The Hellenistic Harbour of Amathus. Underwater Excavations. 1984-1986. Volume 1: Architecture and History 9782869582934, 9782869584464

Table of contents :
Cover
Copyright
Table of contents
Preface and acknowledgements

Citation preview

’

ETUDES CHYPRIOTES

XIX

THE HELLENISTIC HARBOUR OF AMATHUS UNDERWATER EXCAVATIONS, 1984-1986 VOLUME 1. ARCHITECTURE AND HISTORY Jean-Yves EMPEREUR, Tony KOZˇELJ with Olivier PICARD and Manuela WURCH-KOZˇELJ

É

C

O

L

E

F

R

A

N

Ç

A

I

S

E

D

’

A

T

H

È

N

E

S

RK VERKA

40

OQ

KAMBOS

PK

OR

PL

Aqueduct North N orth G Gate e QE

Sakieh Well 3

H House

QF

50

88.60

Well 1

B silica Basilica

40

Well 2 Agio gios gios os Tychonas ycho s

Temple Tem Te emp em ple lle e off A Aphro Aphrodite hrrrodit h Theatre atre 80

Sondage 8

Mo

Sondage 5

lamis to Sa e t u Ro

50

le

Agoraa Agora

Sondage A

Ayia Varva Sondage 16 Sondage 17 Sondage 2

20

Sondage 6 Sondage 4

Basilica B Bas Ba asilica licaa

IInterior Basin In

Sondage 7

a Are d e lor exp n U

We W West Wes estt G Gate ate

st

Kallinikos Ka lin ko oss

QY Q

th M

ole

Basilica lic iica ca

Z

Ha

Sou

rbo

ur

East Tower

En

tra

South Tower N PK

100

200 m

PL

PU

PV

QE

QF

N

nc

Sondage 15 Sondage 18

e

Sondage 12 Sondage 14

Sondage 3 Sondage 13 Sondage 1

Hellenistic Harbour

0

QP

Ea

Palace

East Gate

30

60

West M

70

Postern Gate

QO

ole

Tunnel Tun un u

Sondage 10

0

QO

10

QP

50

100 m

QY

QZ

THE HELLENISTIC HARBOUR OF AMATHUS

É

C

O

L

E

F

R

A

N

Ç

A

I

S

E

Directeur des publications :

Alexandre Farnoux

Responsable des publications :

Géraldine Hue puis Bertrand Grandsagne

D

’

A

T

Révision et mise au point des textes : Colin Clement Réalisation : Fatiha Bouzidi Conception graphique de la couverture : EFA, Guillaume Fuchs Photogravure, impression et reliure : n.v. PEETERS s.a. Dépositaire : De Boccard Édition-Diffusion – 11, rue de Médicis, F – 75006 Paris, www.deboccard.com © École française d’Athènes, 2017 – 6, rue Didotou, GR – 10680 Athènes, www.efa.gr

ISBN 978-2-86958-293-4 Reproduction et traduction, même partielles, interdites sans l’autorisation de l’éditeur pour tous pays, y compris les États-Unis.

H

È

N

E

S

’

ETUDES CHYPRIOTES

THE HELLENISTIC HARBOUR OF AMATHUS UNDERWATER EXCAVATIONS, 1984-1986 VOLUME 1. ARCHITECTURE AND HISTORY Jean-Yves EMPEREUR, Tony KOZELJ with Olivier PICARD and Manuela WURCH-KOZELJ

XIX

This book is dedicated to the memory of Louis Bochaton (1930-2009). Diver extraordinaire, heart and soul of the underwater excavations at Amathus

Louis Bochaton during the excavations of Amathus harbour in 1986 (photograph EFA, Ph. Collet)

Table of contents

Preface and acknowledgements (Jean-Yves Empereur)

10

Introduction (Jean-Yves Empereur)

16

A brief history of the excavations The three campaigns 1984 1985 1986 Excavation methods

PART I. THE HELLENISTIC HARBOUR

18 18 18 19 19 19

29

Chapter 1. The architecture of the Hellenistic harbour (Jean-Yves Empereur, Tony Koželj)

31

The West Mole The east side, to the north, Sondage 8 The exterior of the corner of the West and South Moles, Sondages 5 and 6

33 33 37

The South Mole Sondages 4 and 7 The north side, Sondage A The South Mole, south side, the covered channel

38 38 38 45

8

THE HELLENISTIC HARBOUR OF AMATHUS

The East Mole and the entrance to the harbour Sondages 1, 3, 13, south of the bottleneck entrance Sondages 2, 12, 14, East Mole, north side of the harbour entrance channel The East Mole, further to the north, Sondages 15, 16, 17 and 18

45 45 48 57

Construction methods. Typology, modules and installation of the harbour blocks; machinery Typology of the harbour blocks The number of blocks used in the construction of the harbour The modules of the blocks (Tony Koželj, Manuela Wurch- Koželj) Placing the blocks: description of the machinery

62 62 78 83 87

Chapter 2. The stone quarries of Amathus harbour (Tony Koželj, Manuela Wurch- Koželj) The quarry of Amathus [169] The quarry of Vikles

93 96 102

Chapter 3. History of the Hellenistic harbour (Jean-Yves Empereur)

111

The testimonia Fixing the chronology. Historical sources and archaeological evidence Amathus harbour, a military installation

113 114 115

PART II. POST-HARBOUR HISTORY

121

Chapter 4. The archaeological remains

123

The beachrock (Jean-Yves Empereur, Tony Koželj) The beachrock lines The north side of the harbour and the inner basin The question of the shipsheds Sea level. A later rise

125 125 126 126 127

The wells (Jean-Yves Empereur, Tony Koželj) Well No. 1 Well No. 2 The sakieh (waterwheel) The other wells An abundance of wells Noria or sakieh: defining the terms The wells and the history of the port

133 133 135 136 137 141 143 148

TABLE OF CONTENTS

The coins found in Amathus harbour (Olivier Picard)

9

151

Conclusion. Amathus harbour, a remarkable and enigmatic monument (Jean-Yves Empereur) A remarkable state of preservation Looking backwards and forwards. An overview

161 161 162

Bibliography Bibliography of the excavations of Amathus harbour General bibliography

Chronological table

165 167 169

Fig. 1 — Aerial view of the sunken harbour from an altitude of 500 metres, taken from a helicopter of the British base at Akrotiri (photograph EFA, Ph. Collet)

Preface and acknowledgements Jean-Yves Empereur

The ancient port of Amathus was explored over three underwater excavation campaigns from 1984 to 1986 (fig. 1 and inside front cover plan [= plan 1]). The results were initially presented in halfa-dozen reports and articles1. Final publication, which was almost ready some time ago, has been pushed back for several reasons. The manuscript covering the archaeological material was handed in to the publications department of the French School at Athens in 1999, but technical reasons, most notably the disappearance of the original plans, delayed publication. As for the architectural study, it was on the point of completion when new underwater archaeological developments at Alexandria postponed the final version. Despite the long delay between work on the ground and the publication, we believe that the results of this excavation are worthy of presentation in extenso. We have gathered the different studies into two volumes: the first on the architecture and history of the harbour (with a chapter on the coins); and the second on the artefacts found during the excavations – pottery, amphorae, metal, fauna and wood. Our thanks go first and foremost to Olivier Picard, then director of the EFA, for his constant interest in our work, going so far as to dive with our team during each of the three campaigns. The School partially covered the financing of the work. We would like to express our gratitude to Vassos Karageorghis, then director of the Antiquities Service of Cyprus, for having inspired this excavation and for entrusting us with it, and providing the constant support of his institution. Very special thanks must go to Pambos Iannou, mayor (ȀȠȚȞȠIJȐȡȤȘȢ) of Agios Tychonas, the village in whose area the ruins, both terrestrial and underwater, of ancient Amathus lie. He founded and presided over ȈǹȁȆǹ, a local society for the development of the sunken vestiges of the ancient port of Amathus, which provided our project with an important financial and material contribution. As can be seen on the map and aerial picture (see inside back cover plan [= plan 2] and fig. 2), posidonia sea grass covers a large part of the site and huge bunches growing on the ancient blocks had to be ripped off. This proved a long, hard and tedious task for which it was difficult to find the right tool: billhook, handsaw, pneumatic saw, we tried them all but without finding the way to speed up the work. It seemed that we would have to dedicate a great part of our excavation time to this exercise. Many thanks, thus, must go to the Cyprus National Guard who provided us with the services of some 20 of their divers in 1984 and 1986 (for three months) in order to remove masses of this seaweed. The regrowth of these weeds is rapid and they would dramatically reappear from one campaign to the next (fig. 3a-b).

1.

Cf. list infra, p. 165-166.

12

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 2 — Aerial view of the sunken harbour showing the clumps of posidonia covering the East Mole, taken from a helicopter of the British base at Akrotiri (photograph EFA, Ph. Collet)

The commander of the British base at Akrotiri generously provided compressed air in 1986 and gave us the opportunity on two occasions to conduct aerial photographic coverage of the site, at the very beginning of the excavation and once again a few days before the end of the final campaign (10/10/1984 and 7/10/1986). Similarly, when a squadron of the French navy was in port at Larnaca on Tuesday 7 October 1985, we were able to take more aerial photos through the use of one of their helicopters. We are grateful to Manuela Wurch-Koželj for her contribution on the supply of stones from the nearby quarries. Thanks (again) to Olivier Picard, we are able to add the coins (no less than 43 were discovered during the underwater dig), which were cast by virtue of the skill and generosity of Andreas Giorgiades and photographed by Athanasios Athanasiou, photographer of the Department of Antiquities. Our gratitude goes out to Yiannis Violaris, of the Department of Antiquities, and archaeological officer of the Archaeological Museum of Lemessos (Limassol) District, who was a constant help in our work. We are also grateful to Despo Pilides and Marina Solomidou-Ieronymidou, acting directors, Department of Antiquities, for all the facilities they provided for the material studies. The excavation was under the direction of Jean-Yves Empereur*2 for the three campaigns and co-directed by Colette Verlinden* during the campaigns of 1985 and 1986. The underwater excavation required the collaboration of many, and we are grateful to all those who participated in this team endeavour:

2.

An asterisk denotes a diver.

PREFACE AND ACKNOWLEDGEMENTS

13

Fig. 3 a-b — The slabs of the upper courses of the South Mole before and during cleaning of the posidonia (photograph EFA, Ph. Collet)

14

THE HELLENISTIC HARBOUR OF AMATHUS

Mission 1984: 16 September to 13 October: Armand Magania*, director of diving; Louis Bochaton*, technical director; Hervé Duchêne*, EFA member; Philippe Collet*, EFA photographer; Franck Perdrizet*, EFA topographer; Sylvie Hartmann, EFA restorer-illustrator; Tony Koželj*, EFA architect; Catherine Aubert*, archaeologist in charge of the inventory. Mission 1985: In addition to Armand Magania*, Louis Bochaton*, Tony Koželj* and Sylvie Hartmann: Georges Rapsaet*, professor at the Université Libre de Bruxelles; Annick Chêle*, J. Gachet* and Christian Vandermersch*, archaeologists; Françoise Alabe, EFA member, MarieThérèse Rapsaet, archaeologist in charge of the inventory; Edith Magania and Gisèle Réveillac, assistants to the inventory; Jean-Claude Hurteau* (Iraa de Pau, USR 3155) and Gérard Réveillac* (CCJ, UMR 6573), CNRS photographers. Divers: Marcel Brusson*, Francis Egea*, Jean-Claude Ferrand*, Jacques Martin*, José Ramon*, Stéphane Réveillac*, Jean-Claude Roubaud* and Niki Lappa*, architect in Limassol (fig. 4). Mission 1986: 2 September to 8 October: Louis Bochaton*, Tony Koželj*, Annick Chèle*, JeanClaude Hurteau*, Françoise Alabe and Sylvie Hartmann: Francis Egea*, technical co-director, Catherine Aubert*, archaeologist in charge of the inventory, Polydora Parker*, archaeologist, May Touma, archaeologist, and Andrea Phoungas*, restorer. Divers: Marc Guillaume*, Marc Langlois* and Theodoros Theodorou* (fig. 5). Before diving into the heart of the matter, we would like to express our thanks to: Colin Clement for his unflagging patience and his talents as a translator from French to English of the majority of this volume and for rereading the few articles that were written directly in English; to Fatiha Bouzidi for the care and skill she applied to the layout; and also to Guillaume Fuchs, of the EFA publications department, for his valuable advice and for the creation of the cover of this book. We would also like to recognise the great help accorded by Calliopi Christophi and Elpida Chairy, supervisors of the EFA photography and map archives, as well as the technical staff of the Centre for Alexandrian Studies (CEAlex). A special mention is due to David Blackman, our rigorous reviewer, who reread the entire manuscript with his characteristic attention and erudition: his many corrections and improvements are greatly appreciated. My warmest thanks go to the Honor Frost Foundation (HFF) for its financial support of both the material studies (metal and faunal remains, see Volume 2) and associated drawings and photographs, as well as the layout of the publication (http://honorfrostfoundation.org/the-underwater-portof-amathus-limassol-cyprus-a-hellenistic-naval-harbour-jean-yves-empereur-2016/). And last, but not least, our gratitude goes out to Alexandre Farnoux, director of the EFA, for accepting this volume in the collection Études chypriotes.

PREFACE AND ACKNOWLEDGEMENTS

Fig. 4 — Part of the excavation team, 1985 campaign (photograph EFA, Ph. Collet)

Fig. 5 — Part of the excavation team, 1986 campaign (photograph EFA, Ph. Collet)

15

Sondage 19

150 m

Well 6 Well 5

Sondage 1

South Tower

Sondage 13

East Tower

Sondage 14

Sondage 3

Sondage 18

Sondage 12 Sondage 2

Sondage 15

Sondage 17

Sondage 16

Sondage 11b Sondage 10

Well 2

Sondage 4 Sondage 7 Sondage 6

ole

West M

Sondage 5

Sondage 8

Sondage A

Sou th M

ole

Sondage 11a

Well 1

H Harbour Entrance

0

10

50

100

Well 4 Sondage 11c

Sakieh Well 3

ole

st M

Ea

Fig. 1 — Overall plan of the port developed from aerial photographs taken at an altitude of 500 metres. One can make out the three Hellenistic moles as well as the string of Late Roman wells running along the north of the harbour basin and on a line along the beach. Scale 1:2,500 (plan EFA, Fr. Perdrizet, T. Koželj)

Introduction Jean-Yves Empereur

The ancient harbour of Amathus is exceptional for several reasons. First, it is of a remarkable size and dates from the Hellenistic era, going back as we shall see to the end of the 4th century bc. Furthermore, it appears to have never been used: one might even wonder if construction was completed. Lastly, the history of the sea level is quite particular, with a considerable lowering followed by a rise in water level of at least 2 m in both directions. The majority of explorations of ancient ports have been concentrated on harbours that are now found on dry land as a result of land reclamation or seismic phenomena. Such is the case at Marseilles, Naples, Piraeus, Kition, etc. or, in the second case, at Phalasarna on Crete1. Cases of sunken ports that have been the object of underwater excavations are much more rare. One might mention the Archaic port of Thasos, or that of Caesarea Maritima from the Roman period. The excavation of sunken ports is not common due to both the difficulty and the costs of such an undertaking. The lack of depth paradoxically makes the exercise problematic, requiring amphibious operations that make any observation of the stratigraphy complicated, while also requiring a team that is trained in diving, in the safety measures demanded and in the specific techniques of underwater archaeology. A shipwreck covers a limited zone, whereas harbour structures can extend over considerable surface areas. Thus, the harbour basin of Amathus covers more than 5 hectares, with three moles2, each one stadium long. This size therefore entails choices in excavation procedure and obliges one to engage in spot sondages. Over the three campaigns (1984 to 1986), we opened 20-odd sondages, which we will examine in detail in the following pages (fig. 1 and plan 1). The appearance of the port has completely changed since the excavation years. Today, nothing shows of the ruins: an attractive promenade on a wooden pontoon with metal reinforcements sits 3 m above the site. The beginning of the moles can no longer be seen, nor the three wells that we excavated on the beach. The infill of sand and rock to retain the road that runs above the site has hidden the landscape as we saw it in 1986, and the view to be seen on a photo from 1963 (fig. 2) when the moles poked through the surface has gone.

1.

Cf. F.J. Frost, E. Hadjidakis, “Excavations at the Harbor of Phalasarna in Crete”, Hesperia 59 (1990), p. 513-527.

2.

In order to follow Mediterranean usage, we will use the word mole, rather than jetty, to designate the stone construction built to form and protect the harbour basin.

18

THE HELLENISTIC HARBOUR OF AMATHUS

A BRIEF HISTORY OF THE EXCAVATION The three campaigns This excavation was financed by the École française d’Athènes and ƗƆƐƕƆ, the Cypriot society for the development of the sunken vestiges of the ancient port of Amathus, with the constant and efficient help of the mayor of Agios Tychonas, Pambos Iannou, to whom we express our special thanks. Excavations lasted ten months spread over three campaigns from 1984 to 1986.

1984 The first campaign in 1984 lasted only four weeks, from 16 September to 13 October, and involved ten divers. The aim was to observe underwater the monumental ensemble, to comprehend the extent and then define an approach, such that one could understand the organisation of the moles and their method of construction. The aerial photographs taken by Philippe Collet, thanks to the helicopter provided by the commander of the British base at Akrotiri, gave us an idea of the extent of the port (exterior dimensions more than 5 ha.) and allowed our photogrammetry specialist, Franck Perdrizet, along with architect Tony Koželj, to draw up an overall plan that would have been impossible to create in the time allowed using the classic methods of underwater topography. In addition to this first result, the short campaign let us conduct an initial sondage along the internal façade of the South Mole, at about 20 metres from the south-west corner of the monument and to reveal six

Fig. 2 — The moles of the ancient harbour breaking the surface in 1963 (photograph EFA)

INTRODUCTION

19

courses of header stones with abundant pottery and amphorae from the beginning of the Hellenistic era connected to the construction and use of the quay (fig. 1, Sondage A3). In addition, two wells filled with mud and containing Late Roman pottery were recorded and emptied in the northern part of the harbour basin4.

1985 The campaign of 1985 lasted twice the length of the previous, running for eight weeks from 2 September to 25 October 1985. The team was larger, with 24 members, almost all of them divers, and logistics improved thanks to the positioning of two floating pontoons, each one equipped with three water pumps. The National Guard of Cyprus helped us considerably in cleaning the clusters of posidonia that covered the ancient installations. The commander of the British base at Akrotiri continued his support and provided us with the compressed air that we needed for the underwater excavations. A visit by the French navy to the port of Larnaca meant once again that a helicopter was available for a new series of aerial photographs. We conducted eight sondages in two sectors: five sondages at the junction of the West and South Moles, on their south and north sides: three sondages towards the harbour entrance passage, to the south-east, against the East and South Moles. Lastly, we emptied the fill of a basin attached to a sakieh to the north along the beach, revealing abundant archaeological material (pottery, wood, metal objects, faunal remains) from the Late Roman period.

1986 The third and final campaign took place between 1 September and 13 October 1986 with extensive means at our disposal: 16 divers, two pontoons and five pumps. Our supporters also remained faithful: three months of cleaning the posidonia by the National Guard, compressed air provided by the commander of the British base at Akrotiri, as well as a last series of aerial photographs from a helicopter placed at our disposal. Our efforts were directed principally at the two sides, south and north, of the entrance passage to the south-east of the harbour, widening the sondages begun the previous year. In addition to early Hellenistic ceramic material, some 30 coins were discovered. On the shore, the junction of the port with the city wall was examined. Thanks to the geographer Pierre-Yves Péchoux, the lines of beachrock inside the ancient harbour basin were identified as demonstrating the variations of the shoreline over time.

Excavation methods The external plan of the port was linked to the grid system of the acropolis as drafted by M. Schmid and P. Aupert in 1976. The general plan of the harbour was created from aerial photos taken from a helicopter placed at our disposal on two occasions by the commander of the British base at Akrotiri (fig. 3): the topographer Fr. Perdrizet applied photogrammetry procedures to the photos taken from an altitude of 500 m with partial overlap. During excavations, Tony Koželj, architect at the École Française d’Athènes, was in charge of surveying using a total station mounted on the shore (Wild Distomat D1). Taking advantage of the shallow conditions, the divers were able to hold the prism 3.

See infra, p. 38-45.

4.

See infra, p. 133-136.

20

THE HELLENISTIC HARBOUR OF AMATHUS

reflector above the surface of the water and directly above the point to be registered (fig. 4). Drawings were done underwater on a plastic-coated board with a pencil, eraser and a folding wooden carpenter’s ruler (fig. 5a-b). The underwater excavation was conducted from one pontoon in 1984, then two in 1985 and 1986. The wooden pontoon was kept afloat by a series of empty oil drums, on which the divers had three airline “hookahs” (fig. 6a-b), and compressed air cylinders generously provided by the British base (fig. 7). Long supple hoses supplied the diver with pressurised air so that he did not need to carry a bottle, thus facilitating the lifting and transporting of stones to reveal the masonry courses of the moles. With unlimited air supply, each dive lasted two hours working in pairs, at the rate of two dives per day. The use of diving cylinders was restricted to those divers who had to move a distance underwater, principally when performing tasks related to topography and photography. The sketch in fig. 8 shows the operations on the site with a water suction system set on a pontoon. A simple portable motor-pump (A) is fitted with a hose that sucks in seawater (B) and sends it under pressure past an elbow joint (C) connected to another hose held by the diver. The inversion creates suction that removes the sand and reveals the archaeological material, shards, coins etc. (D and fig. 9a-b), while expelling the mud (E and fig. 10). Registering the material was done underwater using bags prepared in advance and carried by the diver, who handed them over to the inventory service at the end of each dive. They were marked with the year, the stratigraphic unit (SU) number, and the individual identity number of the object in the bag, as well as the name and time of the diver in order to avoid errors (fig. 11).

Fig. 3 — A helicopter from the British base at Akrotiri employed in the aerial photographic coverage of the site that was required to create photogrammetry charts as well as record the details of the sondages and the sunken structures. Photographic coverage was conducted during each of the three campaigns (photograph EFA, C. Verlinden)

INTRODUCTION

Fig. 4 — Direct line topography using a Distomat and reflector above the sunken blocks (photograph EFA, Ph. Collet)

21

Fig. 5a-b — The architect Tony Koželj drafting the blocks under water: Koželj with his drawing on a plastic film-laminated board (photograph EFA, C. Verlinden)

22

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 6a-b — Floating pontoon: to the left, three reels holding thin blue air lines to supply the divers with air; to the right, three pumps sending water under pressure through large grey and blue pipes (photograph EFA, Ph. Collet)

INTRODUCTION

23

Fig. 7 — Compressed air cylinders generously provided by the British base to fill the divers’ tanks (photograph EFA, Ph. Collet)

The inventory advanced in parallel with the excavations, as did the restoration, photography and drawing of the material. Lifting blocks was achieved using balloons inflated with compressed air (fig. 12a-c). When the weight exceeded 5 tons, we added strings of empty oil drums filled with compressed air (fig. 13). This latter manoeuvre was only possible because the site was not too deep, rarely more than 4 m. At the end of each campaign, the documentation was completed and ready for examination by specialists. The SU files with bag numbers and the description of each individual object were handed over, giving an assemblage to be found in the artefacts studies (volume 2), notably for the ceramic contexts. Lastly, a report was published annually, one in French and one in English. About ten provisional reports and articles were published during and just after the excavations, especially an annual report in the Bulletin de Correspondance Hellénique (BCH) and the Annual Report of the Department of Antiquities (ARDAC). One can find a bibliography at the end of this volume5. These regular reports already managed to provide an idea of the progress of the excavation and of the discovery of the harbour. Nevertheless, their limited size and illustration do not go into the details that will be developed in this volume. This work represents the fruit of many hours’ discussion between archaeologist and architect, with the input of ceramologists, a numismatist and other specialists who have provided the studies to be read in the forthcoming volume 2 on the material brought to light over three years of excavation.

5.

Bibliography on the excavations of Amathus harbour, infra, p. 165-166.

24

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 8 — Sketch showing the workings of the underwater excavation. On the floating pontoon, a pump (A) sucks up seawater (B) and sends it under pressure towards an elbow bend (C) that reverses the suction allowing the diver to suck the mud (D) which is expelled behind him (E) (drawing EFA, T. Koželj)

Fig. 9a-b — The diver sucks up the mud, retrieving any objects – shards, coins etc. – which he places in a bag that has been numbered before the dive (photograph EFA, Ph. Collet)

INTRODUCTION

25

Fig. 10 — The mud is expelled by the water suction system (photograph EFA, Ph. Collet)

Fig. 11 — Example of a bag containing an individual object, a coin of Demetrius Poliorcetes. The year is marked [19]86, the sondage = 14 (near the entrance to the harbour), the individual identity number = 14, and the date 11/IX/86. In addition there is the Limassol Museum number = AM 1450, and a serial number for the collection of coins found at Amathus, LN 1224/86 (photograph EFA, J.-Y. Empereur)

26

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 12a-c — Three images showing the lifting of small blocks (photograph EFA, Ph. Collet)

Fig. 12b

INTRODUCTION

27

Fig. 12c

Fig. 13 — Oil drums filled with compressed air were used for lifting large blocks (photograph EFA, Ph. Collet)

Part 1 The Hellenistic harbour

Chapter 1

The architecture of the Hellenistic harbour Jean-Yves Empereur, Tony Koželj

We will begin our description of the harbour with the West Mole and Sondage 8. Thereafter we will move south of this mole and to the sondages that explored the south-west corner of the harbour, on the outside with Sondages 5 and 6. Then, following along the South Mole, we will head eastwards to examine Sondage A, then the channel that traverses the South Mole from north to south, before reaching the eastern end of this same mole, with Sondages 1, 3 and 13 along the southern façade of the harbour entrance. From there we go to the north side of the entrance and Sondages 2, 12 and 14, followed by the ensemble 15, 16, 17 and 18 along the East Mole. After this description of the sondages, we will establish the typology of blocks laid as headers to form the wall cladding of the moles, as well as the modules that were employed. We shall attempt to calculate the number of blocks involved in the construction of the three moles, and then we will examine the manner in which the moles were built and with the aid of what particular machinery. We shall also look for the quarries from where the blocs were extracted.

THE WEST MOLE The east side, to the north, Sondage 8 The West Mole shows up as a broken line on aerial photographs. It is composed, on its eastern edge, of a double cladding of headers with a central fill. Sondage 8 was opened in 1985 and it revealed on the inside of the basin the presence of three courses of header blocks set on a bed of pebbles and sand (fig. 1-2). In total, 17 blocks were cleared during the sondage. They measure roughly 2 m in length and are 0.4 to 0.5 m thick. The west side of the mole is built in the same fashion, and the fill is composed of limestone offcuts and debris from the neighbouring quarries. The smaller number of courses is noteworthy – three against six for the South Mole (Sondage A) and the harbour entrance – demonstrating that this area of the basin must have been shallower in antiquity than the rest of the mole, and the absence of large blocks which reinforce the south-west corner indicates that this part of the harbour was less thick because less exposed to storms. The upper side of the headers is roughly 1 m beneath present sea level (0.88 to 1.08 m), though it cannot be said whether this was the upper course in antiquity. Nevertheless, there are clear signs of wear when compared with the lower courses, most probably an indication of a surface position. The lowest course is composed of larger blocks than the two upper courses. It lies at a depth of 2.24 m upon a bed, some 0.20 m thick, of small, generally irregular stones, some of which are squared off, as shown in the section. To

THE HELLENISTIC HARBOUR OF AMATHUS

8

9

N

Elevation A

9 -1.1

6 -1.5

-2.4

5

-2.8

6

-1.2

3

-1.2

8

-1.5

6

-1.5

1

-1.0

-0.9

-1.0

3

-0.8

8

34

Elevation B Isolated block inside the mole

Sea level at time of excavation

-0.88 -1.03

-0.86

-0.99 -1.08

-1.16

-1.85

-2.24

-2.86

Elevation A

Isolated block inside the mole

0

0.50

1

2

Elevation B 3

4m

Fig. 1 — West Mole, Sondage 8, plan and section with three courses of blocks (drawing and CAD EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

D1

D2 -0.88

C1

C2

C3

-1.03

-0.99

-1.56

B1

-1.23

35

C4

C5

D3

D4

-1.08

-1.51

A3

-1.28

C6

A4 -1.56

-1.19

-2.86

N

Elevation A

-2.45

Elevation B Isolated block inside the mole

Sea level at time of excavation East Face -0.88 -1.03

D1

-1.16

-1.08

D2

C3

C2

C1

-0.86

-0.99

C4

D4

D3 C5

C6

B1

-1.85

A1

A4

A3

A2

-2.24

-2.86

Elevation A

Isolated block inside of the mole

0

0.50

1

2

Elevation B

3

4m

Fig. 2 — West Mole, Sondage 8, plan and section with numbering on three courses of blocks (drawing and CAD EFA, T. Koželj)

36

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 3 — South-west corner of the harbour, Sondages 5 and 6, aerial view (photograph EFA, Ph. Collet)

Fig. 4 — South-west corner of the harbour, Sondages 5 and 6, aerial view (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

37

Fig. 5 — Pile of large roughly finished stone blocks, exterior south-west corner of the harbour, aerial view (photograph EFA, Ph. Collet)

the left of the section, one can see the interior fill of the mole composed of this same layer of stones, but thicker up to ca. 40 cm, topped by 0.70 m of larger rocks that are chips and offcuts from the nearby quarry. One can note the similarity of the lifting bosses that project from the small sides of the blocks (Type 3, see infra, p. 63).

The exterior of the corner of the West and South Moles, Sondages 5 and 6 In 1985, in order to determine the nature of the harbour wall on the side open to the sea, we undertook several new sondages to the south of the harbour as well as in the south-west corner. Growths of posidonia created circular patches on the aerial photos (fig. 3) leading us to believe that there was one or perhaps several reinforcing towers. Thus, two sondages (5 to the west and 6 to the south on plan 2) were opened in this zone outside the harbour (fig. 4). The removal of the seaweed showed that, contrary to the illusion created on the images, there were no reinforcing towers. Roughly squared blocks cut from the local limestone quarries were not dressed. They were not set in regular courses but were thrown willy-nilly (fig. 5). The blocks that were placed on the bottom, on the mud, were more regular and laid with greater care. This south-west corner, being the part of the harbour that is most exposed to storms, was reinforced by blocks of greater size.

38

THE HELLENISTIC HARBOUR OF AMATHUS

THE SOUTH MOLE Sondages 4 and 7 Sondages 4 and 7 to the south of the South Mole also did not reveal an ensemble of parallelepiped blocks as on the north side of the mole, but rather debris from the nearby quarries over a thickness of more than 1 m, all of which was covered by a layer of pebbles and sand, and even the occasional dressed block which had accidentally fallen out of the line of headers (see aerial view, south of Sondage A, fig. 8b). The ensemble of these cut stones lie on a layer of pebbles and sand. It lies upon the muddy bottom, as seen above in the description of the row of headers. From north to south, one follows the succession of headers and protecting blocks. This system recalls what can be seen in the modern harbour of Paphos, with a mole that runs from the castle, the interior serving as a mooring quay, while the side open to the sea is a mass of large blocks protecting against the waves (fig. 6-7). This particular treatment is distinct from the other moles of the harbour, and is due to the specific exposure of this part of the harbour to the waves from the south and south-east.

The north side, Sondage A Sondage A was the very first of the excavation. It began on 1 October 1984. The site was chosen after many days of prospecting inside the harbour in order to understand the layout. The selected spot was the northern side of the South Mole. It is the triangular sondage seen on the photos taken from the helicopter on 10 October 1984 (fig. 8a-c). It sits upon the 180 m-long line of header blocks that form the inside cladding of the South Mole. Using a suction pipe, a thick layer of stones (fig. 9) mixed with some pottery shards was lifted to reveal a flat natural formation that resembled the beachrock in the middle of the harbour basin. Quickly, however, it appeared that this was not the case when a second course was revealed, swiftly followed by a third. In a few days, up to six courses of blocks set as headers were discovered in this 5 m-wide sondage oriented east-west (fig. 10). Access to these courses was obstructed by the accumulation of mud that held many stones and plenty of Hellenistic pottery: mushroom lip amphorae and others with a cylindrical foot with a central hollow, from the end of the 4th century bc. The headers appeared to be set in a quincunx pattern as if from the effect of unequal lateral pressure from the south. One can also notice to the right of the section (fig. 11) a sort of gash where one can make out the end of one section of stones and the beginning of another. This phenomenon is most probably due to the process involved in laying the blocks: see infra for a reconstruction of the machinery used for placing the blocks, fig. 51a-e. This machinery will have finished a section at the end of Series of blocks B5, C6, E7, F7, G7 and begun laying the next towards the west. This means that the machinery, which was moving from east to west, first placed the blocks at the furthest extension of its jib towards the east and then laid the following blocks in reverse as it gradually moved towards the west to join up with the section that had been built previously. This explains the (relatively small) miscalculation that led to Block B5 not being lowered down to the lowest course but becoming stuck at the level of the second course, between blocks A3 and A4 (see fig. 12). Series of blocks B5, C6, E7, F7, G7 is composed of only five courses instead of the six of the other series. As a result, one can make out a collapse on the surface at Block G7, below which the headers tilt some 11 to 13° towards the west. The first course of blocks was placed directly on the sandy bottom, as seen in fig 12. A deep sondage into this loose sand revealed that it contained no human artefacts (fig. 13).

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Fig. 6 — The south mole of the modern harbour of Paphos (photograph EFA, J.-Y. Empereur)

Fig. 7 — Graphic reconstruction of the South Mole (CAD EFA, T. Koželj)

39

40

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 8a-b — South Mole, Sondage A, aerial views (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Fig. 8c — South Mole, Sondage A, aerial view (photograph EFA, Ph. Collet)

41

Fig. 10 — South Mole, Sondage A, courses of blocks (photograph EFA, Ph. Collet)

Fig. 9 — South Mole, Sondage A, clearing the roughly finished stones that filled the harbour basin (photograph EFA, Ph. Collet)

42

THE HELLENISTIC HARBOUR OF AMATHUS

-1.14

-1.20

-1.18

-1.06

-3.76 0

0.5

1

2

3

4m

N

Sea level at time of excavation

-1.14

-3.56

-1.18

-1.06

-1.20

-3.76

Fig. 11 — South Mole, Sondage A, plan and section with courses of blocks (drawing and CAD EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

-1.14 A_G1

A_G3

A_G2

-1.20

-1.18

A_G4

A_G5

A_G6

43

-1.06 A_G7

A_G10

A_G9

A_G8

-3.76 0

0.5

1

2

3

4m

N

Sea level at time of excavation

-1.14 G3

G2

G1

F2

F1 E1

-1.18

E2 D1 C1

F4

F3 E3

C2

E5 C4

C3 B2 A1

F6 E6

G9

G8

G7

G6 F5

E4

D2

B1

-3.56

G5

G4

-1.06

-1.20

G10

F8

F7

E8

E7 C7

C5 C6

B3 A2

B6

B4 B5 A3

A4

-3.76

Fig. 12 — South Mole, Sondage A, plan and section with numbered blocks (drawing and CAD EFA, T. Koželj)

44

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 13 — South Mole, Sondage A, the lower courses of blocks, with the sandy and muddy bottom: deep sondage in the sand can be seen at the bottom of the photo (photograph EFA, Ph. Collet)

Fig. 14 — Aerial view of Amathus harbour: the two floating pontoons are placed exactly in the harbour entrance channel; to the right, one can see where the courses of slabs of Sondages 1, 3 and 13 were cleaned, and to the left, those of Sondages 2, 12 and 14 (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

45

Some measurements: the upper course sits between 1.06 and 1.20 m beneath the surface of the sea and the sondage descends to 3.76 m, that is, more than 2.55 m depth for six courses, though without us knowing how many may be missing. Forty-seven blocks were revealed during this sondage. The funnel shape is due to the basin being filled with stones that are most probably debris from the nearby quarries (see fig. 9). This observation fits with a remark made by Françoise Alabe in her study in the volume 2 on the Hellenistic pottery: she is surprised to see rejects from pottery workshops thrown into the harbour as if one intended to fill it in1. The blocks are Type 3, according to the typology described infra, p. 63.

The South Mole, south side, the covered channel Roughly half way along the South Mole, at 92 m east of the western end and 80 m from the inside corner with the East Mole, excavations revealed a channel whose bed had been left clear during the construction of the courses of headers and the placing of blocks on the southern part of the mole. This channel, 36.86 m long, is clearly visible on the aerial photographs (plan 2 and photo fig. 8b) and played a flushing role, facilitating the circulation of water inside the harbour basin and thus avoiding silting. Such a system is usual in ancient harbour installations and can be found in the Archaic era harbour of Thasos2 and in the Hellenistic port of Taposiris Magna some 50 km west of Alexandria3, as well as in modern harbours.

THE EAST MOLE AND THE ENTRANCE TO THE HARBOUR Sondages 1, 3, 13, south of the bottleneck entrance

Sondage 1. In September 1985 we opened a triangular sondage along the north side of the south mole towards the entrance channel of the harbour (fig. 14 and sondage 1 on the map plan 2). Underneath a 10 cm-thick layer of sand containing no artefacts, we reached stratigraphic levels composed of stones and Hellenistic pottery. The SU 3 was more compact, with shards that were not part of the surface debris. We noted the presence of several fragments of basket-handled amphorae. We descended to the seventh course that lay on the muddy bottom (fig. 15-16). Sondage 3. Thereafter, we began Sondage 3 north of Sondage 1, along the north side of the South Mole in the entrance channel of the harbour (fig. 17). Clearing the sand allowed us to get down to the sixth course that was filled by large-sized pebbles. Very fragmentary Hellenistic pottery appeared in a layer of stones within the grey mud, as well as three Ptolemaic coins (AM1443, AM1444, AM14464). The layout of the blocks is interesting. At this important spot of the harbour entrance, the blocks are placed in an alternating fashion: the upper course is set as stretchers, that beneath as headers, the third as stretchers, and the pattern repeats for the three lower courses. To the west (on the right of the section), the blocks are once again as headers following the layout of the east face of the mole (see fig. 16). Sondage 13. This was begun in 1986 and the intention was to explore the inside corner of the same mole at its south-eastern extremity. We revealed the internal cladding of the blocks discovered 1.

F. Alabe, volume 2: “The other hypothesis, that the potters took it upon themselves to dump their rejects into the harbour, would imply such a scandalous negligence on the part of the authorities as to be rather improbable.”

2.

See J-Y. Empereur, A. Simossi, “Le port de Thasos”, BCH 115 (1991), p. 716.

3.

See M.-F. Boussac, “Recherches récentes à Taposiris Magna et Plinthine, Égypte (1998-2006)”, CRAI 151 (2007), p.457-460 with fig. 6b.

4.

See infra, O. Picard, “The Coins found in Amathus harbour”, p. 156-159.

46

THE HELLENISTIC HARBOUR OF AMATHUS

Sea level at time of excavation

South Mole

South Mole - East Face

-2.70

-2.78 -3.43

-2.14

-2.35

-2.50

Harbour Entrance

-1.85 -1.83 -1.82

-1.85 -1.80

-1.93

-2.10

-1.92

-1.98

-2.41

-2.44

-2.75 -3.17

-2.89

-3.23

-3.53

-3.52 -3.80 -4.35 -4.90

-5.05 0

-5.96

0.5

1

2

3

4m

-5.91

Fig. 15 — Section of Sondages 1, 3 and 13, seen from the east outside the harbour: to the right, towards the north, the harbour entrance (drawing and CAD EFA, T. Koželj)

-2.00 -1.71

0

0.5

-1.99

4m

3

2

1

-1.83

Harbour Entrance

-1.65

-2.35 -1.60

-1.60

-2.44

-2.41

-2.45

-2.70

1_G2 1_G1 -3.43 1_G3 -3.53

1_J 2 1_I 2

1_J 1

-3.17

1_K1

-2.50

1_I 1 1_I 3 -3.23

-2.75

-2.35

-2.14

1_K2 -1.85

1_K3 -1.83

-1.82

-2.41

1_K6

1_K4

1_J 3

1_K7

1_K5

-1.99

-1.85 -1.80 -2.10

-2.41 1_J11

-2.00

1_K9 -1.93

-2.89

1_K10

1_K8

-2.44

1_K11

-2.50

-1.92

-2.80

-1.98

-2.52

-2.18 -5.05

Sea level at time of excavation

-4.35

South Mole - East Face

South Mole

-2.50

-2.35

-2.14

Harbour Entrance

-1.85 -1.83 -1.82 -1.85 -1.80

-1.93

-2.10

K6

-1.92

-1.98

-2.41

K8 K9 K10(I1)K11(E1) -2.44 K7 K1 K2 K3 K4 K5 -2.70 -2.75 -2.78 J3 J12(D2) -2.89 J10 J9 J5 J11 J6 J7 J8 J4 J1 J2 -3. -3.1 3..1 3 17 1 -3.17 -3.23 I7 I13 -3.43 -3.43 I12 I8 I6 I9 3.53 -3.53 I11 I1 I5 H2 I10 H3(C3)--3.52 I4 H1 I2 I3 G9 G10 G11 G1 G2 G12 G13 G14 G15(B1) --3.80 G3 G4 F8 F7 G8 G7 G5 G6 F11(A1) F9 F10 F4 F5 F6 F1 -4.35 F3 E2 F2 D3 D2 C8 E1 D5 D4 C7 D6 D1 E3 E4 -4 -4.90 4 0 C1 C2 C3 C4 C5 C6 -5.05 C9 B1 B2 A1 A2 A3 A0 -5.96

0

0.5

1

2

3

4m

-5.91

Fig. 16 — Plan and section of Sondages 1, 3 and 13 with numbered blocks, seen from the east outside the harbour: to the right, towards the north, the harbour entrance (drawing and CAD EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

N

-2.41 3_H7

-2.41

-2.35

3_H6

-2.00

-1.93

-1.71

3_I1(1_K10)

-1.98

-2.52

F1 E1 E

-2.91 -2 2

3_E2 EE2 (1_H2) _H2) _

F1

3_E3(1_H3)

D2 3_C1(1_F11)

3_B1(1_D6 )

C2

-4.51

--5.18 5.18

B1 A2

3_H4 3_G5 H4 H

FF1

F1

H5 G6 F6

E6 6

D4

D3

-3.94

H3 H

E5

E4

-3.47 -3.47

D1 -4.35

F1 F

FF1 1

-3.66

-1.83

H2

G2

F1

-1.60

-1.65

H1

-2.00

3_G1(1_K11)

F1

3_H3

Sea level at time of excavation

-1.60

-2.18

3_G4

3_F5

-2.89

-1.60

3_E4

-1.99

3_E3(H3)

3_F4

3_F3

-5.05

-1.83

-1.65

-1.92 3_G1(1_K11) -2.52 -2.50 -2.44

3_F2

-2.18

3_H1

-2.44

-2.00

3_D1(1_G15)

-4.35

3_I1(1_K10)

3_G2

3_F1

-1.98

-1.99

3_H2

-2.45

3_G6

3_E1

3_H8

3_H5

-2.41

47

E7

D5 5

D6

C3

A3

B2

94 -4.94 4

-6.04

0

1

2

3

4

5m

Fig. 17 — Plan and north section of Sondage 3: inside face of the cladding of the harbour entrance channel (drawing and CAD EFA, T. Koželj)

THE HELLENISTIC HARBOUR OF AMATHUS

-1.70

-1.60

19_F3

19_F1

19_E3 -1.94 -1.98

-1.50

-1.71

19_F2

N

48

D2

19_E1 19_E2

-1.78

-2.44

D1

-2.69

-3.21

Sea level at time of excavation

-1.95

-2.08 -1.93

-2.69

-2.44

-1.68 -1.89

E1

E2 D1

-3.21

-1.73 -1.73

-1.50 -1.64

-1.50

-1.57

F2

-2.02 F3

F1

E3 D2

E4

E5

D3

D4

C2 C1 B1

-2.57

-3.48 -4.18 -4.65

A2

A1

0.5

-2.52

-2.60

B3

B2

0

-2.36

C3

-4.84 North Face

-1.60

1

2

3

4m

Fig. 18 — Plan and north section of Sondage 12: six courses facing the harbour entrance channel (drawing and CAD EFA, T. Koželj)

during Sondages 1 and 3, with the same alternating headers and stretchers. A large rectangular block reinforced the layout, as can be seen on the plan (fig. 16, east of 1_K3 to 6). Some blocks bear wedge marks from the quarries (see fig. 47a-b). Some intact vases were discovered5. Sondages 12, 2, 14, East Mole, north side of the harbour entrance channel The zone to the north of the entrance channel was explored in detail and three sondages were opened: 12, 2 and 14. Sondage 12. Opened on 8 September 1986 in order to clear the south side of the East Mole in the harbour entrance, opposite Sondage 3 (see plan 2 and fig. 18). The posidonia covered a wealth of pottery, wood and coins: a remarkable amount in comparison with the other sondages. Of the seven courses we worked on, we were only able to completely clear three because of lack of time. 5.

See volume 2, F. Alabe, “The Hellenistic pottery of Amathus harbour”.

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

49

Fig. 19 — Sondage 12: one can make out large fragments of Hellenistic vases stuck in the mud (photograph EFA, Ph. Collet)

Of the 31 blocks identified, 17 bear a Type 3 boss, four bear Type 4, while five bosses had been tooled off. Many worn pebbles were evident in the sand, an understandable feature in a channel with tidal currents. This layer, which went down to the third course of the mole, held a lot of pottery (fig. 19) as well as 30-odd coins, including two from the cities of Paphos and Rhodes (4th century bc), some 10 dating to the period from the death of Alexander until the Antigonids, and two that could be attributed to Demetrius Poliorcetes6. We also recorded the discovery of a palmette-shaped element from a piece of gold jewellery7.. A tower. In the harbour entrance channel we discovered dozens of small carefully worked parallelepiped blocks, some with clamp and lever holes, which belonged to a construction that stood above the platform in the south-west corner of the mole, above the blocks of Sondage 12, some of which were broken by the weight of an upper structure (fig. 20a-b). These stretchers were of small size, around 0.20 m on the longer side (fig. 21a-b and 22a-b). Certain of them displayed a curved face suggesting that they belonged to a building with an upper cylindrical storey (fig. 23). This could mean that there was a tower near the entrance channel with an access stairway. Was this a lighthouse or was it the support for the presentation of a statue? In this same sondage, near the north-east corner of the harbour entrance, a block of local limestone, 2.1 m long by 1.38 at its widest, was discovered lying on the sand beneath unworked stones at a depth of 3 m from the present-day surface (fig. 24a-b). The front side of the block features a 20 cm-deep inset: the right-hand width measures 1.10m. Two rectangular mortises on the front side show where metal cramps joined a second block to this one. The block is distinctive because of the hollows that can be seen on the upper surface, 6.

See infra, O. Picard, “The coins of Amathus harbour”, p. 155.

7.

See volume 2, M. Michael, “The metallic artefacts of Amathus harbour”.

50

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 20a-b — Sondage 12, upper course: the slabs were broken by the weight of the tower (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Fig. 21a-b — Stretchers from the tower at the time of their discovery (photograph EFA, Ph. Collet)

51

52

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 22a-b — The stretchers of the tower lying on the upper course of Sondage 12 (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

53

Fig. 23 — In the middle of the picture, a curved block discovered at the foot of Sondage 12 (photograph EFA, Ph. Collet)

which were designed to affix the statue of a human (fig. 24c). The two cavities received life-sized human feet (the complete sole measures 0.29 m) that were slightly turned outwards, the left being just a bit further forward than the right. A third recess, to the left of the left foot, indicates that the figure held an object (a lance or perhaps a trident?). A fourth recess can be seen to the right. It is circular, 3 cm in diameter, with a deeper cavity in the centre. We cannot say whether this statue base was connected to the harbour, but only that it was found beneath a layer of blocks and sand. No fragment of statuary was discovered in the vicinity, and yet perhaps this is the base of a statue of a human-sized figure holding a vertical object (lance, trident etc.) that was erected near to or upon a tower, which stood nearby, on the north side of the harbour entrance. We will go no further in any attempted identification of this character, be it human or divine. Sondage 2. In 1985, Sondage 2 was intended to examine the inside of the south-west corner of the East Mole, following a square shape on a N/S and E/W alignment. Once the posidonia was cleared a homogeneous ensemble of pottery from the Early Hellenistic period was revealed (fig. 25 -26). The courses of 30 blocks that were cleared have a particular appearance: to the south, there are five courses of good sized blocks while to the north, there are nine courses of narrower blocks, sometimes half the size of the first; the deepest blocks lie at a depth of 3.8 m. The bosses on the small sides are split between Type 3 (13 examples) and Type 4 (eight examples). Sondage 14. Opened on 3 September 1986, south of Sondage 2 along the west side of the East Mole. In the south corner, eight courses were explored down to 5.09 m depth without reaching the sand (fig. 27). This was the deepest part of the entire excavation without us knowing the total height of the reinforced platform that stood above the three sondages, 2, 12 and 14, dominating the entrance channel. The section shows the junction of the large blocks, set as stretchers and headers, with the south side of the East Mole (see fig. 26). Of the 33 blocks cleared, 28 displayed Type 3 bosses.

54

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 24a-b — Marble base of a human statue (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

55

2.10

a

0.29 m

a Fig. 24c — Marble base of a human statue (drawing and CAD EFA, T. Koželj)

Fig. 25 — View of Sondage 2, west side (photograph EFA, Ph. Collet)

1.00

1.38

0.20

0.90

56

THE HELLENISTIC HARBOUR OF AMATHUS

H4

H6

H5

-1.63

-1.65

-1.59

-1.73

-1.78

-1.50

H5

H4

H3

H2

H1

-1.60

I1

-2.80

N

Sea level at time of excavation

-1.60 -2.00

-2.54

-1.96 -1.59 I 1 -1.58 -2.80 -2.25 -2.06 -1.63 -1.69 H7 -1.68 H5 -1.78 -1.77 H4 -1.79

G1

G2

G1

H6

G2 D2

D1 -3.13

F1

E1

-4.22

C1 B1

0

0.5

1

D4

-2.91

F4

E4 D4

C3

A1

C2 B4

B5 A3

A2

A1 East Face

F3

D3

B3

B2

-4.81

D3

H5

G5

G4

E3

B1

D2

D1

G3

F2

C2

E2

-3.70

H4

H3

H2

H1

H8

2

-5.09

3

4m

Fig. 26 — Sondage 2, plan and section, east side with numbered blocks in black, west side blocks in red (drawing and CAD EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

57

Sea level at time of excavation

-1.58

-1.53

-1.68

-1.71

-1.98

-1.94

-1.90

-1.86

-1.75 -2.36

-2.40 -3.04 -3.70 -3.90

-5.38

0

0.5

1

2

3

4m

Fig. 27 — Sondage 14, section of the south side (drawing and CAD EFA, T. Koželj)

The East Mole, further to the north, Sondages 15, 16, 17 and 18 Opened at the end of September 1986, these four sondages were intended to clear a part of the East Mole further north than the previous interventions. Alternating between the west (15 and 17) and east sides of the mole (Sondage 18), they led to the clearing of sections of the headers of this East Mole along some 30 m. The depth of the harbour basin seems to lessen towards the north: the mole has only three courses in Sondage 15 and two in Sondage 17 (fig. 28-29), which is not dissimilar to the situation in Sondage 8 that was undertaken on the West Mole. This line of headers breaks off in an abrupt fashion north of Sondage 15 only to resume on a parallel track after a sidestep further to the north (fig. 30a-b). The pottery material is heterogeneous with an apparent mix going from the 4th century bc to the 6th century ad. In Sondage 18, north-east of zone 2+14, several blocks were found naturally cemented into the beachrock along with fragments of amphorae (handles and necks of LR1 amphorae) at 0.8 m below the surface, which would seem to correspond to the depth of the first line of the Late Roman shore. However, beneath a bed of stones some more pottery appeared, including a late 4th c. bc persian bowl. Thus, this was an area of activity during the Early Hellenistic period that was later cemented over by the beachrock when the sea level fell. Here we finish the description of the sondages conducted in the harbour basin. This catalogue was necessary to understand the construction of the moles, the conveyance of the blocks that make them, their positioning, as well as the nearby quarries where these blocks were cut.

58

THE HELLENISTIC HARBOUR OF AMATHUS

--1.48

-1.53

Sondage 17

N

-1.63 -1.66

Sondage 2

Sondage 12

-2.27

-1.59 9 -1.73 -3.58 -1.78

-2.69

-3.21

-1.55-1.53

-3.25

-3.24 -2.58

--2.52 -2.5

-1.98 -1.94

-1.90

-1.80

-1.86

-1.77 -1.71

1 -1.58 --1.53 -1.68

-1.75 -2.40

-2.36 6

-2.67

-3.06

-3.90

Sondage 14

0

1

2

3

4m

Fig. 28a — Plan and section of Sondages 2 and 12 with, to the north, Sondage 17. Scale 1:100 (drawing and CAD EFA, T. Koželj)

59 THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

--4.13 4.1 13

-1.48 -1.53

Sondage 17

Sea level at time of excavation

-3.53 -3 53 -3. 3

0

-1.54

-3.38

-1.51

-1.34

2

3

-3.80 38

-1.42

Sondage 2+12 1

Fig. 28b — Sondages 17, 2, 12, and 14 (drawing and CAD EFA, Tony Koželj)

-1.47

-2.90 -2 2 90 2.9 90

4

-1.57

5m

-1.36 -1.37

-2.51 -2 2.5 51 1

Sondage 14

60

THE HELLENISTIC HARBOUR OF AMATHUS

-1.48 17_B2 17_A2

N

-1.53

17_B1 17_A1

Sea level at time of excavation

17_B1

17_B2 -1.48

-1.53

17_A2

17_A1

-4.13

0

0.5

1

2

3

4m

Fig. 29 — Sondage 17, plan and section of the west side. Scale 1:50 (drawing and CAD EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

61

Fig. 30a-b: Aerial views of the East Mole: one can make out the twin cladding on the left of the mole, and to the right towards the east, blocks that were thrown into the sea to protect it (photograph EFA, Ph. Collet)

62

THE HELLENISTIC HARBOUR OF AMATHUS

CONSTRUCTION METHOD. TYPOLOGY, MODULES AND INSTALLATION OF THE HARBOUR BLOCKS; MACHINERY Typology of the harbour blocks The construction of the moles of Amathus harbour required the quarrying and placing of thousands of tons of stone to fill the interiors of the moles, but also the regular cutting of thousands of parallelepipedic blocks for the double cladding of these moles. See below for an attempt to calculate the numbers involved8. The oblong blocks are of quite different shapes, although all perfectly contemporary. They were created by work teams operating concurrently in different quarries with specific working methods, though we are unable to identify any logic of construction peculiar to a particular requirement9. The typological differences of the blocks lie in their adaptions for lifting, transport and setting. We have presented seven principal types that can be distinguished (fig. 31), in the knowledge that each type can thereafter be regrouped into several modules. We have then drawn up synoptic tables of the distribution of the blocks according to the type of lifting boss.

1

2

3

5

6

7

4

Fig. 31 — The seven boss types on the blocks of Amathus harbour (CAD EFA, T. Koželj)

Type 1 block bears no protruding hold, no lifting boss, and no channel cut into the block itself (fig. 32a-b). The ropes were attached around the body of the block. This simple system must have caused some difficulties when lifting and setting the blocks in place because at the same time there was an increase in complexity to allow for better control when handling the blocks. Another explanation for this unusual shape could be that the bosses were tooled off after installation, but the technical difficulty of such an operation on blocks under water is immediately obvious. Type 2, which is relatively rare, bears two small handling knobs that were left when the block was cut and extend to left and right of the upper part of the small side. The upper and lateral edges follow the corner of the block but the interior surface is slightly curved to make the attachment of lifting ropes easier. This is a rather complicated type that split the weight carried by the rope in two (fig. 33a-b).

8.

Cf. infra, on the placing of the blocks for calculations of the theoretical number of blocks involved, p. 87-91.

9.

Cf. infra, for descriptions of traces of the extraction of these blocks in the quarries, with a distribution table of the blocs by type and module, p. 104-105.

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

63

Type 3 is the most frequent. To use the terminology of R. Ginouvès and R. Martin10, it displays a “canal en U extérieur” (U-shaped external channel), with a projecting central boss, but where the lateral parts of the block have been tooled off. The rope was placed around this lifting boss (fig. 34a-b). Type 4 has the same central U-shaped boss as the preceding type, but the lateral parts between the boss and the edges of the small side of the block have not been cut back to the depth of the base of the boss. Only the lower part has been tooled off (fig. 35a-b). Type 5 is similar to the preceding with a U-shaped boss but the lower part has not been tooled off any further than the lateral parts, thus forming a protecting channel on three sides for the passage of the hoisting rope (fig. 36a-b). Type 6 is the most classic. It bears a very prominent circular boss (fig. 37a-b). Type 7 resembles Type 3, but the rectangular boss covers the entire height of the slab. It was created by cutting away two rectangular strips to the left and right of the face of the block (fig. 38a-b). It is worth noting that the heaviest blocks are Types 3, 4, 5, and 7. Types 1 and 6 are generally lighter. When looking at the accompanying synoptic tables, which analyse the type of blocks in the six sondages conducted on the three moles, one can propose the following findings as regards distribution:

South mole Sondage 8. Of the 17 blocks revealed, seven are broken and of no identifiable type; the others, i.e. ten blocks, are all Type 3.

West mole Sondage A. From a total of 47, of which two are very partially visible, and five blocks either broken

or with no identifiable handling features, the distribution is as follows: one block of Type 1, one of Type 2, 30 of Type 3, and six of Type 4. Sondage 1. From a total of 91 blocks, of which seven are partially visible and two are broken, seven are of Type 1, two of Type 2, 44 of Type 3, 19 of Type 4, 11 of Type 6 and two of Type 7. Sondage 3. From a total of 43 blocks, one is partially visible, 10 are of Type 1, 28 of Type 3, two of Type 4, two of Type 5.

East mole Sondage 2. Of the 30 blocks revealed, nine are broken and thus of no identifiable type, 13 are

Type 3 and eight are Type 4. Sondage 12, eastern face. Of the 33 blocks revealed, one being partially visible, three blocks are Type 2, 27 Type 3 and two Type 4. Sondage 12, southern face. Of the 20 blocks revealed, one being partially visible, two blocks are Type 1, one Type 2, 15 Type 3 (2 Type 3 on front, Type 4 on rear) and one Type 6. Sondage 14. Of the 33 blocks revealed, five are of no identifiable type, five are Type 1, one Type 2, 17 Type 3, four Type 4, one Type 5 and two Type 6. From a total of 159 blocks, once the 34 non-identifiable are discounted, Type 3 is by far the most numerous, representing 58% of the ensemble and three quarters of the identifiable blocks (73.6%). It is followed, but at some distance, by Type 4 with 41 individuals, that is, 7%, or 13.6% of the identifiable 10.

R. Ginouvès, R. Martin, Dictionnaire méthodique de l’architecture grecque et romaine I: matériaux, techniques de construction, techniques et formes du décor, Paris, 1985, p. 122, pl. 33, fig. 2 and 7.

64

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 32a-b — Type 1 boss (CAD EFA, T. Koželj and photograph EFA, Ph. Collet)

Fig. 33a-b — Type 2 boss (CAD EFA, T. Koželj and photograph EFA, Ph. Collet)

Fig. 34a-b — Type 3 boss (CAD EFA, T. Koželj and photograph EFA, Ph. Collet)

Fig. 35a-b — Type 4 boss (CAD EFA, T. Koželj and photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

65

Fig. 36a-b — Type 5 boss (CAD EFA, T. Koželj and photograph EFA, Ph. Collet)

Fig. 37a-b — Type 6 boss (CAD EFA, T. Koželj and photograph EFA, Ph. Collet)

Fig. 38a-b — Type 7 boss (CAD EFA, T. Koželj and photograph EFA, Ph. Collet)

66

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 39 — Detail of the photogrammetry reconstruction of the South Mole, scale 1:1,250 (plan EFA, Fr. Perdrizet, T. Koželj)

blocks. The other types are represented by just a few examples, from one to five. It is noteworthy that certain sondages revealed several types side by side. Only Sondages 8 and 12 South held one single type, whereas there were two types in Sondage 2, three types in Sondages A and 12 East, and the mix rises to five types for Sondage 14. It is also worth mentioning the frequent concurrence of Types 2 and 3 and of Types 3 and 4: indeed, Type 3 was present in all the sondages and was even on its own in Sondage 8. When considering the distribution of types of lifting boss by mole, the results are the following: West Mole: 13 blocks of Type 3 and eight of Type 4 South Mole: one block of Type 2, 30 of Type 3, six of Type 4 East Mole: five of Type 1, 50 of Type 3, two of Type 4, one of Type 5, two of Type 6, one (?) of Type 7 We may conclude from these numbers that the West and South Moles, which are contiguous, present the same types, principally Type 3, followed in a smaller proportion by Type 4. The East Mole, on the other hand, includes a greater variety of types. While Type 3 is still the commonest, Types 1, 4, 5, 6 and even 7 are present. It is true that the development of the massive structure of the harbour entrance to the south-east required a considerable number of blocks and these must have been supplied from all possible sources. For obvious reasons of balance, the lifting system of each block was symmetrical. We verified this correspondence by examining the intact blocks bearing two similar bosses: – Sondage 3: 3__B3 and 3_D5, two Type 3 bosses; 3_C3, two Type 4 bosses; – Sondage 12: 19_E1 and 19_E2, two Type 3 bosses; – Sondage 14: 14_B1 and 14_D2 to D5, two Type 3 bosses; – Sondage A: block A_F3, two Type 2 bosses; A_G9: two Type 3 bosses. However, we also found several examples bearing different types of bosses on their ends: – Sondage 3: 3_E3 and E5, one Type 3 boss and one Type 4; – Sondage12 South: 19_F1, one Type 3 boss and one Type 4; – Sondage 14: 14_C2, 14_D6 and 7, one Type 3 boss and one Type 4. We can therefore note the compatibility of a boss with a simple channel with another boss with a channel boxed in by side projections. Perhaps it was simply a case of not bothering to tool away the side projections since they did not cause any loss of balance.

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Block no.

Block registration no. Sondage ref. no. / row /order no. E.g.: sondage 13 / row A / block in position 13 => 13_A13

Numberless

Blocks not visible in the elevation, dimensions taken (notes T.K.)

L

Length

W

Width

H

Height or thickness

P

After other data Preserved dimensions

67

Smooth face / without boss 1 Two lateral bosses 2 U-shaped boss 3 Rope channel around a U-shaped boss, leaving lateral bands along the full height of the face 4 5 6 7

Squared off rope channel around a U-shaped boss, creating a U-shaped band on 3 sides (base and sides) Circular lifting boss Vertical protuberance along full height

8

Broken, not visible, or different from 1 to 7

X

Small side of numbered block

XX

Two small sides of the numbered block have the same type of boss

Xfront Xrear

When boss type different at either end

Xl

When length of bloc is visible Xl left-hand end

Xr

When length is visible Xr right-hand end

Same block

Number of the face seen in another sondage, corresponding to the opposite end of the block recorded in the sondage. Same block with both ends recorded according to the sondages in which they are visible. Reference numbers of blocks in bold corresponding respectively to the ends of the same block in each elevation. These ends do not necessarily belong to the same “line”, because the block, although originally set on the same course, has subsided on one side (e.g.) or, it is in a course according to the depth of the sondage (not all sondages have the same depth) The ends of this block: 2_A1 and 12_C3

Weight

Real weight

Weight

Minimum weight (from one or several preserved dimensions but not completed)

Number + number

Number of edges seen in the sondage + number of recognisable edges not counting an edge seen in another sondage * blocks without ref. no.

Synoptic Table of block bosses by sondage — Key for tables

68

Block no.

8_A1 8_A2 8_A3 8_A4 8_B1 8_C1 8_C2 8_C3 8_C4 8_C5 8_C6 8_D1 8_D2 8_D3 8_D4 8_ssN 8_ssN Total 15+2*

THE HELLENISTIC HARBOUR OF AMATHUS

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

1.25P 0.85 0.84 0.74 1.75P 0.77 1.30P 0.75 1.32P 0.43 1.25P 0.65 1.05P 0.45 1.06P 0.40 1.00P 0.50 1.80P 0.65 1.07P 0.75 1.05P 0.50 0.55P 0.35 0.93P 0.45 1.64 0.70 0.86 0.76

H

2

0.63 0.50 0.61 0.59 0.35 0.35 0.44 0.45 0.37 0.42 0.35 0.35 0.30 0.15 0.42 0.43 0.44

3

Remark Same Block Weight

4

5

6

7

8 Other

X X X X X X X

1.740t

broken broken broken X

0

0

X X 10

0

0

0

0

broken broken broken broken * * 7 broken

2.100t 887kg 517kg 930kg 553kg 408kg 546kg 1.060t 730kg 409kg 137kg 457kg 1.000t 678kg

Synoptic Table of block bosses by sondage — Sondage 8, bloc 8_A1 to 8_D4

Block no.

A_A1 A_A2 A_A3 A_A4 A_B1 A_B2 A_B3 A_B4 A_B5 A_B6 A_C1 A_C2 A_C3 A_C4 A_C5

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

0.51P 0.78 0.72 0.63P 0.55P 0.54 0.58 0.60 0.76 0.72P 0.06P 0.58 0.64 0.52 0.72

0.25P 0.32P 0.50 0.53P 0.38 0.32 0.38 0.38 0.58 0.60 0.22P 0.34 0.37 0.44 0.44

2

3

Remark Same Block Weight 4

5

6

7

8 Other broken

X X X X X X X X X NVis X X X X

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Block no.

A_C6 A_C7 A_D1 A_D2 A_E1 A_E2 A_E3 A_E4 A_E5 A_E6 A_E7 A_E8 A_F1 A_F2 A_F3 A_F4 A_F5 A_F6 A_F7 A_F8 A_G1 A_G2 A_G3 A_G4 A_G5 A_G6 A_G7 A_G8 A_G9 A_G10 Total 45+2

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

2,59

2,51 1.76P 1.52P 2,35P 2,22P 2,35P 2,82 2,82 2,96 2,71P

W

H

2

0.68 0.65 0.48P 0.50 0.19P 0.69 0.56 0.64 0.67 0.68 0.60 0.72 0.62 0.70 0.68 0.64 0.67 0.54 0.66 0.62 0.66 0.58 0.69 0.60 0.53 0.72 0.45P 0.72 0.60 0.75

0.46 0.40 0.35 0.33 0.31P 0.42 0.42 0.44 0.56 0.48 0.53 0.41 0.40 0.42 0.40 0.37 smooth 0.43 0.35 0.44 0.40 0.35 0.37 0.38 0.32 0.32 0.26 050 0.45 0.40 0.56 1 1

3

69

Remark Same Block Weight 4

5

6

7

8 Other

X X X X NVis X X X X X X broken X X X

1.830t

X X X X X X X

broken broken

X X X XX 28 +2

6

0

0

1

Synoptic Table 2 of block bosses by sondage — Sondage A, block A_A1 to A_G10

broken 2 NVis 5 broken

1.500t 982kg 1.040t 1.200t 978kg 1.100t 1.700t 2.400t 1.900t 3.000t

70

Block no.

1_A0 1_A1 1_A2 1_A3 1_B1 1_B2 1_C1 1_C2 1_C3 1_C4 1_C5 1_C6 1_C7 1_C8 1_C9 1_D1 1_D2 1_D3 1_D4 1_D5 1_D6

THE HELLENISTIC HARBOUR OF AMATHUS

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

1.24 0.75 0.83 0.66 0.70P 0.55P 0.56P 0.55 0.52 0.76 0.46 0.64 0.83 0.70P

1.02 0.83 0.86 0.66 0.45 0.40 0.37 0.52 0.51 0.42 0.50 0.50 0.97 0.90 0.16 0.39 0.50 0.50 0.43 0.51 0.63

0.60P 0.72 0.70 0.82 0.53 0.60 1.07P

1_E1 0.87 1_E2 0.60 1_E3 0.72 1_E4 1.07 1_F1 0.40 1_F2 0.69 1_F3 1.40 1_F4 0.67 1_F5 0.66 1_F6 0.68 1_F7 0.70 1_F8 0.74 1_F9 0.63 1_F10 0.72 1_F11 1.15P 1_G1 1.22P 0.86 1_G2 0.35 1_G3 0.82 1_G4 0.90 1_G5 0.60 1_G6 0.75 1_G7 0.68 1_G8 0.75 1_G9 0.69

0.20 0.60 0.40 0.65 0.45 0.40 0.52 0.38 0.38 0.40 0.75 0.58 0.50 0.45 0.60 NVis 0.72 0.40 0.45 0.63 0.76 0.50 0.38 0.51

2

3

4

Remark Same Block Weight 5

6

7

8 Other NVis

X X X X X X broken X X X X X X NVis X X X X X Xr broken X X X X X X X X X X or X X X X

X

NVis X X X X X X X X

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Block no.

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

1_G10 1_G11 1_G12 1_G13 1_G14 1_G15 1.70P

0.65 0.64 0.85 0.67 0.50P 1.43P

0.45 0.49 0.46 0.46 0.48 0.30

1_H1 1_H2

0.98

0.80 0.50

0.25 0.70

1_H3

3.55

0.95

0.62

1_I1 1_I2 1_I3 1_I4

1,10P 0.58 0.40 0.87 0.59

0.42 0.42 0.62 0.52

1_I5 1_I6 1_I7 1_I8 1_I9 1_I10 1_I11 1_I12 1_I13 1_J1 1_J2 1_J3 I_J4 I_J5 I_J6 I_J7 I_J8 I_J9 I_J10 I_J11 I_J12

1_K1 1_K2 1_K3

0.93 0.71 0.71 0.74 0.80 0.64 0.85 0.80 1.36P 0.65 0.68 0.86 0.72 0.66 0.65 0.61 0.64 0.61 0.56 1.40P Non Vis 2.38 Non Vis 0.72 0.77 0.60 0.77

2

3

4

71

Remark Same Block Weight 5

6

7

8 Other

X X X X X NVis smooth smooth

3_B1 1.1t 3_C2 892 kg 3_E3 5.440 t 697 kg

X X smooth X X

0.54 0.29 0.29 0.30 0.64 0.50 0.50 0.55 0.40 0.96 smooth 0.355 0.81 0.22

X

0.45 0.45 0.45 0.45 0.43 0.34 0.65

X X X X X X

X X X

square

X X X X NVis X X X

NVis

0.50

NVis Xr

0.85P smooth 0.90P 0.97

boss circular + bands

X X

complete 3_D2 2.090 t

72

Block no.

THE HELLENISTIC HARBOUR OF AMATHUS

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

1_K4

0.95

0.36

1_K5 1_K6 1_K7 1_K8 1_K9 1_K10 3.08 Ss N 2.80

0.72 0.57 0.80 0.55 0.55 0.70 0.55

0.86P 0.68P smooth 0.52 0.33 smooth 0.57 0.60 0.45

1_K11 3,25

0.73

0.45

Total 91

2

3

4

Remark Same Block Weight 5

6

7

8 Other

X

side laying

X X X X X

3.400 t 3_I1 1.81t 3_G1 2.780 t

X 7Lis

2

43+1

18-1

0

11

2

2 broken 7 NVis

Synoptic Table 3 of block bosses by sondage — Sondage 1, Block 1_A1 to 1_K11

Block no.

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

3_A1 3_A2 3_A3 3_A4 3_B1

0.20P 0.20P 0.20P 0.65 0.20P 1.70P 1.43P 0.30

smooth smooth smooth

3_B2 3_B3 3_B4 3_B5 3_B6 3_C1 3_C2

2.25 2.04

smooth

3_C3 3_D1 3_D2

1.75

3_D3 3_D4 3_D5

2.20 1.80 1.57

2.20 2.05

2.38

2.72

2

3

4

Remark Same Block Weight 5

6

7

8 Other

NVis smooth

1_G15 1.896t

0.80 0.65 0.65 0.45 0.50

0.30 0.50 0.52 0.50 0.50 0.62 0.62

0.50 0.82 0.71

0.62 0.72 0.45

X X

0.71 0.63

0.45 0.45 0.45

X X XlXr

XlXr X X X smooth smooth XlXr

1.600t 1_H2 1.660t 1.400t 1_J12 2.090t

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

W

H

1

3_D6 3_E1

3,08

0.65 0.90

0.28 0.45

smooth

3_E2 3_E3

2.10 3,55

0.95

0.45 0.62

3_E4 3_E5

1.70

Block no.

3_E6 3_F1 3_F2 3_F3 3_F4 3_F5 3_F6 3_F7 3_F8 3_G1

2.70

0.64 0. 80 0.75 0.82 0.70 0.65 0.60 0.70 0.60 0.65 0.65

3,25 0.73

3_G2 3_G3 3_H1 3_H2 3_H3 3_H4 Total 43

3,30 3,00 2.35 2.40 2.40 2.80

0.65 0.70 0.62 0.54 0.70 0.71

0.40 0.45/ 0.75 0.55 0.73 0.45 0.45 0.45 0.45 0.40 0.40 0.45 Non Vis 0.45 0.45 0.45 0.45 0.50 0.50 0.45

2

3

4

73

Remark Same Block Weight 5

6

7

8 Other

X 1_K11 3.243t

smooth Xr

Xl

Xr X

Xl

1_H3 5.440t 1.100t

X X X X X X X X X Xl

2.200t

1_K11 2.780t Xr

X X X

2.510t 2.460t 1.710t

smooth

10

0

X X 26 +2

2.330t 1 +1

2

0

Synoptic Table 4 of block bosses by sondage — Sondage 3, block 3_A1 to 3_H4

0

1 NVis

74

Block no.

THE HELLENISTIC HARBOUR OF AMATHUS

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

2

3

2_A1 2_B1 2_B2 2_C1 2_D1 2_D2 2_D3 2_D4 2_E1 2_E2 2_E3 2_E4 2_F1 2_F2 2_F3 2_G1 2_G2 2_G3 2_G4 2_G5 2_H1 2_H2 2_H3 2_H4

0.75 1.03 1.11 0.95 0.64 1.02 0.75 0.52 0.35P 0.73 0.96 0.40 0.93 0.66 0.62 0.75 0.93 0.80 0.67 0.70 0.50 0.72 0.80 1.84P 0.75

0.25P 0.30 0.55 0.28 0.32 0.30 0.40 0.35 0.50 0.22 0.26 0.25 0.37 0.30 0.21 0.25 0.28 0.25 0.29 0.32 0.25 0.23 0.30 0.23

2_H5

1.70P 0.80

0.25

2_H6 2_H7

0.75 1.00P 0.55

0.22 0.33

X

2_H8

0.95P 0.56

0.30

X

2_H9 2_I1

0.55 0.90P 0.75

0.25 0.25

Total 30

4

Remark Same Block Weight 5

6

7

8 Other broken

X

12_C3 12_D3

X X broken X X X

12_E3 12_F1 12_F2 12_F3 12_F4

broken X X X X X X X broken

12_G1 12_G2

X X broken X X X X broken broken

broken broken 0

0

13

8

0

0

Synoptic Table 5 of block bosses by sondage — Sondage 2, block 2_A1 to 2_I1

0

9 broken

12_H1 1.340t 12_H2 1.100t 12_G3 12_H4 498kg 12_H5 340kg 12_H3 749kg

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Block no.

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

2

3

12_A1 12_A2 12_A3 12_B1 12_B2 12_B3 12_B4 12_B5 12_C1 12_C2 12_C3 12_D1 12_D2 12_D3 12_D4 12_E1 12_E2 12_E3 12_E4 12_F1 12_F2 12_F3 12_F4 12_G1 12_G2 12_G3 12_G4 12_G5 12_H1

0.80 0.87 0.62 0.50 0.50 0.80 0.80 0.76 0.73 0.33P 0.87 0.72 0.74 0.75 0.83P 0.63 0.63 0.82 0.75P 0.56 0.66 0.77 0.66 0.64 0.70 0.73 0.65 0.60 1.84P 0.70

0.27P 0.43 0.45 0.53 0.40 0.40 0.45 0.50 0.33 0.36 0.35 0.50 0.50 0.45 0.25 0.63 0.33 0.36 0.22 0.40 0.23 0.38 0.30 0.40 0.38 0.35 0.25 0.43 0.40

12_H2

0.35

12_H3

1.70P 0.65 2frag 0.90P 0.80

12_H4

1.00P 0.58

0.33

X

12_H5

0.95P 0.53

0.26

X

Total 33

4

75

Remark Same Block Weight 5

6

7

8 Other

X X X X X X X X X X X

2_A1

X X X X X X X

2_B1

2_C1

X X X

2_D1 2_D2 2_D3 2_D4 2_G1 2_G2 2_H6

X X X X X X X X

2_H4 1.340t 2_H5 1.100t 2_I1 749kg 2_H7 498kg 2_H8 340kg

X

0.40

X

0

3

28

2

0

0

0

0

Synoptic Table 6 of block bosses by sondage — Sondage 12 eastern face, block 12_A1 to 12_H5

76

Block no.

THE HELLENISTIC HARBOUR OF AMATHUS

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

H

2

3

19_A1 19_A2 19_B1 19_B2 19_B3 19_C1 19_C2 19_C3 19_D1 1.80P

0.75 0.90 0.75 0.68 0.64P 0.53P 0.75 0.65 0.70

0.50 0.50 0.40 0.45 0.51 0.50P 0.38 0.52 0.52

19_D2

0.81

0.50

19_D3 19_D4 19_E1 2.28

0.63 0.63 0.78

0.55 0.55 0.61

X X X X

19_E2

0.72

0.72

X X

1.02

Remark Same Block Weight 4

6

7

8 Other

X X X X X NVis X X X

? prof 14_C7 1.704t ? prof 19_C8

X

19_E3

2.25 2frg 1.95

0.65

Xfront

Xrear

19_E4 19_E5 19_F1

1.80P 0.92P 0.50 0.62 0.55 2.30 0.92 0.43

X X Xfront

Xrear

19_F2

1.75P 0.50

0.47

19_F3

1.75P 0.69

0.40

15

0

Total 20

5

14_D4 2.821t 14_D5 3.033t 14_D6 4.100t 2.153t

smooth front smooth 2

1

0

1

0

14_E1 2.100t broken 14_E2 1.100t 14_E3 1.256t 1 NVis

Synoptic Table 7 of block bosses by sondage — Sondage 12, southern face, block 19_A1 to 19_F3

To avoid any confusion, a series of blocks Sondage 12 A1 to 12_H5 already existing (eastern face), we number this series of blocks Sondage 12_19_A1 to A_J1

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Block no.

14_A1 14_A2 14_A3 14_A4 14_A5 14_A6 14_B1 14_B2 14_B3 14_B4 14_B5 14_B6 14_B7 14_C1 14_C2 14_C3 14_C4 14_C5 14_C6 14_C7

Max. Dimensions LxWxH P = preserved

Edges X X X = two edges front / rear Xl Xr = edges left / right

L

1

W

2.25 1.65 1.12 0.60 0.65 0.43 2.47

2

0.20P 0.22P 0.15P 0.20P smooth 0.20P smooth 0.18P smooth 0.65 0.60 0.57 0.45 0.45 0.46

3

4

Remark Same Block Weight 5

6

7

8 Other NVis NVis NVis

XX X X X X X

0.71 0.67 0.71 0.66 0.80

0.585 0.61 0.66 0.63 0.63 0.66 0.67

14_C8 2.20

0.85

0.63

14_D1 1.81

0.94

0.70

14_D2 2.20 14_D3 2.10 14_D4 2.28

0.97 0.74 0.70

0.65 0.70 0.70

XX XX XX

14_D5 2.25 2frg 14_D6 1.95

0.70

0.70

XX

0.95

0.85

Xrear

Xfront

14_D7 1.90P 0.97 14_E1 2.05 0.92

0.45 0.43

smooth Xr

Xl

14_E2 1.75P 0.50

0.47

14_E3 1.75P 0.62

0.35

smooth rear smooth smooth 5 1

Total 31

1.33 1.63

1.27 0.92 0.50 0.62 0.82 1.03 0.79

H

77

1.600t

X Xl

Xr X X X X X X

irregular block

broken Xfront

13 +4

4

1

2

0

? prof 19_D1 ? prof 19_D2 3.100t 3.100t 3.600t 2.830t 19_E1 2.900t 19_E2 2.900t 19_E3 4,100t 2.200t 19_F1 2.100t 19_F2 1.100t 19_F3 987kg

3 NVis 1 broken 1 irregular

Synoptic Table 8 of block bosses by sondage — Sondage 14, block 14_A1 to 14_E3

Remark regarding construction technique: 14_B1 was laid after 14_B2 because the corner of 14_B2 is broken. The series 14_D2 to 15_D5 was laid first.

78

THE HELLENISTIC HARBOUR OF AMATHUS

Sondage No.

Block 1 Extr. No.

2

A

1

1

28

6

0

0

1

1

45 +2 91

7

2

43

18

0

11

2

2 3

30 40

0 10

0 0

0 0

0 0

15 +2* 33 31

0

0

8 1 +1 0

0 2

8

13 26 +2 8

2 NVis 5 broken 7 NVis 2 broken 9 broken 1 NVis

0

0

0

0 5

3 1

28 13 +4

2 4

0 1

0 2

0 0

20 305 +2 +2*

2 25

1 8

15 174 +6

0 39 +1

0 3

1 14

0 3

12 14

19 Total

3

4

5

6

7

8 Other

Weight L Max Max Preserved ton 2,96

2.400

3,55

5.440

1,11P 3,55

/ 2.510

7 broken

1,80P

>2.10

0 3 NVis 1 irregular 1 broken 1 NVis 12 NVis 24 broken 1 irregular

1,84P 2,47

>1.34 3.600

2,30 //

>2.153 //

Synoptic table of the bosses of all sondages

1_I4 is atypical; however, we have classified it as Type 4: it features a cylindrical boss between two vertical protuberances. 1_I8 is classified as Type 6, although it features a cubic and not a cylindrical boss.

The number of blocks used in the construction of the harbour Aerial photography provides an often precise view of the blocks used in the construction of the moles and one can actually count those that are visible and not covered by posidonia (fig. 1 page 10 and fig. 3). The photogrammetry coverage (see plan 2) helps in the counting of the blocks, especially of the South Mole where they appear quite clearly. Along a length of 180 m, one can make out 255 blocks of the north cladding (fig. 39). Extrapolating from this for the three moles with twin cladding gives us the theoretical figure of 520 blocks. These numbers are valid for the upper course; however, by what coefficient must we multiply them to have the total number, taking into account the different number of courses? We have seen in the description of the sondages that some of them revealed up to eight or even nine courses, one on top of the other, particularly in the south-west corner of the entrance channel, while others showed only three. If we accept this lowest figure, the hypothetical total will be 520 x 3 = 1,560 blocks. We can cross-reference this rough data with the examination of the sondages themselves: in Sondage A, 4 m wide along the north cladding of the South Mole, there were 32 blocks overall, in six courses (see fig. 12). This would give us 800 blocks over a length of 100 m. Another example can be taken from Sondage 1 (see fig. 16): 91 blocks over 10 m making 910 blocks for 100 m, i.e. slightly less than 10% difference. Thus, for the length of the three moles – 145 m for the West Mole, 180 m for the South Mole and 130 m for the East Mole – and counting the twin cladding, we would have a theoretical figure of more than 3,500 blocks. We shall not attempt to calculate the tons of stone used in the implectum, the fill between the twin cladding, but if we imagine 445 m for the total length of the three moles, for a width of 10 m on average, and a depth up to 4 m, then the theoretical total approaches 20,000 m3.

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

79

The upper course We must admit our ignorance as to whether the upper course in antiquity is that which we see today. An argument to the contrary could be supported by the discovery near the harbour entrance of a metal dovetail cramp (fig. 40a-b) lying in sand mixed with pebbles. The object is covered in lead and is quite remarkable in its size and weight (fig. 41a-b): it measures roughly 45 cm by 20 cm and is more than 10 cm thick (see exact dimensions in the drawing), with a weight of 54.3 kg. Dr Vassos Karageorgis has pointed out the presence in Limassol Museum of another similar cramp found by chance in Amathus port before our excavations began. Moreover, it bears a close resemblance to another cramp that we found during our underwater excavation of the ancient harbour of Thasos11, weighing 56 kg for a length of 31 cm. We have noted the presence of fixation cavities on some of the blocks underwater on the South Mole near the entrance channel (fig. 42). Other dovetail mortises are visible on blocks lying on the beach at the beginning (north) of the West Mole (fig. 43). Was the uppermost course fixed in this manner? Given the present state of the harbour, and the fact that the study of ancient Greek ports as a discipline is in its infancy, it is difficult to respond to this question. We can, however, mention an oral tradition that is recounted by those living in the neighbourhood of the harbour: in the 1860s blocks from the moles were loaded onto boats that had come from Port Said to be used in the building of the walls of the Suez Canal. We have consulted the literature and the online archives of the Suez Canal Company but without finding anything relevant to this subject12. However, the following indications are certainly interesting: “Stone from Toura, near Cairo, could be transported to the Gulf of Pelusium by the Damietta branch [of the Nile]. However, the shallow water of the boghaz [straits] and irregular flow rate make this resource unreliable. The stone will thus have to be brought from the Asian coast or from Cyprus or Karpathos, the distance probably being compensated by the advantage of a sure harbour in the proximity of magnificent quarries13.” Another mention comes from the 19th-century geographer Élisée Reclus: “The vestiges of the ancient city and its tombs are rapidly disappearing, the rock of the acropolis being used as a quarry for the buildings and the quays of Port Said: it is from Cyprus that the engineers of the canal import some of their stone” 14. The archives are held in the Archives nationales du monde du travail in Roubaix, France. Of the 1.5 km of archives concerning the Canal, 50 m have been digitised and can be found online through the Bibliotheca Alexandrina website (indexed in Arabic), but without any mention of boats being sent to Cyprus. This question of blocks from the ancient port being salvaged for the construction of the Suez Canal remains open and requires a more thorough enquiry. There is one argument that would support the idea that the uppermost course was not taken for the Suez Canal and indeed that only the terrestrial quarries of Amathus were exploited, and that is its poor state of preservation. Damaged by the erosion of the sea as well as by the fauna and flora (especially the roots of the ever-present posidonia), these blocks would not have been considered suitable for re-use in the building of the canal. Moreover, given the general appearance in which we found the ancient harbour, it would not necessarily have been easy to spot the blocks in the midst of such luxuriant underwater vegetation (fig. 44). 11.

See J-Y. Empereur, A. Simossi, “Le port de Thasos”, BCH 112 (1992), p. 742 and fig.12, p. 741.

12.

For example, the book by C. Piquet, La Compagnie du Canal de Suez: Une concession française en Égypte (18881956), 2008. See also http://suezcanal.bibalex.org/Presentation/Home/home.aspx (site visited 15/3/2017).

13.

“Séance du 31 décembre 1855”, Procès-verbaux des études de la Commission internationale en Égypte, in F. de Lesseps (ed.), Percement de l’isthme de Suez, exposé et documents II, no. 8, 1856, p. 122.

14.

É. Reclus, Nouvelle Géographie universelle, IX: L’Asie antérieure (1884), p. 680, quoted by B. Bousquet, P.-Y. Péchoux, “Géographie du site d’Amathonte”, in P. Aupert, “Rapport sur les travaux de la mission de l’École Française à Amathonte en 1977”, BCH 102 (1978), p. 967, note 1. This work is available on the site Gallica of the Bibliothèque nationale de France.

80

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 40a-b — Sondage 2: dovetail fastening in situ under a pile of stones near the harbour entrance (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

81

max 0.456 m

0.153 m

0.155 m

max 0.192 m

0.396 m

upper side

0.440 m

longitudinal face

lower side

Fig. 41a-b — Lead dovetail fastening (photograph EFA, Ph. Collet, drawing and CAD EFA, T. Koželj)

0.077 m

lateral face, left

0.084 m

0.084 m

0.107 m

upper side

82

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 42 — South Mole, dovetail mortise (photograph EFA, Ph. Collet)

Fig. 43 — Blocks on the beach to the north of the West Mole showing dovetail mortises (photograph EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

83

The modules of the blocks (Tony Koželj, Manuela Wurch- Koželj) We have seen previously that the harbour blocks were cut according to different types that can be classified according to the means of handling and lifting. One can also distinguish different categories of blocks according to recurring sizes (fig. 45). We shall first consider the length of the blocks. They can be classified into two large categories: Length: 1) very large blocks with a length of around 3 m. In the attached drawings, one can see blocks of Sondage 3 measuring 2.80 m, 3 m, and even 3.31 m, this last weighing 4.8 tons; 2) an intermediary module of 2.35 m employed in the same Sondage 3; 3) a smaller module, around 1 m. Width: the variation in width is much less than in length, generally from 50 cm to a maximum of 75 cm, that is a maximum difference of 50%. This is explained by the necessity of stacking the courses of the same width in order to ensure the stability of the ensemble. Thickness: thickness varies to a greater extent. As we have seen in the description of Sondage 215, the two columns of blocks beneath H3 and H5 have a thickness of 25 cm while the neighbouring block B2, at 55 cm, is more than double. Nonetheless, if the differences in thickness do not affect the stability of the columns, one should note that the blocks of the same sondage are generally the same size; see, for example, Sondages A, 1 and 8. Where there is a difference, it is generally in the presence of more voluminous blocks in the lower courses placed directly on the sand, with the smaller sized blocks in the upper courses. One can see, for example, in Sondage 8 how playing with different thicknesses served to correct the unequal heights of the courses. Weight: this, evidently, can vary widely. Certain blocks weigh a third of a ton while others are more than 4 tons (fig. 46a-b). This must be a matter of the limits of different quarries and also the planned position of the block, the biggest being intended for the zones requiring reinforcement to the south-west and south-east.

The base measurement We could most probably refine our classification by searching for the base measurement that was used when the blocks were commissioned, while taking into account the realities imposed by the nature and the location of the quarries, as they are presented in the following chapter. It seems that the reference measurement employed was the foot of 29 to 30 cm, though the state of the blocks after erosion in a marine environment does not allow us to be any more precise. We can make the following claim: the attached tables demonstrate that certain blocks measure, on their small sides, 1 foot, and several others measure 1.5 feet (see fig. 46a-b). All of the blocks fit into this standard of 1 foot, half a foot and quarter of a foot, and this is most probably not by chance. We need to extend this enquiry and compare the harbour blocks with those of the town of Amathus, particularly of the Hellenistic city wall. There is no doubt that the forthcoming publication of this wall will provide precious information for our understanding of the base measurement of the blocks used in the construction of the Hellenistic harbour of Amathus.

Headers or slabs? While we are using the generic term of block for the cladding of the moles, technically-speaking we should use slab: a piece of cut stone less thick than wide for paving a surface or a pathway16, as we 15.

See supra, p. 53 with fig. 26.

16.

See Ginouvès (1992), p. 146.

84

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 44 — South Mole, the upper course before removal of the posidonia (photograph EFA, Ph. Collet)

Fig. 45 — Variety of blocks discovered during the excavation of the harbour (drawing and CAD EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

85

4

70

1 half - feet 4

300

21

1 half - feet 2

70

1 half - feet 2

300

4

sondage 3

19

1 half - feet 2

71

sondage 3

4

1 half - feet 2

280

18

1 half - feet 4

62

4 half - feet

235

sondage 3

15

1 half - feet 4

Fig. 46a — Large size blocks, 3 to 4 tons (drawing and CAD EFA, T. Koželj)

86

THE HELLENISTIC HARBOUR OF AMATHUS

0.76

0.45

0.44

1 foot 1/2

1 foot 1/2

0.50+

0.65

2 feet 1/2 Sondage 13, block 1_G10

2 feet 1/8

2 feet 1/2

0.68 t

0.78

0.66

2 feet 3/8

0.49

1 foot 5/8

1 foot

0.50+

0.64

Sondage 13, block 1_G11

2 feet 1/8

0.50+

0.69

Sondage 13, block 1_J2

2 feet 1/4

Block alone, from the west mole near the sondage 8

1.05

0.79

0.585

1 foot 7/8

0.44

0.51

1 foot 5/8

0.76

1.5 t 1.33

2 feet 5/8

4 feet 3/8

Sondage 14, block 14_C1

Fig. 46b — Small size blocks, less than 1.5 tons (drawing and CAD EFA, T. Koželj)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

87

have seen in the examination of Sondage 2. Sometimes, however, the thickness of the blocks is close to the width, and thus to the technical description of throughstone17.

Placing the blocks: description of the machinery The work of placing the blocks of the mole cladding might in part be explained by the traces that the blocks themselves bear, and we can try to reconstruct the machinery used in the operation.

Traces on the blocks: the machinery First, we shall examine the traces of tools used for placing the blocks. As the illustrations show, several blocks bear traces of placement by means of wedges and crowbars (fig. 47a-b), which allowed for precise positioning. In certain cases, these notches might also be traces of clamp holes for lifting and the correct setting of the blocks (fig. 48-49). We might try to imagine the look of the machinery involved in placing the blocks and carrying them from the nearby quarries. This lifting and positioning machine gradually advanced as the moles were constructed. It must have been as wide as the interior cladding stones of the mole, that is, little more than 2 m, and was equipped with an articulated jib operated by block and tackle, following methods developed by Greek engineers and which can be found in writings about siege engines and the writings of Hero of Alexandria, the 1st-century ad inheritor of the Hellenistic engineering tradition. The machinery advanced on rollers as the mole it was building was extended. The reconstruction of the machine envisages that the blocks to be positioned in a new section of the mole would be carried upon the bed of the machine. These blocks would serve as counterweights. For a block to be placed by the jib, there would need to be a counterweight proportional to the distance of the forward positioning of the block: one block at least as counterweight on the rear bed of the crane for another block placed immediately in front of the machine, and four or even five blocks for another block set at the farthest extension of the jib, in order to avoid any risk of the machinery toppling forwards. We may also reconstitute the hypothetical length of the jib by observing the sondages. Sondage 1 is particularly informative, as can be seen in fig. 50. The three breaks, marked by blue dashes, are clear. They delimit sections of 5 m in length, which indicate the minimum size of the moveable jib of the lifting and positioning machinery (fig. 51a-e). Such machinery is known from ancient texts, with equipment for reducing force exerted by way of pulleys and blocks. These lifting techniques appear to be well mastered and need only be seen within the technological context of the Hellenistic period. We might think of the Pharaonic obelisks of the 18th century bc that were transported, lifted, then lowered, re-transported and re-erected, such as the obelisk of Seti I placed in Alexandria by Ptolemy II in front of the funerary temple of his sister-wife Arsinoe II shortly after her death in 260 bc, and then later carried off to Rome. Such monoliths weighed more than 750 tons, which makes the blocks of Amathus harbour seem relatively light. Hero of Alexander, the 1st century bc engineer who inherited the inventions and inventiveness of the Hellenistic era, has left us a number of manuscripts, with sketches describing such machines of force reduction. When considering the work undertaken for the construction of Amathus harbour, we can be impressed by the extent of the project and the obvious technical skill, but we know that such know-how had long been mastered.

17.

See Ginouvès (1992), p. 105.

88

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 47a-b — South Mole: traces of the use of wedges for positioning a block (photograph EFA, Ph. Collet)

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

89

Fig. 48 — South Mole: traces of the use of a crowbar for positioning a block (photograph EFA, Ph. Collet)

Fig. 49 — South Mole: a series of blocks on the upper course with the same traces of positioning using a crowbar (photograph EFA, Ph. Collet)

G2

E1

--3.43 3.43

F1

G3

-3.53 3.53

E3

F2

G4

I1

-3. 3.17 -3.17 -3.1 1

B1

E2

-2.70

G5

I2

E4

B2

G6

I3

-3.23

A0

C1

I4

J1

-2.50

G7

C2

F3 D1

G8

I5

J2

-2.75

A1

C3

F4

G9

I6

J3

D2

-2.35

A2

C4

D3

F5

A3

C5

I7

J4

G10

K1

-2.14

C6

D4

F6

G11

J5

-1.85

I8

F7

-5.96

J6

-1.83

D5

K2

South Mole - East Face

I9

K3

C7

J7

-1.82

I10 F8

K4

C8

J8

K5

-1.85

F9

G12

I11

K6

-1.80

0

F10

J9

J0

0.5

G13

I2

K7

-2.10

1

G14

-2.41

Fig. 50 — South Mole, Sondage 1: sketch showing the placing of a section of blocks (drawing and CAD EFA, T. Koželj)

G1

-2.78

South Mole

Sea level at time of excavation

H1

2

I13

J11

K8

K9

-1.93

3

H2

4m

-5.05

F11(A1)

G15(B1)

H3(C3)

K11(E1)

-1.98

J12(D2)

K10(I1)

-1.92

C9

D6

-5.91

-4.90

-4.35

-3.80

-3.52

-2.89

-2.44

Harbour Entrance

90 THE HELLENISTIC HARBOUR OF AMATHUS

THE ARCHITECTURE OF THE HELLENISTIC HARBOUR

Fig. 51a-e — Reconstruction of the machinery used for placing the blocks (CAD EFA, T. Koželj)

91

Chapter 2

The stone quarries of Amathus harbour Tony Koželj, Manuela Wurch-Koželj

The foundations of the harbour moles1 are made of a twin cladding of blocks set as headers enveloping an implectum (a fill of irregular blocks and masonry refuse)2. The construction technique of this cladding is original in that it employed specialised lifting machinery3 specially designed for this task to position blocks one above the other, thus creating two parallel linear braced structures for each mole. The appearance of these blocks is quite particular: they have an elongated form, with lifting bosses4 to assist in handling and positioning The blocks seem to have come from quarries in the chora of Amathus5 (plan 1 and fig. 1), certain of which, judging from the quality of the rock (compact limestone), could have allowed for the extraction of large-sized blocks. Five quarries can be considered as supplying the stone used in the construction of the harbour moles. Nevertheless, we lack the relevant data to confirm that they produced these specific blocks. In fact, the quarry fronts have changed, some through the effects of erosion, others because of continuing quarrying activity, and yet others through new land developments. The vertical quarry front of [168]6 has been partly transformed by erosion into a steep cliff where some vestiges still appear here and there. Its eastern sector has been destroyed by road works and housing development. The quarrying sectors to the south-west of the town, [170], have experienced several phases of exploitation connected to the construction and repairs of the rampart walls of the Western Harbour, thus modifying the fronts of these quarries. Early Christian exploitation reduced the quarry front of [167] by extracting blocks for the building of the basilica of Agios Tychonas. Only a few traces in [169] could be significant evidence of exploitation related to the port.

1.

We would like to express our warmest thanks to Catherine Petit-Aupert and Jean-Yves Empereur for entrusting us with the study of these quarries.

2.

See above: p. 33-35, Sondage 8.

3.

See above: p. 91.

4.

For different types of lifting bosses see above, p. 62, fig. 31.

5.

See T. Koželj, M. Wurch-Koželj, “Les carrières de la Chôra d’Amathonte”, in C. Petit-Aupert (ed.), Prospection de la Chôra d’Amathonte, in preparation.

6.

During prospection, inventoried sites were numbered and we have retained that system. For simplicity we will use [168] for Site 168; likewise for all sites. Cf. Fr. Briois, C. Petit-Aupert, in collaboration with P.-Y. Péchoux, “Histoire des campagnes d’Amathonte. I. L’occupation du sol au Néolithique”, Études Chypriotes 16 (2005), p. 21.

96

THE HELLENISTIC HARBOUR OF AMATHUS

167- quarry 153- kiln 116- sanctuary 154- lime kiln 115- necropolis 155- quarry 114- cistern and reservoir 113- quarry

y

hwa

Hig

Agios Tychonas

VIKLES

North Rampart

168- quarry th country pa

112- necropolis

Agios Tychonas

AMATHUS Acropolis

Pentakomo

Agora

ANEMOS 170-X

160lime kilns 160-2 160-3

23 m

16 m

25 m

170- quarries

170-B

170-A 13mm 13

13 m

West entrance

Harbour

path

169-d

88 m m

m 88m

West Rampart

Germasogeia

n way pedestria l shore actua

169-a 169-b

169-c

N

010

50

100 m

169- quarries

Fig. 1 — General plan of the quarries of Amathus, scale 1:10,000 (topographic mapping of the sites and CAD EFA, T. Koželj, M. Wurch-Koželj)

The only working face that shows undeniable traces corresponding to negatives of blocks extracted in the same format as the blocks used in the harbour construction, and with similar wedge holes7 indicating the use of the same extraction technique, are to be found in the quarry of Vikles [113]8. This is certainly one of the quarries that produced blocks for the port of Amathus.

THE QUARRY OF AMATHUS [169] The southern edges of Amathus town are now beneath the sea and are composed of rocky outcrops with traces of the site of the southern rampart and elements linked to its construction, as well as vestiges of all sorts of work, levelling, extraction of blocks etc. The effect of the sea has contributed to the degradation of these vestiges, as have tourist developments, notably the construction of a pedestrian walkway from Amathus to Limassol.

7.

Wedge hole: cavity to receive a wedge (struck by a mallet). The force of the strike will cause the split along the line of the cavity-hole and will detach the block from the bedrock. See: T. Koželj, M. Wurch-Koželj, “Les traces d’extraction à Thasos de l’Antiquité à nos jours”, Thasos, Matières premières et technologie de la préhistoire à nos jours, Actes du Colloque International, Liménaria 26-29.9.1995 (1999), p. 49-55; and “ƌ
uƩƯơƷƫ
ƷƼư
ƯƥƷƲuƩƣƼư
uƥƴuƠƴƲƸ
ƷƫƵ
ƍƠƶƲƸ
ƷƭƵ
ƷƩƯƩƸƷƥƣƩƵ
ƨƩƮƥƩƷƣƩƵ”, AEMTh 20 (2006) [2009], p. 155-167.

8.

Site 113 is mentioned in C. Petit, P.-Y. Péchoux, “La campagne de prospection”, in Travaux de l’École Française à Amathonte en 1991, BCH 116 (1992), p. 784-791, in particular the map p. 785 and table p. 790 (excavations of the Department of Antiquities of Cyprus).

THE STONE QUARRIES OF AMATHUS HARBOUR

97

Signs of preparations for large-sized block extraction and the presence of a workshop for tool maintenance (gourna-reservoir, striking area, hearth)9 are sufficient clues to identify quarrying activities. During prospection this sector was registered as [169] in the catalogue of vestiges of Amathus’ chora, partly presented here, in relation with the port.

Fig. 2 — Remains of the quarry 169a (photograph EFA, T. Koželj)

0.42

0.83

0.18

0.36

0.14

N

0.50 0.20 Depth

0.65

0.70

± 2.00

Fig. 3 — Drawing of the tool maintenance workshop 169a (drawing and CAD EFA, T. Koželj)

9.

Gourna: basin in a quarry, used for the repair of tools. A typology of basins has been proposed: there are fixed basins (carved into the rock) and movable basins (hollowed blocks) for specific uses: basin-hearth (for heating tools) and basin-reservoir (for steeping, i.e. cooling metal). See T. Koželj, M. Wurch-Koželj, “ƘƲ
uƠƴuƥƴƲ
ƷƫƵ
ƍƠƶƲƸ
ƮƥƷƥ
Ʒƫ
ƨƭƠƴƮƩƭƥ
ƷƫƵ
ƥƴƺƥƭƿƷƫƷƥƵ”, ƌ
ƍƠƶƲƵ
ƶƷƫư
ƘƴƲƺƣƥ
ƷƼư
ƆƭǁưƼư (2014), p. 154-168, p. 164, fig. 123a-b; ˦Ɣƭ
ƧƲǀƴưƩƵ
ƶƷƥ
ƯƥƷƲuƩƣƥ
ƷƫƵ
ƍƠƶƲƸ”, Actes du Colloque Thasiaka, Thasos, 17 to 19 September 2011, forthcoming.

98

THE HELLENISTIC HARBOUR OF AMATHUS

The quarry front stretches over 400 m from west to east, of which the section out of water currently reaches a maximum of around 20 m in width. At several points, the outlines of a block are visible, as well as signs of the beginning of channelling (eroded by the sea), evidence of preparatory work for extraction, or Phase 1. Other traces, such as wedge holes, indicate earlier extractions. Sector [169 a]: Circular cavity 0.50 m in diameter and 0.20 m deep. Three blocks set in pi formation, 0.65/0.45 x 0.83 x 0.30 m high. Interior fill composed of small pebbles and several hard stones of large size. Next to the northern block is a more irregular depression, roughly 0.70 m in diameter (fig. 2-3). Below, traces of lines of wedge holes are visible (dim. 0.14 and 0.18 x 0.08 m, at intervals of 0.36 m).

The circular cavity appears to be a gourna-reservoir and this alone is evidence of tool repair activity. We can recognise the pi-shaped structure as a striking area: the hard stones would have served as an “anvil” and the damaged depression would have held a hearth for heating tools. These three elements together—a gourna-hearth, a “striking table” and a gourna-reservoir—indicate intense tool repair activity. A reconstruction of this maintenance installation can be seen in the drawing by T. Koželj (fig. 4). The position of this installation was well chosen, at the western edge of the extraction zone. It may have functioned over a long period.

Fig. 4 — Reconstruction of the quarry maintenance installation 169a (drawing and CAD EFA, T. Koželj)

Sector [169 b]: Several cuts, maximum 2.50 m long, mark the intervals of 0.10 and 0.12 m corresponding to the channel widths defining bands of 0.30 and 0.60 m, that are the widths of the blocks.

Wedge holes of different shapes, rectangular 0.12 m, others trapezoidal 0.08/0.12 m. An irregular edge on the quarry front with a rectangular hollow 0.20 x 0.35 x 0.10 m deep. Parallel with this hollow at 0.15 m, a rectangular cavity 0.08/0.10 x 0.12 x 0.10 m deep. At 3.50 m from the first hollow, and set perpendicular, another hollow, dim. 0.20 x 0.37 m preserved x 0.10 m deep (fig. 5-6).

These vestiges of wedge holes are evidence of the line of extraction in this sector, from south to north. The cuts are the first preparatory steps towards extraction, defining the width of the channelling to be carved and the width of the blocks. In this exact case, the blocks on order measured 0.60 x 2.50 m, the thickness not yet determined at this stage. As for the hollowing, this could be channelling in the process of being carved, but the inner surfaces are very well finished and besides, the presence of the small cavity, as well as the edge on the workface (almost running along it) and the perpendicular carving, would all suggest the position of a lifting machine of tripod type. It would most likely have been used to load blocks onto carts 10 (fig. 7). A reconstruction of this type of tripod has been proposed (fig. 8).

10.

One should remark the coherence between the direction of extraction and the position of the cart(s) as suggested by the traces of the tripod emplacement.

THE STONE QUARRIES OF AMATHUS HARBOUR

99

Fig. 5 — Remains of the quarry 169b (photograph EFA, T. Koželj)

Fig. 7 — Reconstruction of the tripod site with quarry cart 169b (drawing and CAD EFA, T. Koželj)

Fig. 6. — Drawing of the remains of the quarry 169b (drawing and CAD EFA, T. Koželj, M. Wurch-Koželj)

Fig. 8 — Reconstruction of the quarry transport system 169b (drawing and CAD EFA, T. Koželj)

100

THE HELLENISTIC HARBOUR OF AMATHUS

0.95

0.70

0.65

2.50

N SEA

0

Fig. 9 — Remains of the quarry 169c (photograph EFA, T. Koželj)

0,5

1

1,5

¬

¬P

Fig. 10 — Drawing of the remains of the quarry 169c (drawing and CAD EFA, T. Koželj)

Sector [169 c]: Several longitudinal bands, from 0.65 to 0.95 m wide, are separated by channelling of varied width, from a groove, to the beginnings of a channel, to a V-shaped channel in the process of being carved some 0.15 m wide (max. depth measured 0.15 m), all somewhat worn by the sea (fig. 9-10). The length of the future blocks, roughly 2.50 m.

This quarry front bears preparatory traces of blocks of similar sizes to those extracted and employed in the harbour mole structures, as well as the effects of the beginning of extraction work. It is noteworthy that the work was not followed up. Sector [169 d]: The workface ranges over several levels. Numerous traces show the negatives of blocks, both small and large, channelling and some few badly damaged wedge holes. The dimensions of the negatives vary from 0.50 x 0.60 x 0.30 m thick (bearing in mind that the level of the surrounding rock corresponds to the upper level of the extracted block) to 0.60 x 1.00 m (fig. 11). Several blocks were abandoned during extraction (dim. 0.60 x 0.70 m, 0.60 x 1.00 m): the peripheral channelling was carved (from 0.30 to 0.40 m deep), but the blocks were not detached from the bedrock (fig. 12). The smallest undetached block measures 0.25 x 0.40 x 0.15 m thick (fig. 13). Near the western mole are bands of more compact limestone (finger quarry type), see fig. 14, from where small blocks were extracted (dim. 0.30 x 0.35/0.40 x 0.30/0.35 thick) using a vertical wedge as can be seen from the wedge holes openings (0.14 x 0.035/0.04 m) (fig. 15a-b). The workface was disturbed by the digging of a channel and the installation of a construction, some remains of which are still in situ (see fig. 11).

This small form of block was used essentially in “masonry”. Nonetheless, such blocks were used in the construction of the interior cladding of the moles (fig. 16a-b). The sectors surveyed were those where the traces of exploitation were best preserved. Extraction will have been as a function of the quality and veins of the limestone along the rock face11 that had been turned into a quarry [169]. The presence of a tool maintenance workshop indicates that quarrying activity was intense. The tripod emplacement is not surprising and other hoists would have been indispensable for shifting and transporting the blocks from the workface to the specialised lifting machinery used to position the blocks on the harbour moles.

11.

The quarry front includes the currently visible edge and also all the submerged part (sea level having fluctuated greatly since construction of the harbour), where the traces have been smoothed by the actions of the sea.

THE STONE QUARRIES OF AMATHUS HARBOUR

101

Fig. 11 — Remains of the quarry 169d (photograph EFA, T. Koželj)

Fig. 12 — Blocks abandoned during extraction in sector 169d (photograph EFA, T. Koželj)

Fig. 13 — Blocks abandoned during extraction in sector 169d (photograph EFA, T. Koželj)

Fig. 14 — Reconstruction of the extraction method using only one wedge in the working face of a finger quarry (drawing and CAD EFA, T. Koželj)

Fig. 15a-b — Sector of small format blocks extracted using a wedge (photograph EFA, T. Koželj)

102

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 16a-b — Small format blocks of the interior cladding of the Eastern mole (photograph EFA, T. Koželj)

THE QUARRY OF VIKLES [113] The Cypriot Archaeological Service conducted a number of sondages and clearance work on the slopes of Vikles hill. One area was immediately identified as a quarry, and indeed as one of the quarries that produced blocks for the moles of the ancient port (fig. 17). This sector was included in the catalogue of remains [Site 113] of Amathus’ chora, a study sector of the prospection team (fig. 18), under the direction of Catherine Petit-Aupert. Here, however, it is presented in relation to the reconstruction of the harbour. The surface area of this rocky outcrop that was cleared covered 65 m2 and revealed numerous negatives of elongated blocks, narrow channelling and wedge holes (fig. 18-19). This, however, was only a “sample” of the quarry, the workings of which will have extended as a function of the quality and veins of the rock. The exploited material was a grey compact limestone of relatively fine grain. The limestone veins of this zone are oriented from 9 to 11° north. Precise surveying of this zone led to a recognition of how extraction work progressed and of the formats of blocks employed in the construction of Amathus’ harbour mole (fig. 20-21). The western sector of the cleared zone is a levelled off rock outcrop with few apparent traces of extraction. Space A: An isolated wedge hole 0.18 x 0.12 m deep. A line indicating a difference of level. Fissure running west to east, extending through a part of the area. 1: Presence of two wedge holes to the north (contributing to the extraction of 1), bottom of a channel carved along the entire length of 2 and 1, and of a fissure to the south, and of a “line” allowing an estimation of the maximum dimensions of the block, 0.95 x 0.70 m. 2 Negative: To the south, on the small side of the block 0.70/0.72 m, two wedge holes, 0.17 x 0.10 m deep at an interval of 0.06 m, which contributed to the extraction of the block; minimum length, 1.70 m.

The northern sector displays signs of methodical extraction from west to east of elongated blocks, using wedges and narrow channelling, and exploiting the natural fissures.

THE STONE QUARRIES OF AMATHUS HARBOUR

68 m

103

80 m

76 m

168 - quarry 153 - kiln dr oa d as ph alt e

116 - sanctuary

72 m

pottery

chemin rural

60 m

154 - lime kiln

68 m

115 - necropolis

155 - quarry 64 m

56 m

52 m

VIKLES small kouros

ath untry p

114 - cistern and reservoir

oned co

aband

new country track

60 m

56 m

asphalted road

52 m

113 quarry

48 m

44 m 44 m 40 m

asphalted ro

ad

40 m

36 m

112 - necropolis (excavation of the Department of Antiquities of Cyprus)

36 m 32 m

Amathus 28 m 24 m

28 m

24 m

20 m

N

16 m

20 m 12 m

0 5 10

20

30 40

50 m

Fig. 17 — Remains on Vikles hill. Scale 1:200 (topographic mapping and CAD EFA, T. Koželj, M. Wurch-Koželj) 3 Negative: Dim. 2.80 x 0.70/0.90 m, bordered to the north by a thin channel 0.08 m wide and to the south by a natural fissure. To the west, the channel is the remains of the extraction of the western block. It facilitated the positioning of the wedge holes on the western length: five wedge holes of 0.17/0.22 x 0.12 m deep. 4 Negative: Dim. 2.05 x 0.75/0.90 m, bordered to the north and to the east by a thin channel. To the northeast, a mark 0.35 x 0.40 m with two wedge holes. The extracted block would be rectangular with an extended protrusion running the height of the block. 4b Negative: Dim. 0.80 x 0.70 m. A channel to the north visible along 0.50 m. One single wedge hole was cut 0.12/0.14 x 0.12 m deep. Extraction was done from south to north, whereas all the others of the northern sector are from west to east. 5 Negative: Dim. 2.92 x 0.45/0.55 m, bordered to the north and to the south by a thin channel 0.08 m wide. A fissure determined the depth of the extracted block. To the west, four wedge holes of 0.20 x 0.12 m deep, at intervals of 0.25 and 0.40 m.

104

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 18 — Overall view of the cleared sector of the quarry in 1991 (photograph EFA, T. Koželj)

unexcavated area

N

0

0.5

1

Fig. 19 — Plan of Vikles quarry in 1991 (drawing and CAD EFA, T. Koželj)

1.5

2 

3 m

THE STONE QUARRIES OF AMATHUS HARBOUR

105

6: Dim. 3.00 x 0.45/0.70 m (curved rear face), bordered to the north by a thin channel 0.08 m wide and to the south by a fissure. On the west side, four wedge holes 0.20 x 0.12 m deep, at intervals of 0.30 and 0.40 m. 7: Along a curved line, four wedge holes, one eroded, of 0.18/20 x 0.14 m deep, at intervals of 0.25 and 0.55 m. In this latter interval, a slight levelling (0.30 x 0.12 m) joins the thin channel situated to the east. It measures 0.08 m wide and 2.80 m long and extends a fissure 0.80 m long.

In fact, Traces 6 and 7 correspond to the original dimensions (3.10 x 0.90/1.08 m) of a block that was incorrectly extracted. The fracture surface was not right on the channel, hence the curved “line”, even though it joined the fissure (see the notch). The remaining limestone was removed using wedges to obtain a “vertically” cut surface. 8 Negative: Dim. 3.05/2.90 x 0.65/0.70 m, bordered to the north and to the south by a thin channel 0.08 m wide. A fissure determined the depth of the extracted block. To the west, three wedge holes of 0.16/0.23 x 0.12/0.20 m deep, Fig. 20 — Blocks 8, 9 and 10 from Vikles quarry with at intervals of 0.32 m. 9 Negative: Dim. 2.85/2.75 x 0.60/0.65 m, details of wedge holes (photograph EFA, T. Koželj) ringed by a thin channel of 0.10 (north) and 0.08 m (east, south) wide. To the west, three wedge holes of 0.17/0.20 x 0.14 m deep, at intervals of 0.75 and 0.72 m, were enough to extract this block. 10 Negative: Partially visible negative of a block. Length 2.70 m, bordered to the north by a channel cut along some 10 cm, 0.70 m wide, to the right of a fissure to the south. To the east, a slanting fissure indicates a shrinking of width (reconstruction: 0.55 m). Three wedge holes, 0.12/0.18 x 0.12 m deep, at intervals of 0.80 and 0.55 m. 11 Negative: Only the southern part is visible, length 1.00 m. The block is bounded by slanting fissures, indicating a flaring from south to north of 0.58 to 0.90 m (reconstructed width). One wedge hole, 0.14/0.18 x 0.14 m deep and a mark 0.20 x 0.14 m, apparently broadening seems to have contributed to extraction.

Fig. 20 shows the state of the vestiges in 1991 (Negatives 8, 9, 10 and details of the wedge holes). The central sector is composed of two parts: the west, where extraction was also carried out from west to east, but on two fronts; and the east, where extraction was not standardised, both in regard to the block dimensions and the various orientations of extraction. The many fissures in the rock are most probably the reason behind the dimensions of the extracted blocks, but the orientation of extractions does not seem, at first sight, to follow any apparent logic. There is a progression from north to south and towards the east, and another, coming from south to north and towards the west. There is also a negative with two wedge holes on its opposite faces, north and south, evidence of an unusual extraction regarding the dimensions of this block.

106

THE HELLENISTIC HARBOUR OF AMATHUS

39 40

unexcavated area legend: extracted block

N

reconstituted block crack excavated area side of the extraction 0 _ = 11°

0.5

1

1.5

2 

3 m

_ = 9°

Fig. 21 — Plan of Vikles quarry showing operational features (blocks, extraction orientation), (drawing and CAD EFA, T. Koželj, operational features, M. Wurch-Koželj)

We could suppose that this was a sector where two teams of quarrymen met, but it is difficult to prove, given the 12 m2 of this part. Or perhaps it was the site of preparatory work for the extraction of a large format block (see 16ab, 30 and 42, dim. 4.10 x 1.10 m) 12 Negative: Very eroded traces of a wedge hole and of a channel, indicating the orientation of extraction from west to east, and giving the width of the block as 0.60 m. The length runs between the south channel and the fissure that crosses the rock outcrop: 1.55/1.65 m. 13 Negative: Only a slight difference in height allows the delineation of the block’s width: 0.65 m. Length 1.55/1.65 m defined by the channel to the south and the fissure to the north. 14 Negative: As for No. 13, the length of the block is defined: 1.65/1.50m. Pick marks on the surface might indicate the bottom of a channel and the width of the extracted block, 0.50/0.65 m. 15 Negative: A thin channel borders the negative to the east. Reconstructed width taking No. 14 into account would be 0.55/0.60 m. The channel cut along some 10 cm to the south indicates a length of 1.40 m for the extracted block; however, it was interrupted by some sort of snapping forming a curved line.

Many micro-fissures cross the following negatives, making it impossible to determine the length of the block or blocks extracted (see 16 and 30). The direction of extraction of 26 to 29 is given by the trace of a wedge hole on the west side of 29. This second extraction face is criss-crossed by many fissures, and the traces are badly damaged.

THE STONE QUARRIES OF AMATHUS HARBOUR

107

26: Despite the narrow channel (of 12), a fissure 1.40 m to the south, and, to the east, a fissure and traces of two channels (signs of a resumption of work), this place cannot really be read as the negative of a block, all the more so because to the west there are no traces. The surface is badly eroded. In theory, a misshapen block of 1.40 x 0.40 m could have been extracted, or rough blocks from this volume of stone. 27 Negative: Dim. 1.45 x 0.45/0.65 m, bordered by a thin channel of 0.10 (north of 13 and 14) and the start of a channel 0.50 m long and 0.07 m wide, extended by a fissure, and a fissure to the south. The surface of the negative is very cracked and it is possible that extraction was not successful. 28 Negative: Dim. 1.35 x 0.20/0.25 m, bordered by a thin channel of 0.10 (north, channel of 14) and a channel 0.08 m wide, and the fissure to the south. 29 Negative: Dim. 1.30/1.25 x 0.55/0.65 m, bordered by a fissure to the east (the fissure that extends along the limit of 5/6, confirming the angle of the limestone vein); to the south, the fissure is widened by the carving of channelling (0.60 m long and 0.08 m). To the west, one wedge hole is visible, 0.17/0.20 x 0.06 m deep.

These two faces must have been cleared to extract 16. 16 Negative: Delineated by two almost parallel fissures (one of which widened by channelling to extract Block 15), crossed by slanting south-west/north-east fissures, this space corresponds to several negatives since extraction did not happen as planned. It appears that without the two fissures, it would have been possible to extract a very large block of 4.20 x 1.10 m (16ab+30+42). However, the transverse fissures prevented this and the fracture surface stopped at the first fissure, forming a little hollow (16a). The surface between this fissure, in which a wedge hole has been cut, and the wedge hole on the southern edge, seems homogenous and one could identify a block, dim. 3.10/3.35 x 1,10 m (16b+30+42), unless the second transverse fissure affected extraction and reduced the block to 2.90/2.80 x 1.10 m (i.e. 30+42). The missing volumes were certainly used: 0.70/0.90 x 1.10 (16a) and 0.40/0.35 x 1.10 m (16b). It seems logical that to be able to extract correctly the block (16a+16b+30+42), its sides had to be previously quarried.

The traces of wedge holes indicate the orientation of extraction of each block. It appears that work on this part followed all sorts of directions, from north to south, but also from west to east, and in the southern section from south to north and from east to west. The northern sector (Negatives 3 to 11) having been cut, extraction continued with 18 and 19, then on either side, 17 and 20, 21, then 22. 18 Negative: The block was extracted from the north, according to two wedge hole traces, one cut from the channelling of 9, 0.18/0.25 x 0.10 m deep and to 0.40 m, the other: 0.16 x 0.08 m preserved. Edged by traces of wedge holes used in the extractions of other blocks, 17, 19 and 20. Negative dimensions: 1.65 x 0.55/0.60 m. 19 Negative: Two wedge holes to the north, 0.17/0.19 x 0.08 m deep. Bases of preserved channels as well as a fissure, and the slanting channel mark towards the front face gives the dimensions: 1.00 m long, 0.70/0.90 m wide. 17 Negative: Dim. 1.20/1.15 x 0.48/0.70 m, bordered to the south by thin channel, 0.10 m, and fissure to the west. Trace of a wedge hole, 0.17 x 0.06 m, can be seen. 20 Negative: Extraction was from the west: two wedge holes 0.14/0.18 x 0.10 m, 0.18 m apart. Dim. 0.65/0.50 x 0.60/0.80 m, bordered to the north by a fissure (see 11); to the east, a slight difference in level and a wedge hole to extract 22, a fissure and the end of channelling 0.10 m wide. 21 Negative: Dim. 0.85/0.80 x 0.70/0.75 m. To the west, a wedge hole 0.14/0.18 x 0.10 m deep. To the east, channelling bands, 0.10 to 0.14 m wide. Length is estimated despite the lack of traces to the south, because of the channelling bands, and the position of the wedge hole, which would almost be centred, 0.32 and 0.30 m. 22 Negative: Only the southern part of the negative has been cut out leaving a wedge hole of 0.18 x 0.12 deep. Two other wedge holes were used in the extraction of blocks to the east (24) and south (23), giving us the width of the block as 0.60 m, and minimal length of 1.30 m. If we imagine that the north channel bands, bordering 8, 9 and 10, continued, the reconstructed length would be 3.10 m. This block could only have been extracted after removal of 18 and 20.

108

THE HELLENISTIC HARBOUR OF AMATHUS

Blocks 24 and 25 could only have been extracted after 23. The negative of 23 shows two wedge holes to north and to south, proving that the two spaces had been cut into (38, 37 and 36 and Sector B). 24 Negative: A wedge hole indicates the direction of extraction of this block of at least 1.50 m (to the north it has not been excavated) and a thin channel gives a width of 0.35 m. 25: A preserved surface of 0.45 x 0.50 m, indicating that exploitation stretched eastwards. 40: A surface of 0.40 x 0.20 m. 39 Block: Corner extremity of a block, 0.40 x 0.50 m visible, with levelled upper surface. 38, Sector B and 45: The damaged state of the rock outcrop (fissures, hollows from erosion etc.) has destroyed all traces. 37: Dim. 0.50/0.30 x 0.80 m. A wide wedge hole indicates the direction of extraction. No. 37 must have been extracted to give access to 23, and then to 24. 36 Negative: Dim. 0.70 x 0.70/0.55 m; delineated to the east by a fissure (Block 36); to the north by position of the wedge hole for 23, and bordered to the west by a channelling band, visible along 0.35 m and 0.15 m wide. 23 Negative: Dim. 0.65 x 1.05 m, two worked faces, negative bordered by two channels, to the east by an extension of that of 37, and to the west by that of 36 extending to join that of 21. The block was detached using two wedges placed to either side. The wedge holes are clearly distinguishable: 0.15 x 0.10 m deep (north) and 0.15 x 0.15 m deep (south).

45 must have been removed before 35 and 34 (in order to extract 33, 32 and 31). 35 Negative: Dim. 1.15/1.10 x 0.60/0.70 m, bordered to the west by a thin channel, 0.08/0.10 m wide, and to the north by a thin channel, 0.08/0.10 m wide, doubled with a wider channel, 0.10/0.15 m. 34: Dim. 0.85 x 0.70/0.55 m. Bordered by side channels, two wedge holes 0.15 x 0.10 m deep and 0.15 m apart. The damaged front part might correspond to a third wedge hole. 33 Negative: Dim. 1.40/1.55 x 0.90 m, partially delineated by channels. Extraction from the east to west, as indicated by the wedge hole identified for the extraction of Block 31. 44 Negative: Dim. 1.30 x 0.30/0.60 m. Trace 0.40 x 0.10/0.12 m to the east, indicating direction of extraction, east to west. To the north, the negative is bordered by a channel 0.80 x 0.10/0.15 m wide. 43 Negative: Dim. 1.40/1.35 x 1.50/0.95 m, delineated to the west by the fissure that crosses the rock and naturally separates Blocks 6 and 8. To the east, three wedge holes, 0.14/0.18 x 0.08 m, at intervals of 0.07 m, and to the south, marks of two wide (or double) wedge holes, 0.30 x 0.12 m deep, 0.45 m apart, indicating that the fracture surface came in from both sides. To the north, only the south channel of 33 runs along some 15 cm. It is possible that the block was not detached correctly and that levelling towards the fissure (in order to extract 42+30+16) was necessary, hence the trace of the slanting wedge hole. 32: 0.60 x 0.50 m. 31 Negative: Two wedge holes, one to the east, 0.20/0.25 x 0.15 m deep, the other to the south, 0.20/0.25 x 0.17 m deep. Negative delineated to the north by the channel (of 17) and to the west, the fissure. Dim. 0.90/1.00 x 0.60 m.

The traces left by tools are the wedge holes and the narrow channels. The wedge holes are similar: dimensions 0.16/0.18 m and 0.18/0.20 m with a depth varying from 0.08 to 0.18 depending on the state of preservation. One should imagine the use of a relatively wide wedge. The wedge holes and the channelling were cut using chisels. The wedge hole traces recorded on the harbour blocks are similar. The table “Comparative measurements: quarry negatives of extracted blocks and blocks used in the harbour moles” (fig. 22) gives us, on the one hand, a more synoptic presentation of the negatives’ dimensions (more or less corresponding to the rough-hewn blocks) and, on the other, allows us to identify the blocks employed in the construction whose measurements are approximately similar. These are, however, merely estimations and propositions, since one needs to take into account the

THE STONE QUARRIES OF AMATHUS HARBOUR

109

rough-cut work of the blocks coming from the quarry and the trimming of the blocks before their employment in the port. Specific block modules will certainly have been ordered from the quarry; however, as we have seen (e.g. traces 6 and 7), the veins and fissures of the limestone adversely affected any methodical and standardised extraction. Thus, one should perhaps suggest that the quarrymen had to adapt and find a compromise between the blocks ordered and the possibilities of the limestone quarry front, the end product being after all simple architectural blocks for the foundations of the moles. In conclusion, it is from studying the port and the revealing of the originality and particularities of the cladding construction of the moles, using large-sized blocks with lifting bosses that we were able to recognise among the identified quarries, those where the working faces bore the negatives of similar blocks. Two quarries [113] and [169] most certainly provided some blocks. It is not too astonishing that a tool maintenance workshop and a tripod have left traces, but they are nonetheless evidence of the different professions involved in the creation of the military and architectural development that the port represents.

Block no.

Length (m)

Width (m)

Harbour blocks ref.

Dimensions of harbour blocks (L x W x Ht)

1

1.70

0.70/0.72

16_D6

1.67c x 0.95 x 0.85

14_A2

1.65 x ? x 0.22c

2

0.95

0.70

3

2.80

0.70/0.90

3_D5

2.72 x ? x 0.45

4

2.05

0.75/0.90

3_B3

2.04 x ? x 0.50

4b

0.80

0.70

5

2.92

0.45/0.55

3_D5

2.72 x ? x 0.45

6

3.00

0.45/0.70

3_D5

2.72 x ? x 0.45

7

/

/

8

3.05/2.90

0.65/0.70

1_K10

3.08 x 0.70 x 0.60

9

2.85/2.75

0.60/0.65

3_D5

2.72 x ? x 0.45

10

2.70

0.70

3_D5

2.72 x ? x 0.45

11

> 1.00

0.58/0.90

12

1.55/1.65

0.60

3_A3

1.57 x ? x 0.20c

13

1.55/1.65

0.65

3_C4

1.60 x ? x 0.43

14

1.65/1.50

0.50/0.65

14_C2

1.63 x ? x 0.61

14_A2

1.65 x ? x 0.22c

14_C2

1.63 x ? x 0.61

14_A2

1.65 x ? x 0.22c

waster

15

1.20 1.40 rest

0.55/0.60 rest

16a

0.70/0.90

1.10

failure

16b

0.40/0.35

1.10

failure*

17

1.20/1.15

0.48/0.70

18

1.65

0.55/0.60

19

1.00

0.70/0.90

20

0.65/0.50

0.60/0.80

21

0.85/0.80

0.70/0.75

110

THE HELLENISTIC HARBOUR OF AMATHUS

Block no.

Length (m)

Width (m)

22

1.30 3.10 rest

0.60

23

1.05

0.65

24

>1.50

0.35

25

> 0.45

> 0.50

/

26

1.40

0.40

shapeless

27

1.45

0.45/0.65

28

1.35

0.20/0.25

29

1.30/1.25

0.55/0.65

30+42

2.90/2.80

1.10

31

0.90/1.00

0.60

32

0.60

0.50

33

1.40/1.55

0.90

34

0.85

0.70/0.55

35

1.15/1.10

0.60/0.70

36

0.70

0.70/0.55

37

0.50/0.30

0.80

38

Dimensions of harbour blocks (L x W x Ht)

1_K10

3.08 x 0.70 x 0.60

? 3_A3

? *

//

39

0.40

0.50

40

0.40

0.20

/

41

//

42

*cf30

43

1.40/1.35

1.50/0.95

44

1.30

0.30/0.60

45 16+30+42

Harbour blocks ref.

// 4.20 rest

1.10

Fig. 22 — Table presenting comparative dimensions: quarry negatives of extracted blocks and blocks used in the harbour moles (M. Wurch-Koželj)

Chapter 3

History of the Hellenistic harbour Jean-Yves Empereur

THE TESTIMONIA The testimonia concerning Amathus have conveniently been collected by P. Aupert and M.-Chr. Helmann in volume 1 of the collection Amathonte, published in 19841. Despite the great care taken by the authors in their attempt to trace the slightest mention of the town, the collected documents on the port of Amathus are rare and tell us little. The oldest document is attributed to Pseudo-Skylax of Caryanda. In his Periplus, written at the end of the 4th century bc, he gives a list of the ports of Cyprus: ȥȽɒᙼɁᚓȥɇɉɇɈɜȽɋᚌɐɒᚷɋ᚞ɐɍɑȥɠɎɏɍɑǡɈȽᚷɎɟɉɂɇɑᚌɋȽᛅɒᚪȽᚰɁɂ.ȭȽɉȽɊᚷɑᘙɉɉɄɋᚷɑǡɉɇɊɚɋȽᚍɖɍɓɐȽ Ɉɉɂɇɐɒᛂɋ ɖɂɇɊɂɏɇɋᛂɋǡ ȥȽɏɎəɐɂɇȽǡ ȥɂɏɠɋɂɇȽǡ ȦɛɎɄɅɇɑ ȰɍɇɋɜɈɘɋǡ ȭɟɉɍɇ ȋɈȽᚷ ȽᛊɒɄ ɉɇɊɚɋȽ ᚍɖɂɇ ɖɂɇɊɂɏɇɋɟɋȌǡȧəɏɇɍɋᘙɉɉɄɋᚷɑǡᗸɊȽɅɍᛒɑȋȽᛅɒɟɖɅɍɋɚɑɂᚮɐɇɋȌǤȽᛌɒȽɇɎᙽɐȽɇɉɇɊɚɋȽɑᚍɖɍɓɐȽɇᚌɏɛɊɍɓɑǤ

“Opposite Cilicia sits the island of Cyprus and the following towns: Greek Salamis, which has a closed harbour in winter, Karpasia, Kyrenia, Lepethis of the Phoenicians, Soloi (which also has a winter port), Greek Marion, Amathus ([whose inhabitants] are natives). All these towns possess a deserted harbour”.2 The intriguing aspect of this text lies in the description of Amathus harbour (and that of Marion) as deserted, though the date of the Periplus must be just before or in fact contemporary with the construction of the naval harbour, according to our chronology. Therefore, should we accept Aupert when he reads ᚍɏɇμɍɑ as Ʌᚒɏɇɋɍɑ, the deserted harbour thus becoming a harbour open only in summer, as opposed to the port of Soloi which Pseudo-Skylax qualifies as ɉɇμɚɋȽ ɖɂɇμɂɏɇɋɟɋ, a harbour usable in winter? We might propose other hypotheses: 1) Pseudo-Skylax is copying a source two centuries older and in the 6th century bc several Cypriot cities did not have harbours that were functional in bad weather; 2) Pseudo-Skylax is writing before the beginning of construction of the naval port, circa 306 bc; 3) We have seen that the harbour was most probably never used or, if it was, it was rapidly abandoned, and so Pseudo-Skylax is writing after this abandonment, since on the basis of the inner basin still functioning at the end of the 4th century, surely the description of Pseudo-Skylax would mention it. Elsewhere, a passage from Strabo’s Geography is of interest3. Writing at the end of the 1st century bc, he mentions Amathus among other Cypriot towns: ȥɜɒɇɍɋǤǤǤᘗɖɂɇɁᚓɉɇɊɚɋȽɈɉɂɇɐɒɟɋȉᚌɋɒɂᛒɅɚɋᚌɐɒɇȡɛɋɘɋɒɂᚾɒ᚞ɑɐɒɘɇɈ᚞ɑȽᚯɏɚɐɂɘɑᙳɏɖɄɀɚɒɄɑɈȽᚷ ᗸɎɍɉɉɣɋɇɍɑᚫȽɒɏɟɑȉᚌɋɒɂᛒɅɂɋɂᚫɑȝɄɏɓɒᛂɋɐɒəɁɇɍɇɖɜɉɇɍɇɎɂɋɒȽɈɟɐɇɍɇǤȠᚮɒƲᗸɊȽɅɍᛒɑɎɟɉɇɑɈȽᚷɊɂɒȽɌᛑ ɎɍɉɜɖɋɄȫȽɉȽɇᙼɈȽɉɍɓɊɚɋɄǡɈȽᚷᚼɏɍɑɊȽɐɒɍɂɇɁᚓɑᙇɉɓɊɎɍɑȉɂᚮɒȽȥɍɓɏɇᙼɑȓᙴɈɏȽȔɖɂɏɏɍɋɄɐɣɁɄɑǡɂᚫɑ ᚚɋᙳɎᛂȣɏɟɋɘɋɐɒəɁɇɍɇᚏɎɒȽɈɟɐɇɍɇǤȠᚮɒȽɎɟɉɇɑȥɍɠɏɇɍɋᚿɏɊɍɋᚍɖɍɓɐȽǡᗸɏɀɂɜɘɋɈɒɜɐɊȽǤ

“(Citium) has a closed harbour. It is the birthplace of Zeno, the chief of the Stoic sect, and of Apollonius the physician. Thence to Berytus are 1500 stadia. Next is the city Amathus, and between Citium and Berytus, a small city called Palæa, and a pap-shaped mountain, Olympus; then follows Curias, a promontory of a peninsular form, to which from Throni are 700 stadia; then the city Curium, with a mooring, founded by Argives”. 1.

Aupert, Hellmann 1984.

2.

24 Pseudo-Skylax, Périple, 103 (ed. Müller K., GGM, 1855), reproduced in Aupert, Hellmann 1984, p.13-14, text no. 6 with note 14. Pseudo-Skylax from the end of the 4th century bc draws on Skylax, two centuries earlier.

3.

Aupert, Hellmann 1984, p. 14, text 7: Geography, XIV, 6, 3.

114

THE HELLENISTIC HARBOUR OF AMATHUS

Strabo does not talk of a harbour for this city though he does mention a closed port at Kition and a mooring at Kourion. The use of the outer basin had ended long before the end of the 1st century bc, but what of the inner basin? That is another question….

FIXING THE CHRONOLOGY: HISTORICAL SOURCES AND ARCHAEOLOGICAL EVIDENCE As is shown by the abundant pottery and the coins found in the underwater excavations, the harbour was under construction during a period when there was a serious struggle for the possession of Cyprus between Ptolemy Soter, ruler of Egypt, and Antigonus Gonatas, known as Monophtalmos (the one-eyed), who was decisively supported by his son, Demetrius Poliorcetes. Ptolemy had seized Cyprus in 316, getting the better of Antiochus and the petty kings of Cyprus who had stood by his side. Ten years later, Antigonus Gonatas dislodged Ptolemy in what was, for this latter, one of the most devastating defeats of his career at the naval battle of Salamis, which Plutarch described in great detail in his Life of Demetrius. Ptolemy with his brother and general Menelaus threw 210 ships into the confrontation, but Demetrius swiftly seized 70 of them with their crew. Ptolemy managed to flee with eight ships while Menelaus surrendered unconditionally ȧɂɒᙼɁᚓɒ᚝ɋɋȽɓɊȽɖɜȽɋɍᛅɁǯᚾȧɂɋɚɉȽɍɑᙳɋɒɚɐɖɂɋǡᙳɉɉᙼɒɛɋɒɂȭȽɉȽɊᚸɋȽɎȽɏɚɁɘɈɂɒ᛫ȟɄɊɄɒɏɜᛠ ɈȽᚷɒᙼɑɋȽᛒɑɈȽᚷɒᛂɎɂɃɟɋǡᚯɎɎɂᚸɑɒɂɖɇɉɜɍɓɑɈȽᚷɁɇȽɈɍɐɜɍɓɑɈȽᚷɊɓɏɜɍɓɑɈȽᚷɁɇɐɖɇɉɜɍɓɑᚾɎɉɜɒȽɑԝͶ.

“After the sea-fight, Menelaus also made no further resistance, but handed over Salamis to Demetrius, together with his fleet, and his land forces, which comprised twelve hundred horsemen and twelve thousand men-at-arms”. In the aftermath of this resounding victory Antigonus has himself and his son proclaimed king. Ptolemy follows their example, and in 305 bc he adopts the title of king of Egypt. But Ptolemy could not forget Cyprus, so rich in wood and metal, the very raw materials of which Egypt was so short. Nor could he allow one of his most bitter enemies to threaten Alexandria and the north coast of Egypt so easily. Ptolemy did his utmost to reconquer the island and, in 294, when Demetrius was otherwise engaged in the seizure of Athens, he finally succeeded. Ptolemy’s success meant that for the next three centuries the island would remain a possession of the dynasty that he founded. The acropolis of Amathus has been the object of annual excavations by the French School at Athens since 1975 and they have revealed the remains of a large building, known as the “palace”. In the destruction levels of this palace the archaeologists have found seven coins of Antigonus Gonatas5. These coins have been identified as “monnaies… frappées à Salamine, donc après que Démétrios se soit emparé en 306, de cette ville [coins struck at Salamis, thus after Demetrios had seized this town in 306]”6. Françoise Alabe has shown7 that the archaeological material found in the destruction level of the palace resembles very closely that which was found in the harbour. This concordance sits well with the dating that we had arrived at elsewhere and tends to confer even greater precision. Might we link two very different events, such as the destruction of the palace and the filling of the harbour basin, as the chronology may suggest? One could suppose that they were both connected to one single historical event, and the presence of the seven coins of Antigonus dated to

4.

Plutarch, Parallel Lives, vol. 9, §16. See infra, Note 18, for the comment by Diodorus: he gives different but nonetheless large numbers (40 warships and 100 merchantmen).

5.

Cf. P. Marchetti, BCH 102 (1978), p. 948 and 949 fig.19. Attribution to the father or to the son is discussed by Bruno Helly (in favour of Antigonus) and by Newell (for the latter).

6.

Cf. P. Aupert, F. Burkhalter and A. Queyrel, BCH 107 (1983), p. 967.

7.

Fr. Alabe see volume 2.

HISTORY OF THE HELLENISTIC HARBOUR

115

after 306 encourages me to recognise here archaeological traces of the violent seizure of the town by Ptolemy in 294 and the consequent destruction of the palace. As a first result, it would seem that we must attribute the construction of the harbour to Demetrius Poliorcetes. After his naval victory at Salamis in 306 he needed to develop a site to shelter his warships and the many vessels he had seized from Ptolemy. Likewise, he needed to create maritime defences to ensure his possession of the island and he would also have needed to find the means of continuing his war against Ptolemy. The extent of the works demonstrates a grand design, most likely to establish a secure site where he might gather a large fleet, and what better location than one that directly threatened Alexandria. One should mention here the successful raid that took place some 1,600 years later, when Pierre I de Lusignan attacked Mamluke Alexandria. In 1365, he seized and ravaged the town before escaping untouched to Cyprus loaded with booty. As a second fact, the abandoning of its construction is linked to the return of Ptolemy in 294. We have explained the archaeological reasons why we consider that the use of this harbour was limited; if indeed it was ever finished. In reality, the study of the pottery alone is sufficient to show the lack of use over a long period, without forgetting the discovery of roughly 30 coins in the harbour entrance: three have been attributed to Demetrius Poliorcetes8. On the other hand, there are no traces of destruction in the harbour, but this absence might well signify the fact the harbour was not operational. That the reconquest of Ptolemy in 294 should mark the end of the harbour works is hardly surprising. One can easily imagine that Ptolemy did not wish to run the risk of seeing this port used once again by one of his enemies if he was ever to lose control of the island. It is worth remembering that Amathus presented a real threat to Alexandria, sitting opposite the capital of the Ptolemies, a mere 250 nautical miles distant (roughly 475 km) along a north-south line, that is, two days and nights of sailing9. In brief, Ptolemy seized Egypt in 323. He took Cyprus in 316-315, doing away with four of its kings: Androcles, king of Amathus continued until 312-311. Ptolemy installed his brother Menelaus who ruled the island in 306, the year that Demetrius arrived and engaged in the naval battle of Salamis, inflicting a devastating defeat on Ptolemy that destroyed and/or captured two thirds of his fleet. Antigonus’ son Demetrius Poliorcetes reigned alongside his father until the latter’s death in 301, and then alone until 294. He left the island and Ptolemy grabbed it, and so it remained in Ptolemaic hands until 58 bc. According to P. Marchetti’s excavations on land, “The destruction layer is dated to the period around 296-294. 10” Ptolemy wished to eliminate any signs of Antigonid power by destroying the palace and abandoning the harbour.

AMATHUS HARBOUR, A MILITARY INSTALLATION A naval plan. The surface area of the harbour basin and the impressive length of the moles reflect a specific need that is way beyond the simple trade requirements of a town like Amathus. More than 400 m of moles and a tight bottleneck entrance only 20 m wide in order to control the entry of shipping are architectural features that show the specific character of a port installation 8.

O. Picard, infra, p. 154-155.

9.

See P. Arnaud, Les routes de la navigation antique: Itinéraires en Méditerranée, Paris, 2005, p. 216-217. “Strabo estimates the distance between Paphos and Alexandria as 3,000 stadia, whereas Agathemeros counts 3,800 stadia, with the wind from the north (that is, respectively 291 and 370 nautical miles for a stadium estimated at 180 metres). If a journey of 24 hours covered 1,000 stadia (p. 79), the crossing would take between three to four days.”

10.

Cf. Aupert 1996, p. 53-54.

116

THE HELLENISTIC HARBOUR OF AMATHUS

designed to shelter ɎɉɍᚸȽ μȽɈɏᙻ, the streamlined warships, and not ɎɉɍᚸȽ ɐɒɏɍɀɀᛐɉȽ, the round-hulled merchantmen. The shipsheds. Nonetheless, a standard feature of a naval port is missing11. We have not been able to identify any ɋɂᛝɏɇȽ or ɋɂᛝɐɍɇɈɍɇ, those boathouses whose presence is connected to warships, providing shelter, dry dock and a place of repair. This is a notable absence: either there were none or they had yet to be built at the moment work was abandoned, or else they were built on a site that we did not excavate. It is difficult to place them on the northern line of the harbour, along the beach where the ridge of beachrock stretches, in that we do not know the junction between the moles and the inner basin, which has not yet been excavated12. If there were neoria, one might suppose that they were placed in as sheltered a spot as possible, perhaps indeed around the edges of the inner basin. One can also add that the moles played a role in protecting against the sea, wind and storms, but they did not in all likelihood serve as moorings. Indeed, when first discovered, almost all the lifting bosses on the blocks of the moles were preserved. These had not been tooled off after positioning and they pose a real danger to a ship’s hull, which they could puncture with the slightest breath of wind. Although, once again, we might just have another indication that the harbour, as we found it, had not been finished. Military construction method. The entire construction of the harbour speaks of a systematic order of work, a sustained rhythm and efficiency. We have discovered a single phase of construction operations, with no reworking and no repairs, and this has been backed up by studies of the archaeological material revealed during excavation. The pottery belongs to one period, its homogeneity and lack of chronological duration fits with a monument that was constructed at one go, with a regularity of process and in a style that leaves no perceptible interruption in the almost mechanical rhythm of the positioning of the blocks of the moles. The efficiency and the desire to economise is reflected first of all in the use of local resources, ensuring the supply of the most easily accessible stone, and in trying to avoid distance and to reduce the cost and time of transportation. The exploitation of nearby quarries, including on the beach just to the north of harbour, has left perfectly identifiable traces along with negatives of blocks used in the construction of the moles13. Moreover, the surrounding limestone is of good quality while also being quite easily worked. The regularity of the block shape is obvious, with rectangular slabs of varying size depending on where they were to be placed: the largest sizes are found at points most exposed to the sea and winds, especially in the south-west of the harbour. The system of lifting bosses is standardised, with the predominance of a particular type (our Type 3), which is by far the most common even though other variants are in evidence14. Metrology has been carefully respected, with the Attic foot of 0.294 m chosen as the standard and used in the different constituent parts of the harbour (fig. 1). A desire for standardisation in the dimensions of the blocks can be seen from the quarry to the length of the moles. Most of the blocks measure 2.5 feet wide, with bosses of 1 foot wide, the length varying in multiples of feet or half a foot. As for the South Mole, it reaches a length of 600 feet, i.e. one stadium (it was necessary to add 11.

See infra, p. 127.

12.

See supra, p. 126.

13.

See supra, p. 100-106.

14.

See supra, p. 66, the boss types and the distribution tables by sondage, p. 66-78.

117

Sakieh Well 3

100 0

10

50

N

180 m =

Well 4

1 sta

dium

Well 5

150 m

Well 6

HISTORY OF THE HELLENISTIC HARBOUR

ium

m

80 m

=1

Well 2

1

180

m=

70 m 11.0

8m

3.00

162 .50 m 3.00

18m

L = 205

m

Well 1

1 st

adi

um

Harbour Entrance

20

sta d

m

tadiu

0m= 1s

18

Fig. 1 — Plan of Amathus harbour with standard measurements indicated. Scale 1:2,500 (plan EFA, Fr. Perdrizet, T. Koželj)

118

THE HELLENISTIC HARBOUR OF AMATHUS

a few metres to reach right up to the East Mole in order to respect the symmetry of the harbour entrance channel). The three moles are laid out according to a simple geometry, following a scheduled progression along a work calendar that one supposes will have been implemented in order to complete the work within the desired time period. The draught at the moment of construction. At present, the upper preserved course of the ancient harbour blocks varies from one sondage to the next, but on average it lies between 1 and 2 m beneath the surface of the sea, while the lower courses set upon the sand have been excavated down to -5 or even -6 metres (see fig. 1). If one supposes that the upper course stood about 1 m above sea level at the time of construction, that would give a harbour depth of about 3 to 4 m. This fits with the average of the few examples of known harbour draughts, such as Marseille 15, even when taking into account the fact that the tide at Amathus has a mean variation of slightly less than half a metre. South Mole Sondage A = -3.76 m (upper course = between -1.06 and -1.20) (see fig. 11-12, p. 43) Sondage 1 (near the entrance) = -5.96 m (upper course = between -1.80 and -1.85) (see fig. 15-16, p. 46) Sondage 3 = -5.18 m (upper course = between -1.60 and -1.98 (see fig. 17, p. 47) East Mole Sondage 2 = -3.80 m (upper course = between -1.34 and -1.42) (see fig. 26, p. 56) Sondage 12 = -4.84 m (upper course = between -1.50 and -1.95) (see fig. 18, p. 48) This depth of circa 3 m would allow the entrance of warships without too many problems, the draught of a trireme being estimated at about 1 m. As for the entrance channel, at 20 m wide it provided good manoeuvring space for these ships, which were less than 5 m wide: even with all oars deployed there would still be a good ten metres16. Massive expense. Despite the desire to standardise construction through uniform modular elements, and despite the military discipline that one can believe ruled over the operations, the building of Amathus harbour clearly involved a huge investment that sprang from a strong commitment to pursue this great project through to completion. Ȯɍᛒ Ɂǯ ᚌɋ ᚾɉɈəɐɇ ɎȽɏɍɏɊɍᛒɋɒɍɑ ᚼɖɉɍɓ ɅɂɏȽɎɟɋɒɘɋ ɈȽᚷ ɔɜɉɘɋ ɈȽᚷ ɀɓɋȽɇɈᛟɋǡ ᚍɒɇ Ɂǯ ᚿɎɉɘɋ ɈȽᚷ ɖɏɄɊəɒɘɋ ɈȽᚷ ɊɄɖȽɋɄɊəɒɘɋǡ ᙷɎɉᛟɑ ɍᛅɁᚓɋ ᚌɌɚɔɓɀɂ ɒᛂɋ ȟɄɊɛɒɏɇɍɋǡ ᙳɉɉǯ  ᚍɉȽȾɂ ɎəɋɒȽ ɈȽᚷ ɈȽɒɛɀȽɀɂɋɂᚫɑɒᛂɐɒɏȽɒɟɎɂɁɍɋǤ

“But of the throng of attendants, friends, and women which lay in ships of burden close at hand, and further, of all Ptolemy’s arms, money, and engines of war, absolutely nothing escaped Demetrius, but he took everything and brought it safely to his camp”17.

15.

Nantet 2016.

16.

P. Gille, “Les navires à rames de l’Antiquité, trières grecques et liburnes romaines” Journal des Savants (1965), p. 36-72, especially p. 60 for the different dimensions. As for merchant ships, their draught is examined by E. Nantet: from 1.20 m to 3.75 m for the largest of the Madrague, with a table of 10 examples out of 13 at least of 2 m; see Nantet 2016, p. 226.

17.

Plutarch, Parallel Lives, vol. 9, §16 ; Pimouguet-Pédarros 2011, p. 173-176.

HISTORY OF THE HELLENISTIC HARBOUR

119

This operation, both ostentatious and of immediate strategic utility, can be linked to the success of the Antigonids, and to Demetrius Poliorcetes’ victory on land and then sea over Ptolemy and his brother Menelaus at Salamis in 30618. This tremendous victory made Demetrius and his father Antigonus the new masters of Cyprus and of the wealth of the island’s ancient kingdoms, so recently in the possession of Ptolemy. They found themselves at the head of a vast fleet made larger by the vessels captured during the battle. They proclaimed themselves kings of Asia and Demetrius dedicated his flagship to Apollo in the Neorium, the Monument of the Bulls, which he had built at the same moment in the sanctuary of the god on Delos19.

18.

Diodorus of Sicily, Library of History, XX, 50-53, describes the two battles, on land and then on sea, in detail. He gives an account of the impressive loot: at sea, 40 boats with crew and 100 transport vessels with 8,000 soldiers on board.

19.

Guide de Délos, GD24. He also offered the weapons from his terrestrial victory, as is recorded in the inventories of the Athenian administrators, including helmets, diadems, tack for horses, etc. see J. Tréheux, “Sur le Néôrion à Délos”, CRAI 1987, p. 168-184.

Part 2 Post-harbour history

Chapter 4

The archaeological remains Jean-Yves Empereur, Tony Koželj

The beachrock Jean-Yves Empereur, Tony Koželj

Lastly, in the second part of this work, we will look at Sondage 10, along the lines of beachrock, in the centre of the harbour basin, and at the Late Roman wells to the north near the beach (see plan 2 at the end of the volume for the location of the sondages).

THE BEACHROCK LINES During the first excavation campaign in 1984 images obtained from the helicopter flights revealed the existence of a line running roughly 60 m south of the actual beach and about 100 m north of the South Mole. This line can be followed along some 100 m east to west. Aerial photography and underwater examinations revealed large blocks of beachrock, a natural formation of sedimentary sandstone, engendered by the meeting of salt and fresh water (see plan 1, south of the boats, and fig. 1a-b). This sandstone is hard to cut into and crumbles when broken off in pieces. In 1984, we opened Sondage 10 in order to determine the nature of these plates of rock and to establish their date (fig. 2). The line lies at -1/-1.5 m below present-day sea level, with blocks of irregular thickness, around 0.2 m, sloping to the south on a muddy bottom at a depth of -1.70/1.80 m. This line measures about 2 m from north to south: it is not continuous and is cut by channels that delineate irregular plates up to 4 m in width from east to west. In order to be sure that there were no built courses of masonry beneath this natural concrete, we excavated down through it. By cutting down through these plates we ascertained that the beachrock lay upon the mud and sand that filled the harbour basin. Another line, parallel to the above but shorter, appears about a dozen metres further south (see plan 1 and fig. 1a-b). This other beachrock formation lies at a greater depth and must date to a later period than the first, when the sea level was even lower, though we cannot date this progression. In 1986, the geographer Pierre-Yves Péchoux joined us in our examination of the beachrock in an attempt to determine the date of its formation. The large plates of beachrock visible on the aerial photographs in the north-east of the basin cover a part of the East Mole and are thus more recent than this latter (see fig. 30a-b, p. 61). We opened a second sondage in this beachrock. Sondage 18 is located east of Sondages 2+14, towards the entrance of the harbour. Several blocks of the mole were fused into this beachrock at -0.8 m below the surface, which corresponds to the depth of the first Late Roman shoreline. Fragments of pottery were also cemented into this formation: a handle and a neck with handle from two local Late Roman 1 amphorae. On the west side of the harbour, one

126

THE HELLENISTIC HARBOUR OF AMATHUS

can make out two different sorts of beachrock: one with no shards, and a second, deeper one with shards directly upon the limestone that was used as a quarry1. At the time of the excavations, geomorphology in archaeology was a new discipline in its infancy and so we did not have access to dating through physical and chemical analyses. In the absence of coins, the pottery material held in the sandstone was the only chronological indicator and it dated back to the 4th century ad at the earliest.

THE NORTH SIDE OF THE HARBOUR AND THE INNER BASIN To the north of the harbour that was constructed in the Hellenistic period, a semicircular depression stretches inland which has been regarded for several decades as the inner basin of the port (see plan 2 inside the back cover and the aerial view fig. 3). This zone extends south from the ancient agora, some 100 to 150 m, stepped back from the present coastline along roughly 220 m running east/west. In 1978 specialists in archaeometry conducted resistivity tests in the interior of this basin and determined that it was free of any structures2. Pierre Aupert regarded it as the inner part of the ancient harbour and, in an attempt to confirm this hypothesis, he dug a trench to the north, on the edge of the depression, and a sondage in the centre down to below sea level, to a depth of -2.5 m, revealing sandy mud with sea shells (fig. 4)3. In 2014 the geomorphologist, Antoine Chabrol, examined the natural remanent magnetisation and electrical resistivity of the substratum of the inner basin4. His conclusions showed a fine sedimentation, with no traces of construction and a fill 4 to 5 m thick. He then conducted a number of cores that revealed among other information, clays mixed with “organic matter of marine origin” that would indicate that the construction of the outer harbour would have broken the marine dynamic that could have led to the filling in of the inner basin. The author adds that his observations require closer examination and confirmation using laboratory tests and radiocarbon dating. The numerous models of boats from the Archaic era discovered at Amathus5 have demonstrated that the town possessed both a merchant and a naval fleet which required a port (fig. 5-6). One might presume that the inner basin served this purpose before the construction of the outer harbour and that, if we follow the Chabrol’s reasoning, the building of the external harbour caused the silting of the inner harbour basin. One can easily accept this proposition, but also pose another question: could it be that the two basins, inner and outer, became silted up at the same moment?

THE QUESTION OF SHIPSHEDS A crucial point in the study of the naval harbour of Amathus is to decide whether or not shipsheds existed on the north side of port, as has sometimes been mentioned and as one would normally expect in a harbour designed for warships.

1.

See also the remarks of B. Bousquet and P.-Y. Péchoux, BCH 102 (1978), p. 968, with the caption to fig. 43 regarding the “grès de plage historique du rivage d’Amathonte à proximité de la muraille côtière [ancient beachrock on the Alexandrian shore near the coastal wall] (LO. MH 361.380)”.

2.

Chr. Parhas, I. Spahos, “Amathonte: deux ans de prospection géophysique”, BCH 103 (1979), p. 756-762.

3.

P. Aupert, BCH 103 (1979), p. 726-728; Aupert 1996, p. 168-9; “22. Le port interne” (summarised in Aupert 2000, p. 96).

4.

http://www.academia.edu/15601564/First_report_on_the_geomorphological_and_palaeoenvironmental_study_ of_the_harbour_of_Amathus_Cyprus, site visited on 19/08/2017.

5.

Basch 1987.

THE BEACHROCK

127

This is a tricky matter where caution is required. We should not be influenced by the likelihood or even the necessity of there being traces somewhere of neoria: we must respect the archaeological evidence. The argument for the presence of shipsheds is backed by parallels with the neighbouring harbour of Kition, where seven such structures were discovered. Thanks to pottery, the construction of these shipsheds has been dated to the end of the 5th century bc and they were abandoned one century later at the end of the 4th century bc, that is, at the same period when Amathus harbour was developed and then abandoned6. Perhaps the smooth stones bordering the northern part of the moles belong to these structures built to shelter vessels. Their appearance differs from the rest of the moles, with alternating stretchers and headers. However, there is no incline that would allow one to conclude that there is a change in the configuration of the mole. It all remains perfectly horizontal, without the slightest slope that is so characteristic of a shipshed. It is indeed tempting to place shipsheds along the north side of the harbour constructions, as has been seen in several drawings, one of which has been published (fig. 7a-b)7. But this would be to separate arbitrarily the harbour construction from the inner basin, thus supposing that the internal part of the port was no longer in use, even though we have no proof of this. This lack of proof has led the authors of a reference work on shipsheds to reject the notion8. If Amathus harbour did nonetheless possess the shipsheds required to shelter warships, perhaps they will be discovered one day at the back of the inner harbour basin, adjacent to the agora.

SEA LEVEL. A LATER RISE According to our findings, the wide variations in sea level are difficult to explain. Indeed, we have seen that the harbour moles must have stood at least 1 m above the surface of the sea, whereas at present the upper courses are under 1 to 2 m of water. At a certain moment after the construction of the harbour that is difficult to determine, the sea level fell considerably, to such an extent that it engendered the formation of beachrock, which can be found on the East Mole, and following a line parallel to the shore at -1.5 m, with freshwater wells whose bottoms are between -2.22 m and -3.36 m in relation to the present surface, which gives an amplitude of 2 to 3 m more in relation to the water level when works on the ancient harbour were finished, i.e. 5 to 6 m difference. This twin movement of a sea level drop during the Hellenistic period (circa -6 m) and a rise (circa +7 m) are difficult to explain. Perhaps geomorphologists and seismologists will recognise tectonic phenomena that affected the area of Amathus in a particular manner in relation to the rest of the island and the wider geographical context.

6.

M. Yon, J.-C. Sourisseau, “Le port de guerre de Kition”, in D. Blackman, M.C. Lentini (ed.), Ricoveri per navi militari nei porti del Mediterraneo antico e medievale, Atti del Workshop Ravello, 4-5 November 2005, Bari, 2010, p. 57-67.

7.

T. Koželj, M. Wurch-Koželj, “The ancient harbours in Greece and Cyprus” (in Greek), in ȉȡȓIJȠ ȆĮȞİȜȜȒȞȚȠ ȈȣȞȑįȡȚȠ ȁȚμİȞȚțȫȞ DzȡȖȦȞ, Athens (2002), p. 337-357. For Amathus harbour, see p. 348, with fig. 11.

8.

D. Blackmann, B. Rankov, Shipsheds of the Ancient Mediterranean, Cambridge, 2014, p. 586. The authors cite, in note 147, T. Theodoulou, “Nautical activity in Classical Cyprus and the harbour network at the end of the 4th century bc” (in Greek), an unpublished (2007) thesis from the University of Cyprus that also rejects this reconstruction of shipsheds at Amathus.

128

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 1a-b — To the left (= north), the line of irregular slabs of beachrock running parallel to the South Mole on the right of the image (photograph EFA, Ph. Collet)

THE BEACHROCK

129

Fig. 2 — Sondage 10, slabs of beachrock (drawing EFA, T. Koželj)

Fig. 3 — Aerial view of the inner basin: the excavations of the Agora can be seen to the north (photograph EFA, Ph. Collet)

130

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 4 — Location plan of the resistivity measurements made in the inner basin in 1978, reproduced from P. Aupert, BCH 103 (1979), p. 726, fig. 1. Recognition of the archaeological site extension. Apparent resistivities, Longitudinal Wenner 3x2 m

THE BEACHROCK

131

Fig. 5 — Terracotta model of a warship with steering oar and shields: the prow has the shape of a dolphin’s head; the stern curves up and towards the prow. Cypro-Archaic period (750-480 bc), Amathus Tomb 429/5, LM 1002. Museum of Limassol. Courtesy of the Department of Antiquities, Cyprus. Photograph © DAC

Fig. 6 — Terracotta model of a merchant ship. Cypro-Archaic period (750-480 bc), Amathus Tomb 521/2, LM 1143. Museum of Limassol. Courtesy of the Department of Antiquities, Cyprus. Photograph © DAC

Fig. 7a-b — Certain imaginary reconstructions dating from the time of the excavations were refuted by the facts on the ground: a) the harbour bottom is seen as a shoreline up which boats could be hauled onto dry land (unpublished); b) a series of shipsheds is repeated along the northern line of the built harbour. No evidence has been found on site to support this reconstruction, which was published in 1993 (T. Koželj)

The wells Jean-Yves Empereur & Tony Koželj

From the first campaign of 19841, we excavated quadrangular structures that appeared in the northern part of the basin and even on the surface of the beach. We cleared a rectangular well and a second, square well, and then in 1985 an oblong cavity that had held a waterwheel2. Given that the results of these three excavations corresponded both as regards architectural appearance and the chronology provided by material found in the fill, we chose not to excavate the three other similar-looking wells situated on the beach to the east of the port. The position of these six structures can be seen in the plan, fig. 1, p. 16.

WELL NO. 1 Discovered on 19 September 1984, the first well, rectangular in shape (fig. 1a-b), was excavated over a week down to its foundations. It measured 0.68 m x 1.06 m. The upper preserved masonry course was between 0.73 m and 0.84 m beneath present sea level. The structure was composed of small cut stone of varying sizes in six to seven courses according to size. The constructed walls of the well varied between 2.21 m and 2.32 m depending on the state of preservation of the upper course. The well

Fig. 1a-b — Well 1 during excavation (photograph EFA, C. Verlinden)

1.

Empereur 1985, p. 998-999.

2.

Empereur, Verlinden 1986-1, p. 906-907.

134

THE HELLENISTIC HARBOUR OF AMATHUS

was set upon an anchorage of four wooden beams 0.10/0.12 x 0.12/0.14 m in section. The narrow sided beams (south and north) lay upon the broad sided beams (east and west) and the upper beams were placed on a bed of cut stones and the lower were directly on the mud. The bottom of the well was reached at 2.03 m below sea level (fig. 2). The first SU of soft black mud held some shards from LRA 1 amphorae. Lodged beneath this initial layer was a one-handled plastic bottle containing a large stone (most probably a buoy weight belonging to one of the fisherman who still work in the port area). This loose layer covered a second, white and compact layer on the top of which were 20-odd terrestrial snail shells, demonstrating that this was a freshwater well. The white layer contained a good quantity of pottery, pieces of wood including a pulley block, fragments of glass and grey amphorae of local manufacture. On 22 September, the white layer became darker with a lot of pottery. On 24 September, we reached the bottom of the well, where there were terrestrial snail shells and shards of LRA 1 amphorae as found higher up on 22 September. In the western wall, a re-used block with dentils (fig. 3) was found that resembled another fragment discovered on the beach. Fresh water sprang from the bottom of the well. It was quite different from seawater: it was clear and cold and rose to the surface being lighter than seawater because it contained no salt. It was easily recognisable by sight and then by taste. A small control sondage descended 30 cm beneath the foundations but found only virgin sand with no archaeological material.

Sea level at time of excavation

-0.73

-0.66

-0.73

-0.84

-2.03

-2.29

A

B -0.72

C

D

B

N A

C

0

0.5

1

1.5

2 

3 m

-0.70

D

Fig. 2 — Plan and section of Well 1. Scale 1:50 (drawing and CAD EFA, T. Koželj)

THE WELLS

135

WELL NO. 2 A second, square-shaped well was also excavated in 19843 (fig. 4a-b). Well 2, of similar workmanship to Well 1, was located 85 m west of the first well on the same north-south line, some 30 m north of the line of beachrock. Its sides were 1 m long. The upper course lay from 1.06 m to 1.15 m beneath present sea level. The seven courses of small blocks were set upon wooden beams, two resting on the other two, following the system described for Well 1. The bottom of the construction was at 3.36 m below actual sea level. Beneath a layer of stones, in the fill were half-a-dozen LR13 amphorae that had broken in situ. The stones mixed with the shards were evidence that the well was no longer in use at that moment. Under this blockage, the fill layers revealed LR13 amphorae, three trilobate jugs with a single handle, bowls, lamps of the same local manufacture as the amphorae, and all seemingly from the Late Roman period. It is noteworthy that the amphorae and jugs held a lot of organic material with animal bones. From sack to sack, the shards corresponded. Two coins were found, one dating to Constantius II (348-361) and the other to Arcadius (395-401)4. At the bottom of the excavation, we reached a SU of sand that appeared to correspond to the use of the well (fig. 5). A thin stream of cold, sweet water rose from the bottom (fig. 6).

3.

Cf. Empereur 1985, p. 988-989 with fig. 35-36.

4.

Coins AM 1395 and 1396, cf. infra, the study by O. Picard, p. 160.

Fig. 3 — The western wall of Well 1 with a re-used block with dentils (photograph EFA, Ph. Collet)

Fig. 5 — Well 2, view towards the bottom; two wooden beams can be seen beneath the courses of stones (photograph EFA, Ph. Collet)

Fig. 4a-b — Well 2 during excavation; a: view to the east, b: view to the north (photograph EFA, Ph. Collet)

136

THE HELLENISTIC HARBOUR OF AMATHUS

B -1.06

N

C

-1.06

A D 0.00

Sea level at time of excavation

-1.04

-1.12

-1.06

-1.15

-3.36

A

0

0.5

B

1

1.5

C

D

2 m 

Fig. 6 — Plan and section of Well 2. Scale 1:50 (drawing and CAD EFA, T. Koželj)

The two wells had been filled with very homogeneous pottery. Filling had been done quickly, the surface fragments overlapping chronologically with those found at the very bottom. Well 2 was richer in material and held some 30 LRA 1 amphorae. See Volume 2, article on Late Roman amphorae by J.-Y. Empereur.

THE SAKIEH (WATERWHEEL) Discovered on 27 September 1984 and excavated in 1985 (fig. 7a-b)5, Well 3 was the object of a rather difficult amphibious excavation (fig. 8). Located on the beach, 40 m north-east of Well 2, the sea would flow into the reservoir as it was emptied. It is quadrangular in plan, with the following dimensions: exterior short sides, north-east 3.06 m and south-west 3.12 m; long sides, south 8.09 m and north 8.18 m; interior dimensions: 1.10 m by 6,11 m to the south, 6.20 m to the north. Under the weight of the earth, the long side walls have tended to curve inwards giving the interior space a double dovetail shape. The construction method is the same as for the wells, with irregularly cut blocks in six to seven courses (fig. 9), depending on the area and as a function of the size of the chosen 5.

Empereur, Verlinden 1986-1, p. 906-907 with plan and sections p. 906.

THE WELLS

137

Fig. 7a-b — The sakieh before excavation, a) view to the east; b) view to the west (photograph EFA, Ph. Collet)

stone, resting on beams that fit together with mortises. The beams of the three wells are of the same section: 0.12 to 0.14 m, and 6.10 m long for the long sides, and 0.48 m for the three beams laid width-wise. One should note a 3rd small beam, resting on the two long beams in the middle of the well 1.84 m below the upper course. The bottom of the reservoir at the western end was not entirely built up with masonry from the third course down (fig. 10): a middle cavity through which water entered thus giving us the direction of the flow and the rotation of the waterwheel, from west to east. The reservoir is very thick, with walls of 1 m width along the long sides (fig. 11). The upper course is now at present sea level and at certain points plus-0.52 m, while the foundation beams are at minus-2.10 m. It is clear that this Well 3 is different from the other two in both its shape and its usage. Number 3 is a sakieh, a reservoir for a waterwheel. Apart from its oblong shape and the large opening in the lower part of one of its short sides to allow the water to flow6, one must note the discovery in the fill of wooden planks and sections of timber that would have been part of the wheel (fig. 12a-b and 13)7. Given the (preserved) height of the reservoir, the wheel would have had a maximum diameter of 5 m to fit the length of the tank (between 6.11 m and 6.20 m, the width being 0.81 m at the centre). The wood itself shows a curvature that also allows an estimated diameter of 5 m for the wheel. Sakieh pots were among the artefacts found in the fill. These are a sort of flared bucket with a rolled lip and a prominent knob that are attached to the wheel by cords tied around the neck and knob. Some of these pots were found intact8, dating most probably to the very last functioning of this device. A typo-chronology of sakieh pots is not an easy thing to establish9; however, given the pottery and amphorae (LR1 and LR13) that filled the tank, as well as the pottery studied by May Touma10, one might propose that this well was abandoned and filled around the 6th century ad.

THE OTHER WELLS Three other wells were located on the dry land of the beach outside the harbour basin, running along a line towards the east (fig. 14). Situated at the same level as the sakieh described above, Well 4 was some 170 m away, Well 5 was 50 m further east and Well 6, 24 m further on. These wells strongly resemble those that we excavated underwater and we did not empty them. They are square in layout, 6.

Empereur, Verlinden 1987-2, p.17 fig.13.

7.

For identification of the species, cf. the study by L. Lorentzen, S. Manning, volume 2.

8.

Cf. volume 2, chapters by J.-Y. Empereur and M. Touma.

9.

Cf. D. Dixneuf, in Hairy 2011, p. 420-421.

10.

Cf. volume 2.

138

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 8 — The sakieh during excavation: view to the east (photograph EFA, Ph. Collet)

Fig. 10 — The sakieh, with the water inlet cavity to the west (photograph EFA, Ph. Collet)

Fig. 9 — Masonry structure of the sakieh; to the west, initial clearing of the water inlet (photograph EFA, Ph. Collet)

THE WELLS

139

Sea level at time of excavation A 0.09

B

C

0.41

0.27

D

0.35

0.31

0.52

0.01

0.00

-0.44

-1.96

-1.96

2.10

-2.10

A 0.12 0.27

D

0.28 0.21 -1.84

0.01

-1.96

N 0.67

-0.12

0.42 0.35

0.41

B 0 0.5

1

1.5

0.05

C

Fig. 11 — Plan and section of the sakieh (drawing and CGI EFA, T. Koželj)

Fig. 12a-b — Pieces of wood from the sakieh wheel (photograph EFA, T. Koželj)

0.12

140

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 13 — Piece of wood from the sakieh wheel (photograph EFA, T. Koželj)

N

0

0.5

1

1.5

2 

3 m

Fig. 14 — Plan of Well 4. Scale 1:50 (drawing and CGI EFA, N. Lappa)

THE WELLS

141

all roughly the same size and the same construction of small local limestone blocks, Well 4 being 3 m each side. There are certainly other wells, which we were not able to identify because the beach is largely covered with dead seaweed. We can only hope that the modern installations involving the placing of large stone blocks and a walkway over the site have not hidden them definitively.

AN ABUNDANCE OF WELLS The presence of so many wells may be surprising. Four wells and a sakieh over a distance of less than 200 m raises questions as to the need for so many fresh water supply facilities. And what meaning can be given to the presence of animal bones, especially sheep and cattle, in the final fill of these wells, as has been identified by Angelos Hadjikoumis in his study of the faunal material11? If we look at the published reports of land excavations conducted from 1977 until 1991 in the agora of Amathus by M. Loulloupis of the Cypriot Archaeological Service, and then from 2000 to 2004 by J.-P. Prète, I. Tassignon, T. Koželj and M. Wurch-Koželj, we will understand that the wells of the port are part of a much bigger series. In the report published in 2002, the excavators note two wells: the first in the west stoa12 appears on the plans in the form of a dovetail, recalling the central narrowing of the sakieh tank at the port. Certainly, the size is smaller (0.90 to 0.95 m by 1.90 m, and 0.55 m wide in the middle, taking the shape of a butterfly or a double dovetail, under the weight of the earth, as we have seen with the sakieh 3 on the beach), but there is little doubt that this is another sakieh well, dated on the map as “d’époque tardive” (late period) (fig. 15). The frame of a second well is indicated on the plan of the south stoa13, north-east of the building. The report notes that sakieh pots were retrieved in the body of this well. Already, during his excavations at this same place in 1985, M. Loulloupis had indicated unearthing another sakieh14. During the excavation campaign of 1989, M. Loulloupis mentions for the first time the presence of a tannery15. The following year, he is more positive: “The workshop, dyer’s or tannery, from the beginning of the Byzantine period, located some time ago has been completely excavated. Unfortunately, its south side has been destroyed by pillagers of stone and by the construction of a small bridge below the former Limassol-Nicosia road. We discovered two other small stone reservoirs with terracotta ducts for expelling water. In this workshop, the flooring of a small irregular-shaped room was lifted and we discovered beneath the slabs of the Agora courtyard. This is further proof that the workshop was built upon the remains of the Agora.16”

11.

Cf. volume 2.

12.

Cf. Prête et al. 2002, p. 561 with fig. 7a, p. 556 and fig. 10, p. 561: “Devant le passage dans le mur, se trouve un puits tardif B14, dont la fouille n’est pas terminée”.

13.

Further on, p. 567, with plan, fig. 12, p. 563: “Sur le devant de la stoa (= i.e. la stoa Sud, secteur C), mais plus à l’Est, nous avons fouillé un deuxième grand puits tardif qui, d’après le matériel qu’il contenait, devait être un puits à noria, noria où étaient accrochés des vases à large embouchure dont nous avons retrouvé des fragments. (…) Le fond du puits communique avec la nappe phréatique.”

14.

V. Karageorghis, “Chronique des fouilles et découvertes archéologiques à Chypre en 1985”, BCH 110 (1986), p. 865: “Le mur nord du bâtiment (en forme de temple) [= sans doute le nymphée] n’a pas encore été retrouvé, étant donné qu’un puits à roue et une citerne avaient été construits à cet endroit à une époque plus récente”.

15.

V. Karageorghis, “Chronique des fouilles et découvertes archéologiques à Chypre en 1989”, BCH 114 (1990), p. 970, mentions a tannery: “Du côté sud de l’agora, les fouilles sont arrivées à un niveau de l’ancienne route Limassol-Nicosie; (…) Selon toute vraisemblance, l’atelier serait une teinturerie ou une tannerie.”

16.

V. Karageorghis, “Chronique des fouilles et découvertes archéologiques à Chypre en 1990”, BCH 115 (1991), p. 818.

142

THE HELLENISTIC HARBOUR OF AMATHUS

THERMAE

Sector E

1 Sector G

EASTERN PORTICO

NORTHERN PORTICO

Sector A

Sector F APSIDAL BUILDING

Sector Da

BYZANTINE WALL

B-a

B-b EASTERN TEMPLE Sector D b SQUARE BUILDING

2

V

Sector B

B-c Eastern monumental entrance

B-d DOUBLE STOA B-e

Late Roman well

B-g

unexcavated area B-f

B-h

3 r cto

e

ll co

SOUTH STOA

Sector C

N Entrance street

θύρωμα BATHS

0

5

10

15

20 m

unexcavated area

Fig. 15 — Plan of the Amathus Agora, with in blue the sakieh wells, numbered in red 1, 2 and 3 (drawing and CGI EFA, T. Koželj)

We cannot, however, say much more about this workshop. One must await the publication of the author of this discovery as the vestiges have been dismantled in order to reveal more ancient phases. The best description is provided by P. Aupert (ed.), Guide d’Amathonte, 1996, p. 79: “The same applies for the house-workshop, for fullers or dyers, with hypocausts, reservoirs, built-in pithoi above curious cylindrical vases, all established at the beginning of the Byzantine period in the southern zone of the square and nowadays removed to reveal earlier phases17.” This description recalls the heating required for a tannery (hypocausts), the steeping (reservoirs, built-in pithoi) for mixing the 17.

Aupert 1996 indicates, p. 71, note 53 (this is not translated in the English edition, Aupert 2000): “À partir des informations communiquées par Michael Loulloupis, le fouilleur, directeur honoraire du Département des Antiquités de Chypre. Les données fournies ci-dessous sont provisoires, en l’attente de la publication.” Nevertheless, a part of these workshops may still survive in place, as would appear to be indicated in the report of the 2001 campaign in Zone A (north buildings) of the Agora, cf. Prête et al. 2002 p. 553: “Ce secteur, délimité au Sud par le ‘portique

THE WELLS

143

alum and then for rinsing. This polluting craft, which produced nauseating smells, was set up in a then-deserted district of ancient Amathus, where animals were slaughtered and cut up, as shown by the bones in the fill of the sakieh at the beach, and most probably by other finds noted here and there in excavation reports. While the date when tanning activity was abandoned can be fixed at the end of the 6th-beginning of the 7th century, it is more difficult to establish its beginning. Such a chronology could only be specified through an overall study of the installations of these house-workshops, with their hydraulic facilities, basins etc. For the time being, we can simply state that the oldest pottery from the wells dates to the end of the 4th century ad, as do the two coins found in Well 2, but we cannot with any certainty estimate the activity of these artisans as extending over two centuries. The elliptical shape of the sakieh wells was frequently used, alongside the classic square wells, where a simple vertical, sometimes with a counterweight, a shaduf, was sufficient to raise water in a bucket. The sakieh was infinitely more efficient than a well, and one can well imagine its use by craftsmen requiring large quantities of water for their activities. The sakiehs found at Amathus are most probably the oldest known on Cyprus. Their use would thereafter spread in the middle ages and continue into the modern era. An excavation at Potamia led to the examination of a sakieh in the Kastro of the Lusignans (end of the 14th century) and the exploration of a sakieh well installed near two water courses that supplied reservoirs and irrigation channels. The installation was dated to the 16th century. The author of the report notes that this sakieh belonged to a “type observed in numerous cases of noria in the region” 18. They are called ƆƯƥƮƠƷƭư, an Ancient Greek word that is now used only on Cyprus19, and in Greece as ƑƥƧƧƥưƲƳƢƧƥƨƲ, manganopigado, literally winch well. This ƥƯƥƮƠƷƭư model was very widespread until these days in Cyprus and it has left a trace in place names20. It is made of two teethed wheels, meshed together, one vertical with pots or cylinders and the other horizontal with a mortise above, into which a pole would be slotted in order to move the whole machinery by turning the pole above the vertical cog (fig. 16).

NORIA OR SAKIEH: DEFINING THE TERMS Norias and sakiehs are well known in antiquity and have survived up into modern times (fig. 17ab). This is not the place to write a history of such devices, but one might just recall the superb painted representation on a tomb wall in Alexandria’s western necropolis21, or mention the sakiehs discovered during excavations in the capital of the Ptolemies22 and in the Alexandrian chora. The

Nord’, au Nord-Ouest par une salle rectangulaire d’époque tardive (cf. secteur F) et au Nord par un mur de soutènement circonscrivant la zone d’ateliers byzantins, qui n’avait pas été fouillée par M. Loulloupis”. 18.

Cf. N. Lécuyer et alii, “Potamia” in “Chypre”, “Rapport sur les travaux de l’École française d’Athènes en 20032004”, BCH 128-129 (2004-2005), p. 1081-1082, with fig. 3 and 4.

19.

This very Cypriot word was loved by the poet Seferis. A very old word, dating back to Homer, in whose writings it means distaff, recalling the central axis of a winch well (‫ݟ‬ȜĮțȐIJȘ, The Distaff, see e.g. Odyssey 4.135).

20.

A district to the west of Nicosia, a village near Paphos.

21.

M. Rodziewicz, “Painted Narrative Cycle from Hypogeum No. 3 in Wardian, Alexandria”, BSAA 45 (1993), p. 281-290. The date is still disputed: from the 2nd century bc to the 3rd ad.

22.

Th. Gonon, “Puisage de l’eau: la sakieh de Terra Santa”, in Hairy 2011, p. 408-419.

144

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 16a-b — Modern Alakatin (=sakiehs) in Cyprus (photograph, Internet)

Fig. 16c-d — Modern Manganopigado (=sakiehs) in Pigadia, Karpathos. The broken path for the animal can be seen on the right and the machinery with the two wheels on the top. On 16d, the chain with the buckets are still in place (photograph J.-Y. Empereur)

advantage of excavations in a rural setting is the better preservation of the ensemble and especially of the circular path where the animal turned while working the machine23. One should define the terminology, which in current usage can appear ambiguous. Noria is both employed as a generic term signifying all machines for lifting water, and in a more specific way as a wheel designed to lift water, using pots or drums, propelled by the force of running water. As for sakieh, however one chooses to transcribe it (saqieh, sakia, sakeya etc.), the word designates a machine for lifting water by means of one or two vertical wheels coupled with a horizontal wheel usually driven by animal power, whether ox, donkey or camel24. This term is particularly, but not exclusively, applied to Egyptian devices. 23.

V. Pichot, in “Les actions du Centre d’Études Alexandrines”, Rapport Ifao 2014-2015, p. 278-280 with fig. 159161 (consulted at www.ifao.egnet.net, 27/3/2017).

24.

I. Hairy, “Les machines de l’eau en Égypte et à Alexandrie”, in Hairy 2011, p. 554-575 and especially on the difference between sakieh and noria, p. 562-566 related illustration.

THE WELLS

145

Fig. 17a — A sakieh painted on the wall of a hypogeum, Wardian, West Necropolis of Alexandria. Hellenistic to Roman Period (photograph CEAlex, A. Lecler)

There is little doubt that the “norias” discovered in the agora of Amathus and called thus by the excavators should be recognised more specifically as sakiehs. We can propose here two possible reconstructions (fig. 18a-b): the first, with an animal turning a horizontal wheel whose cogs mesh into those of a vertical wheel fitted with pots. The lack of archaeological evidence around the wells keeps us from excluding a second version, with two vertical wheels joined by an axle. However, comparisons with the traditional machines of Cyprus lead us to prefer the first solution, which is the easiest to set up25. 25.

We thank Valérie Pichot for the enlightening discussions regarding the reconstruction of the sakieh of Amathus harbour.

146

THE HELLENISTIC HARBOUR OF AMATHUS

The elements of the sakieh The sakieh pots Ceramic pots were attached by rope to the horizontal cross-pieces of the wooden wheel: when the wheel turned they scooped up water from the bottom of the well and then poured the water out at a higher position into an outlet channel, thus raising the water to a height corresponding to the diameter of the wheel, estimated in our case as being around 5 m. These receptacles held about 2 litres and were shaped in a specific form: without handles, with a strongly projecting rim and a pronounced button foot in order to attach ropes to either end. The shape of these sakieh pots was unchanging and makes dating difficult. These receptacles break often and the fragments fall to the bottom of the well. In the bottom of the tank of Well 3, we recovered 257 fragments of sakieh pots as well as two complete examples (see volume 2, chapter on Late Roman Pottery). As for their name, we have chosen sakieh pot, though I have sometimes come across the term “sāqiya pot” (Donald Bailey) and “sāqiyah pot” (R. Tomber), as well as, more often, the Arabic word qadus.

Local Sakieh pot from Well 3. Ht. 24; Max. L. 13.8; Diam. 13. C43, see volume 2, chapter on Late Roman pottery. Drawing S. Hartmann, EFA archives.

Wood from the sakieh Wood from the sakieh was found thanks to its remarkable preservation in a compact layer of mud that had protected it from the marine fauna and flora, especially wood-boring shipworm. Conservation. In 1985, there were hardly any methods for the conservation of wood. As an experiment we had thought of using glutanol or even freeze-drying, but such costly measures were beyond our means. Thus, it was decided to treat the wood by simply soaking it in baths of progressively less salty water, so that the wood would slowly rid itself of salt entirely. Natural desiccation then preserved the wood up until today. Thanks to the attention of Yiannis Violaris, these pieces are now held in the special laboratory for underwater finds in Larnaka District Museum The species of wood were identified by Brita Lorentzen and Sturt W. Manning, whose study can be read in volume 2. They showed that the wheel was made of local wood, probably cypress, which can be found not far away. They measured the 14C of the fragments in the laboratory at Cornell and were able to date them to the 7th century ad, which concurs with the study of the ceramic and amphora material.

THE WELLS

147

The half-a-dozen fragments of wood revealed by the underwater excavation belong to two types: 1. straight type, which may correspond to the spokes of the wheel; a preserved fragment measures 0.75m. 2. curved type, which may belong to the rim of the wheel. From the size of the tank (6 m long), one can conclude that the wheel would have had a diameter in the neighbourhood of 5 m.

Piece of wood from the sakieh wheel (photograph EFA, T. Koželj)

The shaduf The two square wells were probably fitted with a counterweighted system, or shaduf, such as is known from the beginning of the second millennium bc in Mesopotamia, in Egypt, and then Greece and China. The system is simple, consisting of a pole pivoting on a support with a counterweight – such as the pierced stones found near the wells – which can lift a bucket or perhaps a sakieh pot, six fragments of which were found at the bottom of Well 2. We found numerous pierced stones during the excavation. The majority were interpreted as anchors, which is normal in a port, but it could be that some of them found closer to the wells were the counterweights, such as the one weighing 26 kg.

0

5 cm

Pierced stone lying on a block from the well near where it was found. Shaduf counterweight? Max. W. 42; Hole Diam. 7; Weight 26 kg. Drawing and photo T. Koželj, EFA archives.

148

THE HELLENISTIC HARBOUR OF AMATHUS

Fig. 17b — Modern sakieh in operation at Dendera (Egypt). Photo R. Collet, author’s archives.

THE WELLS AND THE HISTORY OF THE PORT The depth of the wells is relatively constant. Their highest parts vary between -0.73 m to -0.84 m (Well 1), -0.66 to -0.73 m (Well 2), and +0.52 m (Well 3), the maximal spread being 1.36 m, but this difference is countered by the height of the tank, given that the level of their foundations are practically the same: Well 1 = -2.21 to 2.32 m; Well 2 = -2.01 m; Well 3 = -2.10 m. These fresh-water wells were dug on dry land, not in the sea, which would mean that the port was at least partially out of service at this time. The period of their construction dates to the 4th century. They thus date to a period when the harbour basin was dry, at least in this part. These three wells were filled with late Roman material. Abandonment probably precedes the Arab invasion of 649 (too late for the pottery chronology), but the cause is not to be explained by a historical event, but rather by a natural phenomenon, such as a rise in sea level, which would have made the water brackish. This was followed by a swift filling-in, when pottery was probably thrown in all at one moment as is shown by the similarity between the shards at the bottom and those at the top of one of the wells26. The pottery demonstrates that all the wells must have been tainted together by the seawater and filled in at the same time.

26.

Cf. chapters on pottery, volume 2.

THE WELLS

149

Fig. 18a-b — Reconstruction of the beach sakieh with two solutions: a = with 2 wheels, one vertical, one horizontal; b = with 3 wheels, two vertical, one horizontal. The first proposal is the most probable, looking like the traditional Cypriot alakatin (drawings and CGI EFA, T. Koželj)

Bibliographic references On sakiehs, see, L. Ménassa, P. Laferrière, La saqia, technique et vocabulaire de la roue à eau égyptienne, Ifao, Cairo, 1974. Th. Gonon, “Puisage de l’eau: la sakieh de Terra Santa”, in Hairy 2011, p. 408-419. - On sakieh pots/qadus, see D. Dixneuf, “L’eau versée: les godets de noria”, ibidem, p. 420-421. - On shadufs, see I. Hairy, “Les machines de l’eau en Égypte et à Alexandrie”, ibidem, p. 554-575 and especially p. 556-559. See also the film of Hugues Fontaine, La roue and on www.cealex.org the film by Raymond Collet, La dernière sakieh du Fayoum, 2017.

The coins found in Amathus harbour Olivier Picard

The 43 coins, all of them bronze, discovered by the excavation team of the harbour of Amathus during the three campaigns of 1984 to 1986 are in a very poor state of preservation. This is most probably explained by the long period spent in a saline environment. Identification of the coins was not straightforward, as the state of preservation could vary from extremely poor (Cat. no. 42) to quite good (Cat. no. 18). They can be divided up into five types of coinage: – two chalkoi from the cities of Paphos and Rhodes, dating to the 4th century bc – nine bronzes of Alexander type and six of the Antigonids, up to the recapture of Cyprus by Ptolemy 1, around 295 bc – twenty-three Ptolemaic bronzes, struck in Egypt and (only four) in Cyprus, up to the reign of Cleopatra VII – one bronze of the imperial period, Antonine or Severan era – two Roman bronzes of the 4th century ad Nota bene: All the coins are reproduced at the scale 1:1

THE CHALKOI OF THE CITIES Paphos Obverse: Head of Aphrodite to left, wearing a crown. Reverse: Dove to right, Cypriote syllabary signs above. C1. AM 1471: 10 mm Sondage 14 - 08/ X /86 - 86.14.88.1

The dove on the reverse confirms identity of the piece. I. Nicolaou, The coins from the House of Dionysos, Paphos II, Nicosia (1990), nos. 1-3, p. 6. A. Destrooper, “Le monnayage de Paphos au ive siècle, nouvelles perspectives”, in Proceedings of the XIIIth International Congress of Numismatics in Madrid 15-20 September 2003, Madrid (2005), p. 245-252, nos. 246-8, fig. 5-8.

152

THE HELLENISTIC HARBOUR OF AMATHUS

Coins of Cyprus from the Collection of the Bank of Cyprus Cultural Foundation, Nicosia (2008), nos. 18-20. E. Markou, Coinage and History, The case of Cyprus during the Archaic and Classical periods, Nicosia (2015), p. 113.

Rhodes Obverse: head of Rhodos to right. Reverse: Rose with bud to right. R – O on either side of stalk. Issue mark to left? C2. AM 1469: 11 mm Sondage 14 - 07/X/86 - 86.14.84.1

The orientation of the head, which is certainly to the right, and the absence of the kalathos of Aphrodite in the characteristic form, prevents us from recognising here a coin of Paphos and confirms the attribution of this chalkous with rose to Rhodes. The presence of the bud and the ethnic allow attribution of the piece to the second half of the 4th century bc. The small number of these chalkoi gives a good idea of the weakness of the bronze mint in the Cypriot kingdoms before being taken over by Alexander. R. Ashton, “The coinage of Rhodes, 408 – c. 190 bc”, in A. Meadows & K. Shipton (eds.), Money and its uses in the ancient Greek world, Oxford (2001), p. 79-115.

FROM ALEXANDER TO THE ANTIGONIDS The minting of bronze coinage in one or more workshop on Cyprus from the last years of Alexander’s reign and immediately afterwards, up until the moment when it is replaced by coinage of the Antigonids is directly reflected in the finds. There is no good study of these bronzes and the very damaged state of our pieces will not add very much to our knowledge. All of the pieces are types showing Heracles’ head and the hero’s weapons. However, the usual means of describing these are varied, if not incoherent, resulting from a lack of agreement on the interpretation of these types. I consider that in the analysis, the type is of more importance than the legend, which is subordinate to it. Contrary to the usual practices of modern numismatists, and unlike other workshops, for example that of Thasos, it is most likely that the bow and club, as well as the legends, are represented vertically. (cf. M.-Chr. Marcellesi, “Les monnaies grecques et provinciales romaines”, in O. Picard et alii, Les monnaies de fouille du Centre d’Études Alexandrines, ÉtAlex 25 (2012), p. 173-177 [hereafter: Bronzes à Alexandrie]) The pieces can be divided into chalkoi, dichalkoi or quarter obols, and hemiobols (not among these finds), within a system of eight chalkoi to the obol.

Heracles type bronzes Chalkous: 9-13 mm Obverse: Heracles’ head to right wearing lion skin. Reverse: Heracles’ weapons: bow to right, wood to the exterior, most often in a gorytos (bow-case), sometimes with quiver (arrow-case) which is difficult to distinguish here, and club, handle to the top, to left; legend variable.

THE COINS FOUND IN AMATHUS HARBOUR

C3. AM 1476: 10 mm Sondage 14 - 9/ X /86 - 86.14.95.1 ƆƐƊƓƆƒƉƖƔƙ between the bow and the club

C4. AM 1448: 9 mm Sondage 15 - 16/ IX /86 - 86.15.10 ƆƐƊƓƆƒƉƖƔƙ under the club

C5. AM 1466: 13 mm Sondage 14 - 3/ X /86 - 86.14.76 Legend indiscernible

C6. AM 1477: 10 mm Sondage 14 - 10/ X /86 - 86.14.96.1 Legend indiscernible

153

Dichalkous: 16-18 mm Obverse: Heracles’ head to right wearing lion skin. Reverse: Bow in gorytos, club. ƆƐƊƓƆƒƉƖƔƙ C7. AM 1442: 17 mm Sondage 2 - 5/ IX /86 - 86.2.1

C8. AM 1454: 18 mm Sondage 14 - 24/ IX /86 - 86.14.51 Monogram illegible

C9. AM 1470: 17 mm Sondage 14 - 7/ X / 86 - 86.14.84.2

C10. AM 1474: 18 mm Sondage 14 - 8/ X /86 - 86.14.89.2

B A between bow and gorytos C11. AM 1451: 12 mm Sondage 14 - 23/ IX /86 - 86.14.43

It is likely that the great majority, if not all, of these pieces were struck in a Cypriot workshop. As has been pointed out by H. Troxell, Studies in the Macedonian coinage of Alexander the Great, NumStudies 21 (1997), p. 97-98, and G. Le Rider, Alexandre Le Grand, Monnaie, finances et politique (2003), p. 99-100, the legend ƆƐƊƓƆƒƉƖƔƙ is earlier than that giving the title ƇƆ ƶƣƯƩƼƵ , which was probably adopted shortly after the death of the great man.

154

THE HELLENISTIC HARBOUR OF AMATHUS

Bronzes types with Macedonian shield and helmet Obverse: Macedonian shield with gorgoneion in the centre. Reverse: B A to either side of a Macedonian helmet with crest and cheek-guards. Symbol or letter below. Chalkous C12. AM 1460: 12 mm Sondage 14 - 29/ IX /86 - 86.14.63 Issue marks illegible

Dichalkous C13. AM 1458: 17 mm Sondage 14 - 29/ IX /86 - 86.14.61 Issue marks illegible

C14. AM 1462: 16 mm Sondage 14 - 1/ X /86 - 86.14.69

The presence of the gorgoneion in the centre of the shield connects these pieces to a well-attested group on Cyprus, as has already been noted by E.T. Newell, The Coinages of Demetrius Poliorketes, London (1927), p.18-20, and confirmed by the excavations of Kourion and of Salamis. Two small hoards have been published by B. Helly, “Les monnaies”, in V. Karageorghis (ed.), Excavations in the Nekropolis of Salamis, III, Nicosia (1970), p. 250, and by O. Callot, Salamine de Chypre XVI, Les monnaies (2004), p. 13-14. It is agreed that the mint was located in Salamis, cf. K. Liampi, “Zur Chronologie der sogenannten anonymen makedonische Münzen des späten 4. Jhs. v. Chr.”, JNG 36 (1986), p. 41-65. Liampi, p. 50, notes the existence of two values, which she interprets as halves and quarters. However, to see here chalkoi and dichalkoi would underline continuity with Macedonian coinage. Most commentators date this coinage to the years 306-301, after the recapture of Cyprus by Demetrios Poliorketes. The find context did not provide sufficiently precise indication for a definitive date. Liampi, however, has shown that minting this coinage began quite soon after the death of Alexander and that production was very large. D.H. Cox, Coins from the Excavations at Curium, 1932-1953, Numismatic Notes and Monographs 145 (1959), p. 4-5 and 91-3, places the introduction to the initial domination of Cyprus by Antigonos the One-Eyed, before the total conquest of the island by Ptolemy in 311 bc, and this hypothesis is worthy of attention.

Bronzes of Demetrios Poliorketes Chalkous: 12 mm Obverse: Head to right of a warrior wearing a Corinthian helmet. Reverse: Prow to right, BA above, monogram below.

THE COINS FOUND IN AMATHUS HARBOUR

155

C15. AM 1473: 12 mm Sondage 14 - 8/ X /86 - 86.14.89.1 Monogram illegible

Dichalkous: 16-18 mm C16. AM 1461: 16-17 mm Sondage 14 - 30/ IX /86 - 86.14.65

C17. AM 1478: 18 mm Sondage 14 - 10/ X /86 - 86.14.88.1 Monogram illegible

The type with prow, that features on silver coins with the name of Poliorketes struck after the defeat at Ipsos, ensures the attribution of these bronzes. One might wonder whether the warrior on the obverse represents the king himself. Whatever the case, it is generally recognised that they were struck at Salamis, which Demetrios held until 294 bc, when Ptolemy I recaptured Cyprus.

PTOLEMAIC COINAGE With 23 pieces, Ptolemaic bronzes count for more than the half of the material discovered. It is, however, noteworthy that only four were definitely struck on Cyprus, while the other 19 seem to have come from Egypt, and I have classified them according to the system proposed in Bronzes à Alexandrie. It is true that with examples so worn and oxidised, when the types and modules are close, the distinction between Cyprus and Alexandria is often difficult to establish.

Alexandria workshop Series 02 Hemiobol

Obverse: Head to right of Alexander with short hair wearing the mitra and ram’s horns. Reverse: Standing eagle to left, wings spread. ƆƐƊƓ. Possibly symbol or monogram. and helmet, Bronzes à Alexandrie, nos. 56-59. C18. AM 1450: 17 mm Sondage 14 - 11/ IX /86 - 86.14.14

C19. AM 1465: 13 mm Sondage 14 - 7/ X /86 - 86.14.73

156

THE HELLENISTIC HARBOUR OF AMATHUS

C20. AM 1457: 20 mm Sondage 14 - 29/ IX /86 - 86.14.61

C21. AM 1475: 17 mm Sondage 14 - 8/ X / 86 - 86.14.91.1 Bronzes à Alexandrie, no. 60-61

C22. AM 1404: 20 mm Sondage 1 - 6/ IX /85 - 85.19

A

Issue mark illegible, Bronzes à Alexandrie, no 31-59 C23. AM 1405: 21 mm Sondage 1 - 6/ IX /85 - 85.19

C24. AM 1444: 21 mm Sondage 3 - 5/ IX / 86 - 86.3.1

C25. AM 1452: 13 mm Sondage 14 - 24/ IX /86 - 86.14.49

C26. AM 1455: 15 mm Sondage 14 - 24/ IX / 86 - 86.14.52

C27. AM 1459: 17 mm Sondage 14 - 24/ IX / 86 - 86.14.62

THE COINS FOUND IN AMATHUS HARBOUR

157

Diobol C28. AM 1472: 27 mm, fragment Sondage 14 - 8/ X / 86 - 86.14.88.2

The diameter of the piece and the shape of the blank, cast in a shallow depression with bevelled-cut walls, correspond exactly with the diobols of Series 2 and Series 3. The difference between the two is in the reverse type and the central cavity, but the state of preservation of the surface does not allow any precision.

Series 05, 1st phase Module 7 (triobol)

Obverse: Head of Zeus Ammon to right wearing ram’s horns. Reverse: Standing eagle to left, wings spread, head facing forward, horn of plenty to left. ƕƘƔƐƊƑƆƎƔƙ ƇƆƗƎƐƊƝƗ from left to right. C29. AM 1403: 35 mm Sondage 1 - 6/ IX / 85 - 85.19 Central cavity, Bronzes à Alexandrie, p. 53

Series 05, 2nd phase Triobol

Obverse: Head of Zeus Ammon to right wearing ram’s horns. Reverse: Standing eagle to left, head turned backwards, wings closed, horn of plenty on shoulder. Issue mark between talons. ƕƘƔƐƊƑƆƎƔƙ
ƇƆƗƎƐƊƝƗ from left to right.

158

THE HELLENISTIC HARBOUR OF AMATHUS

C30. AM 1449: 34 mm Sondage 18 - 1/ X / 86 - 86.18.2 , Bronzes à Alexandrie, nos. 364-365

C31. AM 1406: 32-33 mm, 24g Sondage 1 - 16/ X / 85 - 85.285 Issue mark illegible

Series 09: blank with casting tangs - central cavity 40 units

Obverse: Head of Zeus Ammon to right wearing ram’s horns. Reverse: Two standing eagles to left on a ground line, wings closed, horn of plenty in front. ƕƘƔƐƊƑƆƎƔƙ ƇƆƗƎƐƊƝƗ from left to right. C32. AM 1441: 18 mm Sondage 12 - 4/ IX / 86 - 85.12.1

C33. AM 1443: 17 mm Sondage 3 - 5/ IX / 86 - 86.3.1

C34. AM 1445: 17 mm Sondage 3 - 5/ IX / 86 - 86.3.1

C35. AM 1453: 17 mm Sondage 14 - 24/ IX / 86 - 86.14.49

C36. AM 1456: 21 mm Sondage 14 - 25/ IX / 86 - 86.14.55

THE COINS FOUND IN AMATHUS HARBOUR

159

Paphos workshop Pentadrachm (?) Obverse: Head of Zeus Ammon to right with laurel. Reverse: Zeus standing to left, head to right, sceptre in left hand and ears of grain in the right held against the body. C37. AM 1446: 13 mm Sondage 3 - 5/ IX /86 - 86.3.1

C38. AM 1447: 17 mm Sondage 12 - 8/ IX /86 - 86.12.5

C39. AM 1464: 11 mm Sondage 14 - 2/ IX / 86 - 86.14.73

C40. AM 1467: 13 mm Sondage 14 - 3/ X / 86 - 86.14.78

The weight, diameter and the irregularity of the blank correspond to those pieces collected by I. Nicolaou, The coins from the House of Dionysos, Paphos II, Nicosia (1990), nos. 469-509, p. 58-62, pl. XVI. These might be close to the pentadrachm pieces struck at the same period in Alexandria, Bronzes à Alexandrie, nos. 872-1007, p. 100-104.

CITY OF THE ROMAN EMPIRE Termessos of Pisidia Obverse: Bust to right of the hero Solymos with beard, wearing a breastplate and Corinthian helmet with crest . Reverse: The hero Solymos wearing a knee-length tunic, standing face-on, head to left, right leg vertical, left leg bent, holding a lance to the right and sword to left. ƋƔƒƝƒ circular from left to right. C41. AM 1463: 22 mm Sondage 14 - 2/ X /1986 - 86.14.72

SNG Copenhagen, Pisidia, Termessos Major, 315-318: from the Antonines to Gallienus. SNG von Aulock, III (1964), 5342: mid 2nd–3rd century ad.

160

THE HELLENISTIC HARBOUR OF AMATHUS

BRONZES OF THE LATE EMPIRE These two pieces were found in well no. 2 of the late imperial period.

Constantius II (348-361) Cast bronze Obverse: Diademed, draped and cuirassed bust of the emperor to left. Beading. Reverse: Virtus standing to left, shield on left arm, spearing a horseman who falls with arms raised to the skies. Beading. [FEL TEMP REPARATIO]. Exergue illegible. C42. AM 1395: 23 mm Late Roman well 2 / 84 84.3529

Arcadius (395-401), Antioch (?) Centionalis Obverse: Diademed, draped and cuirassed bust of the emperor to right. Legend illegible. Reverse: Emperor standing face-on, head to right, holding a lance in right hand, resting left hand on a shield; to his right, standing Victory crowns him. Beading. [VIRTVS – EXERCITI]. On the exergue, ANT[…]? C43. AM 1396: 17 mm Late Roman well 2 / 84 84.3529

Conclusion Amathus harbour, a remarkable and enigmatic monument Jean-Yves Empereur

In 1877, a German traveller descended from the Troodos Mountains towards the sea and passed by Amathus. He described his arrival at the sunken harbour: “… here at my feet the ancient harbour of Amathus, of which nothing remains but its natural basin, formed by rocks which extend some distance into the sea1”. The harbour basin was clearly visible to his eyes, but he reckoned that it was bounded by natural rock. Indeed, how could one doubt that these dark masses covered with seaweed and just breaking the surface of the Mediterranean were not natural formations, but rather the work of man on a grand scale? In fact, these “rocks” need to be observed underwater in order to see that beneath the posidonia they are cut blocks placed systematically by human hand.

A REMARKABLE STATE OF PRESERVATION The archaeological excavations led us to this observation: the harbour of Amathus is so remarkably well preserved as to be quite exceptional. It appears at roughly 1 m below the surface of the sea, but as one explores it at depth, courses of almost intact blocks are revealed, protected by the mud that had quite clearly filled the harbour basin rapidly, avoiding any erosion to these stones which, as we can see in the photographs in this book, have never suffered from contact with the marine environment. One of the major features of Amathus harbour is that after having spent more than 24 centuries beneath the Mediterranean it now reveals the freshness of its original state. Below the uppermost course of blocks, eaten away by the fauna and flora, especially the ubiquitous posidonia, the archaeological divers were astonished to find perfectly intact blocks that appeared to have just arrived from the nearby quarries a few dozen metres away2. This very appearance suggests that the port has a strange history, and one might even ask the question: did it truly have a history? The homogeneity of the ensemble of structures demonstrates that it was constructed at one go, without stops and starts and without repairs, at a sustained pace 1.

Franz von Löher. Cyprus: Historical and Descriptive (1878), p. 344.

2.

No signs of lithophaga: the bivalve molluscs that bore into limestone have not touched these blocks, given the rapid protection provided by the silting process.

162

THE HELLENISTIC HARBOUR OF AMATHUS

that led to its completion in a few years, no more. One can only conclude that it was abandoned rapidly and then wonder as to the reasons. Were these political, man-made or perhaps due to one or more natural phenomena? Natural causes. There could have been several natural causes or even a combination of several factors: 1. Unforeseen silting? The narrowness of the entrance channel due to its naval role might explain a miscalculation by the builders, since the art of harbour engineering, involving the movements of water, currents and the avoidance of silting, is a difficult discipline to master. More than one modern port has suddenly found itself silted up and transformed into a sandy beach. The open channel revealed in the middle of the South Mole seems rather long at 18 m and probably too high to play the efficient flushing role that one might want to ascribe to it. 2. A change in sea level, with a reduction in draught that made the harbour basin inaccessible to shipping? Such a movement is perhaps observable in the appearance of the lines of beachrock and the installation of freshwater wells in the north of the harbour basin. Can geomorphology explain a phenomenon that we can only suggest? Might we be witnessing a drop in sea level and/or the land and seabed rising? 3. Were the moles badly set upon the sea bottom and did they sink under the weight of the blocks? Given that the courses of headers were placed directly on the sand, did the harbour suffer from a sinking phenomenon: can we talk here of subsidence? But if this was the case, then might one not have added extra courses of blocks to counter this effect? Were such repairs made impossible by the abandonment of the harbour? Is the submersion of the moles related to the flooding of the coastal quarries by the tide3? Man-made causes. Rather than geophysical causes, was the harbour abandoned for political reasons? The discovery of several coins of Antigonus Monophthalmus and of his son Demetrius Poliorcetes have prompted us to attribute construction to these two, which is also confirmed by the abundant pottery from the end of the 4th and beginning of the 3rd centuries bc. As we have noted, these fragments of pottery, found at different depths but connected with each other, point to one single phase: they do not indicate a lengthy chronological use of the harbour but very much the contrary. When Ptolemy I reconquered Cyprus in 294 bc, did he close and fill in this naval port, which he considered as a direct threat to Alexandria, just as on land he destroyed the palace, ending all signs of Antigonid occupation4? Perhaps the two phenomena, natural causes and political decisions, combined to create an unexpected fate, such that this grand-scale endeavour was destined never to be used. Further development of geomorphological research and a resumption of excavations would most probably provide answers to these questions.

LOOKING BACKWARDS AND FORWARDS. AN OVERVIEW The excavation of Amathus harbour took place between 1984 and 1986, more than 30 years ago. We were equipped with the best possible tools of the time: a total station for topography, a photogrammetry plotter, regular helicopter flights for aerial photography, a large well-trained team and all the technical means that we needed (suction hoses, air pipes etc.). Our divers were experienced in difficult underwater conditions and tasks, whether removing the thick carpet of posidonia and tons of rocks and pebbles or manipulating the balloons full of compressed air that were used to shift the 3.

B. Bousquet and P.-Y. Péchoux, “La géographie du site. Premiers jalons”, in P. Aupert, M.-Chr. Hellmann 1984, p. 121-146 and especially on the sea level p. 145-146.

4.

See Aupert 1996, p. 53-54.

CONCLUSION. AMATHUS HARBOUR, A REMARKABLE AND ENIGMATIC MONUMENT

163

largest of blocks, an operation that was all the more complicated because of the shallow depth. The photographs, drawings, plans and sections are testimony to the enormous amount of work achieved during the three campaigns, amounting to five months of excavation. Now, 30 years after, we must ask ourselves whether the mission was accomplished.

Aerial and underwater photogrammetry Technological developments of recent years have included new underwater survey techniques that are infinitely more precise than hand-drawn work, and this precision is married to remarkable speed. Of course, the procedure is old and was already in practice on the Madrague de Giens shipwreck in the 1970s5. However, at the time it required all sorts of installations for underwater photography, with a framework of metal bars along which a conventional film camera in a waterproof housing could slide, while taking care that the photos were taken absolutely parallel to the lens, and then were sent to the plotter, which filled a whole room and was manned by a specialised operator (a full-time job), for several months6. Photogrammetry is now used systematically on underwater excavation sites, and automatic pixel matching of overlapping photos using free access processing software has led to a great gain not just in time but also and especially in the accuracy of measurement and in georeferencing, such that we have even been obliged to recommence all the surveying of a site like the sunken Pharos of Alexandria that began in 1994. Another advantage lies in the acquisition of images in 3D, which, through the massing of data, can provide a global view of a building, whatever its size. We can thus produce an overall image of the ensemble of the underwater site of the Pharos across its surface area of 1.3 hectares. This method could well be applied to Amathus harbour, at least to obtain a more precise plan than that which we reproduce here. One could envisage an underwater photographic coverage along the 430 m total length of the three moles, despite the shallow depth. This would require an increased number of photos, but the procedure might be simplified and automated using a ROV. Given the clear water, one might also imagine an aerial photographic coverage using a drone in calm weather when the sea is flat. On the other hand, as regards the elevations, it would be difficult to improve upon the results presented here. If one wished to apply modern survey techniques, it would be necessary to clear once again the tons of rock that we lifted from the sondages conducted along the sides of the moles during our excavations. The sea currents have filled in the cavities: one would have to do it all again.

Geomorphology and Geophysics Outer basin The application of modern geomorphology would almost certainly lead to decisive developments in our understanding of the outer harbour basin, through an examination of variations in sea level, the amplitude and the chronology of these movements, the speed of formation and the dating of the beachrock, as well as a study of phenomena due to seismic activity and any subsequent subsidence. Hydrological studies would also be appropriate, with an examination of the extent of sedimentation

5.

See A. Chéné, “Techniques spéciales”, La photogrammétrie en archéologie, Dossiers de l’archéologie 13 (1975), p. 104107; A. Tchernia, P. Pomey, A. Hesnard, L’épave romaine de la Madrague de Giens (Var), Gallia Suppl. 34 (1978), p. 11-12 and pl. XXXVII-XXXVIII.

6.

The time required for the reconstruction of a shipwreck like the Madrague de Giens by A. Carrier.

164

THE HELLENISTIC HARBOUR OF AMATHUS

from nearby watercourses that would have contributed either occasionally or continually to the silting up of the harbour basin. Inner basin In order to explain the ensemble of the harbour installations at Amathus, it would be essential to apply geophysical survey techniques to discover the limits of the inner basin and determine the nature and date of this basin’s use, as well as its relationship with the outer basin and the possible junction point. It would also be important to explore the edge of this basin to see whether this was simply a beach or a constructed layout with an access to the agora of the city. The harbours of Amathus One of the big questions to be asked concerns the nature and site of the Archaic and Classical harbour, whose existence is attested by the many terracotta models of ships7. Was this the inner basin, as one might reasonably suppose, or should one look elsewhere, which seems less likely? One should also try to locate the Hellenistic and Roman era harbour. Perhaps it was situated east of the ancient port, directly south of the agora. The Late Roman wells We excavated three Late Roman era wells on the beach to the north of the sunken harbour. As indicated on the plan (fig. 1, p. 16, Well 4 to Well 6), we located three others outside the harbour, to the east. We did not excavate these latter, having concentrated our efforts inside the harbour. These wells are filled with Late Roman archaeological material and there is potential future interest here. We shall present pottery and amphorae from the first three wells in the second volume and we have touched upon a first ensemble of faunal material, which is a pioneering study for this era on Cyprus. Further examination of the unexplored wells would help to improve knowledge of the fauna and of the activities of the inhabitants of the era, most probably tanners with hydraulic installations connected in some way with the agora of the town. Should a new excavation be launched, this would be yet another episode in the discovery of what this magnificent construction, unique of its kind, hides beneath the waters of the Mediterranean. We hope that the work of archaeologists to come will increase our knowledge of this harbour, will explain and protect it better and will help draw the attention of future generations.

7.

See supra p. 126 and p. 131, fig. 5 and 6.

Bibliography of the excavations of the Amathus harbour

Chronological order

AUPERT 1979 P. Aupert, “Les activités sur le terrain. Le port d’Amathonte”, BCH 103 (1979), p. 725-733. EMPEREUR 1985 J.-Y. Empereur, “Le port d’Amathonte”, BCH 109 (1985), p. 984-989. KARAGEORGHIS 1985 V. Karageorghis, ARDAC 1984 (1985), p. 53-54. EMPEREUR, VERLINDEN 1986-1 J.-Y. Empereur, C. Verlinden, “Le port d’Amathonte”, BCH 110 (1986), p. 899-907. EMPEREUR, VERLINDEN 1986-2 J.-Y. Empereur, C. Verlinden, “Le port antique d’Amathonte à Chypre”, Archéologia 215 (1986), p. 32-37. KARAGEORGHIS 1986 V. Karageorghis, ARDAC 1985 (1986), p. 48-50. EMPEREUR, VERLINDEN 1987-1 J.-Y. Empereur, C. Verlinden, “Le port d’Amathonte”, BCH 111 (1987), p. 755-759. EMPEREUR, VERLINDEN 1987-2 J.-Y. Empereur, C. Verlinden, “The underwater excavation at the Ancient port of Amathus in Cyprus”, International Journal of Nautical Archaeology 15-4 (1986), p. 1-13.

166

THE HELLENISTIC HARBOUR OF AMATHUS

KARAGEORGHIS 1987 V. Karageorghis, ARDAC 1986 (1987), p. 49-51. ALABE 1988 Fr. Alabe, La céramique non amphorique sub-classique du port d’Amathonte, Essay presented to the AIBL, Paris, (1988), 104 pages, typed document. EMPEREUR 1995 J.-Y. Empereur, “Le port hellénistique d’Amathonte”, in V. Karageorghis, D. Michaelidis (ed.), Proceedings of the International Symposium “Cyprus and the Sea” (1995), p. 131-138. EMPEREUR 1996 J.-Y. Empereur, “Le port d’Amathonte”, in P. Aupert (ed.), Guide d’Amathonte (1996), p. 164-168 (also English and Greek version).

General bibliography

ARNAUD P. Arnaud, Les routes de la navigation antique, Itinéraires en Méditerranée (2005). AUPERT 1996 P. Aupert (dir.), Guide d’Amathonte, ÉFA, Coll. Sites et Monuments 15, Athens - Paris (1996). AUPERT 2000 P. Aupert (ed.), Guide to Amathus, The Bank of Cyprus Cultural Foundation, Nicosia (2000). AUPERT, HELLMANN 1984 P. Aupert, M.-Chr. Hellmann, Testimonia, Auteurs anciens, monnayage, voyageurs, fouilles, origines, géographie, Amathonte 1, Paris (1984). AUPERT, BALANDIER en préparation P. Aupert, Cl. Balandier, avec la collaboration de P. Leriche ; relevés et dessins : Tony Koäelj, Les fortifications d’Amathonte. L’enceinte urbaine, les défenses tardives, les constructions associées et avoisinantes, Études chypriotes (en préparation). BASCH 1987 L. Basch, Le musée imaginaire de la marine antique, Athens (1987). BLACKMANN, RANKOV 2014 D. Blackmann, B. Rankov, Shipsheds of the Ancient Mediterranean, Cambridge (2014). BLANDIN, PETIT, TASSIGNON 2008 B. Blandin, Th. Petit, I. Tassignon, “Fouilles récentes au Palais d’Amathonte (2004-2007)”, Cahiers du Centre d’Études Chypriotes 38 (2008), p. 129-142. BOUSQUET, PÉCHOUX 1984 B. Bousquet et P.-Y. PÉCHOUX, “La géographie du site. Premiers jalons”, in Aupert, Hellmann 1984, p. 121146.

168

THE HELLENISTIC HARBOUR OF AMATHUS

GILLE 1965 P. Gille, “Les navires à rames de l’Antiquité, trières grecques et liburnes romaines”, Journal des Savants (1965), p. 36-72. GINOUVÈS, MARTIN 1985 R. Ginouvès, R. Martin, Dictionnaire méthodique de l’architecture grecque et romaine, I, Matériaux, techniques de construction, techniques et formes de décor, Collection de l’EFR 84 (1985). GINOUVÈS 1992 R. Ginouvès, Dictionnaire méthodique de l’architecture grecque et romaine, II, Éléments constructifs : supports, couvertures, aménagement intérieurs, Collection de l’EFR 84 (1992). GINOUVÈS 1998 R. Ginouvès, Dictionnaire méthodique de l’architecture grecque et romaine, III, Espaces architecturaux, bâtiments et ensembles, Collection de l’EFR 84 (1998). HAIRY 2011 I. Hairy (ed.), Du Nil à Alexandrie, Histoire d’Eaux, Alexandria, 2nd edition (2011). HELLMANN 1999 M.-Chr. Hellmann, Choix d’inscriptions architecturales grecques, traduites et commentées, Travaux de la Maison de l’Orient Méditerranéen 30 (1999). Disponible en ligne HELLMANN 2006 M.-Chr. Hellmann, L’architecture grecque : Tome 2, Architecture religieuse et funéraire (2006). HELLMANN 2010 M.-Chr. Hellmann, L’architecture grecque 3- Habitat, urbanisme, fortification (2010). MICHAELIDIS 1988 P. Michaelidis, “The ancient port of Amathus, its construction, foundation conditions and final abandonment”, Engineering Geology of Ancient Works, Rotterdam (1988), p. 1597-1603. NANTET 2016 E. Nantet, Phortia, le tonnage des navires de commerce en Méditerranée, du VIIIe siècle av. l’ère chrétienne au VIIe siècle de l’ère chrétienne, PUR, Rennes (2016). PIMOUGUET-PÉDARROS 2011 I. Pimouguet-Pédarros, La cité à l’épreuve des rois. Le siège de Rhodes par Démétrios Poliorcète (305-304 av. J.-C.), Rennes (2011). PRÊTE et al. 2002 J.-P. Prête, I. Tassignon, T. Koželj, M. Wurch-Koželj, “L’Agora”, in “Rapports sur les travaux de l’École française. Amathonte en 2001”, BCH 126 (2002), p. 551-562.

Chronological table

323

Androcles, the last king of Amathus, sides with Ptolemy, satrap of Egypt.

313

Execution or arrest of the last kings of Kition, Lapithos, Kyrenia and Marion; Androcles dies in this same year or shortly thereafter. Ptolemy names the king of Salamis, Nicocreon, as governor of Cyprus.

311

After having come to terms with Antigonus, Ptolemy forces Nicocreon to commit suicide. Menelaus, brother of Ptolemy, is named as governor of the island. Ptolemy chooses Paphos as the capital and increases the population by forcing the inhabitants of Marion, who had remained faithful to the Macedonians, to leave their city and settle there.

306

Demetrius Poliorcetes enjoys a spectacular naval and land victory over the forces of Ptolemy and his brother Menelaus at Cypriot Salamis. The victorious flagship is dedicated at Delos in the Monument of the Bulls, which was purpose-built for the occasion. Antigonus Monophthalmus and his son Demetrius are masters of the island, and it will remain under Antigonid control for 12 years. They proclaim themselves kings of Asia.

305

Ptolemy I claims the title of king of Egypt.

301

Antigonus dies at the battle of Ipsus.

294

Demetrius Poliorcetes leaves for Greece and Ptolemy I retakes Cyprus; destruction of the palace and abandonment of the naval harbour. The island will remain under Ptolemaic control until 58 bc.

283

Death of Demetrius Poliorcetes (captured in 285 by Seleucus) and of Ptolemy I.

Cet ouvrage a été imprimé et relié en quatre cents exemplaires par l’imprimerie N.V. Peeters S.A. à Hérent (Belgique) ISBN 978-2-86958-293-4 Imprimé en Belgique