Version classiqueVersion mobile

La géoarchéologie française au xxie siècle

 | 
Nathalie Carcaud
, 
Gilles Arnaud-Fassetta

Partie I. Paléoenvironnements, biogéographie et paysages/Section I. Palaeoenvironments, biogeography and landscapes

Chapter 4. Geomorphological dynamics and the occupancy of the Letoon shrine (Xanthos plain, Turkey) during the last 5000 years

Émilie Écochard, Éric Fouache, Jacques des Courtils et Laurence Cavalier

Texte intégral

1A common theme, found often in romantic literature, holds that the remnants of past civilisations arouse our imagination of the present, their presence being a timeless source of wonder and curiosity. Situated in the deltaic plain of the Eşen Çayı River (or Xanthus during the Antic Period) and located in south-western Anatolia, the archaeological sites of Xanthos and Letoon attract many visitors; men and women whose fascinated by the mysteries of history and archaeology.

2Xanthos dominates the apex of the deltaic plain. Located on the leftbank of the river that divides the plain, the citadel was inhabited from the 8th or 7th c. BC to the 12th c. AD, i.e. throughout the Lycian, Hellenistic and Roman periods, leading up to the dawn of the Middle Ages (Fig. 1; des Courtils, 2003). Further south, 5 km from the citadel and towards the sea, is located the Letoon shrine. Built on the right bank of the river, the three temples are dedicated to Leto, Artemis and Apollo, thus underlining their importance during the Hellenistic period. There are a few additional archaeological sites in this wealthy plain, notably the Patara Roman harbour or citadels posted at the bottom of the slopes that surround the plain. In 1988, Xanthos and Letoon were first listed by UNESCO. Their cultural and historical importance comes from the unusual mix of Lycian and Hellenistic cultures, with special reference to their funerary architecture. Both sites were surveyed for the first time in 1838 by Charles Fellows (Fellows, 1852; Metzger, 1998). From 1950 to 2011, a French archaeological team, lately managed by Jacques des Courtils, Professor at Bordeaux University, worked on these sites under the aegis of the French ministry of Foreign affairs.

3Architectural remains on the sites reflect a rich history of cultural exchange. However, the spatial and architectural evolutions, as well as the history of these sites, can only be understood in light of the environmental context (Jacob, 1994). Archaeologists who study these sites cannot presume that the landscape they see has not changed since the arrival of the first inhabitants in the 7th c. BC. To understand societies that occupied this space, and particularly their relationship with their environment, there is a need to reconstruct the landscapes they have inhabited and developed. This implies understanding the environmental conditions and geomorphological dynamics at play (Fouache and Rasse, 2009). Typical questions raised by archaeologists include: In what kind of environment was the Letoon shrine constructed? How was the shrine accessed? How did people come from Xanthos to Letoon? Answering these questions is a geomorphologists task as the major geomorphological dynamics of the deltaic plain during the last few thousand years have to be identified.

The plain of Xanthos and Letoon: A Mediterranean deltaic plain

A karstic catchment supplying water to the valley of Xanthus and its deltaic plain

4The plain of Xanthos and Letoon is the final section of the valley of the Eşen Çayı River, constituting the central and terminal parts of the catchment, between the area of the Araksay’s confluence and the sea. This section, in which the river follows a relatively rectilinear course, is highly asymmetric. The main tributaries come from the eastern mountain range, rising at a height of about 2000 m, fed by pluvial (meteoric) waters. To the north-east, there are highlands where the Eşen Çayı River spring can be found, at a height of about 1100 m. Both Xanthos and Letoon are located at the end of this valley controlled by a grabben perpendicular to the shoreline. The main geological features of the catchment reveal a complex geological history at the heart of the tectonic dynamics of the Aegean arc. The Anatolian uplift, in the Middle Miocene (Kayan, 1999; Alçiçek, 2007), explains the presence of mountain ranges in the Lycian Peninsula. The overthrusting phase, from north to south, was repeated by tectonic extension during the Late Pliocene. These events largely explain the structural organisation evident in the catchment (Graciansky, 1972; Bousquet and Péchoux, 1981; Collins and Robertson, 1997; Alçiçek, 2010). Three outcomes of this geological history can be identified: firstly, the sensitivity of materials to erosion dynamics (overthrusted rocks are mainly composed of carbonate formations); secondly, the presence of huge limestone formations with chemical weathering sensitivity; and finally, the organisation of the river system. The river network is highly asymmetric, with tributaries flowing principally from the left side of the river due to the presence of highlands to the east and north-east. In this context, the erosion processes are more evident in the east, where steeper slopes combine with a highly developed river system and a spectacular endokarst (underground karst). A holistic understanding of the environmental occupation of human societies needs to take into account this topographical asymmetry.

Fig. 1. A: Location map and map of the boundaries of the catchment areas, with a focus on the Eşen Çayı catchment (based on SRTM datas interpreted with QuantumGIS software). B: Geomorphological map of the plain of Xanthos, drawn from the topographical map (scale 1:25,000), the geological map (scale 1:100,000), maps from Bousquet and Péchoux (1984) and Öner (1999), and from field observations. Coordinates are given in UTM (zone 35). a: topographical information. 1: spot height. b: landforms. 2: permanent streams; 3: intermittent streams; 4: palaeochannels; 5: Holocene alluvial deposits; 6: current costal sand deposits; 7: deflation surface; 8: swamps; 9: recently drained swamps; 10: wooded dunes; 11: grazed wooded dunes. c: formations. c. 1: geological formations. 12: micro-fractured limestone; 13: undifferentiated limestone; c. 2: superficial formations. 14: Pliocene lacustrine deposits; 15: Mio-Pliocene conglomerates: old terraces; 16: visible fault; 17: hidden fault; 18: Pleistocene dejection cones; 19: dejection cones of the end of Pleistocene; 20: recent dejection cones with their active margin; 21: active talus aprons. d: artificial features. 22: archaeological sites. C: View from Xanthos (photograph taken by E. Ecochard, 2006). D: View from Letoon (photograph taken by A. Çiner, 2007).

An interface between the Taurus Mountains and the Mediterranean Sea

5Besides the hinterland dynamics, we must also consider the areas littoral properties in order to comprehend the plains evolution. Sea level rose during the Holocene period, but at a rate that gradually decreased (Vouvalidis, 2005). Relative sea-level variation studies in the Eastern Mediterranean reveal a three phase evolution (Kayan, 1991; Kayan, 1996; Kelletat, 2005; Fouache et al., 2011). Until 6000 years BP, the sea level probably rose progressively, between 5000 and 3500 BP, it probably decreased, as indicated by the draining of the lagoons in western Anatolia (Kayan, 1996). The final phase, from 3500 BC to the Roman period, is when the sea level likely rose to its current level. Marine dynamics continue to shape the Lycian coast. The shoreline dynamics are under the control of prevailing winds from the west (Heikell, 2010). In the Eastern Mediterranean region, the long-shore drift follows a counter-clockwise path (Dominey-Howes, 2004; Heikell, 2010). In this context, it becomes vital to distinguish between the erosion of the sands caused by the sea, from those caused by the wind.

6The plain belongs to the Mediterranean climate zone and is therefore characterised by a large contrast between the warm and wet seasons (Sensoy, 2008). These climatic characteristics impact both amount and nature of plant cover and the rhythms of river discharges. Together, these two factors largely explain both the current instability of the catchment and the landscape instability on short time scales. At the scale of the Holocene or indeed the Quaternary period, the Mediterranean area is particularly interesting for its transitional position between the northern wet climates and the southern arid climates. European climatic components (but also North African and Asian) indeed influence the Mediterranean climate (Bar-Matthews et al., 1999). The Holocene is generally characterised as a period of rising temperatures, though this tendency was at times disturbed by cold intervals (at 8200, 6000-5000, 4200-3800, 3500-2500, and 1200-1000 years cal. BP; Mayewski et al., 2004; Wanner et al., 2008).

A plain inhabited since the Lycian period

7The Teke peninsula (a region corresponding to the ancient Lycia) was settled by Epigravettian groups during the Upper Paleolithic (Yalçinkaya et al., 2002; Kozlowski, 2005). The Neolithic Revolution (ca. 9000 BC in the Fertile Crescent) probably reached southern Anatolia ca. 7000 BC (Guilaine, 2001). Human societies began to transform landscapes: they developed fields and pastures, modifying large areas of forests. These groups selected and promoted plant and animal species based on their needs (Louboutin, 1990). War also became a present and tangible reality. Prior to antiquity, literary sources and Egyptian inscriptions refer to Hittites and Lukkas (ancestors of Lycians) settlements in the plain (Bryce and Zahle, 1986). However, no archaeological trace has yet been found of permanent settlements of these peoples. In fact, permanent settlement is recorded in the plain only from the 7th c. BC (des Courtils, 2003) when societies began to develop and exploit their environment. Under Persian, Greek and Roman dominations, periods of political and economic stability and instability follow each other, alternately promoting or discouraging the intensive exploitation of resources (Le Roy, 1998). The 7th c. was characterised by the wars between Persians and Byzantines; the plain and hillsides becoming gradually abandoned. Temporary settlement became the dominant form of occupation until the 20th c. Permanent occupation only came back in the second half of the 20th c. when the cultivation of cotton and tomatoes became possible, due to the drainage of wet clay soils, which allowed reclamation of the plain.

Methodology used to reconstruct the geomorphological and stratigraphy evolution of the deltaic plain

8The aim of this work was to follow a geoarchaeological methodology that would supplement and clarify the studies of Bousquet and Péchoux (1981) and Öner (1999). It quickly became apparent that the western part of the plain, near the Leto shrine, was the most unstable sector during the Holocene. It is also in this area that fluvial palaeochannels were identified on satellite images and aerial photographs. The hypothesis that these palaeochannels had been functioning during the occupation of the sanctuary had to be verified. Thus, we chose to focus our research in this area. The geophysical exploration consisted of 14 electrical resistivity transects, followed by the collection of 21 samples using a hydraulic piston coring device. These provided the raw data and sediment samples necessary to undertake the geoarchaeological study (Figs. 2 and 3).

9Sediment samples were then analysed in the laboratory. Three types of analyses were carried out on the most representative core samples: variations in magnetic susceptibility, organic matter, and carbonate contents. The measurement of the magnetic susceptibility can be used to characterise the materials (in particular, their response to the magnetic excitation to which they are submitted) and to correlate the sedimentary units from one core sample to another, while also helping to identify process formation and transport of materials. This makes it possible to assess the contribution of different rivers or streams to the sedimentary fill of a region (Lowe and Walker, 1997; Dearing, 1999; Ghilardi, 2007; Dearing, 2011). This method has the advantage of not being destructive and it is usually undertaken before other sediment analyses (Dearing, 1999). The magnetic susceptibility system used was a Barlington MS2 connected to a MS2C sensor. The sensor is moved along the core and a measurement is taken every centimetre. The determination of organic matter content is very important for any palaeoenvironmental study, as it indicates ancient biologic activity and organic sediments inputs. It is particularly useful in the context of work carried out in continental aquatic environments, such as lakes and marshes (Lowe and Walker, 1997). The methodology that we applied was that of loss on ignition (LOI) and used on samples taken every 20 cm in the core. The results of these analyses are shown in Fig. 2. In addition, we sent 15 samples to obtain14C dates. For the sediments that contain insufficient organic matter, it was necessary to use accelerator mass spectrometry (AMS) datings. Finally, analysis was made of a deposit of volcanic debris present in core L–XII (at about 9 m below the topographic surface) in order to identify which volcanic eruption it came from. At the same time, we analysed ancient texts and stories from past travellers. As well as archaeological data, they highlight and can sometimes confirm and supplement geomorphological data.

From the marine bay to the present deltaic plain: Geomorphological evolution during the Holocene

Geomorphological evolution before human occupation: The delta’s construction

10Data from electrical resistivity transects and probes were not sensitive enough to accurately reconstruct the morphology of the terminal area of the Eşen Çayı valley at the beginning of the Holocene. Marine sediments have never been reached by drilling, and electrical transects give only a partial image of the topography of the limestone bedrock. However, we still propose a hypothetical reconstruction based on two types of data. Firstly, the beginning of the Holocene is characterised by a rapid marine transgression. According to the curve of absolute marine level changes proposed by Vouvalidis et al. (2005), the sea level was at least 29 m below the current one. In addition, at Xanthos, remains of an alluvial fan with fluvial pebbles are visible. This alluvial fan is now eroded into terraces (Bousquet and Péchoux, 1981, 1984; Öner, 1999). These first sediments might be dated from the end of Pleistocene, according to the hypothesis of Bousquet and Péchoux (1981). It is possible to date this first delta using data from the Shuttle Radar Topography Mission (SRTM). A theoretical shoreline is then obtained by extending this alluvial fan to the sea level at 10000 BP, approximately 30 m below the current sea level (Vouvalidis et al., 2005; Fig. 4, step 1). The palynological studies conducted by Bottema and Woldring (1984) provide two pieces of information concerning the evolution of the terminal part of the valley up to 2000 BC. Around the 5th millennium BC, the area was exposed to sun and drought for sufficient time to corrode pollen grains. According to the authors, this event may be correlated with local environmental evolution–and not with a climatic change–related to a fluvial change. It may be explained by a tectonic event, which would have raised the eastern part of the plain, isolating it from the fresh river water and the marine saltwater. This part of the plain then became marshy. According to this pair of observations, the two parts of the plain evolved quite differently: the eastern part was isolated from marine dynamics and quickly stabilised with marshy conditions; on the contrary, the western part is characterised by a fluvial sedimentation during the entire Holocene.

Fig. 2. Core locations, logs of the main cores and results of sedimentary analyses. Coordinates are given in UTM (zone 35). 1: clay; 2: silty clay; 3: silt; 4: silty sand; 5: coarse sand; 6: gravels; 7: pebbles; 8: ash; 9: organic matter; 10: fauna; 11: pottery.

Fig. 3. Location of resistivity transects and results obtained in transects nos. 4, 10, and 11. The electrical resistivity was measured using the ABEM Terrameter SAS 300 C device, with the non-automated multi-electrodes method. Data were processed using the RES2DINV software.

11We can say that a lagoon existed at 3295± 30 BC, i.e., between 1633 and 1500 BC, calibrated age (14C dating was obtained from the L-XII core). Indeed, the dated level is a thin black plastic clay deposit that is situated at a depth of between 9.01 and 9.05 m from the ground surface. It covers a homogeneous thin black sand, with faunal remains (a whole Cerastoderma glaucum and a broken gastropod Potamides conicus), which points to a lagoon environment. The preservation of the tephra just above these sands implies a quiet period of deposition.

12In the eastern part of the plain, the pollen analysed by Bottema and Woldring (1984) in the Ova gölü swamp show that at 6500± 380 BP (at 7.3-7.4 m under the ground surface), there existed a fresh water environment. This hints at an early closed lagoon in the eastern part of the plain, which was closed by the delta progradation. In the western part of the plain, the environment was more opened to marine dynamics, with the presence of marine lagoons. To draw a first outline of the delta at about 2000 BC, we must again use the SRTM data, choosing the data, which best corresponds to the western lagoon evolution. Line no. 4 is the best candidate along the lagoon, as it includes the L-XII core. Also, in terms of indications given by Bousquet and Péchoux (1981), the first sedimentary horizon in the plain outcrops at the Letoon hill. Referring to the curve of absolute sea-level change given by Vouvalidis et al. (2005), the theoretical sea level at 2000 BC was 5 m below the current level. If we extend the delta slope obtained by the STRM picture to reach the-5 m altitude, we obtain a theoretical outline of 2000 BC that isolates the eastern part of the bay and corresponds to the lacustrine environment given by Bottema and Woldring (1984; Fig. 4, step 2).

13Around 1500-2000 BC, the studied area was exposed to the Minoan eruption of Santorini. On the one hand, the L-XII core includes ash deposits between 8.98 and 9.01 m (under the topographic surface). These deposits are identified as coming from the Minoan eruption of Santorini, an eruption whose volcanic explosivity index was estimated between 6 (colossal) and 7 (mega-colossal; McCoy and Heiken, 2000; Fig. 2). On the other hand, this tephra is covered by a sandy deposit, which is itself the result of the tsunami initiated by the volcano’s collapse (Stanley and Sheng, 1986; Sullivan, 1988; Hamann et al., 2010; Aydar et al., mimeo). Lagoons were affected by the tsunami between 30 minutes (Minoura et al., 2000) and two and half hours (Goodman-Tchernov et al., 2009) following the collapse of the volcano. Numerical simulations for western Turkey suggest that the tsunami wave would have reached more than 5 m high. If today the studied area is situated at 4 km from the sea, it probably was less than 1 km behind the shoreline during the volcanic eruption (Aydar et al., mimeo). The difference is due to the delta progradation. Lagoons were gradually filled in. In most cores, the bottom lagoonal sands are covered with a thick clayey unit, sometimes reaching up to 4 m in thickness. These plastic clays are grey, sometimes dark-brown, and with a massive structure. They correspond to a quiet depositional environment. As there is no coarse material that would have indicated the proximity of the river or slopes, and as the tephra is preserved in the L-XII core, this unit has the characteristics of a marshy environment, completely different from the previous dynamics influenced by marine connections. The sedimentary fill was created in part by the closing of the offshore bar inlets, shaping the plain through marine dynamics – except in the case of the offshore bar breakdown. The grey clay that is in these cores contains little organic material (below 5%), however, the organic matter indicates a marshy environment, and greater rates of organic matter were expected and Bottema and Woldring (1984) had already noted this characteristic in the cores they had studied. Two interpretations are possible: (i) marshes of the plain, both in the west and the east, have a low nutrient content, preventing the development of significant vegetation. If the plant cover was at that time similar to the current, it is conceivable that the marshes were only bordered by reedbeds. This oligotrophy would be explained by alluviation and transported sediments. The main part of the deposit would be allocthonous (Lowe and Walker, 1997); or (ii) it was an environment characterised by the presence of salty or briny waters, preventing the deposition of organic matter.

Fig. 4. Phased reconstruction of the evolution of the end part of the Eşen Çayı valley. Coordinates are given in UTM (zone 35). a: topographical information. 1: spot height. b: forms. 2: permanent streams; 3: intermittent streams; 4: palaeochannels; 5: Holocene alluvial deposits; 6: current costal sand deposits; 7: deflation surface; 8: swamps; 9: recently drained swamps; 10: wooded dunes; 11: grazed wooded dunes. c: formations. c. 1: geological formations. 12: micro-fractured limestone; 13: undifferentiated limestone. c. 2: surficial formations. 14: Pliocene lacustrine deposits; 15: Mio-Pliocene conglomerates: old terraces; 16: visible fault; 17: hidden fault; 18: Pleistocene dejection cones; 19: dejection cones of the end of Pleistocene; 20: recent dejection cones with their active margin; 21: active talus aprons. d: further information. 22: archaeological sites; 23: cores; 24: marine transgression; 25: progradation of the delta; 26: sea-water intrusions; 27: Letoon shrine; 28: small harbour?; 29: karstic spring; 30: lagoons and marshes filling; 31: access way to the shrine; 32: main artificial channels; 33: dykes; 34: vegetated dunes.

Geomorphological evolution during human occupation: The story of a plain dominated by water

14Although Bottema and Woldring (1984) mention agricultural activity ca. 4000 BP, no Bronze Age sites have been discovered in Lycia (Mellaart, 1968). Des Courtils (2003) was indeed surprised that only a bronze axe, dating from the Bronze Age (3rd millennium BC), was found in the site of Patara, while the rest of Anatolian peninsula saw the development of many sites at this time. We also know, from Hittite sources, that the Lukka people had settled in the western Anatolia. Those people without a known king, often took part in maritime raids (particularly against Egypt) and founded cities. This is highlighted by a Bronze inscription recently discovered in Pisidia, which mentions the Lycian cities of Ptarr (Patara) and Arnna (Xanthos; Bryce and Zahle, 1986). Thus, the first remains of the Lycian people in the plain dates from 700 BC. In Xanthos and Letoon, potteries were found proving the occupation of these sites. Architectural remains come from the 6th c. BC. At that time, Xanthos was under the control of Persians (des Courtils, 2003). The Greek domination, following the conquest of Asia Minor by Alexander the Great in 334 BC has dominated the architectural history of Letoon and Xanthos (des Courtils, 2003). Around that time, it becomes possible to link history to the environmental context.

15A few centuries after the foundation of the sanctuary, an event throws into temporary confusion the geography of the Xanthos plain. A channel of the river beginan to flow in the western area. The L-I Bis and L-VI cores both contain at their bases channel deposits. In the L-I Bis core, for nearly 3 m of thickness there are coarse sand, gravels and smooth pebbles (some of them are broken, due to the passage of the coring device; Fig. 2). The deposit is thinner in the L-VI core, which is located in the north of the Letoon hill, but with a similar facies. This facies is characteristic of a lag deposit in the bottom central part of a channel. It indicates that a channel has flown, for a few years at least, in the western part of the plain. The analysis of the electrical resistivity transects no. 4, 10 and 11 provides more information on these deposits (Fig. 3): (i) they are located between 4 m and 12 m below the ground surface; (ii) they are made of coarse materials, which is revealed by high resistivity values (Beresnev et al., 2002); (iii) there were at least two palaeochannels, functioning at the same time or successively; (iv) the channel(s) came from north-east and flowed to the south-west direction; and lastly (v) the width of one of the channels was between 40 m and 60 m, which corresponds to the current mediumbed width of the river. Archaeological remains date the flow of Eşen Çayı in the western part of the plain to between the 6th and the 5th c. BC when sedimentation was rapid (Le Roy, 1984). The western entrance of the sanctuary was built at that time and it remained there until being destroyed in the 4th c. AD. The western channel(s) would have flowed during almost the entire period of the shrine’s occupation. From the data given above, it is possible to propose a reconstruction of the plains environment when the Letoon shrine was occupied (Fig. 4, step 3). This reconstruction is different from the one given by Öner (1999), which did not mention the channel in the western part of the plain.

16During the Byzantine Age (from the 4th c.), the plain was flourishing, indeed, peace was maintained and business grew. The advantageous location of Lycia between the Aegean Sea and Egypt allowed cities located near the coast to flourish and develop (Sodini, 1998) as in the case of Xanthos (Foss, 1994). Christianity, which progressively entered the region between the 4th and the 6th c. AD, transformed both the city of Xanthos and the shrine (Sodini, 1998). The city was experienced a new period of greatness from 5th c. and for two centuries (Manière-Lévêque, 1998; des Courtils, 2003; Musée de Saint-Romain-en-Gal, 2009-2010). The prosperity of the plain during this period can be explained by the flow of trade as Xanthos was always accessible by waterway. The geography of the plain might have been similar to the precedent era in its spatial configuration. Though as the Letoon slowly lost its importance, and since the river followed a course similar to the current course (in the centre of the plain), the progressive abandonment of the site could also be due to the spread of Christianity. The revival of the Pidnai citadel, which controlled the entrance to the Xanthus valley, seems to indicate that the western part of the plain was an area of frequent visitation. As sabotage was intense in Lycia (Sodini, 1998), the small harbour near Pidnai, whose remains were discovered by the archaeologists of Xanthos and Letoon, would have served a prominent role in the areas security. The war, by the Persians in the 7th c. and the following expeditions led by Arabs, are reflected by the decline of the plain (Foss, 1994; Sodini, 1998). Bousquet and Péchoux (1981) highlighted that during the Byzantine period, the subsidence of the plain returned, causing or intensifying flooding and siltation. Populations suffered from the effects of interfluve erosion, which had its origin in the intense economic development of earlier periods. A delay of 300-500 years might be then observed, between the period of high activity responsible for the drainage basin erosion, and that characterised by flooding and siltation in the context of rural abandonment. In addition to historical reasons, the abandonment of these sites might be considered as a cause and consequence of changing environmental conditions as all topographically depressed areas become wetlands. At the end of the Byzantine period (15th c.), there is no trace of activity in Xanthos (des Courtils, 2003). People occupied the plain only temporarily (des Courtils, 2003) and in Patara, nomads settled in spring with grazeing herds and collected harvests (Texier, 1839-1849). In the 19th c., the small river in the western part of the plain, the Özlen Çayı, was still navigable by small boats (Fellows, 1841). This access way from the sea still used to reach Xanthos, a symbolic continuity of a bygone golden age.

17In the 20th c., the context radically changed. Many improvements were made to facilitate intensive agriculture. To develop the plain, two types of infrastructures have been constructed: (i) dykes to protect the land from river flooding, and (ii) channels to drain and irrigate farmlands. In the eastern part of the plain, the construction of a large channel allowed nearly half of the marsh to be drained. Conquered lands, which were once used as pastures, are today covered by greenhouses for growing winter tomatoes. These greenhouses represent the main economic activity in the plain today. The profusion of dunes, however, threatens the plain, and to deal with the problem of a dune fixation, policy was initiated in 1960 (Bousquet and Péchoux, 1981; Fig. 4, step 4).

Consequences of environmental history on the Letoon history

The Letoon shrine: The choice of solitude

18During the permanent settlement by the Lycians in the plain, the hill on which the Letoon shrine was built was enclosed on the west and south sides by swamps. These swamps were probably supplied by a karstic resurgence. On the east side, one or two channels combined with abrupt slopes to made it difficult to access to the shrine. A now invisible spring gestures or signifies to the sanctuary’s foundation and origin. Antonius Liberalis, Ovid, and Servius are the three authors giving evidence of the arrival of Leto in Lycia (Bryce, 1983). Ovid, in his “Metamorphoses”, relates the myth that the inhabitants of the plain, deemed inhospitable by the goddess, had been transformed into frogs. Edifices built to use the spring for religious purposes (artificial grotto and then a Roman Nymphea) highlight the cultural importance of the water source (Balland, 1998; des Courtils, 2003). These constructions also helped to dry the site by channelling excess water. The spring not only served as the inspiration for the temple’s foundation, but also created a prime environmental constraint, which required the construction of specific architectural features.

19This ideal site for the construction of a shrine had been doubtlessly noticed and appreciated by Lycians. The spring, a symbol of life and fertility, conferred no the place a mystical and mysterious dimension well suited for a shrine (Le Roy, 1984). In addition, the hill was a dry place on which buildings such as theatres and temples (consecrated to Leto and his children, Apollon and Artemis) could easily be erected. The central part of the 36 steps of koilon belonging to the theatre was actually carved into the hill’s limestone and the upper parts were built using blocks extracted on site (Moretti, 1998; des Courtils, 2003). The shrine location, in the middle of a delta characterised by dynamic fluvial channels, transmitted at the same time the loneliness and solitude described above. This configuration can be found on other sites in the upper part of the Xanthe catchment. Religious sites took symbolic advantage of the high dynamics of landscapes to challenge the laws of nature (Le Roy, 1984). However, this choice of shrine location became a problem when the river began to gradually flow nearer.

Coming to the shrine and Xanthos: Plural but fragile communication ways due to high mobility of the environmental context

20At the local geographical scale, archaeologists consider there to have been two access points to the shrine, by the west and by the north. The access from the west should have been the main access, at least from the 4th c. BC to the 4th c. AD. Indeed, during the 4th c. BC, a bailey of north-south orientation was built at the west side of the shrine. In Roman times, the wall might have included a monumental entrance leading to a west-east orientation way, paved and embellished with statues. A Roman stele, with the rules of the sanctuary inscribed on it, was also found here (des Courtils, 2003). This entrance, in the middle of the plain, has no meaning unless it is related to a communication way. A channel is actually the source of interest as the river would have been the link between the shrine and the sea (downstream) and Xanthos (upstream). Indeed, Strabo, in his “Geography” (book 14, III), wrote about light buildings (hupêretikoi) used for transportation to go up the river from the sea to the shrine and the city of Xanthos. The other entrance, on the north side of the site, may have been used by inhabitants of Xanthos who travelled by foot, as well as when the western entrance was not practicable. It was probably a secondary entrance as the ornamentation is less developed (des Courtils, 2003). If visitors wanted to come by foot from Xanthos to Letoon, they had to cross the river twice, at Xanthos and near Letoon. The poses the issue of crossing the river by bridges – the one of Xanthos was built in Roman times – or using boats.

21At the scale of the plain, the Letoon was probably reachable by waterway from Xanthos and from the sea. Indeed, the Pseudo-Skylax indicates that the river was navigable during the 4th c. BC to the city of Xanthos. The Stadiasm of the Great Sea, which is a compilation of several sources and is therefore difficult to date, probably dates to the 5th c. BC. Regarding this issue, Arnaud (2009) also attests to the navigability of the river. However, if some of the rivers of Asia Minor (rivers Melas, Eurymedon or Sangarios) were reachable by large boats from the sea, the Xanthe River was itself only navigable by small boats. Maritime traffic reached Patara after which small boats were used to reach the Letoon and Xanthos (Roelens-Flouneau, in press). Appian, in his account of the Roman civil wars, seems to consider Patara the harbour of the Xanthians. This would indicate a close relationship between the two cities, and most likely also with the sanctuary.

22At the catchment scale, the gentler slopes on the right bank of the valley and the instability of the lower slopes, where ancient environmental legacies such as alluvial terraces or lacustrine sediments are better preserved, might have worked in favour of communication in the western part of the valley. The bridge built at Xanthos during Roman times would then make Letoon approachable (downstream) and also the other cities of the valley (upstream). To cross the river at this point previously required the use of other lighter bridges or boats. On a smaller scale, the Lycia played a major role in trade relations between the Aegean Sea and the Eastern Mediterranean Sea (Childs, 1981; Lemaître, 2007; Arnaud, 2009). In this context, the reachability of the shrine from the sea was very important. It is possible to consider at that time an access at the mouth of the Özlen Çayı, the small river at the northwest edge of the plain. Small boats could pass the shoreline and reach the shrine through marshes and/or by river. Visitors might also come via a small port near the citadel of Pidnai, often used as a stage to reach the Letoon. Indeed, remains are visible today in this area that could indicate a small, minor local harbour.

Conclusion

23This geoarchaeological study carried out in the plain of Xanthos allowed the reconstitution of the main stages of the geomorphological evolution in the area, from the marine bay up until the current alluvial plain. One of the major stages of the geomorphological and human story of the plain is that the Eşen Çayı River flowed in the western part of the plain, near the shrine of Letoon while it was active. The recognition of this event allows a better understanding of the choices made by ancient societies in the location of the federal shrine. The environment allowed and was selected in order to strike a balance between the symbolic loneliness and solitude necessary at a holy shrine, and the accessibility necessary to allow for worshippers to visit.

Bibliographie

References

Alçiçek M.C., «Tectonic development of an orogen-top riftrecorded by its terrestrial sedimentation pattern : The Neogene Esen basin of southwestern Anatolia, Turkey», Sedimentary Geology, no 200, 2007, p. 117-140.

Alçiçek H., «Stratigraphic correlation of the Neogene basins in southwestern Anatolia: regional palaeogeographical, palaeoclimatic and tectonic implications», Palaeogeography, palaeoclimatology, palaeoecology, no 291, 2010, p. 297-318.

Arnaud P., « Notes sur le Stadiasme de la Grande Mer: La Lycie et la Carie », Geographia Antiqua, no 18, 2009, p. 165-194.

Aydar E., Ecochard E., Çiner A. et Fouache E., «Phreatomagmatic Signs in Plinian Distal Ashes Originated from Minoan-Thera of Santorini Volcanic Eruption», mimeo.

Balland A., «Le nymphée du Létôon», Dossiers d’archéologie, no 239, 1998, p. 56-57.

Bar-Matthews M., Ayalon A., Kaufman A., et Wasser burg G., «The Eastern Mediterranean paleoclimate as a reflection of regional events: Soreq cave, Israel», Earth and Planetary Science Letters, no 166, 1999, p. 85-95.

Beresnev I., Hruby C. et Davis C., «The use of multielectrode resistivity imaging in gravel prospecting», Journal of applied geophysics, no 49, 2002, p. 245-254.

Bottema S. et Woldring H., «Late quaternary vegetation and climate of Southwestern Turkey. part II», Palaeohistoria, no 26, 1984, p. 123-149.

Bousquet B. et Péchoux P.-Y., «Évolution géomorphologique des escarpements du Taurus lycien et des bordures du fossé du Xanthe (Turquie)», Méditerranée, no 1, 1981, p. 3-11.

Bousquet B. et Péchoux P.-Y., « La plaine du Xanthe (Turquie), contraintes physiques et aménagements », Études géographiques sur les Pyrénées et questions diverses, no 106, 1984, p. 33-43.

Bryce R., «The arrival of the goddess Leto in Lycia», Historia : Zeitschriftfür Alte Geschichte, no 32, 1983, p. 1-13.

Bryce T. et Zahle J., The Lycians in literary and epigraphic sources, vol. 1, Copenhague, Museum Tusculanum, 1986.

Childs W., «Lycian relations with Persians and Greeks in the fifth and fourth centuries re-examined», Anatolian Studies, no 31, 1981, p. 55-80.

Collins A.S. et Robertson A.H.F., «Kinematic evidence for late Mesozoic-Miocene emplacement of the Lycian Allochthon over the western anatolide belt, SW Turkey», Geological Journal, no 38, 2003, p. 295-310.

Des Courtils J., Guide de Xanthos et du Létôon, Istanbul, Ege Yayınları, 2003.

Dearing J.A., Environmental Magnetic Susceptibility. Using the Bartington MS2 System, Oxford, Bartington Instruments Ltd., 1999.

Dearing J.A., Magnetic susceptibility in landscape dynamics, Oxford, Bartington Instruments Ltd., 2011.

Dominey-Howes D., «A re-analysis of the late Bronze Age eruption and tsunami of Santorini, Greece, and the implications for the volcano-tsunami hazard», Journal of volcanology and geothermal research, no 130, 2004, p. 107-132.

Fellows C., An account of discoveries in Lycia : being a journal kept during a second excursion in Asia Minor, Londres, J. Murray, 1841.

Fellows C., Travels and researches in Asia Minor, more particularly in the province of Lycia, London, J. Murray, 1852.

Foss C., «The Lycian coast in the Byzantine Age», Dumbarton Oaks Papers, no 48, 1994, p. 1-52.

Fouache E. et Rasse M., «Archaeology geoarchaeology and geomorphosite management: towards a typology of geoarchaeosites», in Reynard E., Coratza P., Regolini-Bissig G. (éds.), Geomorphosites, Munich, Verlag DR. Friedrich Pfeil, 2009.

Fouache E., Kelterbaum D., Brückner H., Lericolais G., Porotov A. et Dikarev V., «The Late Holocene evolution of the Black Sea: a critical view on the so-called Phanagorian regression», Quaternary International, 2011, doi: 10.1016/j. quaint. 2011.04.008.

Ghilardi M., Dynamiques spatiales et reconstitutions paléogéographiques de la plaine de Thessalonique (Grèce) à l’Holocène récent, Thèse de doctorat, Université Paris-XII-Val-de-Marne, novembre 2007.

Goodman-Tchernov B., Dey H., Reinhardt E., McCoy F. et Mart Y., «Tsunami waves generated by the Santorini eruption reached Eastern Mediterranean shores», Geology, no 37, 2009, p. 943-946.

Guilaine J., « Changeons d’échelles: pour la très longue durée, pour de larges espaces », Études rurales, 2001, URL: http://etudesrurales.revues.org.

de Graciansky P.-C., Recherches géologiques dans le Taurus lycien occidental, Thèse de doctorat, Université de Paris-Sud Orsay, 1972.

Hamann Y., Wulf S., Ersoy O., Ehrmann W., Aydar E., et Schmiedl G. ., «First evidence of a distal early Holocene ash layer in Eastern Mediterranean deep-sea sediments derived from the Anatolian volcanic province», Quaternary Research, no 73, 2010, p. 497-506.

Heikell R., Guide nautique et de navigation: Turquie-Chypre. De la mer Noire à la frontière syrienne, Paris, Vagnon, 2010.

Jacob J.-P., «Le site archéologique. Un dialogue entre la culture et la nature ?», in Patrimoine culturel, patrimoine naturel, colloque 12-13 décembre 1994, Paris, La Documentation française, 1994.

Kayan I., «Holocene geomorphologic evolution of the Beşik Plain and changing environment of ancient man», Studia Troica, no 1, 1991, p. 79-92.

Kayan I., «Holocene coastal development and archaeology in Turkey», Zeitschrift für Geomorphologie, supplément no 102, 1996, p. 37-59.

Kayan I., «Holocene stratigraphy and geomorphological evolution of the Aegean coastal plains of Anatolia», Quaternary Science Reviews, no 18, 1999, p. 541-548.

Kelletat D., «A Holocene sea-level curve for the Eastern Mediterranean from multiple indicators», Zeitschrift für Geomorphologie, supplément no 137, 2005, p. 1-9.

Kozlowski J., « Paléolithique supérieur et mésolithique en Méditerranée : cadre culturel », L’Anthropologie, no 109, 2005, p. 520-540.

Lemaître S., « Bulletin de la société française d’archéologie classique. La Lycie et les échanges commerciaux dans l’Antiquité. Les données du mobilier amphorique. Premiers éléments de réflexion », Revue archéologique, no 1, 2007, p. 155-221.

Le Roy C., « Le Létôon de Xanthos en Lycie (Turquie). La lutte d’un sanctuaire contre l’évolution du milieu naturel », Bulletin de l’Association des Géographes français, no 499, 1984, p.  41-44.

Le Roy C., « Histoire des Lyciens », Dossiers d’archéologie, no 239, 1998, p. 10-17.

Louboutin C., Au Néolithique : les premiers paysans du monde, Paris, Gallimard, 1990.

Lowe J. et Walker M., Reconstructing Quaternary environments, Harlow, Pearson, 1997.

Manière-Lévêque A.-M., « Les maisons de l’acropole lycienne », Dossiers d’archéologie, no 239, 1998, p. 64-73.

Mayewski P., Rohling E., Stager J.C., Karlen W., Maasch K., Meeker L.D., Meyerson E., Gasse F., van Kreveld S., Holmgren K., Lee-Thorp J., Rosqvist G. Rack F., Staubwasser M., Schneider R.R. et Steig E.J., «Holocene climate variability», Quaternary Research, no 62, 2004, p. 243-255.

McCoy F. et Heiken G., «Tsunami generated by the late Bronze Age eruption of Thera (Santorini), Greece», Pure and Applied Geophysics, no 157, 2000, p. 1227-1256.

Mellaart J., «Anatolian trade with Europe and Anatolian geography and culture provinces in the late Bronze Age», Anatolian Studies, no 18, 1968, p. 187-202.

Metzger H., « Histoire de la recherche. Les premières étapes de l’exploration de Xanthos et du Létôon », Dossiers d’archéologie, no 239, 1998, p. 4-9.

Minoura K., Imamura F., Kuran U., Nakamura T., Papadopoulos G., Takahashi T. et Yalciner A., «Discovery of Minoan tsunami deposits», Geology, no 28, 2000, p. 59-62.

Moretti J.-C., « Les théâtres de Xanthos et du Létôon », Dossiers d’archéologie, no 239, 1998, p. 52-55.

Musée de Saint-Romain-en-Gal, Xanthos, architecture d’une civilisation (catalogue d’exposition), Département du Rhône, 2009-2010.

Ovide, Les Métamorphoses, tome 2, livre 6, Paris, Belles Lettres, 1928.

Öner E., «Letoon ve çevresinde (Eşen çayı deltası) paleojeomofolojik araştırmalar», Ege Coğrafya Dergisi, no 10, 1999, p. 51-82.

Roelens-Flouneau H., «Remarques sur la navigabilité des fleuves d’Asie mineure dans l’Antiquité», in Lebreton S. et Dan A. (éds), Études des fleuves d’Asie Mineure. Une première approche, Presses Universitaires de Rennes, sous presse.

Sensoy S., Türkiye iklimi, 2008, URL : http://www.dmi.gov.tr/iklim/iklim.aspx.

Sodini J.-P., « La Lycie à l’époque byzantine », Dossiers d’archéologie, no 239, 1998, p. 62-63.

Stanley D. et Sheng H., «Volcanic shards from Santorini (upper Minoan ash) in the Nile Delta, Egypt», Nature, (320), 733-735, 1986.

Strabon, The geography of Strabo, volume 6, Londres, The Loeb classical library, 1917-1932.

Sullivan D., «The discovery of Santorini Minoan tephra in western Turkey», Nature, no 333, 1988, p. 552-554.

Texier C., Description de l’Asie mineure : Beaux-arts, monuments historiques, plans et topographie des cités antiques, par Charles Texier, Gravure de Lemaître, 3 vol., Paris, Firmin Didot Frères, 1839-1849.

Vouvalidis K., Syrides G., et Albanakis K., «Holocene morphology of the Thessaloniki Bay : Impact of sea-level rise», Zeitschrift für Geomorphologie, Supplément 137, 2005, p. 147-158.

Wanner H., Beer J., Bütikofer J., Crowley T., Cubasch U., Flückiger J., Goosse H., Grosjean M., Joos F., Kaplan J., Küttel M., Müller S., Prentice C., Solomina O., Stocker T., Tarasov P., Wagner M. et Widmann M., «Mid-to Late Holocene climate change : an overview», Quaternary Science Reviews, no 27, 2008, p. 1791-1828.

Yalçinkaya I., Otte M. et Kozlowski J., La grotte d’Ökuzini : évolution du paléolithique final du sud-ouest de l’Anatolie, Universite de Liège, 2002.

Table des illustrations

Légende Fig. 1. A: Location map and map of the boundaries of the catchment areas, with a focus on the Eşen Çayı catchment (based on SRTM datas interpreted with QuantumGIS software). B: Geomorphological map of the plain of Xanthos, drawn from the topographical map (scale 1:25,000), the geological map (scale 1:100,000), maps from Bousquet and Péchoux (1984) and Öner (1999), and from field observations. Coordinates are given in UTM (zone 35). a: topographical information. 1: spot height. b: landforms. 2: permanent streams; 3: intermittent streams; 4: palaeochannels; 5: Holocene alluvial deposits; 6: current costal sand deposits; 7: deflation surface; 8: swamps; 9: recently drained swamps; 10: wooded dunes; 11: grazed wooded dunes. c: formations. c. 1: geological formations. 12: micro-fractured limestone; 13: undifferentiated limestone; c. 2: superficial formations. 14: Pliocene lacustrine deposits; 15: Mio-Pliocene conglomerates: old terraces; 16: visible fault; 17: hidden fault; 18: Pleistocene dejection cones; 19: dejection cones of the end of Pleistocene; 20: recent dejection cones with their active margin; 21: active talus aprons. d: artificial features. 22: archaeological sites. C: View from Xanthos (photograph taken by E. Ecochard, 2006). D: View from Letoon (photograph taken by A. Çiner, 2007).
URL http://books.openedition.org/editionscnrs/docannexe/image/21948/img-1.jpg
Fichier image/jpeg, 510k
Légende Fig. 2. Core locations, logs of the main cores and results of sedimentary analyses. Coordinates are given in UTM (zone 35). 1: clay; 2: silty clay; 3: silt; 4: silty sand; 5: coarse sand; 6: gravels; 7: pebbles; 8: ash; 9: organic matter; 10: fauna; 11: pottery.
URL http://books.openedition.org/editionscnrs/docannexe/image/21948/img-2.jpg
Fichier image/jpeg, 419k
Légende Fig. 3. Location of resistivity transects and results obtained in transects nos. 4, 10, and 11. The electrical resistivity was measured using the ABEM Terrameter SAS 300 C device, with the non-automated multi-electrodes method. Data were processed using the RES2DINV software.
URL http://books.openedition.org/editionscnrs/docannexe/image/21948/img-3.jpg
Fichier image/jpeg, 457k
Légende Fig. 4. Phased reconstruction of the evolution of the end part of the Eşen Çayı valley. Coordinates are given in UTM (zone 35). a: topographical information. 1: spot height. b: forms. 2: permanent streams; 3: intermittent streams; 4: palaeochannels; 5: Holocene alluvial deposits; 6: current costal sand deposits; 7: deflation surface; 8: swamps; 9: recently drained swamps; 10: wooded dunes; 11: grazed wooded dunes. c: formations. c. 1: geological formations. 12: micro-fractured limestone; 13: undifferentiated limestone. c. 2: surficial formations. 14: Pliocene lacustrine deposits; 15: Mio-Pliocene conglomerates: old terraces; 16: visible fault; 17: hidden fault; 18: Pleistocene dejection cones; 19: dejection cones of the end of Pleistocene; 20: recent dejection cones with their active margin; 21: active talus aprons. d: further information. 22: archaeological sites; 23: cores; 24: marine transgression; 25: progradation of the delta; 26: sea-water intrusions; 27: Letoon shrine; 28: small harbour?; 29: karstic spring; 30: lagoons and marshes filling; 31: access way to the shrine; 32: main artificial channels; 33: dykes; 34: vegetated dunes.
URL http://books.openedition.org/editionscnrs/docannexe/image/21948/img-4.jpg
Fichier image/jpeg, 467k

Auteurs

Doctor, University of Paris-Est Créteil (Paris 12), Mixed Research Unit (UMR 8591) CNRS/Universities of Paris 1 & Paris 12/INRAP (Laboratory of Physical Geography: Present and Quaternary Environments – LGP), Meudon, France (emilie.ecochard@gmail.com).

Professor, Michel de Montaigne (Bordeaux 3) University, Mixed Research Unit (UMR 5607) CNRS/University of Bordeaux 3 (Ausonius – Research Institute for Antiquity and Middle Ages), Pessac, France (courtils@u-bordeaux3.fr).

Associate Professor, Michel de Montaigne (Bordeaux 3) University, Mixed Research Unit (UMR 5607) CNRS/University of Bordeaux 3 (Ausonius – Research Institute for Antiquity and Middle Ages), Pessac, France (laurence.cavalier@u-bordeaux3.fr).

Le texte et les autres éléments (illustrations, fichiers annexes importés) sont sous Licence OpenEdition Books, sauf mention contraire.

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search