi, Minister of Heritage and Tourism, for commissioning this book and for his patience and support through the fieldwork that preceded it. I also speak for our team in appreciation of Biubwa Al-Sabri, Rahma Al-Farsi, Sultan Al-Bakri, Khamis Al-Asmi, Sumaya Al-Busaidi, Mohammad Al-Waili, Khalid Awad Al-Swafi, Ibrahim Al-Maqbali, Khalil Masood Al-Nadabi, Sheikha Khalifa Al-Rasbi, and Qais Al-Mazrouai. In Salalah, we thank Salim Al-Amry, Abdul Aziz Ahmed Suhail Al-Mushaikhi, and Khalid Al-Abry, and especially appreciate all the help and hospitality from Ali Salim Al-Mashaanī. I also recognize the many unnamed Omani hosts and guest workers who supported us, from tea to trenches. This research has been principally funded by the United States National Science Foundation (Grants BCS-0624268, BCS-0957179, BCS-1617165), Ministry of Heritage and Culture (now Heritage and Tourism) of the Sultanate of Oman, National Geographic Society, American Institute of Yemeni Studies, and The Ohio State University. The opinions, findings, and conclusions or recommendations expressed are those of the author and do not necessarily reflect the views of the National Science Foundation. xv

Finally, I thank the colleagues who have inspired me in Oman and beyond, especially Juris Zarins, Jeff Rose, Paul Yule, Rémy Crassard, Andrew Anderson, Darach Lupton, Annette Patzelt, Birgit Mershen, and Tony Miller. Kyle Olson, Emma Lagan, and Benjamin Baaske did invaluable work on illustrations.

In appreciation for all they taught me of Oman’s history and cultural traditions, I dedicate the book to Thuwaiba and ‘Ali.

Glossary notes

There are challenges to the consistent transliteration of Arabic place names and terminology into English/ Roman orthography. In this book, there are also place names in modern South Arabian languages—Shehri, Mehri, and the vernacular Jibali. To guide the text I used the following rules: Where a transliteration already exists in common use in archaeological and geological literature, I use it, even if the transliteration follows neither Library of Congress nor Deutsche Morgenländische Gesellschaft convention. For example, “Wadi Suq” and “Umm er-Radhuma.” Where researchers use non-standard transliteration throughout other books in the AHO series, I have opted for consistency, thus, “Nejd.” Where transliteration appears on official signage and maps in Oman, such as road signs leading to “Ayboot,” or “Mudayy,” I have opted for that use. In the case of “Mudayy,” the Arabic clearly indicates that “Muḍayy” represents the correct consonant. Likewise, I retain transliterations from road signs using Mehri, Shehri, and Jibali place names. Where a person’s name appears in published form and especially as an author, I retain the transliteration as published. In other cases, including the names of wadis and mountains, I follow Library of Congress transliteration rules.

xvii

Chapter 1

A Story and Its Meaning

As one crests a rough track near the springs at Muḍayy, the desert pavement comes into view, a vast mosaic of taupe bedrock and black, dolomitic limestone, polished by sun and wind and broken only by the sentinels of long-dead ancestors. These sentinels are cairns, dry-stone walls that served an ancient purpose. Away from one – or perhaps toward it – leads an enigmatic line made of low heaps of black stone. Each heap is one pace further, and the line falls away at a precipice overlooking the wide Wādī Ghārah. Behind you is another cairn, this one topped by a silver sail and emitting periodic puffs of desert dust. Drawing closer one can see it teems with archaeologists. A dusty pair of workmen in shalwar-khameez lean rhythmically to and for clasping small screens, and the rasping of metal on rock accompanies patter in Bengali alongside sylvan Jibali and ineloquent English. Under the sunshade a curious covey of visitors watches the expert, a small and quick woman in a white headwrap. She is removing human bone, calling out to her graceful peer the bone elements, their positions, and the details of their ages, sexes, and life sufferings. In a short time, she and her colleague have lifted and documented the burials that lay undisturbed here for nearly five thousand years.

Figure 1.1. D001-001 with the Mudayy springs in background left (photograph by T. Steimer-Herbet).

1

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Archaeological desecration demands a story, that of these long-ago Dhofari bedouin and their persistent pastoralism. The excavation of this tomb, D001-001, brought to light the stories of those buried therein and some details of the kin who buried them. By beginning with their lives and experiences, we honor their places in Dhofar’s pastoral prehistory. And by telling their stories, this book populates the archaeological surveys and excavations documented in subsequent chapters of data, analysis, and data patterns in the archaeological record. All told, this record describes what we know of Dhofar’s pastoral peoples from the monuments and settlements they left behind, and it explains why pastoralism persisted for thousands of years at the core of Dhofari life. “Individual A” (let us call him “ʿAwal”) was the first person buried in D001-001, sometime between 6000-5500 years ago. He was elderly when he died at about age 55, and he had lived a painful old age after a tough life. He’d broken a few toes and survived a serious back injury, albeit with consequences. ʿAwal couldn’t move for some time, and part of his spine was fused below his neck, attesting to the pain that some movements caused him. He also had bone spurs of old age on his vertebrae (which probably did not hurt) and had lost a molar in his jaw long before death. ʿAwal had none of the dental caries and decay that come with a diet of carbohydrates (cereals and dates), which he clearly did not eat regularly. Yet his teeth troubled him. Without the attentions of a dentist, the pain from a bone abscess under another jaw tooth was deeply distracting, and he bore this pain in his last years, even to his death. ʿAwal had probably made his home near the twin springs at Muḍayy. True, there were other springs not far away in the Wādī ʿAybūt system, but his goats and perhaps the few cattle he tended needed frequent watering, and the springs kept them tethered close for most of the year. Only when summer mist-drip or the rare tropical cyclone shed substantial moisture inland could ‘Awal and his animals forage southwards down the Wādī Ghārah toward the coastal mountains. In his later years, he probably walked less, and there were others nearby who were faster at catching game and chasing away predators. But old ʿAwal carried with him the wisdom of years, and his tribe relied on his memories to know who and what and when and where. At his death, a small group gathered, surely just about everyone who used the springs and the limited browse in this dry region. Perhaps there were two or three families in all. They wrapped his body into a fetal bundle until it stiffened, and then they carried him up to the skyline in view of the green spring. They laid him on the bare desert pavement. Around him they built a strong stone tomb with a ring wall of flat-laid rough limestone slabs, the kind one could prise from nearby bedrock and shift in teams of three or four. They faced the inside and outside with these slabs, then packed the wall core with cobbles and pebbles. Someone in the group had done this work before, copying and teaching a style learned in another tomb-building event, perhaps for a relative near another spring or stream channel. Tombs like ʿAwal’s were beginning to appear on horizons across the mountain back-slopes and coastal ridges of Arabia, using new building styles learned from one family to the next. By corbelling the interior face, ʿAwal’s kin narrowed the top of the tomb, covering the last gap with a capstone. Then we may simply guess that they slaughtered a goat in ʿAwal’s honor and to share a hearty meal at the end of their labors. At least three hundred years passed by. No one alive still remembered ʿAwal directly of course, but some memory of his importance lived on, retold at feasts. After all, nearly everyone who buried him was likely related to him, and his tomb had been on the horizon all the lives of his many times great-grandchildren, who traced their lineages back to him and knew which of their cousins was permitted in marriage. He was an anchor for people on the move. Through ʿAwal, or The Ancestor he had become, they knew this as a place they belonged.

2

A Story and Its Meaning

Figure 1.2. D001-001 Individual A (‘Awal). In this and all subsequent images, we masked human bone to provide illustration respecting the dead while showing the archaeological context of excavations (photograph by T. Steimer-Herbet and E. Lagan).

Figure 1.3. D001-001 Commingled remains, including neonatal remains (photograph by T. Steimer-Herbet and E. Lagan).

3

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

“Individual B” (let’s call him “Bir’am”) died in his prime near the Muḍayy springs. Perhaps a wolf pack had grown bolder in attacking goats and children, and Bir’am had faced down the wolves and lost. Although his throat was torn out and his belly eaten, his corpse was otherwise untouched. (Let’s suppose that) his kin opened The Ancestor’s tomb. They found ʿAwal’s dessicated corpse not yet completely disarticulated. Carefully they pushed aside his remains, setting a stone boundary around him and covering what remained with rough slabs. Nearby they placed Bir’am on the soft bed of desert loess that had seeped between the chinking in the winter windstorms. Maybe food was scarce that year. (Possibly) Bir’am’s mate, already pregnant at his death, died in childbirth, but people said there wouldn’t have been enough to feed her and the infant anyway. She died far from home, but in another season they retrieved the bones of “Individuals C and D” (let’s call her “Chafat” and her fetus “Delot”), bundled them in a leather carry-cloth and put them into The Ancestor’s tomb with Bir’am. On the way back, they passed the place where Bir’am’s father (“Individual E”) El-ram was left. They gathered up his remains also, and with Chafat and Delot, El-ram went into The Ancestor’s tomb, their remains and their ornaments commingled over that of their kin. Time passed, and our understanding of The Ancestor’s tomb is even less clear. When the old people retold the stories, when people stopped at the spring, they recalled something, and they pointed to The Tomb. Was there ever an Ancestor Eve, before the Ancestor Star appeared? Or a Feast Day of The Ancestor? (We will never know) On some occasions people gathered. True to tradition, they sacrificed a goat and placed a meaty offering inside The Tomb. It lay there in darkness, gently cushioned by the soft desert loess and dusting over in the winter storms. And that is how we found them – sealed tomb, goat bones, commingled remains, and two partially articulated male skeletons at the bottom. Behind the Story: Archaeology and the Lives of Pastoralists Archaeology and the Lives of Pastoralists The story told here is based on archaeological data from excavations conducted in 2009 at a 3rd millennium BCE high circular tomb (henceforth HCT). This tomb, D001-001, was built on the bedrock plateau overlooking one of Muḍayy’s two springs. Tara Steimer and Kimberly Williams conducted the excavations, aided by Masʿūd Al-Kathīrī, Masʿūd Al-Hādhari, and two Bengali workmen (Williams et al. 2014; Chapter 6 this volume). Theirs was the first of many small-scale stone monuments excavated and documented through systematic archaeological survey between 2009 and 2019 (McCorriston et al. 2014, Everhart et al. 2014, Harrower et al. 2014). This book is about what we know. Although we will never know the names of people buried in D001001, nor the name of the tomb itself, archaeological research has provided important insights into the lives they lived. Through a partnership between Oman’s Ministry of Heritage and Culture (now Heritage and Tourism) and foreign archaeologists, we present new evidence of the lives of Dhofar’s ancient inhabitants – the pastoral people who herded cattle and goats in the green enclaves at the edge of Arabia’s desert landscapes. Even from one tomb, the first excavated in this research program, archaeologists understood some essential parameters of Dhofari pastoral life and death 5000 years ago. Pastoralists sometimes died near where they were buried, and sometimes their bones were transported after their bodies decomposed. They were buried in family-type groups, and they depended on the springs. In arid regions like Muḍayy, their livelihood and diet was goats and game without the compliment of farmed cereals or dates. They did some limited trading to get sea shells and frit beads; they needed gatherings like burials and commemoration to 4

A Story and Its Meaning

Figure 1.4. Sheikh Suha‘il Al-Rujdi (left) and an aide visit an HCT excavated by Drs. Tara Steimer and Kimberly Williams (both right) (photograph by M. Harrower).

share in community, find marriage partners, and exchange livestock and crucial information. Their beliefs included burying the dead, conjoining people into social groups in death, revisiting and sacrificing near tombs, even when burial was not involved, placing prominent burials near springs and water, and including personal ornaments that marked people even in death (Figure 1.5). But as with all good research, our first excavations brought up new questions that require more research to answer. The AHSD Project (Arabian Human Social Dynamics Project) set out to understand how people maintained their social attachments across landscapes often empty of people. Where people build houses and settle down in permanent agricultural populations, they maintain households, the core social unit in most societies today. But Dhofar’s people maintained a persistent pastoralism, and on the move, they maintained social communities that supported each other in life and in death. By opening a tomb to conjoin the dead as in the case of D001-001, people attached themselves to a social community, sometimes one spanning centuries. What defined social communities in Dhofar’s prehistory and how were they maintained? Were they always based on ancestors? How long did communities and identities persist, and what were the continuities and ruptures in the mobile pastoral groups that shaped Dhofar’s prehistoric landscapes?

5

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 1.5. View of the springs at Mudayy from D001-001 HCT (photograph by T. Steimer-Herbet).

Theoretical Frameworks Social Cohesion and the Material World As early as the fourteenth century CE, Arab historian Ibn Khaldun attributed human success in harsh environments to ʿaṣabiyyah, or “group feeling.” In the lengthy introduction to a study of world history, Ibn Khaldun sought to explain history in terms of abstract social principles. He differentiated between people of the civilized world and those living in desolate lands, describing the processes that strengthened social cohesion among desert tribes and made them the frequent conquerors of cities. Ibn Khaldun’s major point was that social dynamics explain history (and prehistory). And by linking these social dynamics – that is, the attachments people have to each other and the changes in those attachments – to the different environments in which people lived, Ibn Khaldun fundamentally tied social dynamics to material conditions, a critical point in understanding ancient social dynamics in Dhofar (McCorriston 2013a).

6

A Story and Its Meaning

Material conditions include wealth and access to it, environmental resources, and their fluctuations. These are sources of data that archaeologists can collect, even while social behavior of ancient people can no longer be observed. Social cohesion, social bonds, social lives, and social dynamics are invisible in the archaeological record, but the material remains that people left behind – even the small constructions, simple funerary remains, and food scraps from Dhofar’s ancient people – provide rich indications of people’s behaviors in space and time. These archaeological remains were the target of the AHSD Project research, and they were the source of the data – that is observations of the archaeological record – described in this book.

7

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 1.6. D001-001 HCT on plateau before excavation (photograph by J. McCorriston).

Tribes, Territory, and Landscape: The Spatial Implications of Monuments Dhofar’s landscape is rich in small-scale stone monuments. These are not everywhere but cluster in some places and are absent from others. We suspect that mobile pastoralists placed bodies in tombs and built monuments across Dhofar to mark the territories of mobile social groups. But bodies and burials are not the only way in which people shape vistas and create a landscape rich in social memory and cultural meaning. Prior research in Arabian archaeology and adjacent regions has demonstrated that mobile pastoralists constructed stone monuments, some of which held burials, and some of which did not. There were monuments throughout regions and through time periods in which no settlement appeared. Serge Cleuziou (2002, 2007) and Jessica Giraud (2010) have suggested that the placement and contents of 3rd millennium BCE tombs are material reflections of social accommodations to the opportunities and constraints of oasis life. In western Yemen, Alessandro de Maigret (1996) also suggested that tombs reflect social behavior by marking the territory of mobile peoples refusing integration into urban enclaves. The assumption that tombs mark the territory of a socially cohesive group is widely asserted across the Syrian-Arabian region (e.g., Porter 2002; Steimer-Herbet 2004; Yule 2018). Because they may anchor people in place through kinship with fixed dead, tombs may be hallmarks of tribes and their landscapes. Groups socially constituted as tribes generally maintain historical, genealogical narratives of contingent and contextual kinship (segmentary lineages) that regulate their economic and political activities (Tapper 1990; Evans-Pritchard 1940). Such tribes exist as political and loosely-bounded social entities. Archaeologists widely accept that tombs offer a material link to the social communities – like tribes – that buried their dead within them. This link between tombs and societies suggests that spatial frameworks for communities of practice in ancient Dhofar are identifiable through their hallmark tombs. While not every group that practiced the same tomb construction would have traced a common genealogy as a tribe, those groups that practiced different tomb constructions almost certainly identified themselves as distinct groups. 8

A Story and Its Meaning

Figure 1.7. Diagram of a (tribal) segmentary lineage kinship system in which contingent and contextual affinity is measured by descent from common ancestors, usually through patrilineal lines. Triangles denote male descent; females not shown (image by K. Olson).

And some groups built stone monuments that did not contain burials. These monuments are also markers of communities, at least in cases where they commemorated events shared by many practitioners. As markers of communities, stone monuments shape landscapes, which are the constellations of places significant in people’s lives. When did Tribes First Appear? The Chronological Implications of Monuments Most studies of tribes as social communities lack the long-term perspective afforded only through archaeology, in which material correlates of behavior rather than behaviors themselves are available from the archaeological record. Therefore a study of monuments, including tombs, across different time periods reveals dynamics and conservatism in social behaviors. Because social dynamics are essentially behavioral, it is difficult to monitor them over long time frames like the archaeological past. Nonetheless, tribes are a heritage feature of modern-day Oman. Tribal societies have been largely studied using historical methods, ethnography, and other tools of social and political science (Khoury and Kostiner 1990). Such studies are a microcosm of efforts to document and analyze long-term trends in human behavior that cannot be directly observed in the short span of a research project (or even one researcher’s lifespan). But archaeology offers an important tool in its focus on material remains of human behavior. Tombs are but one category of small-scale stone monument; pastoralists used many types, through time. Through an archaeological spatio-temporal understanding of small-scale stone monuments we expected to document and reconstruct the historic and prehistoric human social dynamics, including the emergence and dynamics of tribes in ancient Dhofar. We explored whether the use of monuments for burials in Dhofar can be explained by significant correlation with the forces of change that elsewhere have been implicated in the emergence of tribes (Alizadeh 2010; Frachetti 2009; Rowton 1975).

9

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Forces of Change and Socio-ecological Dynamics As everywhere, human social dynamics are finely tuned to the environmental circumstances in ancient Arabia. Anthropologists have long understood that environmental conditions play powerful shaping roles in the maintenance and reproduction of social groups and that in turn, social and cultural exigencies shape the ways in which people exploit available resources and environments. Dhofar offers a broad temporal and spatial framework that encompasses both environmental and cultural changes affecting pastoralists. Dhofari pastoralists experienced both the dynamic forces of change and the legacies of history. We already know of several inflection points for pastoral life during the past 7000 years (e.g. Avanzini 2008; Lézine et al. 2010; Lézine et al. 2017). Major aridification (5200 calendar years ago), the appearance of oasis cultivation, and the development of far-away urban enclaves (2900 years ago) offered profound environmental and cultural changes. Such changes rippled the social networks of adjacent communities and perhaps even those in Dhofar. These changes should have constrained and impacted social relationships of herders in the arid hinterlands of Southern Arabia.

10

Chapter 2

Dhofar’s Pastoral Landscape

While they adapted to shifts in regional climate, environments, and the developing cultural contexts of their neighbors, Dhofar’s persistent pastoralists ranged across a topography and vegetation unique on Arabia’s southern coastline. With their different seasonal flushes and resources, four major ecosystem zones in Dhofar provide an underlying and dynamic environmental structure to pastoral movements and lifeways. In addition to a brief description of these four zones and their spatial parameters, this chapter also discusses perturbations and long-term dynamics within them. Four Vegetative and Climate Zones Dhofar lies just beyond the rain shadow of the Horn of Africa so that summer monsoon circulation crosses an expanse of Indian Ocean, forcing upwelling of cold waters and pushing moisture-laden ocean air onshore. As this humid air rises and cools with altitude, moisture precipitates into fog, nurturing vegetation longadapted to such conditions (Hildebrandt and Eltahir 2006; Patzelt 2015).

Figure 2.1. Image map of four major ecosystems in Dhofar: coastal plain, wooded escarpment, grassland plateau, and Nejd desert (image by by K. Olson).

11

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 2.2. Cloud forest of Terminalia dhofarica across the seaward-facing Dhofar escarpment (photograph by J. McCorriston).

Winter rainfall is minor, if any falls at all. Summer monsoon fog determines plant growth in Dhofar. Consequently, Dhofar’s ocean-facing mountains naturally support a cloud forest that captures mist on its leaves, which drip precipitation and enhance available moisture for plant growth (Hildebrandt et al. 2007). Genetic studies indicate that this cloud forest has been contiguous along the escarpment throughout recent millennia (Kürschner et al. 2004). The tropical tree Terminalia dhofarica dominates with a rich association of other trees – Blepharispermum hirtum, Gymnosporia dhofarensis, Euclea schimperi, and in higher, drier locations, Vachellia etbaica and Commiphora spp.. Four major vegetation zones lie in concentric bands from the coast to the inland deserts (Figure 2.1). A Coastal Plain is sabkhah flats and limestone outcrops dissected by shallow drainages and brackish inlets (khors). With the exception of a few freshwater springs and access to a high water table, few locations have sufficient soil and moisture to support vegetation. Coastal regions today use high groundwater for coconut plantations and infield cultivation. Marshes green the edges of khors, and the colluvial fans of the foothills support a Vachellia-Commiphora woodland. The seaward face of the mountain or Escarpment zone is dissected by short, sometimes deep drainages and draped in a dense vegetation of cloud forest dominated by Terminalia dhofarica (Figure 2.2). Today most of this forest has been cleared for grazing at middle elevations so that only the steepest slopes retain the more dense cover typical of escarpment vegetation. Revered for its shade and persistent with deep tap roots, in the upper escarpment Ficus sycamorus characterizes many vistas otherwise virtually barren of trees. 12

Dhofar’s Pastoral Landscape

Figure 2.3. Parkland of the Dhofar plateau with modern short grassland (photograph by J. McCorriston).

Figure 2.4. Native palms in the sandy channel of Wādī Dhahabūn (photograph by J. Everhart).

13

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

On the high Plateau, in some places only a few hundred meters wide, traditional grasslands alternated with woodland. Today one finds a parkland of Vachellia etbaica and Commiphora kataf / kua trees, unpalatable shrubs and herbs, and short grasses (Setaria pumila, Apluda mutica) that have recently replaced the tall grass savannah dominated by Themeda quadrivalis (Patzelt 2015, 2011) (Figure 2.3). Finally there is the Nejd, a stony backslope draining toward the desert interior and sands of the Empty Quarter. Once one crosses the continental divide, fog rapidly dissipates and provides limited moisture. The near (southern) Nejd is famous as the habitat for the finest frankincense (Boswellia sacra), but a rich array of other trees, shrubs, and herbs survive on gravel terraces, colluvial fans and slopes. Cadaba farinosa, Grewia erythraea, and Vachellia tortilis are typical; further into the Nejd one finds Maerua crassifolia and sandy banks of stubby native palm (Nannorrhops ritchiana) in the wadi bottoms (Figure 2.4). Dynamic Forces in Dhofar Even as the regional differences across zones significantly shaped the possibilities for human life, the proximity of different zones could mean that slight shifts within one could provoke change in adjacent zones. Today the boundary between Plateau and Nejd is so well defined that in some places one can stand with a foot in each. But it was not always thus. In addition to seasonal fluctuations and inter-annual variability that may blur gradients, there are also major dynamic forces that influenced the material conditions of persistent pastoral life in ancient Dhofar. Climate Change No one directly can observe past climate. But scientists have become ever better at deducing past climate from proxy records left by past climate events. Proxy records of Dhofar’s climate history include the ratios of heavy and light isotopes of oxygen molecules (different isotope ratios reflect warmer or cooler temperatures through differential evaporation of lighter isotopes) and the fraction of different carbon isotopes (13C and 12C) in calcium carbonates. In the latter case, a higher δ 13C (measured against a standard) value is indicative of tropical grasses and differentiates a former tropical savannah grassland from former woodland vegetation. Another valuable proxy is plant pollen, also reflective of former regional vegetation. All these proxies may be found in a variety of catchments, or traps, that layer and preserve these signatures of ancient environmental and climate conditions in sequences, or archives, that can be assigned chronological dates. Details of Dhofar’s climate proxy data may be found elsewhere (Neff et al. 2001; Hoorn and Cremaschi 2004; Fleitmann et al. 2007, 2011; Lippi et al. 2011; Cremaschi et al. 2015; Lézine et al. 2017); these studies and a wider regional understanding of the dynamics of the Indian Ocean monsoon (Anderson and Prell 1993; Neff et al. 2001; Gupta et al. 2003; Lézine et al. 2014) indicate significant climate inflections and episodes of change in ancient Dhofar. In climate science, researchers have sought to collect regional and local proxy data (including the speleothems, or cave carbonates, from Qunf Cave in Dhofar, Fleitmann et al. 2007) to develop climate reconstructions (Figure 2.5). Such reconstructions can then be used to test and parameterize generalized computational models. Since this chapter seeks to provide a background to human environments, what follows here is not a reconstruction of climate science in Dhofar but an inverse summary. Thus follows an overview of Dhofar’s major climate dynamics organized by regional and global climatic shifts and reflected in local proxy data. 14

Dhofar’s Pastoral Landscape

Figure 2.5. Travertine deposits, like speleothems, form with the evaporation of water that has seeped through limestone, carrying with it an oxygen proxy signal of temperature and a carbon signal of vegetation (photograph by J. McCorriston and K. Olson).

15

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 2.6. Image map of enhanced Early-Middle Holocene monsoon precipitation and paleolacs in Southern Arabia (precipitation data from Brown et al. 2018; Fordham et al. 2017) (image by K. Olson).

Beginning of the Holocene At the beginning of the Holocene period some 11,500 years ago, global climates shifted. From high amplitude fluctuations and overall colder climate in the Pleistocene, global climates developed a warmer, wetter pattern with overall greater stability. The effects across Arabia and the Sahara were profound. Broad shallow lakes formed in the inland deserts, and mountain streams ran further inland for a longer annual season (Figure 2.6). A heavier rainfall than seen in Dhofar today appeared abruptly about 9800 years ago and persisted until about 5000 years ago (Cremaschi and Negrino 2005). This enhanced moisture would have recharged groundwater (Hoorn and Cremaschi 2004), feeding a wider and denser net of springs and seeps for longer periods of each year. Such sustained and augmented water impacts vegetation, and pollen records suggest that similar plant communities to those found today were richly enhanced by denser trees deep into the Nejd (Lézine et al. 2007), thick mountain woodlands, and wider swaths of annual grasses and forbs offering extensive grazing to herbivores domestic and wild. This Green Arabia of the Early Holocene lasted for about 6000 years and permitted the expansion of foragers and herders into regions now too arid to support them today (Zarins 2001; Groucutt and Petraglia 2012). Middle Holocene From deep ocean cores to inland deserts, climate records suggest that the strength of moisture-laden summer monsoon winds began to decline about 6000 years ago (Lézine et al. 2010). With less rainfall, the 16

Dhofar’s Pastoral Landscape

steady evaporation of desert lakes outpaced recharge rates, and Green Arabia’s pasture grasses and desert oases shrank or disappeared. The decrease in precipitation was gradual over the course of centuries, but its effects left abrupt local changes in proxy signatures (e.g., Cremaschi and Negrino 2005; Harrower et al. 2012) as local conditions expressed the long-term trend. In Dhofar, the springs in the Nejd continued to receive groundwater recharge from significant fog-drip in the escarpment and high plateau, probably captured still by leafy vegetation in cloud forest (Abdul-Wahab et al. 2009; Friesen et al. 2018). Yet there were dry years; there were springs that ran dry forcing a longer walk between sources; the fog penetrated inland too seldom for some of the parkland trees on the colluvial slopes of Nejd drainages. In the mountains, the moisture that promoted active soil formation also tapered, so that deposits of desert loess began to accumulate in still places – the interior of caves and abandoned structures (Cremaschi and Negrino 2005). A recent regional study of multiple proxy records across Southern Arabia suggests that aridification began about 5000 years ago but stabilized only about 2000 years ago in truly modern arid conditions (Lézine et al. 2017). Roman Warm Period 250 BCE – 400 CE (Dhofar) Climate scientists and archaeologists alike have become interested in the Roman Warm Period, which potentially lasted from about 250 BCE to 400 CE, followed by a 6th century CE mega-drought (Ljungqvist 2010, c.f., Werner et al. 2019). In Dhofar, the signature of the RWP appears in the isotope records from Qunf Cave (Jabal Qamar) as a “significant drop in precipitation” (Fleitmann et al. 2007: 180) and in the onsite pollen at Sumhuran (Lippi et al. 2011) as an increase in wetlands taxa between 50 BCE–200 CE. Local paleoproxies may be interpreted differently: as aridity (Fleitmann et al. 2007), or a slight modulation with increased moisture between 50 BCE–200 CE in an overall trend towards aridity since the Middle Holocene (Hoorn and Cremaschi 2004; Lippi et al. 2011). The broader regional signal is one of an increased aridity, which could have locally increased wetland taxa in drying khors and greater erosion of denuded ground. With less precipitation, there would be fewer springs, with less flow and more widely spaced. Such a scenario would have implications for vegetation and movements of pastoralists. Although by no means a climate shift as profound or permanent as the Middle Holocene, the Roman Warm Period nonetheless would affect local populations more than any since the downturn in climate 3000 years earlier. The Calamitous 6th century CE In the Near East and in Arabia in particular, historical records portray a 6th century CE of political and social tumult (Smith 1954; Haldon et al. 2014). The great dam at Marʿib broke, and the Sabaeans had not the wherewithal to repair it; Abraha of Ethiopia led an invasion of Yemen and the ‘Asir; new religious orders arose in different oases; and the northern influence of the Byzantine empire waned in an era of drought, crop failures, and plague. Regional climate records suggest a sharp downturn in precipitation (Gupta et al. 2003; Ivanochko et al. 2005; Anderson et al. 2010; Fleitmann et al. 2011; Tierney et al. 2013), prompting crop failures and population instability in the desert margins, where water decline mattered most. Some historians blame the fragmentation of political power structures and decline of empires on climatic factors (e.g., Lamb 1995; Brooke 2014). During the 6th century CE, Dhofar was at the margins of empire and history. Nonetheless, Dhofar experienced this regional climatic shift with declining rainfall evident in the Qunf Cave speleothem isotope data from 500-600 CE (Fleitmann et al. 2011). Even as the changes and significance of climate have become more widely recognized, the local effects of climate are still poorly documented in vegetation proxies from Dhofar. 17

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Medieval Climate Anomaly In the era ~900-~1350 CE, there was a globally recognized warm period (Graham et al. 2011; Xoplaki et al. 2016), with implications for human social and political systems around the world. Norsemen settled Greenland; the Ancestral Puebloans of Southwest USA built Great Houses at Chaco Canyon; and the Mississippian people constructed huge mounds at Cahokia. Nearer to Dhofar, the Rasulid dynasty of Yemen achieved wide territorial integration and control, collecting surplus grain and overseeing commerce among ports along the Indian Ocean (Varisco 1994; Rougeulle 1999). Rasulid governance supported specialization and exchange in products such as horses from the Salalah plain for textiles and other imports. In Dhofar itself (Ẓafār is the term for the coastal plain), mediaeval historians describe port activities and well-watered coastal orchards fed by springs and supporting summer crops of sorghum, millets, and sugar cane as well as irrigated perennial fruits – dates, grapes, pomegranates, citrus, bananas and coconut (Smith and Porter 1988; Newton and Zarins 2017). Of the Dhofar interior, that is, the escarpment, plateau, and Nejd, there are few paleo-proxies of climate and environment. The stalagmite sequences in Dhofar suggest that precipitation feeding groundwater in the escarpment resumed prior levels, until the cessation of most stalagmite formation about 400 years ago (Hoorn and Cremaschi 2004; Cremaschi and Negrino 2005; Fleitmann et al. 2007). Little Ice Age Over Northern Europe from ~1500-1850 CE, cold season precipitation declined, which some historians have invoked as explanations for socio-political turmoil (Lamb 1995). Climate scientists explain the MWP-LIA climate changes as shifts in North Atlantic Ocean and North Pacific Ocean circulation patterns ultimately driven by global changes in the westerlies. These global shifts not only produced variable climatic effects (but everywhere more or less synchronous), but also prompted speculation that changes in Indian Ocean-Southwest Pacific Ocean circulation also might account for a MWP and LIA climate effect (Gupta et al. 2003; Graham et al. 2011). Such drivers would necessarily implicate local shifts in precipitation and temperature patterns in Dhofar, as the Indian Ocean monsoon weakened during the LIA (Gupta et al. 2003; Tejavath et al. 2019). These predictions have local signatures of diminished rainfall in the speleothem records from Southern Oman, where a LIA lasted from ~1320-1660 (Fleitmann et al. 2007) and produced drier conditions, like those noted around the Indian Ocean (Sinha et al. 2011; Norström et al. 2018). Interdependency of Humans and Vegetation The signals of precipitation and temperature from proxy records recovered from sedimentation sequences and stalactite formations in caves largely document the interplay between climate and vegetation in Dhofar, yet the long-term presence of humans and their domesticated herds (Al-Abri et al. 2012; Bayoumi et al. 2016) is attested archaeologically (Zarins 1992) and likely to have substantially affected vegetation locally and regionally. While the dynamics of human ecosystems are still the subject of research, some properties of community ecology suggest profound and sustained human interventions and feedbacks in vegetation, especially of the escarpment and plateau regions. Anthropogenic Burning and Grasslands Studies of genetic distance in modern plant populations suggest that the plant community at higher elevations, notably the juniper-olive woodlands, have long been fragmented and almost certainly retreated to refugia between 1.7 and 1.0 million years ago (Kürschner 2004; Meister et al. 2006). A similar methodological 18

Dhofar’s Pastoral Landscape

Figure 2.7. Archaeological evidence of anthropogenic fires can be seen embedded in sediment terraces in Yemen’s highland deserts, an ecosystem comparable to the backslope of Dhofar’s mountain plateau and near Nejd. No comparable sediment terraces exist in Dhofar to preserve such traces of former activity, but the tall grasslands themselves suggest a comparable antiquity to the practice of burning. Here the terrace surface has eroded to reveal a broad swath of grey ash and micro-charcoal, center (photograph by J. McCorriston).

approach points to the more recent fragmentation of the Terminalia dhofarica forests, which probably extended continuously across the Jebel Fartak of southern Yemen and the Jabals Qamar and Qara of Dhofar through the Early Holocene (Oberprieler et al. 2009). In differentiating the influence of climate dynamics and human management on forest fragmentation, botanists and ecologists look to the broader distributions of plant associations across the East African, South Arabian and South Asian continental expanse; here several factors point to anthropogenic fragmentation of Terminalia woodland and maintenance of tall grassland appropriate for grazing. Comparative ecosystem dynamics in the African savannah show that the Themeda quadrivalis grassland lies at the margin of its natural distribution, with an impoverished range of plants in the grassland community compared to its African counterpart (Patzelt 2011). In this attenuated system, traditional herders in Dhofar encourage the renewal of grassland at the expense of forest by timing deliberate burning of vegetation (Sale 1980). No proxy indicators have been published to establish the time depth of these practices in Dhofar (Figure 2.7). Finally, the sharp spatial transition from grassland to woodland today at the brink of steep precipices and wadi headlands is a persuasive indication of the influence of grazing animals on maintaining grassland over forest. While cattle prefer grasses, they will browse leafy forage when proteins and nutrients decline in dry-season grasses; goats and camels will browse and are effective eliminators of tender shoots and saplings 19

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

(Figure 2.8). Herders tend to direct animals to flatter terrain where energy expended in animals’ range for food is reduced and thereby weight loss minimized. Hence, the forested seaward slopes reflect a cover sharply altered on the plateau. Cloud Cover, Springs, and Mobility That herders direct where and when their animals move should have important effects on localized vegetation dynamics. Cattle require daily watering, which (prior to 1970 and the advent of deep wells) tethered animals within a few kilometers of a water source. During summer months when fog interception by trees in the escarpment resulted in deep moisture penetration into the soils (Hildebrandt et al. 2006), springs on the southward escarpment were re-charged by monsoon moisture. Groundwater studies show that except for rare cyclones, groundwater in the northward facing slopes and Nejd wadis comes from fossil aquifers in the Umm er-Radhuma formation (Strauch et al. 2014; Müller et al. 2016) and is therefore available in fairly permanent quantities and predictable sources, albeit widely spaced. It is plausible that anthropogenic changes in escarpment and plateau forest cover should change recharge in springs, causing reduced quantity and wider spacing of spring flow in the escarpment and its juncture with the coastal plain. If so, then human reduction of forest cover should also impact mobility of herders, particularly in the dry season, a feedback to human modification of vegetation in the Jabals Qara and Qamar but unlikely substantially to affect springs in the Nejd (Figure 2.9).

Figure 2.8. Cattle grazing short grassland on the Dhofar plateau (photograph by K. Pustovoytov and K. Olson).

20

Dhofar’s Pastoral Landscape

Figure 2.9. Hypothetical effect on human transhumance from reduction in forest, fog precipitation, and groundwater recharge. Note the reduced transhumance of (cattle) herders into the Nejd with the replacement of deep-rooted trees by grassland on the plateau. Adapted from Hildebrandt and El-Tahir 2006 (image by C. Hickman and E. Lagan).

21

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

The Outside World No man is an island, and however traditional Dhofari pastoralism remained prior to 1970, there have long been exchanges and introductions across the Arabian Sea and the pilgrimage routes skirting Arabian deserts. Neolithic Arabia saw the introduction of domesticated cattle, sheep, and goats; Bronze Age Arabia linked into a widespread exchange network connecting Afghanistan to Anatolia. Northern Oman participated in this world system with the extraction of copper and the establishment of date palm oases, inland and sea routes along which foreign traders plied exotic ceramics, beads, metals, and less durable textiles, and foodstuffs. Dhofar’s Hidden Exports The occasional copper blade, marine shell, or frit bead in a Dhofar pastoralist’s tomb hallmarks Dhofar’s participation in exchange systems, however attenuated that participation may have been. The landscape of pastoralism persisted across the escarpment, plateau, and Nejd, even as outposts and waystations for traffic emerged (and have become the foci of most archaeological research in Dhofar). No one can say what labor pastoralists may have supplied (who extracted frankincense from the trees and carried it to exchange posts?) or what perishable materials they received. Frankincense remains Dhofar’s most famous export, even as its ancient production is as mysterious as it was in the time of Pliny and Strabo, who wrote of fiery serpents guarding the trees. But there were surely other hidden exports — hidden because they left little or no archaeological trace. Other resins and extracts – myrrh, aloes, dragon’s blood, wood – as well as leather, horn, feathers, pigments, dyes, and hides could have travelled the exchange routes that brought exotic marine shell, metals, textiles, and ornaments into pastoral communities. In the ebb and flow of climate, vegetation, management, and exchange, pastoralism persisted in Dhofar’s ancient pastoral landscape.

22

Chapter 3

Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

Even as far-away historians and geographers of the Classical and Islamic worlds wrote their second-hand and sometimes fantastical accounts of the land of Dhofar, its riches in incense, guardian serpents, and exotic people, Dhofaris themselves left no extensive written history or self-representation, except as un-deciphered in graffiti. It falls to archaeologists and pre-historians to translate the remains of a distant past into narratives of human lives and human landscapes. By relying on the material culture scattered across Dhofar and on the recent interactions of people with their resources, archaeologists reconstruct the behaviors of antiquity. Archaeological studies of cities, towns, fortresses, hamlets, rock-shelters, and the scattered stone tools at campsites and stone-chipping workshops reveal much of the economic procurement and exchange systems in prehistory. Left largely un-explored are the numerous tombs and monuments as rich relicts of social constitution and memory that shaped the ancient Dhofar landscape.

Figure 3.1. Archaeological survey team recording a circular tomb. Left to right: Sarah Fargo, Matthew Senn, Ali Ahmad Al-Kathīrī (photograph by J. Everhart).

23

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 3.2. One of many triliths on a rock terrace, seen across a shallow tributary near Hanūn (photograph by M. Harrower and K. Olson).

Small-Scale Stone Monuments These small-scale stone monuments are the focus of successive research projects by American-Omani and international collaborative teams from 2009-2019 (Figure 3.1). The outlines of Dhofar’s material culture history and its economic implications are well known. Even so, the social meanings of Dhofar’s cultural remains are elusive, along with the identities and constitutive beliefs of ancient Dhofari communities. For example, the rich trove of stone knapping debris, discarded tools, and chert source sites has inspired a generation of researchers to reconstruct the movements and lifestyles of Pleistocene foragers and Early Holocene hunter-herder-collectors who moved across the green Nejd from perhaps 50,000 years ago (Rose et al. 2019a, 2019b) and along Arabia’s resource-rich coastlines (Charpentier 2008; Marrast et al. 2020). Major settlements and urban genesis at Al-Baleed and Sumhuran have attracted multi-national and multidisciplinary teams who detected both the local economic influences and external exchanges shaping these city-scapes (Albright 1982; Avanzini 2008; Newton and Zarins 2017). At the same time, a wide-ranging archaeological survey has shaped the broad outlines of Dhofar’s culture history (Zarins 2001; Zarins and Newton 2013), leaving for more-focused research projects many questions of cultural affinity, chronologies, and the dynamic interplay of social and environmental systems across Dhofar’s varied zones. Throughout this program of archaeological research there has been limited attention to the tombs and monuments that shape certain skylines and beckon at the bends of routes and traverses. Some of the first archaeological research in Dhofar was the 1950s American expedition (AFSM, American Foundation for the Study of Man), which brought archaeologist Frank Albright in search of Ḥaḍramawt Kingdom colonies and outposts ostensibly extracting frankincense. Mention of stone rings and tombs around Sumhuran appear in Albright’s report, but these received no focused investigation until prehistorians with the Italian mission branched out to investigate (Cremaschi and Negrino 2002). Nor did the AFSM team investigate monuments in the vicinities of Hanūn and Andhur, other loci of AFSM study. Dhofar nonetheless acquired a reputation for its triliths, low gravel platforms supporting clustered uprights accompanied by a strategic geometry 24

Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

of hearths and boulders arranged alongside (Figure 3.2). These also appear in adjacent regions of Mahra, Ḥaḍramawt, Central Oman, and the Ash-Sharqiyah (De Cardi et al. 1977, McCorriston et al. 2011), leading to speculation of ethnic and linguistic groups once more widespread and whose descendants perhaps remain in Dhofar today (Yule 2018). This particular focus on triliths aside, the best comprehensive treatment of Dhofar’s monuments lies in the ethnohistoric work of Ali Mahash Al-Shahri (1991), who classified six types of burials and other monuments, linking them to oral accounts and cultural traditions of Shahri speakers in Dhofar. In addition to the triliths, Al-Shahri identified 1) elongated graves of Islamic saints, 2) Ꜫnfo̍ (local Jibbāli term) widely also called “boat-shaped graves” or just “boat-graves” by archaeologists, 3) rock shelter crevices containing burials sealed by dry-stone walls, 4) ꜥasmīn, or circular, dry-stone, chambered tombs, 5) Muruźәd (Jibbāli) (Figure 3.3), and 6) an additional, localized, elongate type, which may ultimately prove to be a local variant of the widely distributed boat-graves or possibly extremely well-preserved platforms from an earlier time. While highly valuable as a typology linked to traditional knowledge and narratives, Al-Shahri’s work lacked systematic archaeological investigation to link these formal types to chronologies and the ancient behaviors associated with them. A wider scope is afforded by the archaeological studies of Arabia’s small scale stone monuments, including regions beyond Dhofar. While this short review will not be comprehensive, it is nonetheless important to recognize broader regional patterns in the distributions of small-scale stone monuments.

Figure 3.3. Muruźәd (Jibbāli), which appear very like the madhaby (Arabic) grilling hearths and sacrificial commemoratives in Hadramawt , Yemen. These examples occur in the Wādī Halūf near D014-003 (see Chapter 6) (photograph by M. Harrower).

25

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Decades of research in northern Oman and the UAE have provided both chronological and spatial contexts for Neolithic cemeteries (Uerpmann et al. 2000), Bronze Age Hafit, Umm an-Nar, and Wadi Suq tombs (Vogt 1985; Giraud and Cleuziou 2009; Williams and Gregoricka 2013) and for Iron Age burials of the Samad and other cultural traditions (Fritz 2010; Yule 2014, 2018). In the margins of Arabia’s arid interior and around its oases, one finds circular tombs of the Bronze Age, sometimes with tails of small stone cairns and a distinctive wall tomb type scattered across Mahra, Ḥaḍramawt, western Arabia (Braemer et al. 2001; Steimer-Herbet 2004; McCorriston et al. 2011), and visible even in the Sinai (Eddy and Wendorf 1998). These are archaeological surveys and excavations drawn upon by Juris Zarins (2001; Zarins and Newton 2013) in his Dhofar archaeological surveys and assignment of monument types to specific chronological phases. Recently the significance of small platforms constructed in the Neolithic has also been emphasized (McCorriston et al. 2012, 2020). These monuments constructed a Dhofar landscape of pastoralists through time as surely as they elsewhere marked the burials of oasis populations, fisher folk, or copper miners more celebrated in Arabia’s archaeological narrative. Research Frameworks The research documented in this book examines pastoral social lives over 7000 years of prehistory in Dhofar. Data described here were collected between 2009 and 2018 as part of the AHSD Project (Arabian Human Social Dynamics Project) and the beginnings of the ASOM Project (Ancient Socioecological systems in Oman Project). Archaeologists used conventional archaeological methods of field survey and excavation augmented by spatial technologies novel at the time and statistical analyses tested and updated since fieldwork ended. Our data collection and field work were driven by research frameworks in anthropology that sought to understand the dynamics of human social systems – such as the cultural means of formation and transmission of social groups across space and time, including the emergence and maintenance of kin-based tribal systems. An archaeologist can leverage the short-term advantages of talking to living people about these issues with longterm chronologies that no other researcher can access. A long-term perspective relies on material correlates of behavior from the archaeological record rather than behaviors themselves. A spatio-temporal framework of the material records of social behavior – in this case, small monuments that mark small-scale communities’ social cohesion – is essential if we are to reconstruct prehistoric and historic human social dynamics. When did Arabian Tribe Systems Emerge? In ancient Arabia, these human social dynamics should include the emergence of genealogical narratives critical to tribal constitution (McCorriston 2013a). With the remarkable record of small-scale stone monuments in Southern Arabia – Dhofar included – archaeological research addresses the social lives of mobile pastoralists in a region sparse in settlement and fixed population centers. In this context, and by inference, in the wider context of arid Arabia, one must rely upon archaeological approaches to study how tribal societies originated, how they manipulated technologies and social means to access and assert claim to resources, and how tribal identities and territories changed because of their interactions with states. Are Monuments Territorial Markers? Issues of Space There is no compelling reason to assume that all pastoralists in Dhofar are or were territorial tribes; indeed, recent anthropological research finds that many societies worldwide – pastoralists included – practice sustainable management of common pool resources without invoking the boundary conditions of territorial 26

Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

regulation (Moritz et al. 2018). At the same time, tribal affinity in Arabia has traditionally and within historical memory conferred specific territorial and resource use rights deniable to members of other groups, whether these been within agreed boundaries (land territory) or exclusion from resources (e.g., wells, springs, pasture, cultivable land, archaeological loot) (Shoup 1990; Lancaster and Lancaster 1999; Gari 2006). Even as it is known that Arabian tribes have asserted exclusive claims to resources in the recent past, one cannot assume this social arrangement (tribal societies) and its mediation of resource access is a timeless and static condition of Arabian prehistory. Quite the opposite is likely, namely that tribes were not always extant, nor were tribes an exclusive mode of social constitution, for pastoralists or for other producers. As a type of social existence, “tribe” has a rich epistemological history, little of it informed by the chronological depth archaeologists can offer. Groups socially constituted as “tribes” generally maintain historical genealogical narratives of contingent and contextual kinship, often called segmentary lineages. Genealogical ties regulate tribes’ economic and political activities and so tribes exist as political and loosely bounded entities. The origins and dynamics of tribes and their relationships to environmental resources invite further study of prehistoric and early historic Arabian societies and landscapes. At stake is an understanding of how humans have manipulated and shaped Arabian environments through its long prehistory via a time-depth study of the dynamics of human landscapes. How Else Can Monuments Serve as Archives of Social Identity? Issues of Time A focus on small-scale monuments as signals of social cohesion exemplifies a larger issue concerning how people develop and maintain social identities and how these identities both shape and respond to environmental circumstances and changes. Across differing theoretical perspectives anthropologists, geographers, historians, and evolutionary biologists have long understood that environmental and cultural frameworks play powerful shaping roles in the maintenance and reproduction of social groups (e.g., Steward 1968; Braudel [1966] 1972; Service 1971; Clutton-Brock & Harvey 1977; Bourdieu 1977; Netting 1977; Bernstein et al. 1979; Laland and O’Brian 2011; Odling Smee and Laland 2011). Human social behaviors are both manifested and reinforced through material symbols and artifacts (e.g., Bourdieu 1977; Appadurai 1986; Schneider and Weiner 1991), which although challenging for outsiders to access nevertheless encode meaning and messages about social identities. Archaeological remains, including monuments, graffiti, and ornaments, offer a unique data set for examining the long span of human social dynamics in deep antiquity, before and beyond the caprice of the written record or purview of a short-lived social scientist. Monuments occupy specific environmental locations. Many believe that monuments physically anchor cultural meanings in these places (McCorriston and Field 2019; Tilley 1992). Monuments were long-enduring. They were meaningful to the people who built them, signaled meaning understood by people who visited them, anchored meanings, stories, and information in specific landscape places, and served as a beacon for re-formulated meanings and new social constitution in later times (Rowlands 1993; Van Dycke and Alcock 2003). As an example, consider the well-studied monument that is Stonehenge in Britain. It was built first as a bank and ditch circling upright timbers or stones, then modified to contain burials, then used to enclose bluestones dragged 200 miles, then saw the erection of the huge triliths iconic today. Nor was this all: the first Neolithic builders 5000 years ago used it as a gathering place to celebrate life, death and social cohesion, but gatherings thereafter marked the burials of high chiefs, the celebration of Druidic beliefs, the picnics of Victorian gentry, and the neo-pagan rites of modern Druids. All these groups have used Stonehenge literally as a touchstone for their community affirmation and beliefs about their relationships to one another and to the world around them. 27

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Monument Patterns As monuments were meaningful, so then their distributions in time and space should be patterned – with discernable changes as meanings embedded in environmental and cultural circumstances changed. It is surely impossible to reconstruct what specific monuments meant to specific people long dead, but the distributions of distinctive and variable monuments erected by Arabians in antiquity map and chronicle shared meaning and the institutions of social behaviors over several thousand years. Broad temporal and spatial frameworks that encompass both environmental and cultural changes provide an appropriate research forum for understanding the influence of such changes on social structures. In the past 7000 years in Southern Arabia, significant aridification (5200 calendar years ago) and the development of Arabian states (3200 years ago) offered profound environmental and cultural changes that could influence social dynamics. Such changes constrained and impacted social relationships of mobile herders who built monuments in the arid hinterlands of Southern Arabia’s renowned incense trading kingdoms. When combined with techniques in spatio-temporal modeling and statistical learning, the tools of Geographic Information Science (including high-resolution satellite imagery) provide exciting means of understanding how people developed and maintained social identities in prehistory and history. By documenting the dynamic spatial boundaries of material markers of social identities (e.g., de Maigret 1996; Steimer-Herbet 2004; Giraud and Cleuziou 2009), archaeologists can physically trace the longterm consolidation and dissolution of group identities as social structures that both created and resulted from change. Patterns gleaned from archaeological fieldwork serve as the basis for an interpretation of social dynamics. The observations that make up this pattern of monuments in space and time are data. Reconstructed patterns, or culture history, as an outcome of data analysis, is only part of the challenge of archaeology. It is the interpretations of pattern that will shape our understanding of persistent pastoralism as social dynamics in ancient Dhofar. Patterns in Space: Survey and Auto-detection To collect data to explore spatial patterns in monument distribution, our AHSD team elected to apply new technologies to specific regions. In the wake of a survey to provide full coverage of Dhofar (Zarins 2001; Zarins and Newton 2013), the AHSD team introduced high-resolution imagery with high-precision GPS technology. Using techniques piloted elsewhere in southern Arabia (McCorriston, Harrower and Crassard 2020; Schuetter et al. 2013), we designed our survey to provide coverage of Dhofar’s four major environmental zones as sample areas defined by high-resolution satellite imagery. Because current population density and development is highest in the coastal zone and coastal access has favored archaeological work in this region, we elected to focus most of our efforts on the plateau and Nejd, with a single strip through the escarpment. Our goal was total coverage of the high-resolution image areas so that we could verify not only the locations of small-scale monuments but also confirm the absence of monuments within known parameters. Because the team sought to target areas certain to have a density of monuments, the selection of expensive satellite imagery did not follow randomizing processes. Concentrations of monuments ensures efficiency in fieldwork commutes and search times. Practical considerations guiding our selection of satellite imagery were: 1) a portion of a major wadi or drainage (testing the hypothesis that passage and access to water was significant in the location of tombs); 2) areas spaced in T-form across Dhofar to provide both an east–west sample and a north–south sample; 3) areas that can be readily reached by existing roads or tracks. Using Google Earth as a regional guide, the AHSD Project team in 2009-2010 selected 17 polygons for purchase from Digital Globe’s Quickbird ™ satellite archive, then among the highest resolution (0.6 m 28

Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

Figure 3.4. High resolution images surveyed across Dhofar (Image 5 not accessed) (image by K. Olson)

Figure 3.5. Development of Computational Algorithms for Automated Detection. The process examines each window of an image with a series of filters; the statistic JB measures the relative difference in intensity between the inner and outer windows; the next filter (Jv) calculates the normalized difference vegetation index (NDVI), and so forth (details in Schuetter et al. 2013) (image by J. Schuetter and K. Olson)

29

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 3.6. Using GPS rover to register ground control point (GCP) for image rectification in Wādī Haylā, south of Mudayy (photograph by J. McCorriston).

pixel) imagery commercially available (Figure 3.4). Experience in Yemen had shown that small-scale stone monuments could be detected both by eye and with the use of computer algorithms, the development of which for automated detection of archaeological sites was a research target for the team (Schuetter et al. 2013) (Figure 3.5). Also drawing upon earlier experience, the team produced image maps for field use from digital satellite images, printing these at enlargement and laminating them with clear plastic. Accurate monument locations registered in fieldwork are essential for developing and training an algorithm for autodetection. Hand-held GPS units sufficiently register coordinates with acceptable error ranges for re-locating an archaeological site on the ground. Yet, the precision needed for image analysis of the characteristic properties of specific pixels required greater GPS accuracy. To rectify the edge distortions of digital satellite images (flat projections of Earth’s curvature), we pre-selected ground-control points (GCP) visually from the satellite images (Figure 3.6). GCP are precise locations clearly identifiable from space, such as the corner of a building, a sharp fork in a road, or the clear bend in the edge of a cliff. GPS points taken at GCP can be used to geo-reference an image so that GPS coordinates taken in the field on monuments accurately overlie the correct image pixels that contain that monument. The team also required high accuracy of GPS registration of monument locations. A GPS in motion, or one that briefly communicates with satellites at a point location, incorporates several sources of error into a GPS reading. To correct for GPS error, the teams used a kinematic GPS configuration overseen by Dorota Grejner-Brzezinska of The Ohio State University’s Department of Civil, Environmental, and Geodetic Engineering. Kinematic GPS consisted of a Trimble 5700 antenna and receiver base station located in Salalah and sometimes another sub-base station located nearer our fieldwork sites, along with a GPS rover backpack with Trimble 5700 receiver, antenna, and data controller. The GPS base station collects data at one point all day long, resulting in a well-established baseline for atmospheric and measurement error. Researchers compare the field location data with the baseline to correct for short-term error introduced in field readings. 30

Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

Figure 3.7. To develop automated detection software, the team visited many computer-generated locations that were false positives for monuments. By ground-truthing these sites, the team refined software accuracy (photograph by J. McCorriston).

Figure 3.8. Matthew Senn and Joy McCorriston documenting a likely Neolithic platform near Andhur (photograph by J. Everhart).

31

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

While this is a more complex configuration than most archaeological applications, it provided better than 30-centimeter accuracy over a 150-kilometer radius. With 60-centimeter satellite imagery, this approach gives high accuracy in the registration of monuments at true location. The accuracy is important not only in the development of detection software (Schuetter et al. 2013), but also in subsequent analysis of monument landscapes and the attributes of different landscape locations (Figure 3.7). Survey teams visited each monument within the imagery to record its GPS coordinates and monument attributes, including the type, dimensions, construction details, relationship to other monuments, artifact scatters, state of preservation, and photographs, all of which have proved useful in post-field data cleaning and assessments (Figure 3.8). The team also recorded monuments encountered on desert tracks while the team travelled to satellite imagery (Figure 3.9). With this data, researchers could use the recorded tracks and monument locations to generate an 800-meter viewshed in GIS (Geographic Information Systems) software with a 3-meter DEM (Digital Elevation Model). The approach provides all known locations where monuments do or verifiably do not exist within a defined area and offers a dataset for landscape analysis of monument locations. Patterns in Time: Excavations With a major objective in exploring chronological patterns, the AHSD needed to probe the individual histories of a sample of monuments in Dhofar. While extensive survey could rely on known monument types and classify new types based on external characteristics, only excavation and the methods of relative and absolute dating can provide a chronology of such monument types. Even so, excavation demands more time and resources per monument than survey, requiring a sampling of surveyed monuments. In 2009, 2010, 2017, and 2018, the AHSD and ASOM teams excavated or tested 28 monuments, a sample of about 8% of all the monuments documented through archaeological survey (n=343). With test excavations, the team sought to generate data meeting the following goals: 1) clarify construction techniques; 2) determine phases of use and re-use; 3) identify uses, such as burials, visitations or offerings, signposting, and fires; 4) obtain samples for absolute (i.e., radiocarbon) and relative dating. The team guided its non-random selection of monuments for test excavations with subjective assessments of preservation and diversity. We used visual criteria from survey observations to identify monuments with a combination of promising features. Our experience in comparable environments in Yemen demonstrated a good probability that a promising monument would yield chronological data (McCorriston, Harrower et al. 2020). To assess what is “promising” for excavation, we relied on our and other archaeologists’ prior experience and on surface characteristics. Considerations guiding our selection of monuments for excavation were: a) Clear conformity to an identifiable type, either previously encountered and belonging to a well-defined class (e.g., trilith, boat-shaped grave), or b) Clearly other type, with attributes not conforming to a known class. c) Signs of human disturbance. Where a chamber entry is open to the sky; or some portion of the monument is missing stones; or a modern military hideout sits atop it; or tire tracks and a recent stone heap suggest quarrying; or a more recent type lies beside an earlier type missing elements (such as boat-shaped graves alongside triliths missing their uprights); then the team could identify disturbances likely to introduce later materials or interfere with interpretations of relative and absolute ages. d) Signs of natural disturbance. Posing comparable problems in interpretation, a fox-hole, tree-growth, termite mound, or hyrax midden can introduce later material and distort stratigraphic sequences important in chronological interpretation. 32

Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

e) Accessibility. In few instances could we justify a long commute, carry, and climb to transport people and equipment. Selecting monuments close to rural routes meant we could spend our resources efficiently. f) Indications of stratigraphic, relative chronology, such as sedimentation, buried stone, underlying sediments, a sealed chamber or intact capstone, diagnostic surface artifacts that might have accrued after construction.

Table 3.1. Excavated Monuments Monument Type

AHSD / ASOM No.

Regional Location

Excavation Detail

Datable Material

Platform (D-shape)

D028-001

Mudayy (Polygon 16)

northwest quadrant

animal bone, lithics

Platform (D-shape/ trapezoidal)

D038-003

Hanūn (Polygon 15)

northwest quadrant

coral beads; bulk organic sediment; burnt landsnail

Platform (D-shape)

D103-001

Wādī Dhahabūn

1 m trench across east interior

none

SCAB

D106-002

Wādī Dhahabūn

surface clearance

pedagenic CaC03

SCAB

D112-001

Wādī Dhahabūn

surface clearance, 1.0x0.5 m external trench

none

HCT

D001-001

Mudayy (Polygon 16)

chamber, cleaned ½ exterior

human bone, animal bone

HCT

D013-001

Halūf (Polygon 3)

chamber, wall sectioned at west

human bone

HCT

D013-002

Halūf (Polygon 3)

chamber

charcoal

HCT

D014-001

Halūf (Polygon 3)

chamber, cleaned ½ exterior, sectioned east-west through tomb wall

indeterminate bone; not datable

HCT

D014-002

Halūf (Polygon 3)

chamber, cleaned ½ exterior

human bone

HCT

D033-001

Hanūn (Polygon 15)

chamber

camel bone

HCT

D036-002

Hanūn regional

chamber

human bone, animal bone

HCT

D006-001

Jibjat (Polygon 2)

chamber, cleaned ½ exterior

bone

HCT

D104-001

Wādī Dhahabūn

chamber

mother-of-pearl ornament

Wall Tomb

D022-005

Mudayy (Polygon 16)

partial chamber, cleaned full exterior

bone

Halūf Tomb

D014-003

Halūf (Polygon 3)

chamber, cleaned exterior, emptied fill of retaining walls

dagger style, charcoal

Halūf Tomb

D001-006

Mudayy (Polygon 16)

chamber, cleaned ½ exterior

none

Trilith

D043-004d

Hanūn regional

0.5x0.5 m test in hearth

bulk organic sediment; charcoal

Trilith

D005-004

Mudayy (Polygon 16)

2 0.5x0.5 m quadrants in platform, ½ hearth

none

Trilith

D101-001

Wādī Dhahabūn

1.0x0.5 m external trench

none

Trilith

D102-001

Wādī Dhahabūn

1.0x0.5 m external trench

none

Trilith

D102-005

Wādī Dhahabūn

0.5x0.5 m quadrant

none

Pecked Boulder

D113

Wādī Dhahabūn

orthophoto

wusum

Chambered Cairn

D001-004

Mudayy (Polygon 16)

chamber, cleaned ½ exterior

human bone

Chambered Cairn

D001-005

Mudayy (Polygon 16)

chamber, cleaned ½ exterior

human bone

Boat-Shaped Grave

D100-001

Wādī Dhahabūn

0.5x1.0 m exterior trench

surface ceramics

Boat-Shaped Grave

D114-009

Halqoot

2.6x1.8 m trench

charcoal

Boat-Shaped Grave

D114-180-A

Halqoot

2.4x1.6 m trench

human tooth

Boat-Shaped Grave

D114-180-B

Halqoot

2.4x1.6 m trench

none

33

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

34

Archaeology in Dhofar: The Arabian Human Social Dynamics (AHSD) Project

Figure 3.9. Aerial view of D113-001, a prominent, immense, flat-topped boulder beside the track through one of Wādī Dhahabūn’s southernmost major tributaries. The elaborate images on its surface can only be seen from above, and there is no nearby overlook. This was a landmark known and shared as cultural memory. It is unlikely that all markings were produced at one time (image by W. AbuAzizeh and T. Everhart).

35

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

In no case did excavators entirely dismantle a monument. Test excavations rarely cut a wall. In the case of platforms, the team usually excavated a pie-shaped quarter, leaving the outer stone perimeter and chock stones intact while exposing two sections, perpendicular one to the other. In the case of SCABs (Stepped Concentric Alignments of Boulders), excavations removed only dislocated stones to reveal the underlying stepped, concentric alignments of boulders and flipped one or two perimeter stones to explore pedagenic carbonates (for radiocarbon sampling). In the case of high circular tombs (HCT) and Ḥalūf tombs, excavators usually cleared rubble to expose one half of the exterior face, examining a section through the accumulated rubble, and the team also removed capstones to excavate chamber sediments, by sectioned halves where cramped conditions permitted. In the case of triliths, the team excavated one or two hearths, usually only quarter or half-sections, leaving perimeter stones intact. In one instance (D005-004), excavators sectioned a trilith platform between uprights, excavating entirely to bedrock without encountering cultural material or disturbance. Finally, the team excavated small test trenches against the exterior perimeter of a boat-shaped grave to flip a stone for pedagenic carbonate sampling. With special permission from the religious authorities in Dhofar (Waqf), the team excavated sections through two boat-shaped-graves without displacing perimeter stones. In all cases, excavations followed natural stratigraphy, using a locus-lot system whereby natural strata are loci with arbitrary sub-divisions as lots within loci. The team sieved all deposits within monuments using ½ cm mesh and ~1 mm mesh flour sieves. Excavators always backfilled monuments using the original backdirt replaced over plastic mesh and modern coins to mark the bottom. Emic and Etic Approaches Recognizing that small-scale stone monuments are an important category of prehistoric cultural remains and had been incompletely documented in Dhofar, an American-Omani team dedicated itself to documenting spatial and chronological distributions of these monuments. On the one hand, we conceived this as a pattern-searching exercise, rooted in culture history. Methodologically, researchers complemented extensive survey with intensive excavations of a sample of monuments. We documented 343 monuments, of which we excavated or tested 27, yielding a rich and precise set of data. On the other hand, and beyond pattern-searching, this research engages social science theories of human landscapes and socially-mediated access to environmental resources. Thereby this research effectively tests a (null) hypothesis that monument distributions in space and time are indistinguishable from random distributions. If patterned, monument distributions should reflect historical accretions of culturally-shared meanings, social identities, and ancient worldviews. While archaeologists cannot expect to reconstruct monuments’ culturally specific meanings (the emics of monument meanings), we can be certain that such meanings did exist and furthermore could be accessed by people in the know. Thus monuments are technologies for archiving such information, and the means to access information are inherited. Both are etic issues to consider further in the next chapter.

36

Chapter 4

Cultural Inheritance of the Dhofar Pastoralists

Pastoral Prehistory in Southern Arabia In the southern highlands of the Arabian Peninsula, including Ḥaḍramawt, Mahra, and Dhofar, pastoralists have been the dominant shapers of environments and economy for at least 8000 years. It remains an open question whether domesticated animals originated from the eastern and northern flanks of the Fertile Crescent (Zeder and Hesse 2000; Peters et al. 2005), moved through interior deserts (Crassard and Drechsler 2013; Makarewicz 2020), or arrived from East Africa (Gifford-Gonzalez and Hanotte 2011). Regardless of their arrival route, domesticated cattle were clearly tended in the highlands of Ḥaḍramawt, Yemen 8000 years ago (Martin et al. 2009) and likely also in the Khawlan region (Fedele 2008). Detected by a few bones from 6th millennium BCE Manayzah, sheep could have arrived just as early (Martin et al. 2009). Both sheep and cattle are grazers (reliant on grasses), and their economic roles in a mixed herding-hunting economy had significant implications for Arabian landscapes.

Figure 4.1. Al-‘Alīy herder with goats in Southern Yemen (photograph by J. McCorriston).

37

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

More than 3000 years before the introduction of agricultural strategies, Arabian pastoralists managed grasslands and constructed ecological niches favorable to herded flocks and to themselves (McCorriston 2006, 2020d). In highland Ḥaḍramawt’s Wādī Sanā, phytolith studies of diverse ancient environmental habitats suggest that Southern Arabia’s pastoralists managed vegetation to optimize the region’s resources for sheep and cattle and that they did so from the first appearance of pastoralists in the region (Buffington 2019) (Figure 4.1). When evolutionary biologists consider the factors that shape an ecological niche favorable to humans, they recognize that inheritance plays a crucial role (Laland and O’Brian 2011). Both ecological inheritance – for example, inter-species relationships and environmental manipulations – and cultural inheritance like knowledge and technologies contribute to fitness of descendant generations (Odling-Smee and Laland 2011). Archaeologists increasingly recognize that pastoralists and their herding behaviors have shaped most of Arabia’s arid ecosystems (Wilkinson 2003; Guagnin et al. 2016; McCorriston et al. 2018), including those of Dhofar. It is therefore critical to recognize the cultural inheritances of persistent pastoralism. No one is born into a cultural vacuum. Persistent pastoralism flourished largely through the ecological and cultural inheritances of pastoral societies. As domesticated herd animals, principally cattle, goats, and sheep, increased in number and significance, they ultimately displaced hunting as an important terrestrial source of protein (McCorriston 2013b). This shift shaped an ecological inheritance for subsequent generations of humans now dependent of pastoral animals. Ecological inheritance included the mutual relationships of humans with domesticated animals and ecosystems managed for grasslands and tender browse. Humans uniquely also rely on cultural inheritance, which includes traditional ecological knowledge transmitted through social engagement (i.e., learning). For pastoral societies in ancient Arabia, cultural inheritance can be further specified from the archaeological record. Cultural Inheritance of Pastoralists The archaeological record is not yet clear whether Dhofar’s first pastoralists migrated in with their herds from other regions or – as many archaeologists working in Yemen and Oman believe – were native to the Southern Arabian highlands and acquired their animals (Al-Abri et al. 2012; Crassard and Dreschler 2013; Rose et al. 2013). There were certainly extensive mechanisms for exchange that left traces in the distributions of East African obsidian across Arabia (Khalidi et al. 2013) and marine shells far inland (Crassard 2008; Crassard et al. 2020). These archaeological traces also signal relationships along which information, perishable material culture, and other ephemerals, such as domesticated animals, could also travel. Because these exchanges occurred across different ecological systems, the cultural know-how of herding domesticated animals was even more critical as domesticates encountered new habitats. Ownership and Cultural Inheritance Wild game belongs to any hunter, but a domesticated animal is owned. To tend an animal from birth is to defer one’s return on an investment of energy – what I save today, I eat tomorrow. This deferral explains herders’ proprietary ownership of their animals (Alvard and Kuznar 2001) and herders’ willingness to risk injury to defend them from predators or raiders. Domesticated animals are a defensible resource even where grasslands and water may be freely accessed and may not be defended or defensible by mobile pastoralists (Cashdan 1983; Kelly1995, Moritz 2016). In Arabian antiquity, defensive strategies toward water and grazing land remain unclear. 38

Cultural Inheritance of the Dhofar Pastoralists

Figure 4.2. Neolithic points from surface collections at Bā Mashnayq, Mudayy (third from left) and in Wādī Ghārah, a tributary to Wādī Ghadūn (photographs by M.Senn and E. Lagan).

Domesticates in Dhofar There are few Neolithic faunal remains (8000-5500 years ago) reported from Dhofar and none clearly identified as domesticates. Bone preservation is poor, sites shallow and small, and there are few excavations. Nonetheless, the finely knapped points appearing in the Arabian Neolithic concurrent with the first domesticated animals (Charpentier 2008; Crassard 2008) may be as much for defense of herds against raiders as for hunting or deterring wild animals (Figure 4.2). Dhofari people certainly had contact with herders elsewhere. The stylistic forms of knapped stone projectile points common to sites like Ad-Dahārīz terrace (Crassard et al. 2020), Hailat Araka (Zarins 2013), Muḍayy, Muṭahafah, and Shisr (Zarins 2001) signal transfer of knowledge linked to broader cultural systems and likely social systems. While not each person with a projectile was by inference a herder, the common cultural inheritance of how to make a certain point conveys with it a common understanding of how to use one and against which targets. Dhofar was no information isolate from its neighbors in Mahra, Ḥaḍramawt, Duqm, and Ash-Sharqiyah. If people learned how to make points from each other, they could also be sharing domesticated animals, instructions for their maintenance, and the cultural proscriptions and penalties associated with ownership. Beyond Dhofar, there is direct faunal evidence for early herders from the site of Manayzah, a rockshelter in the upper Wādī Sanā (Ḥaḍramawt). Faunal remains from 8000 years ago include wild game (gazelles) and domesticated cattle and sheep. Caprine bone could not be identified to species, but some probably came from domesticated goats. The site is small and was only seasonally occupied by small groups who nonetheless left marine shells, obsidian, polished jaspoid adze trimmings, and exotic stone beads as material evidence of wide social and exchange networks. That Manayzah herders also practiced a common style of stone projectile point production signals their shared cultural inheritance with people in Dhofar, who themselves might have owned domesticated herds 8000 years ago. 39

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 4.3. Neolithic ring of skulls from sacrificed cattle at Shi’b Kheshiya, Yemen. The skulls were inserted fresh into soft mud. Later, visitors erected a commemorative stone platform nearby (photograph by M. Harrower).

In time, the strategies for defending domesticated animals developed significantly from a solo spear to a social collective enforcing proscriptions and penalties. The evidence comes from excavations in Wādī Sanā. At the open-air site of Shiʿb Kheshiya, excavators uncovered a ring of Neolithic cattle skulls sunk into a marshland 6500 years ago (Figure 4.3). Analysis of the bone and its context shows that herders convened to sacrifice more than 40 cows for a huge feast that fed groups arriving from distant wadis (McCorriston et al. 2012; Henton et al. 2014; Martin 2020). Even though these rituals took place many thousands of years ago, the practices of gathering, sacrifice, and feasting have been important to social communities ever since across large parts of the Arabian Peninsula (McCorriston 2011). In the middle Neolithic, this brief union of mobile herders facilitated social ties and exchanges critical in maintaining and sharing traditional ecological knowledge of animals, grazing land, vernal pools, and social expectations for accessing them. Behaviors by parents acting within social cooperatives, like the one in Wādī Sanā, affect their children’s access to resources beyond the household. People belonging to a group may sacrifice something but in return may draw upon each other. Social collectives and the rituals to support them are a cultural inheritance of persistent pastoralists. Chapter 5 explores the evidence for such ancient social collectives among herders in Neolithic Dhofar.

40

Cultural Inheritance of the Dhofar Pastoralists

Messaging-While-Absent Another of the cultural inheritances of Arabian pastoralists is the system to communicate across broad social networks, especially networks of mobile, dispersed people discontinuously and infrequently in direct contact. Herders need to direct animals to good pasture, which varies with seasonal and inter-annual rainfall and local pressures from other herds. On the narrow Dhofar plateau, grasslands could develop on extensive soils, but elsewhere, grasses are highly restricted to loamy terraces, fans, and the margins of vernal pools, distant and unpredictable in availability. Information is crucial, and pastoralists messaged others in their absence. Dhofar’s pastoralists built monuments to convey messages. Just as the tomb described in Chapters 1 and 6 (D001-001) archived information about who were buried there and how one related to them, all monuments encode social information accessible as indirect communication (Rowlands 1993; Van Dycke and Alcock 2003). This principle has long applied to Oman’s northern Bronze Age tombs – Hafit and Umm an-Nar styles – suggested to signal territorial rights of the groups whose ancestors lay there (Cleuziou 2002; Giraud and Cleuziou 2009). Whatever the signal in Dhofar, as with direct communication, transmission is subject to interpretation and manipulation, so the “reader” may re-formulate an embedded message. The flexibility of such messaging and reformulation offers a venue for negotiating social relationships, inserting oneself in some, shifting the power in others, and affirming one’s ties throughout. One explanation for messaging-while-absent is that the labor invested in building monuments signals a willingness to invest in other strategies, such as those that protect livestock and ensure access to resources. Just as the energy required to make a projectile point produces a signal that one is willing to defend resources, so the energy invested in cooperative monument construction signals social cooperation to access resources (McCorriston 2013b). For mobile pastoralists usually dispersed, social cooperation was essential for sharing information, acquiring exotic materials, marriage partners, and exchanging livestock. Knowing or interpreting one’s connection to monument builders in a specific place secured social cooperation paths to such benefits locally. Messaging relied on culturally-inherited symbolic repertoires, such as a sacred or funerary aspect comprehensible to other groups. Technologies Even as humans are inventive, they are also learners. Learned technologies are a cultural inheritance that shapes a pastoral niche and promotes the persistence of pastoralism for generations to come. In this capacity, culturally-inherited technologies in ancient Dhofar would accomplish some or all of the following: stabilize the predictability of resources; improve resource abundance and quality; narrow the ranges between patches of resources (like grazing land); diversify the resource base (such as mixing hunting with herding, or extracting milk from herd animals, or farming on the side); or intensify exploitation of resources (such as regrowth after scheduled burning, or raising surplus animals to exchange with traders). To understand the landscape that perpetuated pastoralism, it is critical to recognize the technologies that shaped it. What can archaeologists detect or infer of these technologies in Dhofar? Domesticates First, there are the domesticated animals themselves – cattle, goats, and sheep probably in Dhofar 8000 years ago. Later additions were donkeys and camels, the latter perhaps 2400 years ago. Accompanying technologies included corrals, tethers, milking and milk control with the paraphernalia they require, gelding, hornblunting, and domesticated dogs used in the manipulation of domesticated herds. Dogs are widely depicted 41

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 4.4. Painted images on the rock walls of natural rock shelters in Dhofar’s coastal mountains often include images of domesticated dogs, camels, cattle, and goats. These are at the site of Mthbon, southeast of Jibjat (photograph by J. McCorriston and K. Olson).

on rock images throughout Arabia as early as 8000 years ago (Khan 1993; Inizan and Rachad 2007; Avner et al. 2017), and they appear as painted images in Dhofar’s rock shelters 2000 years ago (Figure 4.4). As a means of protecting herds from predation, dogs serve to stabilize the predictability of resources. As a means of extending human potable sources across arid lands and at fouled water holes, milch cows, nanny-goats, and she-camels improve resource abundance and quality. Projectile Points Weapons used in defense also served to diversify resources when used for hunting. Projectile points were in Dhofar earlier than domesticated animals, insofar as the limited data on domesticates suggests (Hilbert 2013; Cremaschi et al. 2015). Even so, there was a significant change and greater skill in the technology and skill invested in their manufacture 8000 years ago, a time when they may have also been re-purposed to stabilize the predictability of domesticated herd animals. Anthropogenic Burning In many ecosystems, humans set periodic fires to remove old, tough vegetation, release their nutrients as ash, and promote new growth. The timing of such fires will also affect vegetation composition. In Dhofar, anthropogenic burning increased the abundance of grassland resources by discouraging competitor plants like perennial shrubs (Sale 1980; Kürschner et al. 2004). As grasslands expanded, the range between grassland patches decreased while increasing the predictability of tender shoots nutritiously best for herbivores.

42

Cultural Inheritance of the Dhofar Pastoralists

Monument Construction A particular monument construction style engages learned technology. Knowing the correct construction techniques to archive a particular social message can project that message to others. Such messaging can stabilize the predictability of resources. For example, where the monument signals that a social cooperative built it, its existence suggests assertion and recognition of socially-conferred rights. Such rights may limit over-use or unexpected depletion of resources like grass or herd animals. In other words, newcomers may be deterred from raiding where they encounter signs of a contemporary and engaged social collective. By reducing conflict and promoting cooperation, monuments construct a cultural inheritance of persistent pastoralism. Houses, Corrals, and Byres Settlements were few in ancient Dhofar, but where they occur, they were generally the work of pastoralists. Archaeological excavations at Halqoot and Shakeel revealed 2000-year-old sturdy, circular, one-room houses with thick dry-stone walls and center-post roofs alongside cattle byres and goat pens (McCorriston, Buffington et al. 2020, Chapter 7 this volume). Settlement survey on the eastern Jabal Qara detected multiple stylistic forms, suggesting different episodes of settlement construction (Buffington et al. 2019). By settling down even seasonally, herders could both intensify their exploitation of grasslands and stabilize the predictability of springs, pools, reservoirs, and seasonal vegetation. Animal confinement also implies concentration and collection of dung and dung ash, fertilizers for plant productivity in subsequent seasons (Figure 4.5).

Figure 4.5. Traditional style cattle byre in contemporary Dhofar mountain village. Confined animals produce a large accumulation of dung, useful as fertilizer or fuel (photograph by J. McCorriston).

43

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

By confining young animals, pastoralists minimalize risks of predation and of weakening juveniles – animal and human – following full-grown mothers all day. Furthermore, such control of young animals is critical in milking, which diversifies pastoralists’ resource base, promotes human reproduction and population growth, and preserves more herd animals (O’Brian and Bentley 2021). Modern Dhofari people express a high incidence of the -13910*T and -13915*G alleles for lactose persistence (Al-Abri et al. 2012), suggesting many thousands of years for this gene-culture co-evolution developing in place. Materials Destined for Exchange Technologies and tools to prepare materials for exchange can work to stabilize the predictability of resources. Dhofar is a rich – and from endemic species, unique – source for plant adhesives, fish poison, feathers (e.g., steppe eagles, flycatchers), fibers (e.g., wild okra, palms), hides, meat, cheese, horn, resins (e.g, frankincense), and wood. Acquiring and processing these materials provided pastoralists materials to foster economic and social relationships buffering herd depletion and scarcity. Although these non-durable materials rarely survive in the archaeological record, the traces of acquired goods such as exotic stone, amber, marine shell, and metal attest to the networks along which they travelled. Conclusions Dhofar today is a landscape shaped by persistent pastoralism to perpetuate pastoral lifestyles and the success of future pastoralist generations. The environmental imprint of pastoralists is everywhere evident, with anthropogenic tall grasslands historically extensive across the plateau, fragmented forests in the escarpments, routes connecting Nejd watering places, livestock breeds adapted to Dhofar, and a severe reduction in large predators like Arabian wolf, panther, and leopard. At the same time, the cultural inheritance of Dhofari pastoralism is also clear, even where the archaeological record is relatively poor in durable material culture. The strong cultural significance attached today to owning and maintaining livestock, even where no longer nutritionally necessary or efficient, is an inheritance of pastoral traditions. In Dhofar, the monument technologies of messaging-while-absent line the horizons and routes of pastoralists’ passage, and painted images adorn rockshelters in the escarpments. Stone tools outlast the perishable technologies, but their presence attests not only to defense and hunting but also to working wood, hides, and bone. By considering the linkages between material culture and the landscapes they shaped, this chapter casts the archaeological record of monuments, settlements, camps, and artifacts in a broader context of social meanings and Dhofar’s cultural inheritance.

44

Chapter 5

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Alfin – we will call her thus – looked over the growing Witness Place where the sun sank behind the western hill of Bā Mashnayq. With her slender hands she heaved the rough cobble she carried up over the Sentinels, a line of upright limestone slabs facing the sun’s renewal place. Already, after only a day of toil, the Gathering had showed its strength, with 12 herds come to Renewal. So many hands to carry meant the work had gone swiftly. The Sentinels were completed, an enclosure of massive uprights looping back from the Principals, which stood in a line facing east. Alfin imagined herself one of the journey birds who settled at the Muḍayy springs after the water season faded. If one were to pass now (although they wouldn’t come for some moons yet), it would see her, a slight girl in yellow hide-wrap, her massed hair plumed with chuckar feathers, standing in the sunset gust and facing a squat, half-finished stone platform. Defined by a perimeter of limestone uprights, the interior was still a shallow fill of cobbles among an uneven jumble of small boulders, most of which supported uprights balanced on the rock surface. A bird’s eye would also discern fragments of cattle bone and the occasional pierced seashell ornament, remnants of the sacrifice and great feast that took place a year ago. And then a bird would fly on, drawn instead to the fresh water, feathery shade of tamarisk, teeming feast of flies and crawlers, and the shelter of reeds and rushes across the Bā Mashnayq marshlands. Alfin’s eyes swept over the remnants of the Sacrifice. They were a low platform, just one cobble deep, against the south edge of the Sentinels. On top leaned the crumbling crania of a dozen cattle. Bleached, listing, horns splayed, blank-socketed, mute to Alfin’s growing anxiety, they offered none of the bloody bellows of yesteryear. With her usual limp, Alfin turned away, facing instead the steep terrace edge, below which Renewal spread across the marshy flats of Bā Mashnayq. Standing at the cliff-edge, she could look down on half a dozen open hearths, where Renewal folk stirred the embers and blew on their first glimmers after sunset. Children brought armloads of twigs and dry stalks for burning. The last cobbles cluttered into the Witness Place, and people filed away to clamber down the cleft leading to the valley below. Across the marsh, more fires winked in, like new stars against the black cliff, and the bleating of goats culled for slaughter mingled with the plaint of cows heavy for milking. Last to leave, Alfin turned to join Renewal feasting. Alfin limped, and all her 14 years herding cattle and the few family goats, she had known she would never be an Exchange. No one would give cattle for a limping girl! At the Gathering last year, a mate had joined with Alfin’s sister, and Alfin’s herd sent away two heifers in exchange and gave an old cow for Sacrifice. No one spoke for Alfin then, but at this Renewal a new herd had arrived. No one knew them, but they spoke of other Gatherings, and with conspicuous energy, they joined in building the Witness Place. Last night the father spoke for the new youth, who asked for… Alfin. He was an unlooked-for Exchange, especially with three heifers and a nanny goat offered against her departure. With each stone she lifted, she stepped closer to parting and her uncertain future with a new herd… Alfin’s story is fanciful; you may rewrite it yourself, for her or for any other of the Neolithic herders who spread sparsely across Dhofar’s patchy Nejd 7000 years ago. But if you do, make use of what we know of these times from the archaeological remains left by pastoralists in ancient Dhofar. The data come from excavations and discoveries at small-scale stone monuments (Figure 5.1), and they have been interpreted in the context of similar sites published elsewhere (McCorriston et al. 2012; McCorriston and Harrower 2020; Martin 2020). 45

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 5.1. Image map of Neolithic platform monuments located by the AHSD Project survey team (image by K. Olson).

Figure 5.2. Bedrock terrace at D028 overlooking vegetation fed by one of Mudayy’s springs; note the stone monuments on the terrace and accumulated sandy sediments in the drainage below (photograph by J. Everhart).

46

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Neolithic platforms Archaeology marks time by the appearances of new technology, so the term “new lithic,” (Neo-lithic) in Arabia suggests that it was a change in stone tools that ushered in a new era. This terminology only hints at the changes in human society that arrived with the technologies of pastoralism – ownership, deferred resources, broader social collectives, and messaging while absent, to name a few. While one may mark the beginning of the Neolithic with distinctive projectile points, techniques for making them, and domesticated animals (around 8750-8000 years ago), there was a clear change around 7500 years ago. At this time, pastoralists began to build small-scale stone monuments, and the evidence from Shiʿb Kheshiya in Yemen suggests that these commemorated episodes of gathering, sacrifice, and feast (Chapter 4). Alfin’s story is based on a Dhofar Neolithic platform excavated at Muḍayy. D028-001 On a bedrock terrace overlooking Bā Mashnayq and one of two Muḍayy springs sit several prominent small-scale stone monuments (Figure 5.2). One is the squat-shaped platform with a perimeter of unworked limestone slabs and a jumbled stone fill (D028-001, the “Witness Place” of this chapter’s opening narrative). The platform is 7.5 m diameter with one straight (eastern) side, and it stands 0.85 m high. About 15 m to the south is a second monument (D028-002), this one constructed with stepped, concentric alignments of boulders rising from a lower perimeter (6.5 m diameter) to a central peak a meter high. Its uppermost rings seem disturbed, with a hollow of silty sediment at the top. Directly to the east, about 4 m away, is a quadruple set of cobble alignments 3.3 m length, and the terrace also has other, smaller, lower constructed features. At its northeast edge is a low platform that may be the residue of something dismantled to make a modern military blind. An episode in the 1970 Dhofar Civil War occurred here, overlooking the soft, sandy valley infused with charcoals from ancient hearths. Surely some of these hearths, now sadly bulldozed, dated to the Middle Neolithic (7500-6000 years ago) and were the camps of pastoralists who built these monuments.

Figure 5.3. Finely knapped bifacial points from the surface at D028 (image by M. Senn and E. Lagan).

47

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 5.4. Neolithic platform sectioned in one quarter, with exposed fill and large perimeter slabs supported by interior blocks. A second large monument is visible in right background (photograph by T. Steimer-Herbet).

Figure 5.5. Found in the lower fill of D028-001, a pierced shell of Impages hectica, a marine shell likely used as an ornament (image by C. Heyne, J. McCorriston and E. Lagan).

48

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Excavations at D028-001 and a detailed survey of its terrace context revealed that the monument sat on a sterile desert pavement over bedrock and that the finely-knapped Suwayh-type points around it accrued after the monument was built, discarded perhaps several hundred years later (Charpentier 2008) (Figure 5.3). By removing fill in the northwest quadrant to reveal several profiles across the interior of the monument, excavators could determine that a perimeter of upright limestone slabs was braced by heavy chock stones of up to 0.50 m diameter. Subsequent fill included boulders, cobbles, and pebbles, all clastic limestone, and an accumulation of silty sediment among them (Figure 5.4). Only a scant few chert chips and flakes occurred in the fill, and none of these matched the chestnut-brown raw material of tools on the terrace surface. In addition, the fill contained a few unidentifiable fragments of badly weathered animal bone, one within the size range of a large caprine or small cow. At monument’s center and 0.80 m below its top, the fill contained a pierced pendant of marine shell (Figure 5.5). Even as the construction style and form resembled Neolithic monuments known from Yemen, a radiocarbon age obtained on faunal long bone (apatite) in the fill confirmed its Neolithic age, 6610 +/- 44 14 C yr BP (AA95064), calibrated to lie between 5620-5486 BCE. The monument’s connection to and use of a one-cobble deep, trapezoid-shaped platform, about 1.5 m wide and abutting at the south, remains illdefined (Figures 5.6 and 5.7). Given the location on a rock terrace with desert pavement, nothing but stone and buried bone within D028-001 remain of whatever activities occurred here. D038-003 Like D028-001, D038-003 sits atop a low bedrock terrace thinly covered by desert reg, a natural pavement of small, clastic pebbles and cobbles trapping windblown sediment below. The terrace is tongue shaped; it overlooks the confluence of two shallow drainage channels feeding the broader system at Hanūn, with a deep rock pool in a wadi bed, offering reliable water through long dry months. Hard limestone layers at the base of many feeder channels offer widespread seasonal opportunities for vernal pools to water herds, making this a stopping place over the centuries (Figure 5.8). Frankincense trees grow here, just at the limit of monsoon fog, in an environment with more atmospheric moisture than the oasis at Muḍayy. Under one ledge of the terrace, dry-stone walling seals a rockshelter that almost certainly contains burials of the past 500 years (Al-Shahri 1991), and above it, several undisturbed Muslim graves sit along the edge. Along the opposite edge is a small trilith monument; eight more extend along the terrace, with others visible across the wadi. Hanūn lies in the region with the highest density of triliths in all Dhofar and perhaps all of Arabia (Chapter 8). Long before these were built, people constructed two, possibly three, platform monuments along the tongue-shaped terrace. Two are well preserved, with an elongate D-shaped outline of uprights, the straight edge facing east (Figures 5.9 and 5.10). One of the eastern uprights of D038-003 had slumped in antiquity to become covered by rubble collapse spilling outward and by the windblown silt trapped by such debris. This slumped stone measured an astonishing 1.65 m in length and must have been impressive as a focal upright. To the east is a low circular enclosure wall, 2.65 m diameter, built of stones laid along their axes (not uprights) and filled with a few large stones and sediment. One large stone across its center may be a fallen upright, a betyl like those at Neolithic platforms elsewhere (McCorriston 2011: 94, 97, D103-001 below). A later hearth deposit of charcoal-rich sediment and burned land snails overlay one edge of the stone circle. Altogether, D038-003 extends west-east as a 6.80 m platform, plus a 1.65 m fallen upright, plus a 1 m gap, plus a 2.65 m circle, that is, a length of 12.10 m. Directly to the east, D038-011 is less well-preserved but conforms to this layout.

49

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 5.6. Overview of Neolithic platform D028-001 with excavated northwest quadrant (photograph by T. Steimer-Herbet).

Figure 5.7. Plan of Neolithic platform D028-001 (image by T. Steimer-Herbet and K. Olson).

50

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Figure 5.8. Hard Umm er-Radhuma limestone layers harbor rock pools at Hanūn (photograph by J. Everhart and K. Olson).

Suspecting these to be Neolithic monuments, the AHSD team excavated a 2.5 m wide northwest quadrant to bedrock, finding only a few chert flakes and a cow-sized tooth enamel fragment among the clastic limestone rubble of small boulders, cobbles and pebbles that comprised the fill. Among these had also settled fine silt, probably from desert storms that deposit loess in the quiet interstices of piled stone. At present, the monument stands 0.63 m height. In the bottom layer of fill and at the center, excavators recovered two tubular white coral beads, each about a centimeter in length (Figure 5.11). A 1 m excavation along the northern exterior established a short profile against the uprights’ exterior. The profile suggests that the perimeter uprights were chocked in a shallow trench dug into desert reg. Their standing faces today are protected by a spill of rubble fill from the platform interior. None of the fill, neither inside nor the rubble collapse, contain any of the abundant thermally-altered rock that litters the terrace and surface of the platform. This suggests that the fires nearby occurred after monument construction. It is moreover evident that the collapse of the interior of the platform occurred outward after weathering in upper part caused severe degradation of the limestone uprights, which flake and crumble. Their condition is quite unlike the firm uprights of the adjacent trilith. Dates for D038-003: The differential condition of stone uprights weathered in place in platform (Neolithic) and triliths (Iron Age, see Chapter 8) are only one clue to the relative age of the former. There are three organic samples providing radiocarbon ages for this monument, and to understand their implications, some explanation is required. First, the coral bead incorporated at the time of construction implies that the platform must date later than about 4686-4547 BCE (UGAMS40177, 5760 +/- 20 14C yr BP). Using a marine correction factor local to the western Arabian Sea (Southon et al. 2002; Saliège et al. 2005) the median date is 4444 BCE for the terminus post quem of monument construction. A second radiocarbon age, (AA90335, 5370 +/- 36 14C yr BP) around 4331-4058 BCE comes from bulk organics in the hearth overlying the circular structure to the east. It is a terminus ante quem, meaning the circular structure was built earlier. Yet the bulk 51

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 5.9. D038-003 on a low terrace; a trilith in the foreground may have robbed stone from the less preserved Neolithic monument under excavation in the rear (photograph by M. Senn).

Figure 5.10. Plan of Neolithic platform D038-003 (image by T. Steimer-Herbet, J. McCorriston and K. Olson).

52

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Figure 5.11. Twin coral beads from the fill of D038-003 (image by C. Heyne, J. McCorriston and E. Lagan).

sediment was close to modern surface and contains modern rootlets, so it may yield an estimate younger than the deposition of the sediment itself. On the other hand, the (marine) calibration method used for the coral bead results in a date younger than its radiocarbon age would seem. Another estimate for the terminus ante quem of the circle is the burned land snail (AA90334, 2194 +/- 53 14C yr BP) with a calibrated age around 200 BCE. Snail carbonates incorporate older carbon (from limestone) during their lives, so the radiocarbon age may be somewhat older than the death of the snail. This assumes its burned appearance was caused by the fire during the use of the hearth, which could also be considerably later than monument construction. With a Bayesian analysis of probabilities that assumes the platform and circle are roughly contemporary to each other and that the hearth must be younger, the posteriors indicate a range of 57093230 BCE after which the platform was constructed (mean around 4000 BCE, with marine correction). Moreover, the monument construction was completed before revisits began, 4501-2382 BCE. Middle-late 5th millennium BCE is a chronology consistent with platforms elsewhere (Table 5.1). Taphonomic interpretations: Stratigraphic evidence, including the preservation and length of a huge, fallen upright at the east end, suggest that this platform has stood far longer than the neighboring triliths. An older construction than the triliths, the platform experienced thousands more years of exposure and decay in desert thermal and moisture extremes. As uprights weakened, they crumbled, and the fill spilled outward, covering and preserving the bases of perimeter slabs. These are nonetheless fragile, spalling chips as soon as they are uncovered, while uprights of the adjacent trilith show little weathering. The site was also re-visited, not only evident from thermally-altered rock debris and recent burials, but also the numerous triliths in the vicinity. A peculiar small feature on the terrace is a square of uprights with a central stone (Figure 5.12). If utilitarian, it might have served to chock a post; nonetheless, it is possibly symbolic in nature, like some of the deliberate offerings and caches found in northern deserts (Avner 2018). It remains unexcavated. It is conceivable that an act of iconoclasm or convenient re-use would have removed a standing stone from the eastern circular structure (D038-003a), set in place to support one. Such reconstruction is conjecture, based on similar examples elsewhere. Iconoclasm might also have toppled the 1.6 m slab in the middle of the east end of the platform. While its fall is conjecture, it is incontrovertible that a slab this size required a large crew, a minimum of ten strong adults, to hew and carry it. 53

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Table 5.1. Bayesian analysis (Chronomodel software) with 210-year marine calibration per Southon et al. 2002, Saliège et al. 2005.

Phase Before Platform Construction (begin)   Before Platform Construction (end) Revisits (begin)     Revisits (end)

Lab #

14C Age

Unmodelled Calibration BCE/CE (Oxcal)

Bayesian Posterior 2σ

Posterior Median

 

 

 

5709-3230

 

UGAMS40177

5760 +/- 20 

4685-4547

 

4444 (w/marine)

 

 

 

5709-3230

 

  AA90335 AA90334  

  5370 +/- 36  2194 +/- 53   

  4331-4058 389-111 BCE  

4501-2382     4427 BCE- 41 CE

  3759  

One must also note that newer monuments – triliths, graves – did not apparently quarry the older platforms for stone, for the upright stone on these newer monuments is less weathered than the platform uprights. Cultural conventions required new stone and transport to build new monuments. Moreover, these conventions overrode the energetic efficiencies in quarrying older monuments – it was simply more work to bring new stone uphill than pluck adjacent stone. D038-003 is but one example where thousands of years later, new builders let the existing monuments stand.

Figure 5.12. Deliberate stone arrangement several meters to the north of D038-003 (photographs by J. McCorriston and K. Olson).

54

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

D103-001 In an eastern drainage of Wādī Dhahabūn lies a small monument conforming to the style of Neolithic platforms. It sits on a terrace about 3 m above the current wadi channel and at the base of a limestone cliff. Large, regular-spherical chert boulders erode from a distinctive white stratum high in the limestone cliff, and nearby sandy wash supports broad stretches of native dwarf palms (Nannorrhops ritchiana, Griff. Aitch.), with flat-topped Vachellia tortilis in the channel itself. Like the monuments at Muḍayy, D103-001 lies well inland of the penetration of monsoon fog. The terrace on which it sits has a surface of clastic desert reg trapping thin silt over oxidized sediment and rolled limestone cobbles laid as an ancient streambed. There are ample un-modified flakes and shatter from chert knapping, but survey found no formal tools or diagnostic debris on the terrace, with one exception, a limestone plaque chipped into irregular form. At the far east edge of this stretch of terrace, where the current channel clips against the cliff, there is a ring of large cobbles with the distinct cobble fill of a hearth (Figure 5.13). None of these features offer material for absolute or relative dating. The platform monument itself is D-shaped, with an eastern end defined by three large flattened-sphere chert boulders set on their rims in straight alignment (Figure 5.14). The remaining perimeter is made of stream-rolled boulders and measures 3.10 m east-west and 2.30 m north-south. At the surface, the fill is also of stream-rolled limestone cobbles, most around 15 cm in diameter. Outside and to the east is a shattered limestone slab, once about a meter in length and, based on comparable monuments elsewhere, probably once set up as a betyl (McCorriston 2011: 94, 97) (Figure 5.15). We observed chipping and scoring – the latter possibly from knife sharpening – on one of the eastern uprights, but it is unclear whether this activity dates to the initial construction and immediate use of the monument. The AHSD team excavated a small (0.63 m north-south) trench against the exterior north of the monument perimeter stones and another (Trench B) 1.30 m east-west x 4.10 m north-south in the eastern interior. The monument fill was 23-29 cm deep, with larger stream-rolled limestone cobbles and small boulders underlying the upper cobbles, including small boulders as interior chock stones against the eastern uprights. Throughout the fill an active colony of harvester ants had a nest, with substantial bioturbation and introduction of modern organic material into the silty sediments (Figure 5.16). Excavations recovered no artifacts beyond a few undiagnostic and unmodified chert flakes, and there was no burial. An attempt to expose pedogenic carbonates for radiocarbon analysis proved futile, as there are insufficient accumulations under the stones. While it proved impossible to date this monument using absolute methods, the conformity of shape, construction technique, orientation, and exterior standing stone to the east strongly argues for a Middle Neolithic date between 5500-4000 BCE. SCABs (Stepped Concentric Alignments of Boulders) When we started survey in Dhofar in 2009, we encountered unfamiliar monuments, which we registered as “tumuli,” with careful documentation of their locations, dimensions, construction materials, nearby monuments, and any surface artifacts (usually none). Subsequent analysis of 281 monuments shows that 18 of these are low mounds of stone conforming to stepped concentric alignments of boulders, or SCABs. None have been excavated, but from surface indications, they appear to have been constructed by laying a foundation of boulders, probably starting with a tight inner circle and increasing its perimeter, ring by ring. This provided a stable base for the next layer, with its perimeter inside and supported by the outermost basal stones. With repetition, each layer decreased in diameter, achieving finally the shape of a broad, truncated 55

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 5.13. Hearth D103.002 at the east end of the terrace on which sits D103.001. Although the hearth is of indeterminate date, it is one of few signs of subsequent prehistoric visits on this terrace (photograph by J. McCorriston),

Figure 5.14. D103-001 looking east before excavation. Note the alignment of upright boulders facing east (photograph by J. McCorriston).

56

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Figure 5.15. D103.001 viewed from the east, after excavation. Note the shattered limestone betyl in the left foreground and knife sharpening scars on the central upright (photograph by J. McCorriston).

Figure 5.16. D103.001 east-facing interior profile viewed from the east (photograph by J. McCorriston).

57

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

cone. In several cases, a loop of basal cobbles defines a feature abutting the base of the SCAB – these also have not been excavated but might mark activities taking place after the original construction. D106-001 and D106-002 Situated on a low gravel terrace beside the modern channel, these are a pair of SCABs in the easternmost branch of Wādī Dhahabūn, only a few kilometers upstream from the platform at D103-001. But for a spare acacia (Vachellia tortilis), the terrace is all but bare of vegetation. Survey of its surface revealed no chipped stone, apart from another chipped limestone plaque. Both SCABs make use of the flattened-sphere, large chert nodules (ca. 40-60 cm diam) eroding from the exposed cliff face (Figure 5.17). The largest of these, well over 1 meter in length, has been modified by deliberate knapping to taper to a point, and it lies to the east of D106-001, east of a short array of a single limestone slab and chert pillows set on their rims as the easternmost perimeter of the basal level of the SCAB (Figure 5.18). In other words, there appears special attention to a distinctive eastern aspect facing a former standing stone – a betyl – the only modified stone in the construction of D106-001 or D106-002. D106-002 lacks both the distinctive alignment of chert boulders along its eastern perimeter and any remnant standing stone. This monument thereby conforms closely to the shape and features of SCAB construction. Samples of pedogenic carbonates underneath one of the basal perimeter stones from each SCAB remain unanalyzed by radiocarbon, but may eventually contribute to dating the monument. Eastern-facing uprights and an eastern standing stone are elements characteristic of Neolithic platform monuments, and the location of this pair of SCABs on a low terrace also conforms to Neolithic platform locations.

Figure 5.17. D106-001 SCAB with east-facing alignment of larger uprights. Note the meter-long shaped stone fallen at southeast (orthophoto by W. AbuAzizeh and T. Everhart).

58

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Figure 5.18. D106-001 SCAB with east-facing alignment of larger uprights. Note the betyl stone with shaped, tapered end in foreground, approximately 1 meter length (photograph by J. McCorriston).

Figure 5.19. Pedagenic carbonates (white, left) formed by the interaction of soil biota and limestone upright after its placement in the monument. Because the living organisms depositing carbon in soil carbonates postdate the monument construction, a radiocarbon age would secure a terminus ante quem for the construction (photograph by J. McCorriston).

59

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

D112-001 Another unexcavated SCAB lies in a western tributary of Wādī Dhahabūn. Like the monuments D103-001, D106-001, D106-002, this SCAB lies on a low gravel terrace adjacent to the modern channel (Figure 5.20). Apart from a modern shelter crudely constructed of undressed stone walls, the only cultural remains on the terrace are the SCAB and a small trilith. In this region of the Nejd there is no fog penetration, and the few acacia and Ziziphus trees in the surrounding landscape are deep-rooted near and in the modern channel. The channel itself is braided across many kilometers; the steep canyons of the northern escarpment lie to the south. Instead of excavation inside the monument, the team documented surface detail and excavated a 0.40 m deep trench 1 m x 0.90 m against the southwest outer perimeter of the monument (Figure 5.21). Its purpose was to investigate the terrace surface for a possible paleosol upon which the monument was constructed. Furthermore, formal excavation documented the exposure of an outer perimeter stone before turning it to explore for pedogenic carbonates. This experimental program sought to use the radioactive carbon captured through soil biota to establish a date for monument construction. In this case, there were insufficient secondary carbonates to obtain a radiocarbon determination. Nor were there visible signs of a buried paleosol, perhaps unsurprising given the aridity of the location. An ancient paleosol embedded in the channel section of the terrace did not yield datable organic materials and is significantly older than the terrace surface. Nonetheless, the team cleared the SCAB monument of rubble stone to reveal its original constructed form.

Figure 5.20. D112-001 viewed from the south (photograph by J. McCorriston).

60

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Figure 5.21. D112-001 SCAB after clearing to expose rocks insitu. Note the shallow excavation 1 meter wide in the northwest against a perimeter stone (orthophoto by T. Everhart).

Other SCABs Survey in Wādī Ḥalūf and the environs of Muḍayy documented the locations of many other SCABs. At Muḍayy there is a pair of SCABs (D005-2, D005-11) on the gravel terrace downslope and north of D028001 (the monument in this chapter’s opening narrative). Across the broad wadi and on a low terrace against its northern slopes lies another constellation of SCABs (D027-001, D027-003) and a platform (D027-002) (Figure 5.22). On the streamside terraces of Wādī Ḥalūf lie others (e.g., D061-001), and their distribution in similar locations continues across the northern escarpment and southern Nejd. 61

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 5.22. D027-001, D027-003, and D027-002 at Mudayy (photograph by J. McCorriston).

What were SCABs used for? Are these SCABs tombs? We do not know. Are they Neolithic in date? The sample of platforms and SCABs is too small to render a statistically significant analysis of their locations and strength of correlated attributes. Evidence from D106-001 seems to tie the SCAB orientation (east) and standing stone to known features of Neolithic platform monuments, but there is no independent date for SCABs. Nonetheless, their locations, orientation (where notable), their overall scarcity, and their local settings on low terraces suggest some affinities with the platforms. They probably date to the Middle Neolithic. Even without excavation, some differences are apparent, pointing to important differences in energetics and group efforts. The average calculated volumes of platforms (X̅ =22.7 m3, SD 25.4) and SCABs (X̅ =5.28 m3, SD=4.8) suggests a difference in the energy invested in their construction. Platforms took more labor to build, and some of the larger stones required the collective effort of 7-10 strong adults to shift them, implying labor groups. Because uprights require support, platforms are stable when constructed rapidly in one episode.

62

Building Pastoral Communities in the Neolithic (5000–4500 BCE)

Table 5.2. Volume mean and variance by monument type.

Type

n=

Mean volume of stone m3

SD

Platforms

12

22,7

25,4

SCABs

22

5,28

4,8

Ideology and Institutional Meanings In Alfin’s fictive story, the platform serves as a collective commemoration of a prior gathering, sacrifice, and feast; a new gathering brings both those who belong in group and those who wish to induct themselves. Collaborative labor built a memorial, providing also a medium for storytelling, exchanges, and negotiation of social identities. This interpretation comes from detailed analyses of a Middle Neolithic platforms in Yemen (Martin 2020; McCorriston 2011; McCorriston et al. 2012; McCorriston 2020b, 2020c). Elements of the contemporaneous Dhofar platforms parallel such examples. These suggest that Dhofar’s Neolithic platforms are a material manifestation of cultural practices once common across the inland backslopes and desert oases of Southern Arabia. The faunal bone and marine shell beads in excavated platforms at Muḍayy and Hanūn suggest the remnants of sacrifice. The evidence is admittedly meager, but the finds could have come from decorated sacrificial victims. Walter Dostal (1983) describes a modern Mahra tradition of bull sacrifice; the victim was adorned with flowers and ornaments, including shells. During the era of federated kingdoms, the South Arabian urban communities emphasized sacrifice of cattle over other animals, a notable difference from other Arabian groups (Henninger 1946/47 [1981: 234]). Cows and a single bull – a symbolic herd – were the sacrifice at Neolithic Kheshiya in Ḥaḍramawt (Martin 2020). As to the function of the commemorative platform, recall the cultural inheritances of pastoralists (Chapter 4). These include technologies of messaging-while-absent, powerful manipulations of the environment to construct a niche securing advantages for future generations. Dhofar’s pastoralists built monuments to convey messages. Even as the message is lost, archaeologists can understand that its intent was a collective intent in the case of platforms. Each of these monument technologies – platform and SCAB construction – offered the possibility for incorporation of a group through physical collective (the gathering), inscription into that group through participation in commemoration, and definition of social identity through accessing and manipulating the message. Dhofar’s Neolithic pastoralists were seemingly few and thinly dispersed, if we can infer their numbers from the relative numbers of monuments left behind and the increase in numbers of monuments in subsequent periods. Nonetheless, their monuments and the practices they represent were effective means to sustain the necessary interactions and communities that perpetuated pastoral populations in the waning millennium of a green Arabia.

63

Chapter 6

Tribes in the Bronze Age (3200–1500 BCE)

Remember the burials of ‘Awal and his descendants in Chapter 1? This book opens with one of many possible narratives to introduce the Bronze Age tradition of tomb construction, widespread across Arabia and manifest in Dhofar. About five thousand years ago, pastoral herders began to build a new style of monument. The new monuments were highly visible tombs; these held the bodies of some who had died and could serve as touchstones for the memories of ancestors or kin. This chapter explores the diversity of tomb types in Dhofar’s Early Bronze Age and interprets their attributes as hallmarks of a new social compact based on genealogical narratives anchored in tombs. The AHSD survey team focused on one tomb type, designated throughout fieldwork as a “high circular tomb” (HCT). In an effort to refine an algorithm for auto-detection of small-scale stone monuments (Chapter 3), the survey team focused on the attributes of a monument type both numerous and likely to be detectable in satellite imagery. With determined and methodic repetition, ʿAli, Matt, and Jen scaled the crusty slopes around Muḍayy and other high places, sending clastic showers of sun-sintered stone down the

Figure 6.1. Matthew Senn documents the perimeter of an HCT monument using kinetic high-precision GPS (photograph by J. Everhart and K. Olson).

64

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.2. Image map of Bronze Age tombs documented in survey and excavation (image by K. Olson).

track behind them. At each high tomb they paused for a rapid photograph (“…who left the longer photo scale below in the truck?”) and stretched out the metric tape; meanwhile Matt established a center point GPS reading before clambering around the perimeter (Figure 6.1). Often the site was only a heap of stone, but the team became experts at the subtle clues of an HCT – a ring of uprights defining a central chamber, a silty depression, capstone slabs pulled aside, the crude chinking of a recent rebuild. Other days we maneuvered off-road to the fresh locations sent by our USA-based statisticians, Prem and Jared, using precision GPS and rectified image-maps from our engineering colleagues, Dorota and Jihaye. Their instructions led us over torturous tracts to lonely plateaus where the computer had targeted likely sites. Sometimes these groundtruthings pulled us up to a stone heap, but there were many failures, like dark circles of dark limestone in the flat, desert reg and at least one modern garbage dump reeking of dead camel. Yet the false positives led the team’s statisticians to refinement, re-runs, and a recursive program of field survey. While our algorithm was a proof-of-concept project focused on one tomb type (Schuetter 2010; Schuetter et al. 2013, Harrower et al. 2013), the team documented many other monuments (Figure 6.2). This chapter explores the many different types of Bronze Age tombs, not all of them common. Some of these were re-used, some appear to be single use only; some were disturbed after interment and others were left sealed for thousands of years. From few Neolithic small-scale stone monuments, our survey records show a larger number of Bronze Age tombs. From our excavations, we detect nearly two thousand years of tomb-use in Bronze Age Dhofar as pastoralists renewed their use of stone monuments, with some significant changes in social practice after a gap in time.

65

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Table 6.1. Survey results 2009-2018. Bronze Age tomb types, numbers per type, and chronology in Dhofar.

Type

n=

approx. chronological range

High Circular Tomb (HCT) Wall Tombs Halūf Tombs

135 3 3

(? 4700-)-3500-2000 BCE (? 2500-) Pre 1600 BCE 3100 BCE - ?

Tower Tombs (HCT) and Burials Just because the 3rd millennium BCE was called the “Early Bronze Age” does not make it an era rich in bronze in Dhofar. A few copper scraps and an occasional pin turn up in tombs, along with an outstanding example of a copper-alloy dagger, rare for Dhofar (McCorriston et al. 2014). Rare also are formal stone tools; instead people utilized and modified chert flakes without strict standards of form or a highly skilled process. Dhofar lacks irrefutable evidence of settlements in the 3rd millennium BCE; people were still highly mobile, and their principle economic strategies emphasized coastal fishing and marine resources, with herded domestic animals and wild game as the economic mainstay of mountain and desert dwellers. Copper, marine shells, coral, and carnelian beads in the interior attest to widespread exchanges, but the mechanisms, whether population movements, direct or indirect contacts, remain unclear. Apart from small-scale stone monuments, there are few remains from Bronze Age cultural groups in Dhofar, leaving archaeologists to derive from tombs the practices of social and economic life. What follows is a catalogue of excavated tombs, followed by a discussion of their collective importance for pastoral history.

Figure 6.3. D001-001 after excavation (photograph by T. Steimer-Herbet and K. Olson).

66

Tribes in the Bronze Age (3200–1500 BCE)

D001-001 (N 17.472; E 53.357) Set on a high bedrock plateau at Muḍayy, HCT D001-001 straddles a steep and short detour from the main drainage of Wādī Ghārah to one of two springs at Muḍayy. The desert reg is relatively sterile of artifacts, suggesting this high locale saw repeated traffic only in the Bronze Age. It is about 30 meters to the cliff edge, and another half kilometer to the spring. While there is another tomb (D001-006, see below) about 2/3 km away on the plateau to the ESE (112˚), the landscape otherwise is desolate. Built of dry-walled, undressed stone as a double face with a rubble core, the D001-001 circular tomb was constructed on the natural (reg) surface. The exterior spans a diameter of about 5.0 meters with a circular chamber 1.20 meters across. With intact chamber capstones about 0.50 meter above ground surface, this tomb presented excavators an opportunity to investigate burials and burial practices that might hint at the social lives of ancient Dhofari pastoralists. Excavations summarized elsewhere revealed vertical faces to the exterior walls so that the tomb originally appeared cylindrical in form, faced with carefully laid stretchers of unworked limestone slabs (McCorriston et al. 2014; Williams et al. 2014) (Figure 6.3). Originally constructed on the desert reg surface, the base of the chamber interior measured 1.40 meters across and was corbelled to roof the chamber, leaving a narrower opening covered by two, meter-long, limestone slabs (Figure 6.4). Inside, an articulated skeleton of an older male (Burial A, 40+ years) lay on a stone surface. The placement of bone and state of articulation suggested he had been shifted after primary burial and ended at the eastern wall of the chamber, demarcated by a semi-ring of cobbles. A radiocarbon age (AA86371, 4980 ± 140) shows this burial occurred 6000-5500 years ago. Another stone pavement capped this burial. Next, people buried another adult male (Burial B, 20-30 years old at death) (AA86373, 4315 ± 43) 5000 years ago, who was subsequently shifted to the southern

Figure 6.4. Chamber fill of D001-001 with minor collapse of corbelled roof (photograph by T. Steimer-Herbet and K. Olson).

67

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

edge of the chamber before the ligaments had completely decayed. Near the base of the chamber were loose beads, one of steatite and the other of frit, perhaps once sewn to clothing or some other organic material (Figure 6.5). The last burials were three sets of comingled remains: female burial C (20-25 years), an associated preterm or neonate (D), and set somewhat apart, adult male E (40+ years). Burial E may have already decomposed to dry bone so that his remains were scattered by the partial collapse of the upper chamber interior.After each lower burial Figure 6.5. Two beads from the base of D001episode and after the three comingled burials about 5000 years 001 chamber (photographs by C. Heyne and E. Lagan). ago (AA86375, 4452 ± 42), windblown sediment accumulated in the still hollow of the chamber, filling it with clean sandy silt. Atop this someone placed partial remains of a sheep or goat, and again, after the lapse of time, sediment infilling, and collapse of one capstone, another visitor placed more caprine remains (Everhart et al. 2014). Kimberly Williams studied the distinctive traces of lives lived visible on bone. From the healed broken middle and tips of two toes, a healed trauma to the neck, fusion of the upper spine, bone spurs of old age, tooth losses, and a partially healed jaw abscess, one can guess at Individual A’s injuries and pain reconstructed in Chapter 1 (Williams et al. 2014). D013-001 (N 17.340; E 53.993) Both D013-001 and D013-002 sit along the edge of a bedrock plateau overlooking a Wādī Haylāʾ, a wide eastern tributary to the Wādī Ghadūn system (Figure 6.6). D013-001 is an HCT constructed on a surface already littered with the stone debris from Dhofar’s leptolithic chert knapping tradition, which is Upper Paleolithic in date and long pre-dates tomb construction (Hilbert 2013). The view from this tomb is expansive, and from the route below it is prominent on the ridgeline. The original height of the tomb is unknown, and it was evidently looted after its primary use. The southern portion of the tomb was still standing at 0.90 m with remnants of its corbeled roof. Undressed limestone slabs formed the outer facing of the walls, which appeared to be partially collapsed or crudely heaped. The burial chamber measured 1.6 m and contained a single burial (Figure 6.7). Excavations focused on the chamber only, as we hesitated to reveal the outer wall faces for fear of collapse. Excavations showed the following sequence, determined by Kimberley Williams. One flexed and articulated body was placed facing south on the original ground surface, still strewn with its older knapped lithics. Windblown sediment accumulated over the burial in the still space of the hollow chamber. Subsequently someone opened the tomb and dug a single hole dug through the center of the burial chamber, disturbing most of the postcranial human remains. An articulated left tibia, fibula, and foot were in position in the eastern section and an articulated right maxilla and mandible remained in the western section, along with highly fragmented cranial remains (Figure 6.8). Because the third molar had erupted and worn, the remains were of an adult; with robust ramus and tibia, he was probably a male (Figure 6.9). The chamber walls slumped inward, probably protecting the articulated extremities, and where they collapsed over the skull, they crushed bone. The only item included with burial was a single flat ornament of mother-of-pearl in a perfect “O” near the cranium (Figure 6.10). A radiocarbon age determination from human bone apatite showed the burial to be about 5200 years old (AA86372, 4508 ± 49).

68

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.6. D013-001 overlooking Wādī Haylā (photograph by K. Williams and J. McCorriston).

Figure 6.7. Chamber of D013-001 at the end of excavation. Note the dry-wall technique, here consisting of an inner facing and outwardly faced or buttressed with heaped stone (photograph by J. Williams and E. Lagan).

69

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.8. Articulated left tibia, fibula, and foot of the male buried in D013-001 (image by K. Williams and E. Lagan).

Figure 6.9. Right maxilla with erupted third molar from the adult male buried in D013-001 (image by K. Williams and E. Lagan).

70

Tribes in the Bronze Age (3200–1500 BCE)

There were subsequent inclusions of faunal parts from a caprine-sized animal. These partial remains were placed into the upper levels of sediment, suggesting some time had passed for sediment accumulation after the original burial (Figure 6.11). Nonetheless, the animal offering – if it is so interpreted – occurred before the last removal of capstones and disturbance. D013-002 (N 17.339; E 53.994)

Figure 6.10. Mother-of-pearl ornament from D013-001 (image by C. Heyne, J. McCorriston and E. Lagan).

Another HCT, Site D013-002 also sits atop the bedrock surface strewn with older lithics and lies about 200 m from D013-001 (Figure 6.12). The monument is small, about 1.4 m exterior diameter, with a burial chamber no wider than 0.9 m (Figure 6.13). A small ring of stones was arranged on the plateau surface near the tomb. Where the chamber wall was best preserved, it reached 0.6 m height. Just centimeters from the edge of the plateau, the tomb has lost some of its stone to erosion, yet it contained about 0.3 m depth of internal sediment. The excavated chamber contained no human remains, but a pierced Conus shell bead suggested a deliberate interment of some kind – someone carried shell up from the ocean (Figure 6.14). There were also 11 indeterminate fragments of animal bone that excavators observed to be articulated and badly preserved. The sediment fill appeared disturbed, an artificial mix of wind-blown sandy silt, shattered limestone clasts from the chamber walls, and the older knapped debitage from the terrace surface (Figure 6.15).

Figure 6.11. Section D013-001 (image by K. Williams, T. Steimer-Herbet and K. Olson).

71

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.12. D013-002 Overlooking Wādī Haylā (photograph by K. Williams and J. McCorriston).

Figure 6.13. Chamber of D013-002 after excavation (photograph by K. Williams).

72

Tribes in the Bronze Age (3200–1500 BCE)

While difficult to interpret fully, excavation results suggest that if an interment occurred, it was ancient and decomposed, perhaps because of exposure by subsequent damage to the tomb. Like other HCTs, there were animal remains suggesting a later offering. The datable material for this tomb was a charcoal fragment from the underlying surface, providing a date about 6400 years ago (AA85556, 5607 ± 40), sometime after which the tomb was constructed. D014-001 (N 17.313; E 54.013) Both D014-001 and D014-002 lie along an ancient track that skirts the Wādī Haylāʾ bed, by-passing a meander. The track cuts across the neck of a promontory and through a saddle in the hill (Figure 6.16). Figure 6.14. Pierced Conus shell recovered from D013-002 chamber We presume the track is ancient; its worn stones trace the same direct (image by C. Heyne and E. Lagan). route adopted by the high-tension wires between Ittin and Thumrait, and this route links many ancient monuments and sites. Both tombs are prominent on a low terrace without visibility to and from the main wadi. Ḥalūf tomb (D014-3) lies the east. Neither monument appeared particularly well preserved, with overall sine-shaped profiles covered in rubble and no capstones in place. Lichen growth indicated a long stability of the fallen rocks. Possibly some retaining stone (outer perimeters) was robbed or removed. D014-001 is a 6.4 m diameter circular mound of rounded and angular boulders and slabs, with small clastic cobbles.

Figure 6.15. D013-002 Plan and section (image by K. Williams, T. Steimer-Herbet and K. Olson).

73

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.16. Overview of D014-001 (fore and center) before excavation. Behind and left is D014-002 and D014-003 (photograph by C. Heyne and J. McCorriston).

The excavated chamber measured about 1.20 m E-W x 1 m N-S, filled with filled with a silty sediment with decomposing limestone inclusions and an area of organically enriched, greasy fill typical of the residue left from fleshy decomposition of one or more burials (Figures 6.17 and 6.18). Excavation revealed an original 4.6 m diameter between relict exterior large boulders as lower courses of the outer wall. No intact bone remained, but the decomposition deposit directly overlay the bedrock floor of the tomb. Flakes of bone too badly decomposed even for collection suggest that there had been a burial at one time. There was a single, non-formal, chert flake tool in the rubble fill slumped into the chamber. Two ceramic sherds lay on the terrace surface near the tomb. Excavators concluded that this tomb had been disturbed, most likely quarried for construction of a nearby Ḥalūf tomb. D014-002 (N 17.313; E 54.013) This HCT was built on bedrock, possibly before the construction of adjacent tombs, for its low level and robbed exterior may have furnished stone to D014-003 (Ḥalūf tomb). The remaining structure measures 5.80 m N-S and 4.70 m E-W, with a preserved height of 0.70 m. Capstones had been shunted aside, and a central chamber 1 m diameter was filled with small cobbles and collapsed slabs (Figure 6.19). Below about 0.6 m depth of silty aeolian sediment mixed with clastic cobbles, a 6 cm reddish layer revealed the powdery fragments of long bones (a humerus and a radius) no longer in articulation (Figure 6.20). A radiocarbon age from the bioapatite (AA90833, 5821 ± 44) shows the burial to be about 6700 years ago. Neither D014-002 or its neighbor D014-001 contained evidence of subsequent animal sacrifices. The age of the burial – if it indeed was human – remains an ancient outlier in the HCT tradition. 74

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.17. Chamber of D014-001. The original desert reg terrace surface has been cleared to bedrock at the base of the chamber (photograph by C. Heyne and J. McCorriston).

Figure 6.18. D014-001 Plan and elevation (image by C. Heyne, T. Steimer-Herbet and K. Olson). 75

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.19. D014-002 Plan and elevation (image by C. Heyne, T. Steimer-Herbet and K. Olson).

D033-001 (N 17.428; E54.111) In the stony flats of the Nejd where drainage meanders across broad channels, HCT are highly visible on ridges and across broad vistas. An isolate, D033-001 sits on a high terrace overlooking the wadi channel (Figure 6.21). Although the tomb appeared to be in poor state of preservation, the exterior diameter was still clearly defined by lower courses of an outer facing that had been largely removed or collapsed in its upper part. The structure measured 3.90 m N-S and 3.15 m E-W. A circular chamber measured 1.0 m and 0.80 m, with intact height 0.6 m under the sole remaining capstone slab. The inner walls in the chamber on the eastern and western sides were straight, but on the south side where preservation was best, the walls slanted inward, like corbelled-roof construction apparent in other HCTs (Figure 6.22). 76

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.20. Long bone shaft at the base of the chamber in D014-002 (image by T. Steimer-Herbet and E. Lagan).

Figure 6.21. D033-001 Documented by survey team (photograph by J. Everhart).

77

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.22. D033-001 Plan and elevation (image by T. Steimer-Herbet and K. Olson).

The tomb appeared recently reused, with discarded, charred, butchered camel bone on the surface and throughout an aeolian, silty fill. Perhaps as the original base of the chamber, a limestone paver floor of undressed slabs sealed a thin deposit of sediment, which contained no artifacts or bone. Excavators reached bedrock without finding any sign of an ancient burial. Despite the smell of decomposition, a bone apatite sample returned a radiocarbon determination about 6600 years old (AA90829 5759 ± 46). The age is difficult to interpret. This camel bone surely post-dates the period in which camels were introduced as domesticates 2900 years ago (Beech et al. 2009; Curci et al. 2014). Perhaps the bone has undergone significant diagenesis since its deposition; yet bone apatite research clearly shows this to be a reliable measurement in arid preservation environments (Zazzo and Saliège 2011; 78

Tribes in the Bronze Age (3200–1500 BCE)

Zazzo 2014). Our team’s bone bioapatite radiocarbon analyses, including moister contexts in the Dhofar escarpment and using multiple laboratories are otherwise generally consistent with each other and with expectations. Another possible explanation is excavator error in assigning (older) bone to the obvious remains of modern camel at the surface. D036-002 (N 17.389; E 54.100) Situated high on a bluff with prominent overview of Wādī Dhahabūn’s westernmost tributary, this is one of a pair of HCT (Figure 6.23). On the surface, it appears to have undergone subsequent modification, with twin alignments of stone as walls or tails abutting the tomb. These may have served a prosaic purpose (windbreaks, hunting blinds) by subsequent travelers, or they may have belonged to earlier symbolic syntax no longer decipherable by modern minds. D036-002 has a diameter of 3.1 m, and its preserved height is 0.75 m. The upper part of the tomb was removed; some of the original external wall remains intact, consisting of three layers of large, black, dolomitic limestone, clearly visible at distance. Builders transported this facing stone up to the site, for there is no source on the adjacent plateau. Light-colored and locally sourced limestone served in the interior. Because the chamber was deliberately filled with stone as in the case of D014-002, the original chamber facings were hard to ascertain. Excavators determined a chamber diameter of approximately 1 m. Excavation revealed a sequence of burial, disturbance, revisits, and re-disturbance. At the base of the chamber, overlying bedrock at 1 m depth, were three black limestone clastic cobbles arranged together at the west end; upon these lay the fragmented remains of half a human cranium, facing south. There was no intact primary inhumation, but some human bone remained on the bedrock surface, including shaft bone;

Figure 6.23. D036-002 in overview with D036-001 in the background (photograph by T. Steimer-Herbet).

79

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

excavators re-buried the bone in place after documentation. Overlying this deposit was rubble fill of a poorly sorted mix of silt and clastic limestone pebbles and cobbles. A deposit of animal bone, including identifiable bone from both goat and sheep, occurred along with three Conus shell beads. These were placed in the tomb after burial, collapse, and presumably, a deliberate re-opening, for the large slab capstone was replaced over the deposit of animal bone and beads. Subsequent collapse or deliberate rubble in-filling took place, possibly when large facing stones were re-purposed to build two abutting walls. A radiocarbon age on bioapatite from a cranial fragment (AA95063 3624 ± 38) and another on long bone (AA90831 3646 ± 40) on the chamber floor indicate a human burial about 4000 years ago, one that may have been disturbed by collapse or removals. A radiocarbon age on bioapatite from faunal bone (AA90830 2807 ± 38) suggests offerings placed many years later. The sheep bone, goat bone, and exotic beads of marine origin point to Iron Age sacrifices around 900 BCE, a thousand years after the last human burial, and the careful replacement of the capstone suggests a deliberate purpose to their inclusion. As in D001-001, D013-001, perhaps D033-001 and several similar tombs in Yemen (McCorriston et al. 2011, 2020a), animal offerings accompanied subsequent visits to this tomb. D006-001 (N 17.276; E 54.490) At the drier, eastern tributaries of the Nejd, HCT are fewer and dispersed. The site at D006-001 north of Jibjat stands right at the cliff edge of the narrow Jabal Qara plateau, overlooking one of the main tributaries to Wādī Dhahabūn. The remains are highly visible from the plateau itself and from the wadi below, and the overlook is but a half kilometer from the ancient (and modern) aqaba leading into the Nejd (Figure 6.24).

Figure 6.24. D006-001 before excavation (photograph by J. McCorriston).

80

Tribes in the Bronze Age (3200–1500 BCE)

At survey, D006-001 measured 4.15 m E-W and 5.10 m N-S, with a visible central 1.6 x 0.9 m diameter chamber, and large, flat capstone slabs slumped inward. Part of the exterior wall facings remained, but others had been robbed, leading to considerable erosion of the tomb’s core. The remaining cairn stands 1.1 m high. Given the poor condition, excavators focused on chamber fill and only noted the exterior. Excavation revealed a fill of unsorted sediment and clastic limestone cobbles and pebbles. Despite sieving with a 2 mm kitchen sieve, no artifacts, bone, or charcoal appeared. It appears that either that someone emptied the HCT in antiquity or that its chamber collapsed inward before aeolian sediment accumulated. A possible motive for removal of large facing stones could be the construction of later boat graves (D006-002, -003, -004, -005) a few hundred meters away. There are no associated samples to provide an age for construction or use; we assume this to be Bronze Age in date. D104-001 (N 17.443; E 54.470) HCT D104-001 sits in lonely isolation on a limestone terrace spur overlooking a wide meander of an eastern Wādī Dhahabūn tributary (Figure 6.25). The tomb itself measures 3 m across at its base, and tapers to an apex as a truncated cone 0.86 m in height. The terrace offers a rich source of tabular chert, source for a primary knapping site with a few core fragments, primary flakes, and many secondary flakes. There were no formal tools abandoned on the terrace, although one knapper thoughtfully tucked a unidirectional core into the surface rubble of the tomb.

Figure 6.25. D104-001 viewed from southwest (photograph by J. McCorriston and K. Olson).

81

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.26. D104-001 base of chamber with mother-of-pearl ornament in situ (photograph by J. McCorriston).

Excavators removed several large, intact capstones of a central chamber 1.2 m in diameter. To ensure excavator safety, they also removed 28 upper facing stones that formed a corbelled roof to the chamber. No cleaning or excavation of the outer walls ensued. The chamber fill was a homogeneous soft silt and sand, the upper 10 cm of which contained chert flakes. This fill accumulated over a sterile terrace surface, on which there were no chert flakes or signs of human knapping. Although no bone appeared, a mother-of-pearl pendant pierced badge or pendant (DA#46813) lay on the original surface, with two frit or steatite beads, a shell bead, and a biconically drilled bead of carnelian (DA#46810) (Figure 6.26). The sole and minute fragment of charcoal surfaced relatively high in the chamber fill, and it could be a windblown introduction along with sand and silt. Likewise, the upper deposits of chert flakes might have dropped through capstone chinks when a knapping visitor used the tomb as a likely perch. Lacking bone or charcoal, the best material for dating this tomb is the assemblage of ornaments. Although simple, the pendant recalls one from D013-001; the smaller beads match those from D001-001; and the carnelian bead is consistent with 3rd millennium BCE materials in wide circulation (Kenoyer 1997; Ludvik et al. 2015). Communities of Practice in Space Even to the casual eye, there appears to be patterning in the spatial distribution of HCT. One of their salient features is prominent visibility at the edge of cliffs and promontories. HCT dominate the skylines, and when one climbs to the barren plateaus on which they sit, one discovers that tombs were meant to be seen; there are almost no examples invisible from below. Their prominence suggests they signal to those who moved 82

Tribes in the Bronze Age (3200–1500 BCE)

along the wadis, where passage across low terraces and smooth channels required least effort for herders and animals and offered greatest access to shade, grass and tender browse, and water. Sometimes these tombs cluster, as at Muḍayy where one finds them overlooking the springs, and Hagif where 57 HCT overlook the broad backslope of the Dhofar plateau. Elsewhere, HCT may appear isolated, as in the eastern tributaries of Wādī Dhahabūn. Overall, these tombs are features of the Nejd, absent from the Dhofar plateau and escarpment, with a few clusters on the coastal plain. These tombs mark the passage of mobile people whose construction energies lay in small monuments to the dead rather than in permanent housing for the living. Excavations revealed differences in the preservation of human remains and in burial practices. Some inhumations left skeletal elements articulated, or partially so when new burials were added. Other burials, even in the same tomb, included secondary burials. Some tombs contained only a few bones, while others yielded the full skeleton. Proximity to the moist margins of the khareef may have skewed preservation. Deep in the Nejd, D104-001 had no trace of human remains – could this have been a cenotaph? Our excavations of HCT also revealed the inclusion of ornaments like beads, pendants, perhaps clothing appliqué, but nothing beyond what was worn – no ceramic jars, stone tools, copper weapons, no ochre, bitumen, plant silica, or plaster that might suggest baskets of food or other burial practices that interred material goods to accompany the dead, displayed consumption at a funeral, or removed wealth from circulation. Any marker of social status beyond dress lay in the outward tomb itself and in the narratives carried forward and constructed by the living. Wall Tombs and their Western Counterparts In the western Nejd of Dhofar one finds a few examples of Wall Tombs, a distinct form distributed across the arid plateaus of Ḥaḍramawt, Shabwa, and Marʿib provinces in Yemen (Braemer et al. 2001; McCorriston, Harrower et al. 2020), along the inner ‘Asir (Zarins 1981) and as far north as the Sinai (Eddy & Wendorf 1998). Beyond Dhofar, these elongate tombs take the form of a low, straight wall constructed of undressed stone facings and a rubble core. At the midpoint is a chamber too small for a primary adult human burial. Some excavated examples contain (adult) human bone and ornaments in the chamber, but others contain no remains. In Ḥaḍramawt, Yemen, ancient inhabitants built the Wall Tombs near HCTs (McCorriston, Harrower et al. 2020: 451), probably at the end of the 3rd millennium and early 2nd millennium BCE (Vogt 1997; Eddy and Wendorf 1998; Steimer-Herbet et al. 2006; McCorriston and Dye 2020: 500-501). D022-005 (N 17.468; E 53.387) The single Wall Tomb excavated in Dhofar lies on the stony plateau beside an HCT D022-004 and C-shaped shelter at the edge of an eastern cliff overlooking the Wādī Ghārah southeast of Muḍayy. Oriented along a 4˚ axis, the tomb extends 6.9 m with a central chamber 1 m x 0.56 m (Figures 6.27 and 6.28). Before excavation the team cleared the exterior and surface of D022-005 and revealed a foundation of large, flat-lying, undressed blocks, upon which multiple flat courses shaped an external facade. Parallel interior divisions not bonded to the exterior shaped the northern and southern limits of the chamber, and the two ends of the tomb were filled with clastic cobble rubble. Clean, windblow sandy silt probably explains the accumulation in the interstices of this fill. In the upper 10 cm of the fill of the southern arm, a chipped chert knife knapped on a large flake had been discarded at the end of construction. The chamber itself remained sealed by two layers of flat capstones, also embedded in sterile silty sand. Underneath, the chamber contained only silty sand, easily accumulated through the chinks of a stone-deep façade. 83

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.27. View of sealed chamber (center) D022-005 after cleaning the monument, probing construction, and before excavation of the chamber (photograph by T. Steimer-Herbet and K. Olson).

Figure 6.28. D022-005 Plan and elevation (image by T. Steimer-Herbet and K. Olson).

84

Tribes in the Bronze Age (3200–1500 BCE)

Excavation of the chamber revealed animal disturbance in the upper 10 cm, possibly a slim rodent or snake. From this deposit, excavators recovered a two small animal bone fragments. The only other contents were two round balls of silt coated in white powder. When mis-identified as tiny crania, they were left intact on the chamber floor and reburied awaiting an expert’s treatment. Subsequent excavation and assessment suggest these were a thin silica veneer left by the decayed organics of dung beetle egg cases. Otherwise, the chamber contained no remains. There is therefore no material to date the construction and use of this Wall Tomb. Lacking other appropriate samples, the team submitted the bone from upper fill for a terminus ante quem around 1600 BCE (AA90836 3307 ± 41). Although the purpose of this intact, yet empty-chambered monument remains obscure, construction before 1600 BCE is consistent with expectations from other Wall Tombs. Ḥalūf tombs Beside the ancient trail passing D014-001 and D014-002 is an 11 m long monument almost 2 m in height with a sharp contrast between the dark, dolomitic limestone foundation and upper courses of bright stone (Figure 6.29). Also in a heap of rubble, another monument like it (D014-007) lies a half kilometer to the northwest. And another (D001-006) sits atop the plateau at Muḍayy, overlooking the Wādī Ghārah and D001-001. While these monuments might by some be classified as Wall Tombs, they are distinctive in their use of two colors of stone, in the alignments and arrangements of black stone on the surface beside them, and in details revealed only through excavation. D014-003 (N 17.314; E 54.014) Eleven meters in length and oriented north-south, D014-003 faces west toward a pair of HCT. A trail across the promontory saddle cuts between these monuments. In regional context, this trail lies alongside a north-south route across the low terraces of the easternmost tributary leading into Wādī Ghadūn and the inland Nejd. Less than a kilometer away is a similar monument, also buried under rubble (Figure 6.30). The team decided to excavate one and to leave a second example untouched for future archaeologists and technologies. Excavators documented the collapsed state, then removed all rubble, including a heap atop the south end. Thereafter, the exterior façade was cleared to the terrace ground surface, and the interior rubble fill of clastic, mostly unsorted light-colored limestone and silty sand also removed, revealing an oval chamber at the center.

Figure 6.29. Halūf tomb D014-003 after excavation (photograph by J. McCorriston and K. Olson).

85

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.30. D014-007 Unexcavated Halūf monument (photograph by J. McCorriston).

The original builders had constructed an oval, 2 x 1 m wide chamber on the terrace surface, shaping the chamber with large, undressed limestone blocks at least two courses high (Figure 6.31). Due to subsequent disturbance, the upper construction of this chamber has been lost, but it is clear from the sterile silty-sand fill that it was either open or trapping aeolian sediment for considerable time. Several clasts of dolomitic limestone are later intrusions, likely from the removal of an ornamental entrance in the exterior façade. After the chamber, the original builders laid out the 11 m long façade foundation of three lower courses of black dolomitic limestone, locally available on the adjoining terraces. Upper courses, thirteen of them preserved at the south end, brought the height of the outer facing to about 1 m. The front faced west, overlooking trail, several HCT, and the downslope of the terrace. This façade notably had limestone so well fitted on the exterior that at first glance they appear dressed (Figure 6.32). Yet at the rear, the limestone slabs suggest considerably less care in construction, and the interior face is jagged. In the middle of the western front, builders marked an ornamental doorway 1.6 m wide by continuing the black dolomitic limestone façade to the highest preserved courses. To consolidate the façade, builders then dumped an inner rubble fill throughout the interior, save the chamber. A few charcoal fragments came from the lowest fill at the southern and northern ends. If a burial took place, it surely occurred before the final levels of fill, for a wellpreserved ornamental copper dagger with three rivets had been tucked into the lower inner chinking of the south façade next to the door, where it passed undetected (Figures 6.33 and 6.34).

86

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.31. Excavated chamber in Halūf tomb D014-003 (photograph by J. McCorriston and K. Olson).

Figure 6.32. Detail of undressed stone work, Halūf tomb D014-003 (photograph by J. McCorriston and K. Olson).

87

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.33. Copper dagger recovered from inner face of Halūf tomb D014-003 (and shown in situ) (image by C. Heyne, J. McCorriston and E. Lagan).

Figure 6.34. D014-003 Plan with hearths aligned to the west (image by J. McCorriston, T. Steimer-Herbet and K. Olson).

88

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.35. Heat-treated agate bead from the upper fill of D014-003 (image by C. Heyne, J. McCorriston and E. Lagan).

From the start of excavations archaeologists noted a heap of rubble at the south end. Further investigation suggests it was an accumulation of rubble cast aside as subsequent visitors broke open the sealed monument. Likely once a tomb, this monument was manifestly robbed, or at least re-opened. Apart from the dagger and a few chert flakes in the fill, there were no contents, and the black doorway had clearly been punctured. One explanation is the removal of an intact, wrapped burial bundle with valued grave goods. If robbers removed such a bundle and stripped it outside chamber, no bone nor even the telltale tiny beads from ancient burials would remain. Near the top of the upper rubble, excavators did recover one bead (Figure 6.35). Jonathan Mark Kenoyer (personal communication 2021) notes that “this bead is definitely agate... [possibly] banded agate dyed to make brown or black lines... [a technique] practiced as early as the 3rd millennium BCE but continue[d] into the Iron Age and later.” He notes no other known examples from Oman, and like the dagger, this may be a rare, imported crafted item in Dhofar. The copper dagger is a rare example of the disposal of wealth and prestigious items of display in a Dhofar monument. With rolled edges, this dagger was never sharp, nor intended to be. It is consistent with daggers from 3rd millennium BCE contexts in northern Oman and beyond (Giardino 2017: 52-53). Joseph Lehner used metallurgical analysis to assess its composition – an arsenical copper rich with significant traces of nickel. This approach suggests that the dagger has much closer ties to the copper sources in northern Oman than other known regions, suggesting also that it was acquired from there. Furthermore, its hiding spot was missed by robbers, and it hints of other wealth long gone. And finally, the copper dagger by itself is a break from a Dhofar burial pattern in which bead ornaments were the only surviving items. There is some but not conclusive evidence to suggest this Ḥalūf tomb is 3rd millennium BCE in date. Although the association between charcoal and monument remains unclear, radiocarbon assays on charcoal from the basal fill of the north end (AA85558 4293± 42) ca. 3000 BCE and from the same context in the south fill (AA8555 74390 ± 46) ca 3100 BCE establish a terminus post quem for tomb construction. These charcoals may have come from existing hearths on the terrace. One notable feature of the western front is three north-south alignments of what appear to be the cobble rings of small hearths. They are preserved as many as seven in the middle alignment and extend downslope parallel to the orientation of the tomb itself (Figure 6.36). 89

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.36. View of D014-003 (background) from excavated D014-002 (foreground). Note the alignments of cobble rings in front of Halūf tomb D014-003 (photograph by J. McCorriston).

The spatial arrangement indicates a deliberate relationship, either because the tomb was set over multiple lines of hearths (explaining the charcoal in lower fill) or more likely, because the cobble rings represent subsequent activities responding to the presence of the monument. If the latter, these are not haphazard campsites; they align in an evenly spaced grid and relate to the monument and to each other. Furthermore, these hearths apparently reuse dolomitic limestone clasts and limestone slabs from the rubble of the tomb itself and are clear of tomb collapse. D001-006 (N 17.470; E 53.363) This monument is oval in plan, 8.4 x 4.1 m, oriented east-west on the high plateau between the Wādī Ghārah and one of the springs at Muḍayy. To the WNW (292˚) lies HCT D001-001. The desert reg around D001006 has a pattern of black dolomitic limestone cobble heaps like the pattern observed at the Ḥalūf tomb D014-003 (above). Outside the southern midpoint, there is an alignment of these cobble heaps stretching 30 m directly South to the cliff edge; from the northwestern end, another alignment extends 15 m toward the northeast (Figure 6.37). Also on the surface outside the southern façade lies a grid of dolomitic limestone clusters of cobbles, sourced from at least 400 m away at the cliff. Each cluster is about 50 cm diameter, possibly the remains of deliberately spaced hearths. Like the arrangement in front of D014-003, the purpose and chronological association with the monument is undetermined; the alignments with the tomb suggest a subsequent installation for the grid and alignments of stone clusters. In Yemen, such alignments are called “tails” and may reference subsequent visitors or burials (Braemer et al. 2001, Steimer-Herbet 2004).

90

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.37. D001-006 on horizon viewed from the south; black cobble heaps align between tomb and cliff edge (photograph by T. Steimer-Herbet).

The monument has eroded in its upper part to a preserved height of 1 m at the center and only 30 cm at each end. Rubble and collapse have preserved the lower façades, constructed of small (ca. 10 x 10 x 5 cm), black, dolomitic limestone cobbles laid as courses. Upper courses were of larger, undressed limestone slabs bright in color. Although the excavation team left the fill unexcavated, the façade appears to be double thickness, and the presence of black cobbles among the eroded rubble atop the monument suggests that an upper band of black capped the façade (Figure 6.38). Excavation revealed a chamber off-center, 4.3 m from the western façade, 3.3 m from the eastern façade and centered between the northern and southern facades (Figure 6.39). At construction, the interior of the chamber was a visual display of different courses, preserved to a height of 0.8 m and clearly carefully designed. Two courses of black stone form a base on the terrace surface, surmounted by a course of white limestone, then 3 courses of black stone, another course of white, and an uppermost course of black and white limestone intermixed. While the chamber itself measured about 1.3 m diameter with a circular plan, its roof was corbelled at about a 40-degree angle to an opening 0.72 m diameter (Figure 6.40). If the original purpose was a burial, this left no trace, and the evidence suggests that a tomb was opened multiple times. Large limestone slabs lay directly on the northeast part of the chamber floor; these suggest a collapse or reopening shortly after first closure. Atop these slabs, unsorted sandy silt accumulated, presumably through aeolian process, and into this sediment dispersed several chert flakes, presumably dropped by visitors into an open chamber. There were also rodent burrows, suggesting an open chamber. From the mid-depth sediment fill, a single indeterminate animal rib fragment represents the only bone in the tomb.

91

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.38. Halūf tomb D001-006 after clearing one half of the exterior, viewed from the south (photograph by T. Steimer-Herbet and J. McCorriston).

Figure 6.39. Halūf tomb D001-006 Plan and elevation (image by T. Steimer-Herbet and K. Olson).

92

Tribes in the Bronze Age (3200–1500 BCE)

Figure 6.40. View of the chamber of Halūf tomb D001-006 after excavation and exterior cleaning (photograph by T. Steimer-Herbet and K. Olson).

A large limestone slab 1 x 0.8 m atop the eastern surface of the monument was probably once a chamber capstone. After its removal, silt and few chert flakes accumulated, rodents burrowed in, and clastic limestone cobbles slid down to rest on the sandy silt of the chamber. The only material yielding an absolute age (AA90837, 4382 ± 41) ca. 3100 BCE was the animal bone, probably dragged in by a rodent burrowing in the opened chamber, or perhaps a sacrificial cut of meat offered by a human visitor. Given that unburied bone survives only months, a radiocarbon age on this sample dates the terminus ante quem for the accumulation of sediment in the chamber, probably a decade to centuries after construction. The age is consistent with the Ḥalūf tomb D014-003. Although there are differences in chamber construction and exterior design, the Ḥalūf monuments have some important similarities. Both use black and white limestone in contrasting bands, as found also in some unexcavated monuments in northeastern Ḥaḍramawt, Yemen (Wādī Sanā, Wādī Washʿah) (Figure 6.41). Both Dhofar Ḥalūf monuments appear to have been opened – early after interment before sediment could accumulate – and all contents removed (if any were ever there). Perhaps there was consumption of wealth by burying it, conspicuously signaled by the care and design of the exteriors.

93

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 6.41. Exterior detail, Halūf tomb D001-006 (photograph by T. Steimer-Herbet).

Visitations, Offerings, and Tribes Even as there are evident differences in Bronze Age monuments that suggest a divergence from Neolithic practices, monuments continued to serve as touchstones for social life for mobile people who tended domesticated herd animals. The bones of caprines and cattle appear in the tombs. These offerings of animal parts were placed atop windblown sediments; there was a significant time lapse between burial and offerings. Tombs were opened, sometimes bodies added, sometimes removed, and perhaps the patterned stone heaps beside Ḥalūf tombs also attest to visits. The living connected to the dead through these practices, whether through genealogical descent, asserted kinship, or other social convention. There could also have been activities and offerings that left no trace. Like Neolithic platforms, Bronze Age tombs were the only durable structures in which pastoralists invested labor, signifying an importance in the treatment and commemoration of dead ancestors. Not all treatment was equal; few Ḥalūf monuments suggest a few people for whom conspicuous consumption in death (labor, burial goods?) set them apart. The evidence is far too slim to suggest social differentiation based on wealth surplus and power; nonetheless, Neolithic platforms contain no burials nor evidence of social differentiation.

94

Tribes in the Bronze Age (3200–1500 BCE)

Using genealogy and kinship with ancestors as social touchstones suggests an emergence in prehistory – 5000 years ago – of tribal social constitution (Cleuziou 2002; McCorriston 2013a). With the remarkable record of small-scale stone monuments in Southern Arabia – Dhofar included – archaeological research documents the social lives of mobile pastoralists in a region sparse in settlement and fixed population centers. In this context, and by inference, in the wider context of arid Arabia, one must rely upon archaeological approaches to study how tribal societies originated, how they manipulated technologies like messaging and archiving information through monuments, and what this technology means as the interface between humans and environments, both in oases (Khalidi et al. 2018) and in the arid niche around them (McCorriston 2013a, McCorriston et al. 2018). Ultimately, tribes used genealogy to access and assert claim to resources through urf (tribal law) (McCorriston 2020a: 16). There is no compelling reason to assume that all pastoralists in Dhofar are or ever were territorial tribes; indeed, recent anthropological research finds that many societies and economic producers worldwide – pastoralists included – practice sustainable management of common pool resources without invoking the boundary conditions of territorial regulation (Moritz et al. 2018). At the same time, tribal affinity in Arabia has traditionally and within historical memory conferred specific territorial and resource use rights deniable to members of other groups, whether these rights have been within agreed boundaries (land territory, e.g., hima) or access to resources (e.g., wells, springs, pasture, cultivable land, archaeological loot) (Shoup 1978; Lancaster and Lancaster 1999). Even as it is known that Arabian tribes have asserted exclusive claims in the recent past, one cannot assume this social arrangement (tribal societies) and its mediation of resource access is a timeless and static condition of Arabian prehistory. Quite the opposite is likely, namely that tribes were not always extant nor were tribes an exclusive mode of social constitution, for pastoralists or for other producers. Changes in the distinctive style of monuments and the practices around them emphasize a dynamic social history among persistent pastoralists.

95

Chapter 7

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Were it not for Muhammad, our survey team might have missed the ancient hamlet at Shakeel. To rectify projection errors in our satellite imagery, the team drove to pre-selected points, like the sharp fork in the road south of Jibjat. Curious, we continued to the village, where Muhammad waved down our strange car of foreigners winding through his intimate community. Despite our weak Arabic and ignorance (or wisely perhaps because of them), Muhammad graciously guided us to nearby, abandoned habitations of old. We recognized at Shakeel the unmistakable ruins of an abandoned hamlet, with a handful of free-standing stone dwellings bordering a shallow wash. Though persistent pastoralists left tombs and platforms across the Nejd near to the backslope of the Dhofar escarpment, the mountains and plateau are all but devoid of monuments. Instead, archaeologists find traces of occupation in caves and rockshelters, including painted images and text, chipped stone tools, and the charred dung mats from stalled animals. Pastoralists thrived in these richly vegetated zones, but unlike the desert landscapes, few prominent traces endure. In the Late Iron Age, settlements appeared in the upper escarpment. Few have been excavated, and this chapter describes several sites from which excavations and dated materials are available. The evidence points to a singular period in which herders lingered in sturdy shelters and corrals at the lip of a grassy plateau.

Figure 7.1. Image map of Shakeel and Halqoot (image by K. Olson).

96

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Figure 7.2. Occupation site D114 viewed from the east (photograph by J. McCorriston).

Settlements at Shakeel and Halqoot Over two field campaign in 2012 and 2017, the team excavated trenches at a pair of settlements at Shakeel (D069) and Halqoot (D114), south of Jibjat in the eastern Jabal Qara (Figure 7.1). Halqoot lies just north of the modern village of that name. The site straddles a gully with seasonal flow at the headwaters of the Wādī Darbāt system, just below the southern edge of the Dhofar plateau. Although the site is mostly clear of trees today, a relict Acacia-Commiphora woodland shades the rockiest slopes upstream. There is no permanent water for winter occupants, but the site lies in a small bowl, somewhat sheltered from the harshest north wind (Figure 7.2). A few kilometers’ walk across the grassy plateau, pastoralists could access seasonal spring flow after the monsoon. Few sites in Dhofar manifest the density of structures one finds at Halqoot (Figure 7.3). Large, lichengrey limestone slabs protrude from the surface, arcing in alignments that link walls and retain fill. In addition to over a hundred free-standing houses, stone cairns, and boat-shaped graves, the site has eight corrals with adjacent cell rooms opening either into the corral or outwards, sharing a wall. Prior to detailed mapping and excavations, such sites have been loosely attributed to the Bronze Age, ostensibly when aridification of inland deserts forced mobile pastoralists to settle in the mountains (Zarins 2001). Yet excavations and radiocarbon ages from stratified and carefully documented contexts suggest a different chronology. At Shakeel, a few kilometers to the southwest, one finds a limited array of a few free-standing houses and a corral straddling a shallow wash near the headwaters of another gully feeding Wādī Darbāt’s upper tributaries (Figure 7.4). Here remain scant relicts of the Vachellia-Commiphora cover, with Ficus sycamorus and, downslope where the drainage widens and steepens, a thick cloud-forest of Terminalia dhofarica. A detailed site survey detected ten structures on the surface, and test-pits across the site revealed an exterior hearth and the sediments of a corral. Another small hamlet (D111) with about a dozen structures existed about a kilometer to the west; it has been destroyed by bulldozer. Like the excavations at Halqoot, those at Shakeel revealed an unexpected chronology of Iron Age settlement and abandonments over about 1000 years. 97

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 7.3. D114 architecture and graves. Excavated sites indicated in red (image by A. Buffington, W. AbuAzizeh and T. Everhart). Figure 7.4. D069 Map of architectural sites (image by M. Senn and B. Baaske).

98

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Excavations at Halqoot (N 17.224; E 54.512) At Halqoot, the diversity of structures and their re-use histories attest to a sustained importance for people who apparently depended heavily on herded animals and who remained mobile, even as they occupied sturdy stone houses for at least part of the year. Excavations in cells attached to corrals and others freestanding slightly apart revealed several different construction styles. All structures were curvilinear, and most were constructed on a thin soil over bedrock or upon bedrock itself. Some walls were more carefully faced and chinked than others. For example, the remains of D114-016 consists of a 5 x 2 m oval of large boulders set on bedrock, perhaps as a base for upper courses that have slumped, and perhaps for a thorny brush corral (Figure 7.5). In contrast, the inner façade of D114-004 is of undressed upright slabs, carefully chinked with fitting cobbles. The builders similarly faced an outer wall, packing the meter-wide interval with cobbles and pebbles. Huge boulders buttressed the base of the outer façade, and a stone corbelled roof reached at least two meters high. Limestone pavers surfaced the inner floor, a surface that spanned a single, circular room 5 m in diameter (Figure 7.6). As herders abandoned and re-occupied the site, such paver floors were renewed. For example, in D114-004-A intervals of use and abandonment led to four installations of successive paver floors (Figure 7.7). As both the exterior debris accumulated outside and interior paver surfaces accumulated, occupants constructed three stone steps to ease passage through a narrow doorway. At the same time, the doorway became shorter, limited under a pair of heavy stone lintels (Figure 7.8). Ultimately the structure served not as a dwelling but as a convenient discard location, with ash dumps and animal bone in the upper fill. As with all abandoned chambers, the quiet space trapped wind-born sediment, which filled the interior. The example of D114-004-A yielded a series of radiocarbon ages on charcoal and bone that span about 300 years of occupation on the paver floors.

Figure 7.5. D114-016 Boulders form a corral perimeter, possibly once topped by brush (photograph by W. AbuAzizeh and K. Olson).

99

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 7.6. D114-004 A. Interior of a room with paver floor and dry-stone facing to the interior. Steps lead from a courtyard accumulation to the interior living surface (photograph by W. AbuAzizeh and K. Olson).

100

Figure 7.7. D114-004A South section (image by A. Buffington, W. AbuAzizeh, B. Baaske and K. Olson).

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

101

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 7.8. D114-001 corral and D114-004 attached dwelling in overview; lintels-supported entrance intact and stepped-down entrance highlighted (image by W. AbuAzizeh, J. McCorriston and B. Baaske).

Another example from Halqoot is D114-085, a house built into a hillslope with a stone terrace in front. Constructed of the same double-faced, rubble core meter-thick stone walls, the 3.6 m diameter room also had four successive layers of limestone pavers, accumulated over occupation debris. This dwelling yielded radiocarbon ages that spanned 100-300 years, albeit earlier and without overlapping the occupation at D114004. The D114 site with its many structures was clearly re-used, not only repeatedly and probably seasonally in a generation but also across many generations. Other structures in the site show a similar pattern. Some of the cells appended to corrals have only bedrock floors, suggesting that they may have served as stalls and chambers to segregate animals, not as housing. The corrals accumulated thick beds of ash, most likely from the fired dung of confined cattle. Cattle digest seeds and husks so that the ash from their dung contains few, if any, identifiable plant materials. In D114-004-B, the interior of a corral, excavators revealed ashy accumulations over a meter deep, contemporaneous with the successive uses of the attached exterior house. Nevertheless, despite repeated re-occupation, one of the most startling aspects of Halqoot is the lack of portable material remains or their broken, discarded remnants. Charcoal and bone fragments dominate the traces of occupation. Perhaps because the site has no local source of good chert, few flakes and no chipped stone tools occur, either on the surface or in occupational accumulations. There are no ceramics, a few 102

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Figure 7.9. Beads and ornaments from D114: a) mother of pearl, b) green serpentine, c) coral, d-e) dentalium shell (image by J. McCorriston and E. Lagan).

pieces of ground stone, and a handful of shell, coral, and stone beads or ornaments (Figure 7.9). There are no tools or charred by-products of crop processing or domesticated plants, none from spinning or weaving, nothing to suggest craft production in metals, bone, or leather. If a rich inventory of material culture ever existed, it was in organics long decayed – skin bags, frond baskets, dyestuffs and pigments, mats, bark bags, and wooden bowls come to mind. Multiple examples of faunal bone and charcoal from stratigraphically secure contexts yielded radiocarbon ages (McCorriston, Buffington et al. 2020: 161, tab. 1) (Figure 7.10). Although the stratigraphy is not continuous across the site, the deposition within each structure followed a similar pattern: Phase IV is bedrock or the original ground surface; Phase III A-D represented multiple paver surfaces and the occupational debris between them or, in the case of cells and corrals without pavement, the occupational accumulation and immediate abandonment of surfaces. Phase II represents abandonment accumulations, usually in a matrix of wind-accumulated sandy silt and limestone clasts tumbling from the walls. Phase I is the modern surface and upper topsoil horizon developed on the abandoned site. Pre-dating the installation of a paved surface in a circular dwelling, bone apatite from a caprine jaw yielded an age 700-500 BCE (UGAMS29215a) on Phase IV surfaces, likely pre-dating the construction of the house. Of the Phase III occupation, D114-085 represents the oldest series of house interior occupations as paver surfaces, 300-100 BCE (UGAMS29220, UGAMS42261, UGAMS42262, UGAMS42263). Phase III occupation dated elsewhere in the site is probably several hundred years later, 100-400 CE in D114-004 (UGAMS29219, UGAMS29221 UGAMS29216) and D114-028 (UGAMS29217). Phase II abandonment occurred after 400 CE (UGAMS29218) in D114-004, assuming the radiocarbon age on charcoal came from old wood. Elsewhere on site, Phase II abandonment occurred about the same time (UGAMS42264). From the radiocarbon ages obtained, Halqoot structures were intermittently re-occupied in at least two episodes, 300-100 BCE and 100-400 CE. With over a hundred structures untested, this sequence and gap only captures part of the site’s history, which could have been continuous. 103

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 7.10. Radiocarbon calibrations from D069 Shakeel and D114 Halqoot (image by J. McCorriston).

104

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Excavations at Shakeel (N 17.204; E 54.515) Shakeel is the smaller settlement, with fewer structures and a sole corral. Preservation on the surface varied; for one structure (D069-010), the only surface remains are a ring of cobbles and rubble. One can assume that a structure stood here, and people robbed limestone slabs and boulders from the facings to build other, later structures nearby. At the other extreme of preservation, D069-002 retained walls to meter’s height; surface inspection showed the north-facing doorway, and excavations revealed a floor of pavers with a hearth area and the post-hole for a center-post roof (Figures 7.11, 7.12 and 7.13). Against the southwest back wall of D069-002 was a hearth. Excavations revealed shorter occupation sequences that those described for several houses at Halqoot. At Shakeel, pastoralists built the same style of double-faced, rubble-core circular walls, about a meter thick. Likewise, their foundations used large limestone slabs, requiring the cooperative effort of 7-10 strong adults to shift them. People chinked these uprights with boulders and laid courses above to make wall facings inner and outer. Some structures were built on bedrock, others on the thin soils of an existing ground surface. In two structures (D069-001, D069-002) with limestone paver flooring, people used the pavers to re-surface over an accumulation of occupational debris that included ash and bone but no artifacts. In D069-001, excavations revealed a low, 2 m long stone platform built against the wall to the left of the east-southeast facing entry (Figure 7.14). There was also a hearth installed at the rear (in the southwest). When sectioned, the platform was all stone, yielding no artifacts, bone, or other materials. In D069-004, excavators did recover a rough, small limestone handstone, like one recovered at D114 (Figure 7.15).

Figure 7.11. D069-002 Trench A northeast section (image by J. McCorriston and B. Baaske)

105

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 7.12. D069-002 House plan after excavation (image by J. McCorriston, B. Baaske and K. Olson).

106

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Figure 7.13. D069-002 Overview after excavation. Note construction style matches D114 houses (photograph by L. Proctor and K. Olson).

Figure 7.14. D069-001 Quad A Platform or partial re-pavement with house entrance at upper left (photograph by J. McCorriston and K. Olson).

107

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 7.15. Handstone abandoned on the floor of D069-004 (image by M. Senn and E. Lagan).

Like the structures at D114, the houses at Shakeel show a similar pattern of use and abandonment. Phase IV is the original bedrock or ground surface; Phase III includes occupational accumulations and pavers; Phase II is an abandonment mix of sandy silt aeolian sediment and limestone clasts eroding from the walls. Phase I is the upper topsoil development on the abandonment fill. Radiocarbon ages obtained from Phase III samples at Shakeel suggest an occupation span ca. 100 BCE100 CE across multiple buildings and an external hearth. An older Phase III charcoal sample (UGAMS11835) may simply come from old wood, an explanation highlighted by its stratigraphic position posterior to younger charcoal (UGAMS11836). A 200-year occupation span at Shakeel accords well with stratigraphic evidence for fewer re-surfacings inside houses and is shorter than at Halqoot. The Shakeel period 100 BCE100 CE also fills the gap in radiocarbon ages from D114. Further radiocarbon assays and analysis would be needed to refine site chronologies; at present, it appears that these types of settlements span the Late Iron Age (Yule 2014: 64, 77). Not Bronze Age Settlement These sites represent but two of many sites in the Dhofar escarpment with curvilinear structures that from the surface appear to be corrals and adjoining pens and houses. At Ṭāqah, a coastal settlement with similar architecture is Iron Age in date (Zarins 2001: 48 Figure 30). Even as there is a stratigraphically earlier structure there, there is no absolute date to suggest when it was built. Nor is there earlier settlement firmly dated elsewhere in Dhofar. On typological grounds, Cremaschi and Negrino (2002: 342-343) identified several small hamlets of circular houses as Iron Age, associated on the coast with Iron Age ceramics. Caves were long in use as shelters (Hilbert 2013; Cremaschi et al. 2015), but the only absolute dates on curvilinear architectural housing are in the Late Iron Age. On the north-facing slopes of Wādī Kharshīt, Juris Zarins (2001: 72-75) first identified a rich complex of 73 high circular tombs typical of the Bronze Age, but none have been excavated. Adjacent to them are curvilinear structures with double-faced, rubble-core walls, some built with natural rock faces as the back wall. Zarins’ test excavation yielded no suitable samples for absolute dating of construction, so the attribution to Bronze Age rests on association with tombs and with platforms, which themselves are undated (Figure 7.16). When these hamlets of curvilinear walled houses appeared, Hadrami colonists from the federated Kingdom of Ḥaḍramawt to the west also settled in Dhofar. The Hadrami settlements included the walled town at Sumhuran and a series of outposts, such as ʿAyn Ḥamrān, Hanūn, Shisr, and Andhur ostensibly designed to facilitate traffic in frankincense (Albright 1958, Cleveland 1960, Zarins 2001). The architectural design and construction style of Hadrami settlements bears no resemblance to the Iron Age architecture of hamlets in the escarpment, suggesting that different cultural groups used these different settlements in the Iron Age.

108

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Figure 7.16. High circular tombs and houses or corrals on the southern slope of Wādī Kharshīt (photograph by J. McCorriston).

What processes explain sedentism or at least, reduced mobility around 300 BCE-300 CE? First, there is the expansion of the Hadrami Kingdom with colonies, forts, and ports adjusted to shifting Roman trade routes through the seas to bypass Seleucid and Nabataean tariffs (Bowersock 1994; Sidebotham 2011). Second, the episode of settlement in Dhofar coincides with a significant decline in precipitation (Fleitmann and Matter 2009). The consequences are poorly understood but may have had significant impacts on water sources in the Nejd, on Terminalia dhofarica cloud forests in the escarpment, and may potentially have favored the expansion of tall grasslands suitable for grazing. Third, the introduction of camels for caravans and food sometime after 1000 BCE (Magee 2015) added a new element to the grazing systems in Dhofar. Unsuited to grazing with cattle in mixed herds (Russell 1988), camels may have crowded some of the traditional lands cattle herders had used in the near Nejd, crowding cattle herders into the grassy escarpment and forcing the corralling and night-grazing used in the fly-ridden monsoon season even today (ElMahi 2010). Whatever the combination of geo-political and local economies, climate-driven aridification, environmental changes, and the introduction of carrier-camel technology, the forces that drove sedentism shifted again by the 5th century CE. Although persistent pastoralists continued to circulate in the mountains, coastal plain, and near Nejd, they abandoned their Late Iron Age settlements.

109

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Late Iron Age Cattle Keepers of the Dhofar Mountains The singular episode of Late Iron Age settlements and their subsequent abandonment has implications for the interpretation of persistent pastoralism. Cattle pastoralism has structured Dhofari mountain life for 6000 years; nonetheless, the past was dynamic. Even as the construction techniques of Iron Age hamlets resemble those in recent use, traditional Dhofari mountain life offers very limited analogy to the Late Iron Age. In the wake of Dhofar’s 1970-1975 conflict, German ethnographer Jörg Janzen wrote a seminal study of the “process of modern change in the nomadic way of life” in Dhofar (Janzen 1986: 7). Even as his study never claimed a timeless past, a temptation for archaeologists is to use Janzen’s detailed documentation of traditional life as a baseline against which to interpret archaeological remains (ElMahi 2010, 2001). Janzen’s work describes the seasonal movements of cattle pastoralists in the mountains, camel herders between coast and Nejd, goat-herding in western Dhofar, and people in coastal towns who migrated herds seasonally to the mountains. Available pasture was supplemented by sardine fodder transported on pack-camels (Janzen 1986: 13; ElMahi 2001), eventually replaced by imported fodder transported by truck. Limited summer vegetable cultivation and sorghum crops were part of the traditional mountain economy, grown in enclosures to protect crops from grazing. These traditional patterns of Dhofari mountain cattle herders veer significantly from the records of Late Iron Age pastoralists. From the Late Iron Age hamlets at Shakeel and Halqoot, excavations recovered no plant materials to suggest any agricultural engagement or exchanges for agricultural products; there were no seeds, husks, or identifiable plant macro-fossil remains, nor were there tools for crop-processing (McCorriston, Buffington et al. 2020). A study of the ample wood charcoals found a heavy reliance on locally available wood taxa like Terminalia sp. and Vachellia sp.. Furthermore, there were wood taxa like Moringa sp., Tamarix sp., and Ziziphus sp. that inhabitants must have gathered and transported from the Nejd, suggesting a pattern of mobility unlike the 20th century (Buffington and McCorriston 2019). Apart from a few shell beads like those of earlier eras, there is no evidence of transhumance or exchange with the coast. Sumharam, inhabited by Hadrami colonists, lay only a few dozen kilometers to the south. Yet none of the ceramics, cereals, camels (bone), alabaster, whetstones, basalt, mill stones, exotic beads in agate and carnelian, iron, or microlithic knapped stone techniques appear at the hamlets at Shakeel and Halqoot. Archaeology gives a vignette of mountain-dwelling Dhofari cattle herders living quite independent of the frankincense traffickers of the coast and interior (McCorriston, Buffington et al. 2020). Yet there is more that hints at interaction between cattle herders and the caravanners. Painted images and graffiti decorate the walls of rockshelters in the Jabal Qara. These images have no associated absolute, radiometric dates, but the images must postdate the introduction of domesticated camels, which are depicted with mounted riders. The images also depict ships, people cultivating dates, camel herds, riders mounted on horses, riders armed with swords and shields, and armed conflict (Figure 7.17). Numerous associated texts in the yet undeciphered Dhofari script hint at a date between the use of Old South Arabian scripts in the Yemen Kingdoms (from ca 800 BCE) and the introduction of Arabic script in the 9th century CE (no Arabic appears with the images). The painted images and graffiti probably date to the Late Iron Age when coastal colonists brought ships, date cultivation, caravan traffic, and perhaps raided mountain cattle. Although archaeologists do not know who painted the images, the scripts hint at sustained contacts across cultural groups of mobile herders and align within a literacy tradition of the nomad world in Arabia and Northern Africa (MacDonald 2009).

110

Cultivating Pastoralism in the Late Iron Age (400 BCE–400 CE)

Figure 7.17. Images of camels in a rockshelter beside a Late Iron Age site; nearby images show camels, men with swords and shields, and perhaps a ship (photograph by J. McCorriston).

Viewed in broader context, the evidence from hamlets at Shakeel and Halqoot suggest that cattle pastoralists of the Dhofar mountains lived in segregation, sometimes conflict, and yet proximity to colonists, contemporary caravanners, and camel herders of the Nejd. After about 400 CE, when mountain hamlets were abandoned, colonists abandoned Sumhuran and its outposts for desert traffic. Perhaps the rare episode of house and corral construction offered a defense against raiding while others crowded the mountains. Permanent settlement perhaps assured persistent pastoralists ready access to the rich and reliable grasslands of the Dhofar plateau. The dynamism of a settlement and abandonment in the Late Iron Age raises questions about other settlement. There exist other archaeological settlement sites in the Jabal Qara. To establish whether other temporal pulses of settlement occurred, these sites need further archaeological investigation. Nonetheless with multiple, stratigraphic, and absolute dates from several sites, a Late Iron Age settlement episode now stands out sharply defined.

111

Chapter 8

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

Concurrent with Late Iron Age settlement in the mountains, the archaeological record in the Nejd suggests a focus of activity around permanent springs and water sources. Since the 1950s, archaeologists have noted the ruins of forts and outposts of the Hadrami colonists and assumed these were facilities supporting overland traffic of Dhofar’s famed frankincense (Albright 1982; Zarins 2001). These outposts lie near springs like Andhur and Shisr or ample rock pools trapping runoff (Hanūn). At Muḍayy, the traces of Late Iron Age pastoralists are trilith monuments and burial cairns overlooking twin springs that today feed minor date palm plantings. It is 2040, and the tiny oasis at Muḍayy has realized its potential as a tourist attraction. Visitors pause here along the caravan trail leading from Dhofar’s famed Wādī Dawkah frankincense groves toward the ancient Yemen kingdoms. We future visitors park where riding camels wait outside the limited circulation zone. Although there are shorter options, we will take the sunset tour, beginning after lunch in ‘The Nakhil’, a shaded restaurant where a traditional Omani meal arrives beside spring-fed pools. On camelback, we climb to the Neolithic platforms at Bā Mashnayq. The Muḍayy cell phone tower makes it easy to access phone apps describing the Neolithic gatherings, sacrifices, and feasts that took place here 6500 years ago, and we imagine the people of Alfin in the valley below. Descending to the camels, we ride past the leafy oasis of the springs, pausing to enjoy bright weaver birds flitting between the palms. Small puffs of desert silt rise from camel treads as we climb to a stony plateau. Pausing by a Bronze Age tomb, we lean from the saddle to photograph a desolate monument, sentinel on this lonely overlook. Our guide explains that the stone balls lying about are geodes, formed by slow percolation through Eocene cavities in the limestone. The camels know where to go; they plod carefully across the plateau to reach a dramatic black-and-white Ḥalūf tomb, perched on the highest outcrop. We remain mounted at careful distance to avoid damage to the fragile surface; the phone apps explain all and give us the drone view of the empty chamber. Finally, we descend the far pass to the immense Wādī Ghārah, where we stop for a cool drink at a wildlife blind. We see desert shrikes and wheatears; with luck, a few gazelles are at rest under an acacia. Our camels move easily up wadi, headed for home. Before Muḍayy, the long trilith field comes into view, again described through our trusty mobile phones. As we near the springs, our guide pauses outside the Islamic cemetery, where twin medieval domes house early Muslim tombs. One has been recently restored, and both are treasured heritage of the early Islamic era in Dhofar. We pause at the springs to stretch our legs in the Visitors’ Center, enjoying dioramas of Muḍayy through the ages. Here we learn more about the sites we saw before we retire to a cool oasis night at the Muḍayy Ribaba Hotel. Trilith Monuments and Communities of Practice Muḍayy has a long prehistory of persistent pastoralist occupation, none of it in houses. Instead, pastoralists and caravanners left tombs and triliths to mark their passage. The springs surely anchored desert pastoralists for millennia, but these same springs made Muḍayy an attractive, indeed essential stop for overland travelers.

112

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

In the Late Iron Age, trilith monuments at Muḍayy replicate an enigmatic and distinctive syntax found in monument forms from the high deserts of Ḥaḍramawt and coastal Mahra in Yemen across Dhofar, Central Oman, and the Sharqiyya. It remains unclear who built these monuments in the Late Iron Age, whether an expansion of Mehri speakers northward or Samad culture southward (Yule 2014: 74, 2018). There remains also a possibility that trilith design and execution was broadly inclusive of multiple ethnic and cultural groups as a commemoration of shared practice or protocols (Bin ʿAqīl and McCorriston 2009). Triliths have distinctive elements: an axis of multiple, low platforms end-to-end; each platform supports clusters of 1-3 uprights; four boulders form corners of a square at the drainage-downslope side of each platform; in front are multiple shallow pits ringed with cobbles at the surface. These pits contain the heatshattered remnants of grilling hearths (De Cardi et al. 1977; Al-Shahri 1991; Zarins 2001; Bin ʿAqīl and McCorriston 2009; Yule 2014: 74). Reported radiocarbon ages from the hearths generally fall within the Late Iron Age 300 BCE-300 CE, with a few exceptions (Yule 2014: 73; McCorriston, Harrower et al. 2020: 445-446) (Figure 8.1). Archaeologists and travelers have suggested many purposes for the iconic triliths. They mark the way for caravan routes; they are bench supports for seating; they are the supportive backdrop for tents; they signal rituals associated with frankincense or celestial events; or they map segmentary social constitutions through commemoration of gatherings (Bin ʿAqīl and McCorriston 2009). None of these is a readily testable hypothesis, and further quantitative work should explore trilith chronology and construction details. Triliths in Space and Time Between 2009-2013, the archaeological survey team located 116 trilith alignments in Dhofar, with additional sites added in 2017 to a total of 138 examples (Figure 8.2). Some are only one platform in length without any indication there was ever another set of elements; others extend for 100 meters with 26 platforms. With precise spatial documentation using sub-meter accuracy and GPS kinetic mapping, the team generated a dataset suitable for quantitative analysis. For example, to examine possible association with celestial events, Michael Harrower tested whether the long axes of 116 Dhofar triliths are random in alignment. He assumed that a non-random pattern would support the hypothesis that triliths tracked one or more celestial orientations. Harrower found that a significant majority of triliths do align within 30 degrees of north-to-south. They are not random. But possibly the explanation is geographic; most of the Nejd drainage runs south-to-north, and triliths lie parallel to the axis of flow, on low terraces of the wadis (Harrower et al. 2014). While these observations do not address the highly symbolic form of triliths, the data suggest that the choice of monument placement was influenced by visibility from the streambed and adjacent terraces where passers-by on major routes would see them. Targeted for further study, a few triliths yielded chronological indications, relative and absolute. In a few instances, the construction of adjacent monuments manifestly quarried stone from existing triliths. In D005, triliths downslope had intact uprights, but those upslope lacked uprights. Other examples are evident at D041, D043, and D045. At D101 construction of a nearby set of (later) boat graves likely explains the removal of trilith uprights. In another example, graffiti in an ancient South Arabian script is upside down on a boulder re-used in boat grave construction; the adjacent trilith has been similarly quarried. These examples of relative dating complement absolute ages on excavated samples that date trilith use.

113

Figure 8.1. Trilith schema after Yule 2014:73 (image by E. Lagan).

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

114

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

Figure 8.2. Image map of trilith locations surveyed. Details in Harrower et al. 2014 (image by K. Olson).

D043-004 (N 17.348; E 54.361) On a broad gravel terrace where tributaries to the Wādī Dhahabūn drainage system widen and topographic relief becomes less steep, seven triliths form a veritable field of monuments; the shortest is 4.9 m with several platforms, and the longest is 44.6 m. Within a few hundred meters is a natural rock pool retaining surface water through long dry seasons. From surface observations, three trilith alignments east-west suggest a chronological series. At the east, D043-002 seems youngest, with a full complement of uprights in place on the platforms and intact arrangements of four boulders square between platforms and hearths. To the west, neighboring D043-003 and D043-004 in parallel alignments appear older because they are robbed of uprights. D043-003 has short graffiti in ancient Dhofar script. A test trench 0.5 x 0.5 m in the northeast quadrant probed the fourth hearth (D043-004D) counted from the northernmost extent of D043-004 (Figure 8.3). This hearth was 2.5 m in diameter, ringed by cobbles 4-15 cm in diameter and filled with clastic, thermally-altered rock, mostly small cobble and pebble size. These were unsorted in a fine sand and silt matrix of loose yellow-brown sediment, likely windblown and trapped among the cobbles. There were rootlets and insect casings in the upper sediments, but at 0.57 m below the surface, an ashy lens of dark grey sediment yielded a small charcoal fragment and a sediment sample enriched with organic humic acids. The limited excavation suggests that the feature is a pit ringed with cobbles that acquired and retained heat through heating wood on top and subsequent mixing of coals and hot surface stone into the deeper reservoir. Such an operation could leave a relatively clean grilling surface for meats and simultaneously bury packages of food for roasting below. 115

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 8.3. Excavation of trilith hearth D043-004a (photograph by C. Heyne).

Radiocarbon ages on two samples are (AA90333, 2141 ± 42) ca. 230-50 BCE from the charcoal and much older (AA90337, 3230 ± 100) ca. 1500 BCE on the humic acids, possibly because these are contaminated with older limestone. Of these ages, the charcoal sample appears to yield the more reliable one, not only because it accords with the broader literature on triliths, but also because the sample can be reliably linked to the final use of the hearth. One cannot discount the possibility that people constructed and used triliths for longer than the range of radiocarbon assays on charcoal, which manifests the final use of hearths. D005-004 (N 17.475; E 53.332) This monument sits among a trilith field at Bā Mashnayq, Muḍayy, nearly in view of one of the twin springs that define the oasis and close to a diverse set of other monuments. On the bedrock terrace upslope are large Neolithic platforms, and another is D005-008 on the same gravel terrace of the broad wadi (Figure 8.4). Other very large platforms nearby have never been probed, and the sandy sediments at the base of the bedrock contain ashes and thermally altered rock from innumerable fires. Close to water, a fine campsite on the route to and from Yemen, the triliths at Bā Mashnayq suggest even from surface observation that their builders re-visited this site. Triliths upslope lack uprights and defined edges; some are simply a series of aligned hearths with only traces of cobble heaps where platforms and uprights presumably stood. To address persistent questions from residents in Dhofar and because prior excavation has only probed the hearths of triliths, the team designed a small excavation in one of the low platforms. We tested whether these very shallow platforms cover a burial. Our survey results show that trilith platforms are 5 to 40 cm in height without uprights (which exceptionally may reach 95 cm). This is shallow for a grave; we tested for any bone or an underlying cist. The team excavated two 0.5 x 0.5 m quadrants within a platform. No bone or artifacts appeared either among cobbles and silt fill of the platform nor beneath it. There is no evidence of an 116

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

Figure 8.4. Trilith D005-004 (photograph by Joy McCorriston).

underlying pit or cist, nor evidence of burial or artifacts. Our results suggest triliths were built for a different purpose. For D005-004, excavation recovered no samples for absolute or relative dating. D101-001 (N 17.348; E 54.361) This example is one of triliths at the confluence of several tributaries to a braided, broad western channel of Wādī Dhahabūn. The site sits on the low gravel terrace beside a limestone outcrop with overhanging shelves of harder rock and narrow shelters along its length. Multiple groups have left remains, of which the triliths are probably oldest. Subsequently, visitors quarried the trilith uprights, probably to re-use in the construction of a series of boat graves (Chapter 9) only a few meters away. Probably subsequent are a set of individual Islamic tombs, each marked with one or more shahid stones. Finally, bedouin have blocked with a dry-stone wall the narrow overhang in a low cliff (Figure 8.5). This is a known practice for communal burials. These were neither viewed nor disturbed by the survey team. The only monuments we disturbed were the exterior of one platform of trilith D101-001 and the hearth associated with that platform; against the southeast exterior of the platform, the team excavated to a depth of 0.03 m in a 0.90 x 1 m trench. In the southwest quadrat of the circular hearth, the team excavated a 0.85 x 0.75 m trench to remove thermally-altered rock fill. Although excavators recovered charcoal fragments, these have been reserved without completing a radiocarbon assay. Our intent was to recover pedogenic carbonates from the underside of the platform perimeter stones (Pustovoytov et al. 2007). Pedogenic carbonates contain carbon deposited by soil organisms in the soil; we had hoped to constrain the date at which a trilith was constructed and its latest use. Unfortunately, the perimeter stones of the trilith platform had insufficient carbonate laminates for this process.

117

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 8.5. Within recent memory, bedouin used dry walls to seal niches in the cliff face, behind which multiple burials may be found (photograph by J. McCorriston).

Figure 8.6. D102-003 looking north to D102-001 (photograph by J. McCorriston).

118

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

Figure 8.7. D102-001 Excavation to the original ground surface on which D102-001 was constructed. Note the thin pedogenic carbonates on the underside of the stone (photograph by J. McCorriston and K. Olson).

D102-001 (N 17.297; E 54.506) One of several triliths, this example sits on a low gravel terrace in the deeply incised upper channel of an eastern tributary to Wādī Dhahabūn (Figure 8.6). Most of the uprights are toppled or gone. Situated further south and closer to the escarpment, this monument and the surface on which people constructed it should have received greater moisture than D101-001. Lichens signify some fog penetration here, and the current limit for fog-dependent frankincense is several kilometers to the north. The ideal locale is unknown for pedogenic carbonate formations under monuments that contain appropriate radiocarbon for dating monument construction. Is there an ideal location within a fog zone? A pedogenic carbonate collected in the escarpment in 2012 was too young to constrain settlement construction close to its actual date; sites deep in the Nejd have minimal soil activity and no thick pedogenic carbonate laminates. Built on the surface of the terrace, the trilith retains some uprights on the southern end, but many have toppled, and most are missing. An excavation against the northeastern, upslope exterior of one of the trilith platforms showed that a single cobble depth defines the perimeter of platforms filled with smaller gravel. By flipping one perimeter stone, the expert determined that there is insufficient thickness of pedogenic carbonate for radiocarbon determination of the monument’s construction date (Figure 8.7). D102-005 A few hundred meters to the north, a similar trilith on the same terrace also lacked most uprights. We excavated a similar probe 0.6 m x 1.13 m to 0.08 m depth to the exterior base of the upslope perimeter cobbles of the platform. Like D102-001, D102-005 was constructed on the surface of the gravel terrace and had accumulated 119

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 8.8. Excavation beside D102-005 reveals insufficient carbonate formations on underside of perimeter cobbles (photograph by J. McCorriston).

minor sedimentation on its upslope side. When a stone was turned, it also yielded insufficient pedogenic carbonate for the desired radiocarbon determination (Figure 8.8). Excavators also quarter sectioned a hearth, 102-005-C (third in line from the northwest to southeast) (Figure 8.9). The team hoped to constrain monument construction (pedogenic carbonates) and final use (hearth charcoal), but neither probe yielded datable material. No one has determined who built these trilith monuments, nor the duration of each monument’s use. Some triliths accumulated right next to previous trilith monuments in extensive trilith fields. In such locations, builders of newer triliths plausibly robbed uprights and perimeter cobbles to construct newer versions. While we cannot determine whether triliths were built all at once, they everywhere show the same constellation of platform, pattern of boulders, and hearths in front as repeating elements (Figure 8.10). Significantly, triliths are potentially accretive, built as elements added over time in a line. Sometimes one finds a solitary element, sometimes a line of dozens, sometimes a field of aligned elements. They could be built by large groups gathering at once, but nothing in their construction requires one. What is clear is a culturally shared convention that inscribed the builders of each added element into a larger social consensus (Rowlands 1993). Although excavators can no longer capture the specific meaning of the monument, the overall trilith syntax as a messaging-while-absent technology used by mobile people is perfectly clear.

120

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

Figure 8.9. Hearth beside D102-005 sectioned (photograph by J. McCorriston and K. Olson).

Figure 8.10. Syntax of a trilith: platform (rear), 4 small boulders squared (mid), and hearths in front (foreground) (photograph by J. McCorriston).

121

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 8.11. Tombs D001-005 (foreground) and D001-004 before excavation (photograph by T. Steimer-Herbet).

Chamber Cairns at Muḍayy (490 CE) Expecting further examples of Bronze Age high circular tombs, the team excavated a pair of chambered cairns at Muḍayy. Situated on a rock shelf overlooking the historic cemetery and one of Muḍayy’s springs, these tombs looked like many other Bronze Age examples, potentially once with an outer facing of courses since quarried (Figure 8.11). But these turned out to be a surprising set of late examples, either re-use of ancient tombs from which previous burials left no trace, or rough imitations of ancient chambered cairns (e.g., D014-002). D001-004 One of two chamber cairns 25 m apart, D001-004 had a conical shape with intact capstones of unworked limestone slabs and an outer face of limestone slabs and rubble. Heaped over a chamber constructed with limestone uprights and a roughly corbelled roof, this cairn might once have had an exterior of courses laid as dry stone facing (Figure 8.12). Or not. Excavators removed both capstones and about a meter of fine silty sand to reach a fully articulated burial. The burial, an adult male 30-40 years, lay flexed on his right side, facing west, and had no accompanying grave goods (Figure 8.13). Pathological conditions on the skeleton show that he experienced significant stress on his back, had a healed rib, and lacked dental caries typical of carbohydrate-rich diets (dates, cereals, agricultural produce). He was laid on the reg surface of the terrace in a chamber subsequently sealed with meter-long limestone capstones. Aeolian activity most likely explains the accumulation of sediment, which thereafter attracted a burrowing rodent. There is no trace of subsequent activity or visitors (Williams et al. 2014; McCorriston et al. 2014). A radiocarbon age on bioapatite from human bone (AA90835, 1533 ± 36) established an interment sometime between 430-600 CE. 122

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

Figure 8.12. D001-004 Plan and section (image by T. Steimer-Herbet).

D001-005 Only meters away, D001-005 had incurred bulldozer damage, removing part of the exterior of the tomb. Despite the lateral damage, one capstone remained in place, and the construction details match those of D001-004. Excavators expected to find a disturbed interior but instead uncovered a fully articulated, flexed burial. Buried on her right side also facing west, this adult woman had one arm extended and one tight across her chest (Figure 8.14). Like the sole male in D001-005, she had no durable grave goods or ornaments. After burial in an empty chamber, capstones sealed the roof of the tomb, sediment accumulated over her body, and there is no evidence of subsequent visits. She died aged 40 or older, with her front teeth and first molar significantly worn and her back teeth in good condition, again without the dental caries signifying the date-rich diet of a palm oasis. She had suffered a healed trauma on the pelvis that contributed to spinal degeneration over time. Further study of the isotopic composition of her teeth suggest that she drank from different water sources in early childhood to those later in life and that she had a consistent diet throughout her life (Williams et al. 2014). With a bioapatite radiocarbon age on human bone (AA90834, 1607 ± 37) excavators establish her death sometime between 400-550 CE. This marks the latest use of this tomb, which may or may not have been constructed thousands of years previously (Figure 8.15). 123

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

Figure 8.13. Flexed adult male burial in D001-004 (image by K. Williams and E. Lagan).

Figure 8.14. D001-005 Adult female buried in D001-005 (image by K. Williams and E. Lagan).

124

Mobility and Community in the Late Iron Age (300 BCE–500 CE)

Figure 8.15. D001-005 Plan and sections (image by T. Steimer-Herbet).

Mobility and Community in the Late Iron Age While chamber cairns at Muḍayy oasis contain burials later in date than the last use of triliths, they nonetheless seem to contain evidence of mobile pastoralists focused on the oasis. Perhaps the woman married in; neither person ate sugar-rich dates or cereals in quantity, if at all. Whether the handful of trilith locations around Muḍayy also represent the earlier practices of mobile pastoralists tightly tethered to the springs, a waystation for caravanners, ceremonial tribal events, the meeting places of cameleers and goat herdsmen, or some other activity, the presence of people around the springs gives no sign of agriculture or settlement. Throughout the Iron Age, persistent pastoralists plied the Nejd routes and anchored themselves to the sparse and reliable water and grazing sources not dominated by Hadrami colonists, among them springs at Muḍayy, and the desert-scape around Hanūn. Nothing suggests a close association with cattle and caprine herders of the Dhofar mountains, but the evidence to date is too sparse. Nonetheless it is clear that those who constructed triliths did so far from the mountain settlements, probably because a sparse population and greater mobility in the desert relied on messaging-while-absent to shape the human landscape.

125

Chapter 9

Patterns of Monuments and Settlement Shaped Dhofar

With narratives of research and of reconstructed lives, this volume throughout has addressed the formation and maintenance of social life beyond the herder’s household in persistent pastoralist economies. Sometimes these communities were aggregates, as the households who settled, at least part time, in the mountain hamlets of Jabal Qara. Some were fluid constellations of households sharing cultural conventions and abstract systems of messaging. Sometimes these mobile pastoralists defined community through ancestry; earlier, they practiced periodic gatherings commemorated by collectively erected monuments to mark their place. Pastoralism persisted, buoyed by social networks essential to the reproduction of any household or cultural group. Chronologies of Monument Construction But even as pastoralism persisted, the means of establishing and maintaining social communities and social networks shifted over time. Survey and excavation results suggest a chronology of monument construction and use. Between 5500 – 4000 BCE, platform monuments commemorated the gatherings of people and their herds at sources of seasonal or perennial water. Survey and excavation results suggest a subsequent gap in monument construction until about 3200 BCE. While it is certainly possible that future excavation and chronological analysis will expand the chronological range of platforms or the onset of subsequent Bronze Age tombs, the record now suggests a gap of about 800 years. Over this time, many generations of herders used the Dhofar plateau and near Nejd, but in the absence of stone monuments, we have scant trace of them. Projectile points occur in the Nejd; people were there. But pastoralists abandoned the use of permanent monuments as commemoratives of large gatherings. A climatic shift to more arid conditions began about 6000 years ago, ending the green Arabia phase of the Early Holocene. A reasonable assumption is that as the Nejd became more arid, fewer animals and fewer people could be supported and populations thinned. If widely disbursed, people probably no longer convened in large numbers. Whatever technologies pastoralists used to assure the continuity of their social and cultural communities, for about 800 years they did not include building new stone monuments. By 5000 years ago stone monument construction returned, this time in the form of tombs. The arid Nejd was no longer an environment in which cattle could thrive, and it is most likely that pastoralists in the Nejd relied on caprines and hunted wild game to a greater extent than had their predecessors. A desert ecosystem reduced in biomass could support widely dispersed households and offered fewer key water sources for animals. In this environment, people relied on ancestry to tie them to place and to others with claim to the resources of specific places. Tomb construction lasted from 3200-2200 BCE, although visitation and offerings may have continued thereafter. Another gap follows, almost a thousand years without monument construction and with few archaeological signals that the sparse resources of the Nejd were in use at all. Elsewhere in Arabia, the early centuries of the 2nd millennium BCE are similarly quiet. Small hamlets appeared close to important oases and in well-watered regions of Southern and Eastern Arabia beyond Dhofar. 126

Figure 9.1. Chronology of monuments in Dhofar. Yellow bars are labelled by tomb type; the gray bars represent (left to right) Halūf tombs, Wall tomb, Chamber cairns, and Islamic graves (image by J. McCorriston).

Patterns of Monuments and Settlement Shaped Dhofar

127

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

In Dhofar, monument construction began again with the erection of triliths in the Nejd 300 BCE-300 CE. At the same time, settlement appears in the escarpment. While cattle pastoralists in the mountains were adapting to an era of greater aridity than previously or at present, the people moving through the Nejd built trilith monuments. Camels were probably in use, perhaps newly herded in Dhofar as a local response to the arid conditions of the Late Iron Age. A thriving trade in frankincense may have spurred local herders as frankincense producers, laying claim to trees and traffic among them. There is no clear archaeological evidence that documents exactly who produced the aromatic resin. It is also unknown whether trilith monuments can be linked to production and exchange in frankincense. One reasonable surmise is that the extent of the groves shrank. Dependent on the marginal fog where monsoon precipitation ends, frankincense trees are sensitive to change in precipitation and circulation as well as to grazing by camels. If the resource became less scarce, stone monuments serving to tie people to place may have been important for local assertions of access. Overlooking the spring, two tombs at Muḍayy tie their single occupants and their kin to a place of permanent water. Ancestry and kinship as ties to place were perhaps again manifest in these monuments, and it remains unclear how many of the unexcavated circular tombs in Dhofar have late burials or late offerings in the early centuries of the Common Era. What the broader record indicates thereafter is that burial monuments reappear in the vicinity of abandoned settlements (Bonacossi 2002; Zarins 2010); such locales offered a ready quarry where the labor of acquiring and transporting large blocks was minimal. Persistent pastoralists were again on the move in the escarpment, coastal plain, plateau, and near Nejd, all places where burial monuments reappear. Persistent Pastoralism and Cultural Inheritance in Dhofar Even as the practices of monument construction, commemoration, and affinity through monuments changed, all evidence suggests that pastoralism persisted. Genetic and archaeological evidence suggest that specialized pastoralism – whether of cattle, camels, or caprines – remained a mainstay of Dhofari economic life. Punctuated only 2000 years ago by a half-millennium long, hyper-arid episode when pastoralists built one-room housing, corrals, and byres in opportune proximity to water sources, mobility was the normal, persistent practice. Even as settlement, possibly only used seasonally, appeared in the mountains, mobile peoples of the desert still constructed and used monuments (e.g., triliths). In eras when the archaeological record is blank, it is unclear how practices were abandoned or reshaped around existing monuments. For example, archaeology could never recover the reverent gestures of a passerby who left no material remains, if such obeisance was even observed. A lock of hair, a leather bag of yoghurt, a feathered totem – these are all perishables that could have once messaged commemoration and social inclusion through performance of recognized practices. There certainly were transformations over time in monument practice; some are visible as differences in monument form and contents. What is clear is a persistent pastoralism among mobile and dispersed people, with a long-term cultural inheritance of monument technology using materials and bodies. Monument practice served to constitute mobile people into societies throughout Holocene prehistory in Dhofar, even as the specific practices changed over time. Albeit a dynamic framework, thousands of generations transmitted a generalized, inherited knowledge of monument practice, shaping cultural landscape for the generations to come.

128

Patterns of Monuments and Settlement Shaped Dhofar

What does this tell us about Modern Oman? Modern Omani residents in Dhofar are the heirs of this cultural landscape. This book tells the stories of ancient pastoralists because in Dhofar, these are by far the most numerous of ancestors. Urban dwellers, farmers, merchants, and hunters account for a small percentage of Dhofar’s former inhabitants. Beyond genetic bloodlines and an ability to tolerate lactase into adulthood, the cultural traditions of pastoralists persist among modern communities. Cattle and camels imbue prestige beyond their economic value. While possible, farming has never developed as a significant economic practice in mountain settlements. The impact of grazing animals and the pastoralists who tend them has shaped ecosystems, including the traditional tall grasslands of the plateau and the frankincense belt of the near Nejd. The cultural landscape is present in the major routes that still pass by ancient monuments, in urban dwellers’ attachments to rural homelands, and in cultural proscriptions like the injunction on collecting frankincense another has tapped. The archaeological record reveals connections to the past that have both spatial and temporal-historical dimensions. Spatial Dimensions of Persistent Pastoralism and Modern Oman This book has focused on research in Dhofar, but there is abundant evident to suggest that the persistent pastoralists there are heirs to a wider South Arabian tradition. Extensive archaeological research in Ḥaḍramawt, Yemen has produced a similar chronology of monument construction and practices by cultural groups closely linked since the Early Holocene (Crassard et al. 2020; McCorriston and Harrower 2020). Linguistic research on modern South Arabian languages (MSAL) finds clear continuities across the western border with Mahra, Yemen (Simone-Semelle 1991; Holes 2006). Some suggest that the modern range of MSAL speakers is considerably reduced from a prehistoric range of cattle pastoralists across much of Early Holocene Green Arabia (Zarins 1992). Following earlier scholarship, Paul Yule (2014, 2018) associates triliths with Mehri speakers in a range that extends through inland northern Oman and firmly links northern and southern Oman in prehistory and early history. Nonetheless, these regions were not monocultural or mono-linguistic through time. In the Late Iron Age, Samad (early Arabic speakers) apparently were interspersed with MSAL language speakers (Yule 2018). There remains spatial diversity in language dialects and social identities of Dhofar’s modern inhabitants (Janzen 1986, Peterson 2004). What we discern from the archaeological record – most particularly of the Late Iron Age with cattle herders in the escarpment, camel and caprine pastoralists in the Nejd, and Hadrami colonists in coastal and oasis locales – points to a prehistoric experience of diverse populations accommodated to each other and sharing resources in a common landscape. Characteristic of modern Dhofar, the experience of socio-economic accommodations by different peoples is a persistent, or at least recurrent, tradition. Temporal/Historical The temporal dimension of Dhofar’s cultural landscape has important implications for how modern Omanis view the past. Archaeological cycles of monument construction establish a history for pastoralists who otherwise left none. This historical trajectory was constrained by cultural adherence to pastoral lifeways and the maintenance strategies that sustained them. In ecological terms, this cultural inheritance of traditional knowledge maintained a human ecological niche suited to the reproduction of households and herds.

129

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

As landscape history, modifications from the past – such as monuments and the routes that connected people to them – constrained and influenced the practices of living people. A simpler way to say this is that people valued lives lived as herders and organized their social lives accordingly. The dynamic monument cycles with their historical shifts in form and practice moreover serve to caution against traditional pastoral practices as timeless analogies for the past. By recognizing the dynamics of pastoral persistence, Oman enjoys a richer tradition of adaptative strategies. Even as monument practice to link dispersed households is a theme across history, the range of practical adjustments showcases the adaptive strength of Dhofar’s prehistoric inhabitants.

130

Bibliography

ABDUL-WAHAB, S.A., H. Al-Hinai, K.A. Al-Najar and M.S. Al-Kalbani 2009. Fog and Rain Water Collection from Trees in the Dhofar Region in the Sultanate of Oman. The Journal of Engineering Research 6: 51-8. AL-ABRI, A., E. Podgorná, J.I. Rose, L. Pereira, C.J. Mulligan, N.M. Silva, R. Bayoumi, P. Soares and V. Cerný 2012. Pleistocene-Holocene Boundary in Southern Arabia from the Perspective of Human mtDNA Variation. American Journal of Physical Anthropology 149 (2): 291-8. AL-ABRI, A., O. Al-Rawas, S. Al-Yahyaee, M. Al-Habori, A. Al-Zubairi and R. Bayoumi 2012. Distribution of the Lactase Persistence-Associated Variant Alleles -13910*T and -13915*G among the People of Oman and Yemen. Human Biology 84 (3): 271-86. doi:10.3378/027.084.0310. ALBRIGHT, F.P. 1982. The American Archaeological Expedition in Dhofar, Oman, 1952-1953 (Publications of the American Foundation for the Study of Man, Vol. 6). Washington, DC, AFSM. ALIZADEH, A. 2010. The rise of the Highland Elamite State in Southwestern Iran. Current Anthropology 51 (3): 353-83. AL-SHAHRĪ, A.M. 1991. Grave-types and triliths in Dhofar. Arabian Archaeology and Epigraphy 2: 182-195. ALVARD, M.S. and L. Kuznar 2001. Deferred Harvests: The Transition from Hunting to Animal Husbandry. American Anthropologist 103 (2): 295-311. ANDERSON, D.M. and W.L. Prell 1993. A 300 KYR Record of Upwelling Off Oman during the Late Quaternary: Evidence of the Asian Southwest Monsoon. Paleoceanography 8 (2): 193-208. ANDERSON, D.M., C.K. Baulcomb, A.K. Duvivier and A.K. Gupta 2010. Indian Summer Monsoon during the Last Two Millennia. Journal of Quaternary Science 25 (6): 911-7. APPADURAI, A. 1986 The social life of things. Cambridge, Cambridge University Press. AVANZINI, A. 2008. A Port in Arabia between Rome and the Indian Ocean (3Rd C. BC-5Th C. AD) (Khor Rohri Report, Vol. 2). Rome, L’Erma di Bretschneider. AVNER, U. 2018. Protohistoric Developments of Religion and Cult in the Negev Desert. Tel Aviv 45 (1): 23-62. AVNER, U., L. Horowitz and W. Horowitz 2017. The Symbolic Role of Ibex in the Negev Rock Art. Journal of Arid Environments 143: 35-43. BAYOUMI, R., S. De Fanti, M. Sazzini, C. Giuliani, A. Quagliariello, E. Bortolini, A. Boattini, et al. 2016. Positive Selection of Lactase Persistence among People of Southern Arabia. American Journal of Physical Anthropology 161 (4): 676-84. BEECH M., M. Mashkour, M. Huels and A. Zazzo 2009. Prehistoric camels in south-eastern Arabia: the discovery of a new site in Abu Dhabi’s Western Region, United Arab Emirates. Proceedings of the Seminar for Arabian Studies 39: 17-30. 131

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

BERNSTEIN, I.S. and E.O. Smith 1979. Primate ecology and human origins: ecological influences on social organization. New York/London, Garland STPM Press. BIN ʿAQIL, A. and J. McCorriston 2009. Convergences in the Ethnography, Semantics, and Archaeology of Prehistoric Small Scale Monument Types in Hadramawt (Southern Arabia). Antiquity 83: 602-18. BONACOSSI, D.M. 2002. Excavations at Khor Rohri: The 1997 and 1998 campaigns. In A. Avanzini (ed.), Khor Rohri Report Vol. 1. Pisa, Edizione Plus, 29-69. BOURDIEU, P. 1977. Outline of a theory of practice. Cambridge, Cambridge University Press. BOWERSOCK, G. 1994. Roman Arabia. Cambridge, MA, Harvard University Press.  BRAEMER, F., T. Steimer-Herbet, L. Buchet, J.-F. Saliège and H. Guy 2001. Le Bronze Ancien du Ramlāt as Sabatayn (Yémen): Deux Nécropoles de la Première Moitié du IIIème Millenaire à la Bordure du Désert: Jebel Jidrān et Jebel Ruwīq. Paléorient 21 (1): 21-44. BRAUDEL, F. [1966] 1972. The Mediterranean and the Mediterranean World in the Age of Philip II. New York, Harper Collins [English translation of second revised edition] BROOKE, J.L. 2014. Climate Change and the Course of Global History: A rough journey. Cambridge, Cambridge University Press. BROWN, J., D. Hill, A.M. Dolan, A.C. Carnaval and A.M. Haywood 2018. PaleoClim, high spatial resolution paleoclimate surfaces for global land areas. Sci Data 5, 180254. doi: https://doi.org/10.1038/sdata.2018.254 BUFFINGTON, A.F. 2019. Using Phytolith Assemblages to Detect a Pastoral Landscape in Neolithic Wādi Ṣanā, Yemen. PhD Dissertation. Department of Anthropology, The Ohio State University. BUFFINGTON, A.F. and J. McCorriston 2019. Wood exploitation patterns and pastoralist-environment relationships: charcoal remains from Iron Age Šhakal, Dhufar, Sultanate of Oman. Vegetation History and Archaeobotany 28: 283-94. BUFFINGTON, A.F., K. Olson, J. Roe, A. Al-Kathiri and J. McCorriston (no date). Homesteads, Holcms and Villages: Results of 2018 Ancient Socio-Ecological Systems in Oman project’s settlement survey and excavations. ASOR 2019 Annual Meeting in San Diego, California. CASHDAN, E. 1983. Territoriality among Human Foragers: Ecological Models and an Application to Four Bushman Groups. Current Anthropology 24 (1): 47-66. CHARPENTIER, V. 2008. Hunter-Gatherers of the Empty Quarter of the Early Holocene to the Last Neolithic Societies: Chronology of the Late Prehistory of South-Eastern Arabia (8000-3100 BC). Proceedings of the Seminar for Arabian Studies 38: 93-115. CLEUZIOU, S. 2002. Présence Et Mise En Scène Des Morts à L’Usage Des Vivants Dans Les Communautés Protohistoriques: L’Exemple De La Péninsule D’Oman a L’Age Du Bronze Ancien. In M. Molinos and A. Zifferero (eds.), I Primi Popoli d’Europa. Florence, All’insegna del Giglio, 17-31. CLEUZIOU, S. 2007. Evolution toward Complexity in a Coastal Desert Environment: The Early Bronze Age in the Ja’alan, Sultanate of Oman. In T.A. Kohler and S.E. van der Leeuw (eds.), The Model-Based Archaeology of Socionatural Systems. Santa Fe, School for Advanced Research Press, 209-27. CLEVELAND, R.L. 1960. The 1960 American Archaeological Expedition to Dhofar. Bulletin of the American Schools of Oriental Research 159: 14-25.

132

Bibliography

CLUTTON‐BROCK, T.H. and P.H. Harvey 1977. Primate ecology and social organization. Journal of Zoology 183 (1): 1-39. CRASSARD, R. 2008. La Préhistoire Du Yémen. Diffusions Et Diversités Locales, Á Travers L’étude D’industries Lithiques Du Hadramawt (BAR International Series 1842). Oxford, Archaeopress. CRASSARD, R. and P. Drechsler 2013. Towards new paradigms: Multiple pathways for the Arabian Neolithic. Arabian Archaeology and Epigraphy 24 (1): 3-8. CRASSARD, R., J. McCorriston, L. Martin and T. Dye 2020. Chapter 8: Manayzah: A Terminal PleistoceneEarly Holocene Rockshelter Occupation. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt, The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008 (Monumenta Archaeologia Series, Vol. 43). Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 157-211. CRASSARD, R., V. Charpentier, J. McCorriston, J. Vosges, S. Bouzid and M.D. Petraglia 2020. Flutedpoint technology in Neolithic Arabia: An independent invention far from the Americas. PLOS ONE 15 (8): e0236314. CREMASCHI, M. and F. Negrino 2002. The frankincense road of Sumhuram: paleoenvironmental and prehistorical background. In A. Avanzini (ed.), Khor Rori Report, Vol. 1. Pisa, Edizioni Plus, 325-63. CREMASCHI, M., A. Zerboni, V. Charpentier, R. Crassard, I. Isola, E. Regattieri and G. Zanchetta 2015. Early-Middle Holocene environmental changes and pre-Neolithic human occupations as recorded in the cavities of Jebel Qara (Dhofar, southern Sultanate of Oman). Quaternary International 382: 264-76. CURCI, A., M. Carletti and M. Tosi 2014. The camel remains from site HD-6 (Ra’s al-Hadd, Sultanate of Oman): An opportunity for a critical review of dromedary findings in Eastern Arabia. Anthropozoologica  49 (2): 207-22. DE CARDI B., B. Doe and P. Roskams 1977. Excavation and survey in the Sharqiyah, Oman, 1976. Journal of Oman Studies 3 (1): 17-33. DE MAIGRET, A. 1996. New Evidence from the Yemenite Turret Graves for the Problem of the Emergences of the South Arabian States. In  J.E. Reade (ed.), The Indian Ocean in Antiquity. London, Kegan Paul International, 321-37. DOSTAL, W. 1983. Some Remarks on the Ritual Significance of the Bull in Pre-Islamic South Arabia. In R.L. Bidwell and G.R.Smith (eds.), Arabian and Islamic Studies. Articles presented to R.B. Sergeant. London, Longman, 196-213. EDDY, F.W. and F. Wendorf. 1998. Prehistoric Pastoral Nomads in the Sinai. Sahara 10: 7-20. ELMAHI, A.T. 2001. The traditional groups of Dhofar, Oman: a parallel for ancient cultural ecology. Proceedings of the Seminar for Arabian Studies 31: 131-43. ELMAHI, A.T. 2010. Pastoralists’ adjustments to hematophageous flies in Dhofar: an analogy of an ancient adaptation. Adumatu 21: 15-32. EVANS-PRITCHARD, E.E. 1940. The Nuer. Oxford, Oxford University Press. EVERHART, J., J. McCorriston and K. Twiss 2014. The Taphonomy of Mortuary Remains: Arabian Human Social Dynamics (AHSD) Project: Results from Late Neolithic and Early Bronze Age Arabian Tombs. Journal of Oman Studies 18: 175-85.

133

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

FEDELE, F.G. 2008. Wadi al-Tayyilah 3, a Neolithic and Pre-Neolithic occupation on the eastern Yemen Plateau, and its archaeofaunal information. Proceedings of the Seminar for Arabian Studies 38: 153-72. FLEITMANN, D., S. Burns, A. Mangini, M. Mudelsee, J. Kramers, I. Villa and U. Neff 2007. Holocene ITCZ and Indian monsoon dynamics recorded in stalagmites from Oman and Yemen (Socotra). Quaternary Science Reviews 26: 170-88. FLEITMANN, D., S. Burns, M. Pekala, A. Mangini, A. Al-Subbary, M. Al-Aowah and A. Matter 2011. Holocene and Pleistocene pluvial periods in Yemen, southern Arabia. Quaternary Science Reviews 30: 783-7. FLEITMANN, D. and A. Matter 2009. The Speleothem Record of Climate Variability in Southern Arabia. Comptes Rendus - Geoscience 341 (8-9): 633-42. FORDHAM, D.A., F. Saltré, S. Haythorne, T.M.L. Wigley, B.L. Otto-Bliesner, K.C. Chan and B.W. Brook 2017. PaleoView: a tool for generating continuous climate projections spanning the last 21 000 years at regional and global scales. Ecography 40: 1348-58. doi: https://doi.org/10.1111/ecog.03031 FRACHETTI, M.D. 2009. Differentiated Landscapes and Non-Uniform Complexity among Bronze Age Societies of the Eurasian Steppe. In B. Hanks and K. Linduff (eds.), Social Complexity in Prehistoric Eurasia: Monuments, Metals and Mobility. Cambridge, Cambridge University Press, 19-46. FRIESEN, J., M. Zink, A. Bawain and T. Müller 2018. Hydrometeorology of the Dhofar cloud forest and its implications for groundwater recharge. Journal of Hydrology: Regional Studies 16: 54-66. FRITZ, C. 2010. Collective Burials and Status Differentiation in Iron Age II Southeastern Arabia. In L. Weeks (ed.), Death and Burial in Arabia and Beyond (BAR International Series 2107). Oxford, Archaeopress, 101-7. GARI, L. 2006. A History of the Hima Conservation System.  Environment and History  12: 213-28. doi: https://doi.org/10.3197/096734006776680236. GIARDINO, C. 2017. Magan: The Land of Copper (The Archaeological Heritage of Oman, Vol. 2). Muscat, Ministry of Heritage and Culture, Sultanate of Oman. GIFFORD-GONZALEZ, D. and O. Hanotte 2011. Domesticating Animals in Africa: Implications of Genetic and Archaeological Findings. Journal of World Prehistory 24: 1-23. GIRAUD, J. 2010. Early Bronze Age Graves and Graveyards in the Eastern Ja’alan (Sultanate of Oman): An Assessment of the Social Rules Working in the Evolution of a Funerary Landscape. In L. Weeks (ed.), Death and Burial in Arabia and Beyond (BAR International Series 2107). Oxford, Archaeopress, 71-84. GIRAUD, J. and S. Cleuziou. 2009. Funerary Landscape as Part of the Social Landscape and its Perceptions: 3000 Early Bronze Age Burials in the Eastern Jaʾlān (Sultanate of Oman). Proceedings of the Seminar for Arabian Studies 39: 163-80. GRAHAM, N.E., C.M. Ammann, D. Fleitmann, K.M. Cobb and J. Luterbacher 2011. Support for Global Climate Reorganization during the Medieval Climate Anomaly. Climate Dynamics 37 (5-6): 1217-45. GROUCUTT, H. and M. Petraglia, 2012. The prehistory of the Arabian peninsula: Deserts, dispersals and demography. Evolutionary Anthropology: Issues, News, and Reviews 21: 113-125. GUAGNIN, M., R. Jennings, H. Eager, A. Parton, C. Christopher Stimpson, C. Stepanek, M. Pfeiffer, H.S. Groucutt, N.A. Drake, A. Alsharekh and M.D. Petraglia 2016. Rock art imagery as a proxy for Holocene environmental change: A view from Shuwaymis, NW Saudi Arabia. The Holocene 26: 1822-34.

134

Bibliography

GUPTA, A.K., D.M. Anderson and J.T. Overpeck 2003. Abrupt Changes in the Asian Southwest Monsoon during the Holocene and their Links to the North Atlantic Ocean. Nature 221 (6921): 354-7. HALDON, J., N. Roberts, A. Izdebski, D. Fleitmann, M. McCormick, M. Cassis, O. Doonan, et al. 2014. The Climate and Environment of Byzantine Anatolia: Integrating Science, History, and Archaeology. Journal of Interdisciplinary History 45 (2): 113-61. HARROWER, M.J., E.A. Oches and J. McCorriston 2012. Hydro-Geospatial Analysis of Ancient Pastoral/ Agro-Pastoral Landscapes Along Wadi Sana (Yemen). Journal of Arid Environments 86: 131-8. HARROWER, M.J., J. Schuetter, J. McCorriston, P. Goel and M.J. Senn 2013. Survey, Automated Detection, and Spatial Distribution Analysis of Cairn Tombs in Ancient Southern Arabia. In D.C. Comer and M.J. Harrower (ed.), Mapping Archaeological Landscapes from Space. New York, Springer, 259-268. HARROWER, M. J., M.J. Senn and J. McCorriston 2014. Tombs, Triliths and Oases: Spatial Analysis of the Arabian Human Social Dynamics Project (AHSD) Archaeological Survey 2009-2010. Journal of Oman Studies 18: 145-52. HENNINGER, J. 1946/47 Das Opfer in den altsüdarabischen Hochkulturen. [Reprinted in Arabica Sacra: Orbis Biblicus et Orientalis 40, Freiburg/Göttingen 1981: 204-253]. HENTON, E., J. McCorriston, L. Martin and E. Oches 2014. Seasonal aggregation and ritual slaughter: isotopic and dental microwear evidence for cattle herder mobility in the Arabian Neolithic. Journal of Anthropological Archaeology 33: 119-31. HILBERT, Y.H. 2013. Khamseen Rock Shelter and the Late Palaeolithic-Neolithic Transition in Dhofar. Arabian Archaeology and Epigraphy 24 (1): 5. HILDEBRANDT, A., M. Al Aufi, M. Amerjeed, M. Shammas and E.A.B. Eltahir 2007. Ecohydrology of a Seasonal Cloud Forest in Dhofar: 1. Field Experiment. WRCR Water Resources Research 43 (10): W10411. doi: 10.1029/2006WR005261 HILDEBRANDT, A. and E.A.B. Eltahir 2006. Forest on the Edge: Seasonal Cloud Forest in Oman Creates its Own Ecological Niche. Geophysical Research Letters 33 (11): L11401. HOLES, C. 2006. The Arabic Dialects of Arabia. Proceedings of the Seminar for Arabian Studies 36: 25-34. HOORN, C. and M. Cremaschi 2004. Late Holocene palaeoenvironmental history of Khawr Rawri and Khawr Al Balid (Dhofar, Sultanate of Oman). Palaeogeography, Palaeoclimatology, Palaeoecology 213 : 1-36. INIZAN, M.-L. and M. Rachad 2007. Art Rupestre Et Peuplements Préhistoriques Au Yémen. Sanaa, CEFAS. IVANOCHKO, T.S., R.S. Ganeshram, G.J.A. Brummer, G. Ganssen, S.J.A. Jung, S.G. Moreton and D. Kroon 2005. Variations in Tropical Convection as an Amplifier of Global Climate Change at the Millennial Scale. Earth and Planetary Science Letters 235 (1-2): 302-14. JANZEN, J. 1986. Nomads in the Sultanate of Oman. Tradition and Development in Dhofar. Boulder, CO, Westview Press. KELLY, R. 1995. The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways. Washington, DC, Smithsonian Institution Press. KENOYER, J.M. 1997. Trade and Technology of the Indus Valley: New insights from Harappa, Pakistan. World Archaeology 29 (2): 262-80.

135

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

KHALIDI, L., M.-L. Inizan, B. Gratuze and R. Crassard 2013. Considering the Arabian Neolithic through a Reconstitution of Interregional Obsidian Distribution Patterns in the Region. Arabian Archaeology and Epigraphy 24 (1): 59-67. KHALIDI, L., L. Purdue and J. Charbonnier 2018. Introduction. In L. Purdue, J. Charbonnier and L. Khalidi (eds.), From Refugia to Oases (Acts of the 38th Rencontres Internationales d’Archéologie et d’Histoire d’Antibes). Antibes, Éditions ADPCA, 119-33. KHAN, M. 1993. The Prehistoric Rock Art of Northern Saudi Arabia: A Synthetic Approach to the Study of the Rock Art from Wadi Damm, Northwest of Tabuk. Riyadh, Ministry of Education, Kingdom of Saudi Arabia. KHOURY, P.S. and J. Kostiner, eds. 1990. Tribes and state formation in the Middle East. Berkeley, University of California Press. KÜRSCHNER, H., P. Hein, N. Kilian and M. Hubaishan 2004. The Hybantho Durae-Anogeissetum Dhofaricae Ass. Nova-Phytosociology, Structure and Ecology of an Endemic South Arabian Forest Community. Phytocoenologia 34: 569-612. KWARTENG, A.Y., A.S. Dorvlo and G.T. Vijaya Kumara 2009. Analysis of a 27-year rainfall data (19772003) in the Sultanate of Oman. International Journal of Climatology 29: 605-17. LALAND, K.N. and M.J. O’Brian 2011. Cultural Niche Construction: An Introduction. Biological Theory 6: 191-202. LAMB, H.H. 1995. Climate, history, and the modern world. London, Routledge. LANCASTER, W. and F. Lancaster 1999. Identities and Economies: Mountain and Coastal Ras AlKhaimah. Proceedings of the Seminar for Arabian Studies 29: 89-94. LÉZINE, A.M., F. Bassinot and J.Y. Peterschmitt 2014. Orbitally-induced changes of the Atlantic and Indian monsoons over the past 20,000 years: New insights based on the comparison of continental and marine records. Bulletin de la Société Géologique de France 185: 3-12. LÉZINE, A.M., S. Ivory, P. Braconnot and O. Marti 2017. Timing of the southward retreat of the ITCZ at the end of the Holocene Humid Period in Southern Arabia: Data-model comparison. Quaternary Science Reviews 164: 68-76. LÉZINE, A.M., J.J. Tiercelin, C. Robert, J.F. Saliège, S. Cleuziou, M.L. Inizan and F. Braemer 2007. Centennial to millennial-scale variability of the Indian monsoon during the early Holocene from a sediment, pollen and isotope record from the desert of Yemen. Palaeogeography, Palaeoclimatology, Palaeoecology, 243: 235-49. LÉZINE, A.M., C. Robert, S. Cleuziou, M.L., Inizan, F. Braemer, J.-F. Saliège, F. Sylvestre, J.-J. Tiercelin, R. Crassard, S. Méry, V. Charpentier and T. Steimer-Herbet 2010. Climate Change and Human Occupation in the Southern Arabian Lowlands during the Last Deglaciation and the Holocene. GLOBAL Global and Planetary Change 72 (4): 412-28. LIPPI, M., C. Bellini, M. Benvenuti and M. Fedi, 2011. Palaeoenvironmental signals in ancient urban setting: The heavy rainfall record in Sumhuram, a pre-Islamic archaeological site of Dhofar (S Oman). The Holocene 21: 951-65. LJUNGQVIST, F.C.A. 2010. New Reconstruction of Temperature Variability in the Extra-Tropical Northern Hemisphere During the Last Two Millennia. Physical Geography 92: 339-51.

136

Bibliography

LUDVIK, G., J.M. Kenoyer, M. Pieniążek and W. Aylward 2015. New Perspectives on Stone Bead Technology at Bronze Age Troy. Anatolian Studies 65: 1-18. MACDONALD, M.C.A. 2009.  Literacy and Identity in Pre-Islamic Arabia (Variorum Collected Series). Surrey, UK, Ashgate. MAGEE, P. 2015. When was the dromedary domestication in the Ancient Near East? Zeitschrift für OrientArchäologie 8: 252-77. MAKAREWICZ, C.A. 2020. The adoption of cattle pastoralism in the Arabian Peninsula: A reappraisal. Arabian Archaeology and Epigraphy 31: 168-77. MARRAST, A., P. Béarez and V. Charpentier 2020. Sharks in the Lagoon? Fishing Exploitation at the Neolithic Site of Suwayh 1 (Ash Sharqiyah Region, Arabian Sea, Sultanate of Oman). Arabian Archaeology and Epigraphy 31 (1): 178-93. MARTIN, L. 2020. The Kheshiya Cattle Skull Ring. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 275-349. MARTIN, L., J. McCorriston and R. Crassard 2009. Early Arabian pastoralism at Manayzah in Wadi Sana, Hadramawt. Proceedings of the Seminar for Arabian Studies 39: 271-82. McCORRISTON, J. 2006. Breaking the Rain Barrier and the sub-Tropical Spread of Near Eastern Agriculture. In B. Winterhalder and D. Kennett (eds.), Behavioral Ecology and the Transition to Agriculture. Berkeley, University of California Press, 217-36. McCORRISTON, J. 2011. Pilgrimage and Household in the Ancient Near East. Cambridge, Cambridge University Press. McCORRISTON, J. 2013a. Pastoralism and Pilgrimage: Ibn Khaldūn’s Bayt-State Model and the Rise of Arabian Kingdoms. Current Anthropology 54: 607-40. McCORRISTON, J. 2013b. The Neolithic in Arabia: A View from the South. Arabian Archaeology and Epigraphy 24: 68-72. McCORRISTON, J. 2020a. Chapyer 1: Introduction to Research. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 19982008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 3-27. McCORRISTON, J. 2020b. Excavations at the Kheshiya Cattle Skull Ring and Neolithic Monument. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 247-71. McCORRISTON, J. 2020c. Neolithic Stone Platforms, Survey and Excavations. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 349-84. McCORRISTON, J. 2020d. Conclusions. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 533-40.

137

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

McCORRISTON, J., A. Buffington, K. Olson, L. Martin, W. Abu-Azizeh, T. Everhart, A. Al Maashani, A. Ahmad Al Kathiri and A. Al Mehri 2020. Ancient Pastoral Settlement in the Dhofar Mountains: Archaeological Excavations at Shakil and Halqoot. Journal of Oman Studies 21: 152-71. McCORRISTON, J. and T. S. Dye 2020. Chapter 18: A Bayesian approach to chronology of the Southern Jol. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 495-532. McCORRISTON, J. and J. Field 2019. Chapter 11: Building Monuments, Building Society. In J. McCorriston and J. Field (eds.), World Prehistory and the Anthropocene. London, Thames and Hudson, 244-75. McCORRISTON, J. and M. Harrower, eds. 2020. Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press. McCORRISTON, J., M. Harrower and R. Crassard 2020. Topic-Specific Survey Approaches. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 109-18. McCORRISTON, J., M. Harrower, L.A. Martin and E.A. Oches 2012. Cattle Cults of the Arabian Neolithic and Early Territorial Societies. American Anthropologist 114: 45-63. McCORRISTON, J., M. Harrower, T. Steimer-Herbet, K. Williams and J. Everhart 2020. Survey and Excavation of Small-Scale Monuments. In J. McCorriston and M. J. Harrower (eds.), Landscape History of Hadramawt: The Roots of Agriculture in Southern Arabia (RASA) Project 1998-2008. Los Angeles, CA, UCLA Cotsen Institute of Archaeology Press, 407-54. McCORRISTON, J., M. Harrower, T. Steimer, K. Williams, M. Senn, M. Al Hādharī, M. Al Kathīrī, A. A. Al Kathīrī, J.-F. Saliège and J. Everhart 2014. Monuments and Landscape of Mobile Pastoralists in Dhufar: The Arabian Human Social Dynamics (AHSD) Project 2009-2011. Journal of Oman Studies 12: 117-43. McCORRISTON, J., M. Moritz, A. Buffington, K. Pustovoytov, S. Ivory and W. AbuAzizeh 2018. Constructing the South Arabian Pastoral Landscape. In L. Purdue, J. Charbonnier and L. Khalidi (eds.), From Refugia to Oases (Acts of the 38th Rencontres Internationales d’Archéologie et d’Histoire d’Antibes). Antibes, Éditions ADPCA, 119-33. McCORRISTON J., T. Steimer-Herbet, M. Harrower, K. Williams, J. Saliège and A. Bin ‘Aqil 2011. Gazetteer of small-scale monuments in prehistoric Hadramawt, Yemen: A radiocarbon chronology from RASAAHSD project research 1996-2008. Arabian Archaeology and Epigraphy 22 (1): 1-22. MEISTER, J., M.A. Hubaishan, N. Kilian and C. Oberprieler 2006. Temporal and Spatial Diversification of the Shrub Justicia areysiana Deflers (Acanthaceae) Endemic to the Monsoon Affected Coastal Mountains of the Southern Arabian Peninsula. Plant Systematics & Evolution 262 (3): 153-71. MILLER, C.S., S.A.G. Leroy, P.E.F. Collins and H.A.K. Lahijani 2016. Late Holocene vegetation and ocean variability in the Gulf of Oman. Quaternary Science Reviews 143: 120-32. MORITZ, M., 2016. Open property regimes. International Journal of the Commons 10 (2): 688-708. MORITZ, M., R. Behnke, C.M. Beitl, R. Bliege Bird, R. Morais Chiaravalloti, J.K. Clark, S.A. Crabtree, et al. 2018. Emergent Sustainability in Open Property Regimes.  Proceedings of the National Academy of Sciences 115 (51): 12859-67. 138

Bibliography

MÜLLER, T., K. Osenbrück, G. Strauch, S. Pavetich, K. Al-Mashaikhi, C. Herb, S. Merchel, G. Rugel, W. Aeschbach and W. Sanford. 2016. Use of Multiple Age Tracers to Estimate Groundwater Residence Times and Long-Term Recharge Rates in Arid Southern Oma. Applied Geochemistry 74: 67-83. NEFF, U., S.J. Burns, A. Mangini, M. Mudelsee, D. Fleitmann and A. Matter 2001. Strong coherence between solar variability and the monsoon in Oman between 9 and 6 kyr ago. Nature 411: 290-93. NETTING, R.M. 1977. Cultural Ecology. Menlo Park, CA, Benjamin Cummings. NEWTON, L. 2010. Shrines in Dhofar. In L. Weeks (ed.), Death and Burial in Arabia and Beyond (BAR International Series 2107). Oxford, Archaeopress, 329-46. NEWTON, L. and J. Zarins 2017.  Dhofar through the Ages; an Ecological, Archaeological and Historical Landscape (The Archaeological Heritage of Oman, Vol. 1). Muscat, Ministry of Heritage and Culture, Sultanate of Oman. NORSTRÖM, E., G. Norén, R.H. Smittenberg, E.A. Massuanganhe and A. Ekblom 2018. Leaf Wax dD Inferring Variable Medieval Hydroclimate and Early Initiation of Little Ice Age (LIA) Dryness in Southern Mozambique. Global and Planetary Change 170: 221-33. O’BRIAN, M. J. and R.A. Bentley 2021. Genes, Culture and the Human Niche: An Overview. Evolutionary Anthropology 30: 40-9 OBERPRIELER, C., J. Meister, C. Schneider and N. Kilian 2009. Genetic Structure of Anogeissus dhofarica (Combretaceae) Populations Endemic to the Monsoonal Fog Oases of the Southern Arabian Peninsula. Biological Journal of the Linnean Society 97 (1): 40-51. ODLING-SMEE, J. and K.N. Laland 2011. Ecological Inheritance and Cultural Inheritance: What Are They and How Do They Differ? Biological Theory 6: 220-30. PATZELT, A. 2011. The Themeda quadrivalvis Tall-Grass Savannah of Oman at the Crossroad between Africa and Asia. Edinburgh Journal of Botany 68 (2): 301-19. PATZELT, A. 2015. Synopsis of the flora and vegetation of Oman, with special emphasis on patterns of plant endemism. Abhandlungen der Braunschweigischen Wissenschaftlichen Gesellschaft 2014: 282-317. PETERS, J., A. von den Dreisch and D. Helmer 2005 The upper Euphrates-Tigris basin: cradle of agropastoralism? In J.-D. Vigne, J. Peters and D. Helmer (eds.), First Steps of Animal Domestication, New Archaeozoological Approaches. Oxford, Oxbow Books, 96-124. PETERSON, J.E. 2004. Oman’s Diverse Society: Southern Oman. Middle East Journal 58: 54-69. PORTER, A. 2002. The dynamics of death: ancestors, pastoralism, and the origins of a third-millennium city in Syria. Bulletin of the American Schools of Oriental Research 325: 1-36. PUSTOVOYTOV, K., K. Schmidt and H. Parzinger 2007. Radiocarbon Dating of Thin Pedogenic Carbonate Laminae from Holocene Archaeological Sites. The Holocene 17: 835-43. ROSE, J.I., V. Cerný and R. Bayoumi 2013. Tabula Rasa or Refugia? using Genetic Data to Assess the Peopling of Arabia. Arabian Archaeology and Epigraphy 24 (1): 95-101. ROSE, J.I., Y.H. Hilbert, A.E. Marks and V.I. Usik 2019a. The First Peoples of Oman: Palaeolithic Archaeology of the Najd Plateau (The Archaeological Heritage of Oman Series, Vol. 5). Oxford, Archaeopress.

139

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

ROSE, J.I., Y.H. Hilbert, V. Usik, A. Marks, M.M. Jaboob, V. Černý, R. Crassard and F. Preusser 2019b. 30,000-Year-Old Geometric Microliths Reveal Glacial Refugium in Dhofar, Southern Oman. Journal of Paleolithic Archaeology 2: 338-57. ROUGEULLE, A. 1999. Coastal Settlements in Southern Yemen: The 1996-1997 Survey Expeditions on the Ḥaḍramawt and Maḥra Coasts. Proceedings of the Seminar for Arabian Studies 29: 123-36. ROWLANDS, M. 1993. The Role of Memory in the Transmission of Culture. World Archaeology 25: 141-51. ROWTON, M.B. 1975. Enclosed Nomadism. Journal of the Economic and Social History of the Orient 17: 1-30. RUSSELL, K.W. 1988.  After Eden: The Behavioral Ecology of Early Food Production in the Near East and North Africa (BAR international series, 391). Oxford, British Archaeological Reports. SALE, J.B. 1980. The Ecology of the Mountain Region of Dhofar. In  S.N. Shaw-Reade, J.B. Sale, M.D. Gallagher and R.H. Daly (eds.), The Scientific Results of the Oman Flora and Fauna Survey (Dhofar) 1977. Sultanate of Oman, Diwan of H.M. for Protocol, 25-54. SALIÈGE,  J.-F.,  A.M.,  Lézine and S.  Cleuziou 2005.  Estimation de l’effet réservoir  14C marin en mer d’Arabie. Paléorient 31(1): 35-43. SCHNEIDER, J. and A. Weiner (eds.), 1991. Cloth and Human Experience. Washington, DC, Smithsonian Institution Press. SCHUETTER, J.M. 2010. Cairn Detection in Southern Arabia Using a Supervised Automatic Detection Algorithm and Multiple Data Spectroscopic Clustering. PhD Dissertation, Department of Statistics, The Ohio State University. SCHUETTER, J., P. Goel, J. McCorriston, J. Park, M. Senn and M. Harrower 2013. Autodetection of Ancient Arabian Tombs in High-Resolution Satellite Imagery. International Journal of Remote Sensing 34 (19): 6611-35. SERVICE, E.R. 1971. Culture and Evolutionism. New York, Holt, Rinehart and Winston. SHOUP, J. 1990. Middle Eastern Sheep Pastoralism and the Hima System. Chapter 7. In J.G. Galaty and D.L. Johnson (eds.), The World of Pastoralism; Herding Systems in Comparative Perspective. New York, The Guilford Press. 195-215. SIDEBOTHAM, S. 2011. Berenike and the Ancient Maritime Spice Route. Berkeley, University of California Press.  SIMEONE-SENELLE, M.C. 1991. Recents développements des recherches sur les langues sudarabiques modernes. Proceedings of the International Hamito-Semitic Congress, Vol. 2. Vienna, Ver offentlichungen der Institut für Afrikanistik un Ägyptologies der Universität Wien, 321-37. SINHA, A., L. Stott, M. Berkelhammer, H. Cheng, R.L. Edwards, B. Buckley, M. Aldenderfer and M. Mudelsee 2011. A Global Context for Megadroughts in Monsoon Asia during the Past Millennium. Quaternary Science Reviews 30 (1): 47-62. SMITH, G.R. and V. Porter 1988. The Rasulids in Dhofar in the VIIth-VIIIth/XIII-XIVth Centuries. The Journal of the Royal Asiatic Society of Great Britain and Ireland 1: 26-44. SMITH, S. 1954. Events in Arabia in the 6th Century A.D. Bulletin of the School of Oriental and African Studies 16: 425-68.

140

Bibliography

SOUTHON, J., M. Kashgarian, M. Fontugne, B. Metivier and W. W-S Yim 2002. Marine Reservoir Corrections for the Indian Ocean and Southeast Asia. Radiocarbon 44 (1): 167-80. STEIMER-HERBET, T. 2004.  Classification Des Sepultures a Superstructures Lithiques Dans Le Levant Et L’Arabie Meridionale (IVe Et IIIe Millenaires Avant J.-C.) (BAR International Series 1246). Oxford, Archaeopress. STEIMER-HERBET, T., G. Davtian and F. Braemer 2006. Pastoralists’ Tombs and Settlement Patterns in Wādī Washʿah during the Bronze Age (Ḥaḍramawt, Yemen).  Proceedings of the Seminar for Arabian Studies 36: 257-65.  STEWARD, J. 1968. Cultural Ecology. In D.L. Sills (ed.), International Encyclopedia of the Social Science. New York, Macmillan, 337-344. STRAUCH, G., K. Al-Mashaikhi, A. Bawain, K. Knöller, J. Friesen and T. Müller 2014. Stable H and O Isotope Variations Reveal Sources of Recharge in Dhofar, Sultanate of Oman. Isotopes in Environmental and Health Studies 50 (4): 475-90. TAPPER, R. 1990. Anthropologists, historians, and tribespeople on tribe and state formation in the Middle East. In P.S. Khoury and J. Kostiner (eds.), Tribes and state formation in the Middle East. Berkeley, University of California Press, 48-73. TEJAVATH, C.T., K. Ashok, S. Chakraborty and R. Ramesh 2019. A PMIP3 Narrative of Modulation of ENSO Teleconnections to the Indian Summer Monsoon by Background Changes in the Last Millennium. Climate Dynamics 53 (5): 3445-61. TIERNEY, J.E., J.E. Smerdon, K.J. Anchukaitis and R. Seager 2013. Multidecadal Variability in East African Hydroclimate Controlled by the Indian Ocean. Nature 493 (7432): 389-92. TILLEY, C. 1994. A Phenomenology of Landscape: Places, Paths and Monuments. Oxford, Berg. UERPMANN, H.-P., M. Uerpmann and S. Jasim 2000. Stone Age Nomadism in SE Arabia: Paleoeconomic Considerations on The Neolithic Site of Al-Buhais 18 in the Emirate of Sharjah, U.A.E. Proceedings of The Seminar for Arabian Studies 30: 229-34. VAN DYCKE R. and S. Alcock 2003. Archaeologies of Memory. Oxford, Blackwell. VARISCO, D. 1994. Mediaeval agriculture and Islamic science: the almanac of a Yemeni sultan. Seattle, University of Washington Press. VOGT, B. 1985. Zur Chronologie Un Entwicklung Der Graber Des Spaten 5-2 Jtsd. V. Chr. Auf Der Halbinsel Oman. Doctoral dissertation, Universität Goettingen. VOGT, B. 1997. La fin de la préhistoire au Hadramawt. In C.J. Robin and B. Vogt (eds.), Yémen au pays de la reine de Saba. Paris, Flammarion, 30-3. WERNER, J.P., J. Wang, J.J. Gómez-Navarro, N. Steiger and R. Neukom 2019. No evidence for globally coherent warm and cold periods over the preindustrial Common Era. Nature 571 (7766): 550-54. WILKINSON, T.J. 2003. Archaeological Landscapes of the Near East. Tucson, University of Arizona Press. WILLIAMS, K.D. and L.A. Gregoricka 2013. The Social, Spatial, and Bioarchaeological Histories of Ancient Oman Project: The Mortuary Landscape of Dhank. Arabian Archaeology and Epigraphy 24 (2): 134-50.

141

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

WILLIAMS, K.D., T. Steimer-Herbet, L.A. Gregoricka, J.-F. Saliège and J. McCorriston 2014. Bioarchaeological Analyses of 3rd Millennium BC High Circular Tower Tombs from The Arabian Human Social Dynamics (AHSD) Project in Dhofar, Oman. Journal of Oman Studies 18: 153-73. XOPLAKI, E., D. Fleitmann, J. Luterbacher, S. Wagner, J.F. Haldon, E. Zorita, I. Telelis, A. Toreti and A. Izdebski 2016. The Medieval Climate Anomaly and Byzantium: A Review of the Evidence on Climatic Fluctuations, Economic Performance and Societal Change. Quaternary Science Reviews 136: 229-52. YULE, P.A. 2014. Cross-roads: Early and Late Iron Age South-Eastern Arabia (Abhandlungen der Deutschen Orient-Gesellschaft 30). Wiesbaden, Harrassowitz Verlag. YULE, P.A. 2018. Toward an Identity of the Samad Period Population. Zeitschrift für Orient-Archäologie 11: 438-86.  ZARINS, J. 1992. Pastoral Nomadism in Arabia: Ethnoarchaeology and the Archaeological Record: A Case Study. In O. Bar-Yosef and A. Khazanov (eds.), Pastoralism in the Levant: Archaeological Materials in Anthropological Perspectives. Madison, WI, Prehistory Press, 219-40. ZARINS, J. 2001. The Land of Incense. Archaeological Work in the Governate of Dhofar, Sultanate of Oman 1990-1995 (Sultan Qaboos University Publications Archaeology and Cultural Heritage Series, Vol. 1). Sultanate of Oman, Ministry of Information, Project of the National Committee for the Supervision of Archaeological Survey in the Sultanate. ZARINS, J. 2010 Funerary monuments of southern Arabia: The Iron Age - early Islamic traditions. In L. Weeks (ed.), Death and Burial in Arabia and Beyond (BAR International Series 2107). Oxford, Archaeopress, 225-36. ZARINS J. 2013. Hailat Araka and the South Arabian Neolithic. Arabian Archaeology and Epigraphy 24 (1): 109-17. ZARINS, J., A. Al-Jawad Murad and K. Yaish 1981. The Second Preliminary Report on the Southwestern Province. Atlal. Journal of Saudi Arabian Archaeology, 5: 9-42. ZARINS, J. and L. Newton 2013. Atlas of Archaeological Survey in Governorate of Dhofar Sultanate of Oman. Mina al Fahel, Sultanate of Oman, Office of the Advisor to H.M. the Sultan for Cultural Affairs. ZAZZO, A. 2014. Bone and Enamel Carbonate Diagenesis: A Radiocarbon Prospective. Palaeogeography, Palaeoclimatology, Palaeoecology 416: 168-78. ZAZZO, A. and J.-F. Saliège 2011. Radiocarbon dating of Biological Apatites: A Review. Paleogeography, Paleoclimatology, Palaeoecology 310: 52-61. ZEDER, M.A. and B. Hesse 2000. The Initial Domestication of Goats (Capra hircus) in the Zagros Mountains 10,000 Years Ago. Science 287: 2254-57.

142

Index

‘Asir (Saudi Arabia), 17, 83 ‘Ayn Ḥamrān (Oman), 108 Abraha of Ethiopia, 17 acacia, 58, 60, 112 Acacia-Commiphora, 97 Ad-Dahārīz (Oman), 39 Afghanistan, 22 agate, 89, 110 alabaster, 110 Al-Baleed (Oman), 24 aloe, 22 amber, 44 American Foundation for the Study of Man (AFSM), 24 Anatolia, 22 Ancient Socioecological Systems in Oman Project (ASOM), 26, 32, 33 Andhur (Oman), 24, 31, 108, 112 Apluda mutica, 14 Arabian Human Social Dynamics Project (AHSD), 5, 7, 23, 26, 28, 32, 33, 46, 51, 55, 64 Arabian Sea, 22, 51 Arabic script, 110 Ash-Sharqiyah (Oman), 25, 39 Atlantic Ocean, 18 Bā Mashnayq (Oman), 39, 45, 47, 112, 116 banana, 18 basalt, 110 beads, 22, 51, 53, 71, 82, 89 betyl, 49, 55, 57, 58, 59 Blepharispermum hirtum, 12 boat-shaped graves, 25, 32, 33, 36, 97 Boswellia sacra, 14, see also frankincense Bronze Age, 22, 26, 41, 64-67, 81, 97, 108, 112, 122, 126 Byzantine Empyre, 17 Cadaba farinosa, 14 Cahokia (USA), 18 camel, 20, 33, 41-45, 63, 65, 78, 79, 94, 102, 109-112, 126-129 caprine, 39, 49, 68, 71, 94, 103, 125-129 caravans, 110-113, 125 carnelian, 66, 82, 110 cattle, 2, 4, 19-22, 37-45, 63, 94, 102, 109-111, 125-129 ceramics, 22, 33, 74, 83, 102, 108, 110 cereals, 2, 4, 110, 122, 125 Chaco Canyon (USA), 18

chert, 24, 49, 51, 55, 58, 66, 68, 74, 81-83, 89, 91, 93, 102 citrus, 18 coconut, 12, 18 Commiphora kataf, 14 Commiphora spp., 12 Conus sp., 71, 73, 80 dates, 2, 4, 18, 110, 122, 125 dog, 41, 42 donkey, 41 dragon’s blood, 22 druids, 27 Duqm (Oman), 39 eagle, 44 Early Holocene period, 16, 19, 24, 126, 129 Ethiopia, 17 Euclea schimperi, 12 Europe, 18 Ficus sycamorus, 12, 97 frankincense, 14, 22, 24, 44, 49, 108, 110, 112, 113, 119, 128, 129, see also Boswellia sacra gazelle, 39, 112 goat, 2, 4, 20, 22, 37-45, 68, 80, 110, 125 graffiti, 23, 27, 110, 113, 115 grape, 18 Greenland, 18 Grewia erythraea, 14 Gymnosporia dhofarensis, 12 Ḥaḍramawt (Yemen), 24-26, 37-39, 63, 83, 93, 108, 113, 129 Hafit period, 26, 41 Hagif (Oman), 83 Hailat Araka (Oman), 39 Halqoot (Oman), 33, 43, 96-99, 102105, 108-111 Halūf tombs, 33, 36, 66, 73, 74, 85-94, 112, 127 Hanūn (Oman), 24, 33, 49, 51, 63, 108, 112, 125 High Circular Tombs (HCT), 4-8, 33, 36, 64-68, 71-76, 79-86, 90 Horn of Africa, 2 Ibn Khaldun, 6 Indian Ocean, 11, 14, 18 Iron Age, 26, 51, 80, 89, 96-99, 108113, 125, 128, 129

iron, 110 Jabal Qamar (Oman), 17, 19, 20 Jabal Qara (Oman), 19, 20, 43, 80, 97, 110, 111, 126 jasper, 39 Jebel Fartak (Yemen), 19 Jibjat (Oman), 42, 80, 96, 97 juniper, 18, 33 land snail, 33, 49, 53 limestone, 1, 2, 12, 15, 45, 47, 49, 51, 53, 55, 57-59, 65-68, 71, 74, 7881, 85, 86, 90-93, 97, 99, 102-105, 108, 112, 116, 117, 122 Little Ice Age, 18 Maerua crassifolia, 14 Mahra (Yemen), 25, 26, 37, 39, 63, 113, 129 Manayzah (Yemen), 37, 39 Marʿib (Yemen), 17, 83 marine shells, 38, 39, 44, 48, 49, 63, 66, 71, 73, 80, 82, 103, 110 Medieval Warm Period and Little Ice Age climate changes (MWPLIA), 18 Mehri language, 113, 129 Middle Holocene period, 16, 17 milk, 41, 44, 45 millet, 18 Modern South Arabian Languages (MSAL), 129 monsoon, 11, 12, 14, 16, 18, 20, 49, 55, 109, 128 Moringa sp., 110 mother-of-pearl, 33, 68, 71, 82 Mthbon (Oman), 42 Muḍayy (Oman), 1, 2, 4, 6, 30 ,33, 39, 45-49, 55, 61-64, 67, 83, 85, 90, 112, 113, 116, 122, 125, 128 Muṭahafah (Oman), 39 myrrh, 22 Nabataeans, 109 Nannorrhops ritchiana, 14, 55 Near East, 17 Nejd (Oman), 11, 14, 16-24, 28, 44, 45, 60, 61, 76, 80, 83, 85, 96, 109113, 119, 125-129 Neolithic period, 22, 26, 27, 31, 39, 40, 45-52, 55, 58, 62-65, 94, 112, 116 Norsemen, 18,

143

Persistent Pastoralism: Monuments and Settlements in the Archaeology of Dhofar

obsidian, 38, 39 olive, 18 Pacific Ocean, 18 Paleolithic period, 68 palm, 13, 14, 22, 44, 55, 112, 123 pomegranate, 18 Qunf Cave (Oman), 14, 17 Rasulid Dynasty, 18 rockshelter, 39, 44, 49, 96, 110, 111 Roman Empire, 109 Roman Warm Period, 17 Sahara Desert, 16 Salalah (Oman), 18, 30 Samad Culture, 26, 113, 129 Seleucid Empire, 109 Setaria pumila, 14 Shabwa (Yemen), 83 Shakeel (Oman), 43, 96, 97, 104, 105, 108, 100, 111 sheep, 22, 37-41, 68, 80 Shiʿb Kheshiya (Yemen), 40, 47 Shisr (Oman), 39, 108, 112

144

Sinai Peninsula, 26, 83 sorghum, 18, 110 Stepped Concentric Alignments of Boulders (SCABs), 36, 55, 58, 61-63 sugar cane, 18 Sumhuran (Oman), 24, 108, 111 Tamarix sp., 110 Ṭāqah (Oman), 108 Terminalia dhofarica, 12, 19, 97, 109 Terminalia sp., 110 textiles, 18, 22 Themeda quadrivalis, 14 ,19 Umm an-Nar period, 26, 41 Umm er-Radhuma formation, 20 ,51 United States of America (USA), 18, 24, 36, 65 Vachellia etbaica, 12, 14 Vachellia sp., 110 Vachellia tortilis, 14, 55, 58, 97 Wādī ʿAybūt (Oman), 2

Wādī Darbāt (Oman), 97 Wādī Dawkah (Oman), 112 Wādī Dhahabūn (Oman), 13, 34, 55, 58, 60, 79-83, 115 ,117, 119 Wādī Ghadūn (Oman), 39, 68, 85 Wādī Ghārah (Oman), 1, 2, 39, 67, 83, 85, 90, 112 Wādī Ḥalūf (Oman), 25, 33, 61 Wādī Haylāʾ (Oman), 30, 68, 69, 72, 73 Wādī Kharshīt (Oman), 108, 109, Wādī Sanā (Yemen), 38-40, 93 Wadi Suq (Oman), 26 Wādī Washʿah (Yemen), 93 Wall tombs, 26, 33, 66, 83, 85, 127 Waqf (Oman), 36 wild okra, 44 Yemen, 8, 17-19, 25, 30, 32, 37-40, 47, 49, 63, 80, 83, 90, 93, 110, 112, 113, 116, 129 Ẓafār (Oman), 18 Ziziphus sp., 60, 110

OmanMHT

OmanMHT

OmanMHT

OmanMHT