Palaepaphos: Unlocking the Landscape Context of the Sanctuary of the Cypriot Goddess

The cult centre of Palaepaphos was in continuous use since the construction of its megalithic temenos in the Late Bronze Age, but the new political role which the sanctuary was made to perform in the Hellenistic and Roman periods has supressed its original identity. From the 3rd c. BC, when the citystates of the island were abolished by Ptolemy Soter, to the end of the 4th c. AD, when pilgrim visitations and state-endorsed festivals were gradually abandoned under the growing impact of Christianity, the abode of the Cypriot goddess served the colonial politics of the Ptolemaic kingdom and the Roman Empire, respectively. The Palaepaphos Urban Landscape Project (PULP) has shown that the recovery of the sanctuary’s millennium-long primary role depends on the recovery of the almost invisible landscape of its founding polity and the region’s associated settlement structure. With the use of geospatial analyses and advanced documentation and imaging technologies, PULP has been building a diachronic model of the urban structure of the ancient polity and a site distribution model of the Paphos hydrological basin. Current results have unlocked the significance of the sanctuary’s spatial location in relation to a long-lost gateway to the sea that was the foundation kernel of Ancient Paphos.


Introduction
The sanctuary of the Cypriot Aphrodite at Palaepaphos is by all accounts the most well-known sacred monument of Ancient Cyprus (Fig. 1). It is also one of the most poorly understood, not only by laymen but also by scholars outside the narrow boundaries of the archaeology of Cyprus. Despite the fact that it "represents the sanctuary with the longest unbroken cult tradition in Cyprus -from the Late Bronze Age (c.1200 BC) to the Late Roman period (end of the 4 th c. AD)" (Maier, 2000, p. 496), it is viewed as if it had always been a 'pan-Cypriot' cult centre, which attracted pilgrims not only from the rest of Cyprus but also from all over the Mediterranean. In reality, this perception applies only to the character it had in the Graeco-Roman period and, more so, when the island wafs part of the Roman Empire -hence in the latter (and shorter) part of its long life. Besides underestimating the sanctuary's strongly autochthonous identity, this misconception has affected the interpretation of the early history of Ancient Paphos and especially our understanding of the political economy to which it owed its long-term success as an autonomous and prosperous polis-state (on the identification of the Cypriot polities as poleis, not kingdoms, see Demand, 1996, p. 8;Hatzopoulos, 2014). In the present article, we will argue that the sanctuary's original identity was undermined from very early on by (a) the literary visibility, which the sanctuary (but not the polis of Paphos) enjoys in the surviving ancient sources, and (b) the lack of visibility with respect to the urban structure of the polity of which it was an integral component. This imbalance manipulated the archaeological interpretation of the sacred site and the worship of the deity from the start. As a result, the lengthy biography of the sanctuary was compressed into a chronologically and ideologically biased narrative. Today, more than a decade since the initiation (in 2006) of the Palaepaphos Urban Landscape Project (PULP), this anomalous interpretation has been sufficiently remedied: landscape analysis has significantly enriched the visibility of Ancient Paphos from the horizon of its foundation (ca 1700 BC) to the termination of its political independence by Ptolemy I in 310 BC (Diodorus Siculus 20.21.1-3;Michaelidou-Nicolaou, 1976, p. 25). The landscape has been unlocked and the sanctuary is no longer isolated from the urban environment of the ancient polis, nor is it perceived as distant from the navigable coast as it certainly is today. As we will see below, coastal transformations must have deprived Ancient Paphos of its original function as a gateway to the sea even before the 4th c. BC when Nea Paphos was founded as the region's main port of call (cf. Młynarczyk, 1990, p. 23). Since then, the site of Paphos, where the capital centre of the homonymous city-state had grown, began to be referred to as Palaia or Palaeopaphos (cf. Mitford, 1960, p. 198;Masson, 1983, pp. 93-94). Pilgrims had to approach the Ptolemaic-Roman sanctuary of Palaepaphos on foot either from Nea Paphos (Strabo 14.6.3: Näf, 2013, p. 5) or via inland routes (Acta Barnabae 18: Näf, 2013, p. 52).

Literary Visibility, Colonial Narratives and the Epigraphic Evidence
Long before the first official attempt to explore its ruins in 1887 (Gardner, Hogarth, Jones, & Elsey Smith, 1888), the site of the Paphian sanctuary attracted travellers, antiquaries and treasure hunters, and enjoyed the attention of European humanists versed in the literary records of antiquity (cf. Maier, 2004, pp. 31-32). This early interest in the 'temple' (Fig. 2), as the temenos is often described in ancient and modern sources, had little to do with the visible material remains of the sacred site, which continue to confuse visitors to this day -especially since there is no Greek-style temple in Palaepaphos or elsewhere in Cyprus, for that matter (cf. Michaelides & Papantoniou, 2018, p. 282). It was generated by the study of an extensive corpus of Greek and Latin testimonia, which record the myths associated with Cypriot Aphrodite, Venus in Latin sources, and describe her cult at Paphos (Hadjioannou, 1973;Karageorghis, 2005;Näf, 2013). With few exceptions, of which Homer is the most significant (in the Odyssey 8.363, he mentions a sanctuary of Aphrodite at Paphos; in the Iliad 5.330, he identifies the goddess as Kypris), these excerpts date from the Roman period (Näf, 2013, pp. 42-67). As a matter of fact, some of them are Christian polemics struggling to undermine one of the last pagan strongholds on the island (cf. Maier, 2000, p. 505 andNäf, 2013, p. 62 on the fourth-century AD author Firmicus Maternus). Furthermore, as Michael Given has authoritatively argued in "Corrupting Aphrodite" (2002), in the late 19 th and early 20 th century, ancient sources were employed in the construction of colonialist narratives by European authors, whose ultimate goal was to justify the imperial/colonial rule imposed upon a worthless indigenous population: "Aphrodite, according to these writers, had corrupted the Cypriot people, and her influence was still to be seen." (Given, 2002, p. 420; see also, Morris & Papantoniou, 2014, pp. 187-188). Because of the fame, or notoriety, which ancient and modern literature bestowed upon the Paphian goddess and her cult, the sanctuary was the first site that the Cyprus Exploration Trust attempted to investigate less than a decade after the British took over the administration of Cyprus from the Ottoman Turks (in 1879). "The great object of our work", they claimed, "would of course be the plan of this the oldest shrine in the island, from whence much light might be thrown on Phoenician temples generally" (Gardner et al., 1888, p. 152). Disillusioned by the absence of a comprehensible built monument and the desolate state of the sacred site, the British excavators moved on after a three-month campaign (cf. Maier, 2000, p. 495). Apparently, their working methods inflicted further damage to the temenos, which had already suffered gravely during the Mediaeval period ( fig. 3) (consult photograph in Maier, 2004, p. 32, fig. 21 with a troop of unskilled local workmen, "Digging in the Aphrodite Temple at Paphos, 1888"). It is, however, to the credit of the members of the expedition that they published a meticulous and extensive report, which includes an impressive commentary on early travellers and antiquaries, a thorough list of the literary excerpts and, most importantly, all the inscriptions collected from the sanctuary and the environs of Kouklia-Palaepaphos to that day (Gardner et al., 1888, pp. 147-271). This commendable handling of the epigraphic record was continued in the 20 th century by T.B. Mitford (cf. 1960Mitford (cf. , 1961Mitford & Masson, 1983) and, following his death, by Olivier Masson (Masson & Mitford, 1986). Unlike the literary references, whose value is questionable -all of them are exogenous and contain a limited amount of culturally or chronologically specific information -the inscriptions represent primary and locally produced material evidence of a high historical value. Among other, the local epigraphic record confirms that the name of the city-state was Paphos. This is also the name by which it was known to the Neo-Assyrian empire: Paphos and its king Eteandros are commemorated on the royal prism of Esarhaddon (dated 673/672 BC), which carries a list of the Cypriot polities and their rulers (cf. Masson, 1992;Iacovou, 2013a, p. 143).
Even a peremptory spatial and chronological analysis of the epigraphic corpus relating to Paphos would show that, with the exception of a well-known alphabetic inscription in honour of the last king, Nikokles (e.g. Gardner et al., 1888, pp. 187, 239;Mitford, 1961, p. 2), almost the entirety of the inscriptions from the site of the sanctuary dates to the Ptolemaic and Roman eras. The pre-Ptolemaic inscriptional record of Paphos, on the other hand, has been forthcoming from sites beyond the plateau of the sanctuary, especially from the site of Marchello (cf. PULP website: https://ucy.ac.cy/pulp/ for recently produced digital plans of the archaeological landscape; also, Iacovou 2008, p. 285. fig. 5;2013b, p. 277, fig. 1). The Cypro-Archaic and Cypro-Classical inscriptions of Ancient Paphos (cf. Mitford, 1971, pp. 7-11, 373-6;Masson, 1983, p. 192, p. 412;Egetmeyer, 2008Egetmeyer, , p. 1005) and the contemporary coin issues (cf. Markou, 2016), provide a long list of royal names from the early 7th c. to the end of the 4th c. BC. (cf. Satraki, 2012, pp. 391-400). The close relation between royal authority and the goddess is made explicit especially by the basileis of the 4th c. BC, who repeatedly underline their dual role as vasileis (kings) of Paphos and iereis (priests) of the goddess (Maier, 1989;Hermary, 2014), though they never address her as Aphrodite but only as wanassa, the sovereign (Karageorghis, 2005, p. 41). Although Homer identifies the goddess of Paphos by name as Aphrodite, on the syllabic Greek inscriptions of Cyprus she is addressed as wanassa or theos (deity) almost to the end of the 4th c. BC, when the name Aphrodite appears on inscriptions for the first time (Karageorghis, 1997(Karageorghis, , 2005Morris & Papantoniou, 2014, p. 184). The inscriptions of Paphos on which the name Aphrodite occurs date to the Ptolemaic and Roman periods and they are alphabetic.
If we were to compare the socio-political information extracted from the inscriptions issued by, or for, the basileis of Paphos to the inscribed records produced during the Ptolemaic and Roman periods (cf. Mitford, 1961;Cayla, 2003;Kantirea, 2015), we would have to admit that both the sanctuary and the deity must have had two chronologically discrete political identities: the earlier one developed in association with the autochthonous city-state system; the later one under foreign colonial regimes. The ideological transformations that took place when the Cypriot poleis lost their political independence and the island was incorporated into the Ptolemaic kingdom and, later, the Roman Empire, have been meticulously analysed by Papantoniou (2012a;also 2013a). The Palaepaphos Urban Landscape Project (PULP) has focused specifically on landscape transformations and settlement patterning in relation to the region of Paphos (cf. Iacovou, 2014).

Ancient Paphos: An Invisible Urban Landscape
Until 2002, when the Archaeological Research Unit of the University of Cyprus and the Laboratory of Geophysical-Satellite Remote Sensing and Archaeo-environment (Mediterranean Institute, Foundation for Research and Technology-Hellas) initiated the "Palaepaphos Digital Atlas Project" (cf. Iacovou, 2008, p. 266), the great temenos at Palaepaphos was perceived as a site of pilgrimage, which, like a monastery, was isolated from the urban landscape and urban activities. Even an informed visitor was left with the impression that it had been purposefully located at a distance from the coast, on the top of a plateau which, though thickly covered with ruins, was otherwise surrounded by many square kilometres of open fields. In the Mediaeval period, the Couvoucle, the Manor House of the feudal estate of the Frankish kings of Cyprus, was constructed on the plateau of the sanctuary, less than 200m south of the temenos (Maier, 2004, p. 56). In the meantime, the hamlet that took its name from the Couvoucle (today pronounced Kouklia) encroached upon the sanctuary's northern perimeter (Fig. 4). Despite the impressive body of textual records that identified Paphos as the capital centre of a city-state (see above), the material remains of the ancient polity, to which the most celebrated sacred site of Ancient Cyprus owed its existence, remained unrecognised.
With the exception of the surviving megalithic ashlar blocks of the temenos (Fig. 5), the built structures visible today within the sanctuary site date to the Roman period (Maier, 2004, p. 45). The Cypriots, however, knew well that of all their sanctuaries that of Paphos was the oldest: according to Tacitus, in the 1st c. AD, they claimed before the Roman Senate that it was older than the sanctuary of Zeus at Salamis and older than the sanctuary of Aphrodite at Amathous (Annals 3.62: vetustissimum Paphiae Veneri). Their claim has been archaeologically confirmed by excavations carried out in the 20 th century: the earliest of any kind of remains at Salamis (cf. Yon, 1993;Fourrier, 2018) and Amathous (cf. Hermary, 1999;Iacovou, 2002) date to the Early Iron Age. At Paphos, on the other hand, the meticulous investigation of the temenos area by Maier confirmed the sanctuary's establishment ca 1200 BC (Maier, 1985(Maier, , pp. 12-13, 2004. The studies of Maier and other scholars notwithstanding, including that of Webb (1999), which is of decisive relevance as regards the role of the cult centre during the earliest part of its life, no other built monument of the LBA is readily visible at Palaepaphos. Yet, the wealth of the Late Cypriot tombs in the area (cf. Catling, 1968Catling, , 1979Karageorghis, 1990) and the recently published well fillings of nearby Evreti (Rüden, Georgiou, Jacobs & Halstead, 2016) leave no doubt as to the economic robustness achieved by the Late Cypriot polity and its active participation in maritime trade.  In the 1950s, three monuments of the Archaic and Classical periods were found at some distance from the temenos, on different plateaus. They are the rampart on Marchello, the palace on Hadjiabdoulla and the Classical peristyle house at Evreti (cf. Maier, 2004, pp. 59, 74, 77) (Fig. 6). Although all three received the short-term attention of the British Kouklia Expedition, they "were far from being fully excavated" (cf. Maier, 2004, pp. 34-35). In 1966, F.-G. Maier resumed fieldwork at Palaepaphos as director of the Swiss-German Expedition. In the next four decades, with an unparalleled attention to detail, he pulled together all the evidence produced (but left unpublished) by previous expeditions and used it to create a diachronic historical diagram of Palaepaphos. We recently found it necessary to re-examine his interpretation (inherited from Mitford) of the relationship of the Marchello rampart to a city wall (Maier, 2008 and below), which does not seem to exist (cf. Iacovou 2013b, pp. 282-285, fig. 3;Fourrier, 2018, pp. 143-144). Nevertheless, Maier's synthesis of the available material evidence remains the most valuable guide to the cultural horizons of Palaepaphos and the chequered history of its research. Hundreds of tombs dating from the LC period to the end of the Roman era were, and are discovered annually (often illicitly by tomb robbers), not only around the plateau of the sanctuary but also around and beyond the other plateaus (Fig. 7). The radius has expanded considerably since the discovery of the Early Iron Age Skales necropolis to the south-east (Karageorghis, 1983;Karageorghis & Raptou, 2016), and even more so with the totally unexpected discovery of the Cypro-Geometric burial site of Plakes to the west site of the river Dhiarizos (Raptou, 2002;Karageorghis & Raptou, 2014). Admittedly, the spatial distribution of the burial clusters creates an almost complete circle around the sanctuary, and give the impression of an extra muros mortuary landscape. Despite the fact that all these tomb clusters do not necessarily belong to the same cultural horizon, scholars have attempted to estimate the size of Ancient Paphos in the LBA by measuring an imaginary intra muros area. Estimates varied between 65 (Knapp, 1996, p. 61) and 144 hectares (Merrillees, 1992, p. 317), but these site-size exercises (Iacovou, 2007, pp. 1-6) did not make the urban landscape more visible or more comprehensible.

The Palaepaphos Digital Atlas (2002-2003)
In a landscape packed with rich anthropogenic data that has been continuously lived upon since the 2nd millennium BC, one does not expect to recover equally substantial parts of the urban structure of all chrono-cultural phases. In the case of Palaepaphos, the daunting expanses of the still empty fields around the sanctuary offered a potential (for archaeologists seeking to recover special function areas) and a threat (by developers demanding land-use changes). In 2002-2003, given the opportunity of a Greece-Cyprus collaboration programme on the use of advanced technologies, we implemented the Palaepaphos Digital Atlas. The project was primarily designed as a heritage management tool and it had two main components: the development of a GIS linked to an entity-related geo-database and a multi-sensor geophysical survey (cf. Sarris et al., 2006). As a first step, the vast amount of archaeo-cultural information from published and unpublished records dating from prehistoric to pre-modern times was collected and placed under a digital 'umbrella'. Also, a digital land relief of the municipal area of Kouklia-Palaepaphos with current property and land-use status was produced with the collaboration of the Department of Lands and Surveys of the Republic of Cyprus.
The archaeo-data were collected from a widely dispersed area (exceeding three sq. km.); they were analysed into chrono-cultural layers and the next step was their spatial analysis on the property parcels map. Here, we encountered a practical problem especially in relation to tombs that had been excavated in the context of rescue operations. Their geo-location was described in the records of the Department of Antiquities with a regional toponym on the cadastral map but, only rarely was a property plot number noted. Toponyms, as we were to find out, defy exact borders; hence, a full integration with the property parcels map is not possible. Fortunately, we were able to gain a more precise knowledge of the find spot of many tombs from the local custodian, Onisiforos Loukaides. Over his 40 years of service, Onisiforos had not only supervised the excavation of hundreds of tombs during rescue operations; he had also registered them, adding in his ledger hand-made drawings of diagnostic pottery and other artefacts. This invaluable information allowed us to work on a preliminary spatial and temporal landscape analysis of Palaepaphos and to transfer the results onto orthophoto maps.
As soon as we had digitized the private property parcels and had marked the listed plots (the A and B schedule protected parcels of the Department of Antiquities), it became alarmingly clear that only 11 hectares were in the A list (i.e. they had been expropriated), and almost all of them were concentrated on the plateau of the sanctuary (Fig. 8). This confirmed our worst fears: in spite of the fact that almost 60% of the private parcels had been included in the B schedule protected list (i.e. plots of potential archaeological interest), these areas and the remaining urban landscape of Ancient Paphos could have been easily annihilated and for ever lost. All it would take was political pressure on behalf of the local land owners to have their plots taken out of the agricultural register. Once declassified the land could be incorporated into new urban development plans.

The Geophysical Survey (2003, 2007) and the Initiation of PULP (2006)
According to the Antiquities Law of Cyprus, the state's heritage manager can justify the expropriation of a plot when it can be shown that it contains standing monuments. For as long as the urban nucleus of Palaepaphos remained evasive it could not receive a protected status. Hence, the need for a geophysical survey became imminent. Since we were still harbouring the impression that the area of Palaepaphos and its sanctuary were contained within an invisible city-wall, in 2003, we covered 44.178 sq.m. of ground in an effort to define an intra and an extra muros landscape (Sarris et al., 2006). It was in the course of this survey (with magnetic prospection and electrical soil resistance) that the realities of the topography made it clear that this long-established belief was suspect (Iacovou, 2008, pp. 272-273). The assumption that the ancient polis was contained within a full circuit of a defensive rampart had been promoted by the directors of the short-lived Kouklia Expedition (Iacovou, 2013b, p. 281). Despite the fact that T.B. Mitford and J.H. Iliffe had only just arrived in Palaepaphos in 1950, in their first joint report the two of them expressed the hope that "it might be possible eventually to trace the whole circuit of the city walls, and thence obtain some clue to the size and population of Bronze and Early Iron Age Paphos" (Mitford & Iliffe, 1951, p. 57).
We have subsequently described our survey adventure, as "a research strategy based on false impressions" (Iacovou, 2013a, p. 276), but, if it were not for the geophysical survey, we would still have been thinking of an urban centre within a city wall to this day. Instead of giving up on geophysical prospection, in 2007 we increased its extent by another 56.202 sq.m. (cf. Iacovou, 2008, p. 264, n. 3) using a ground penetrating radar in addition to magnetic prospection and electrical soil resistance (Fig. 9). This time we were targeting the three main plateaus located to the north and east of the sanctuary (Marchello, Mantissa and Hadjiabdoulla) in the context of PULP, which had been initiated the previous year. The 'removal' of the city-wall from the urban equation meant that a new research strategy was necessary, which would help us comprehend the structure of the ancient polity. The initial activity of PULP was, therefore, geared towards the plateaus, where the geophysical prospection indicated intense activity and, also, because on two of them (Marchello and Hadjiabdoulla) sections of built monuments had been exposed in the 1950s (above). Landscape analysis was intensified through targeted short-term excavations that allowed us to confirm that all three plateaus were special function zones of the urban nucleus since the LBA (for a short description of field activity from , Iacovou, 2017a. PULP's landscape analysis, therefore, begun with (a) the medium-scale documentation of the plateau of the sanctuary and the three neighbouring plateaus to its north and east. With the initiation of fieldwork in 2006 we developed (b) a micro-scale analysis to document standing monuments and movable finds in each of the excavated sectors (consult Fig. 16). Before long, when the critical question of the polity's foundation in relation to the megalithic temenos became pressing, it was necessary to develop (c) a third scale of documentation: a macro-scale analysis of the hydrological basin of Paphos (Iacovou, 2012(Iacovou, , 2014. The establishment and long-term survival of a Cypriot polity required access to copper deposits as well as to a port of call (cf. on the state model of Cyprus: cf. Iacovou, 2007Iacovou, , p. 18, 2013cKnapp 2013, pp. 36-40). The hydrological basin or catchment of Paphos represents the essential or minimum economic territory that could have provided both copper deposits and a gateway to the sea (Fig. 10). The presence of slag heaps in the Paphos Forest (25 km from the coast) had been noted (Stos-Gale, Maliotis, & Gale, 1998). Also, paleo-environmental studies had shown that due to coastal transformations the main 2nd millennium BC gateways of Cyprus (e.g. Enkomi, Kition, Hala Sultan Tekke) are neither visible nor are they situated near the current coastline (cf. Gifford, 1985;Devillers, 2008). Yet, neither the cupriferous zone within the territory of Paphos, nor the possibility that a Bronze Age anchoring facility might have disappeared inland (cf. Walsh, 2014, p. 50), due to silting and sea level change, had received consideration in relation to the rise of a powerful polity in Paphos.

The Foundation Horizon of Paphos
Although the notion that the sanctuary provided the excuse for Ancient Paphos to grow around it is still occasionally expressed, the fact is that the temenos was constructed 500 years after the foundation of the gateway of Ancient Paphos: the earliest ceramic evidence from the Palaepaphos tombs places the initial establishment of settlers on the site to the transition from the Middle to the Late Cypriot horizon, ca 1700 BC (cf. Crewe & Georgiou, 2018, pp. 55-56); the temenos, on the other hand, was constructed during the transition from LCIIC to LCIIIA, ca 1200 BC. The sanctuary, therefore, was not present during the foundation horizon of the site (ca 1700 BC). Contrary to the statement that "Paphos' claim to fame did not so much spring from political or economic power ... but rather rested on her Sanctuary" (Maier, 2004, p. 12), which continues to appear in the work of scholars (cf. Howitt-Marshal, 2012, p. 116), the rise of a polity and the management of an economically durable territorial state like Ancient Paphos could not have relied on revenues from pilgrim tourism. In any case, during the first phase of its existence, the sanctuary was neither a pan-Cypriot nor a pan-Mediterranean site of pilgrimage; it was the urban cult centre of the city-state of Paphos.
The settlement of ancient Paphos had not grown around a sanctuary; it had grown around a harbour through which Ancient Paphos participated in seaborne commercial activities since the early LC horizon (cf. Crewe & Georgiou, 2018, pp. 64-65). The first to provide convincing evidence regarding the foundation of Ancient Paphos as a gateway to the sea during the transition from the Middle to the LBA was Giorgos Georgiou (2007, pp. 435-437). The number of sites recorded in the catchment in MCIII/LCI is nearly double by comparison to the previous phase (it jumps from 25 to 45), and includes the first evidence of activity at the site where the urban centre of Palaepaphos was to grow (Georgiou, 2007, pp. 425, 477). This transitional phase is the horizon of the foundation of all the main gateways -e.g. Enkomi, Hala Sultan Tekke -through which Cyprus joined the Mediterranean exchange network of the second millennium BC (cf. Keswani, 1996;Peltenburg, 1996, pp. 30-35).
In 2010, when Athos Agapiou undertook the first extensive geospatial analyses (including viewshed, least-cost path, proximities, site catchment analysis, etc.) of all the available Bronze Age settlement data from the hydrological basin of Paphos, his results showed that the majority of LC I sites are concentrated on a north to south axis (Agapiou, 2010); they mark two different routes, one along the river Dhiarizos, the other along the Ezousas river (Fig. 11). Both routes originate in the copper-rich foothills of the Troodos and lead to the coast: the Dhiarizos route terminates at the newly founded Late Cypriot settlement of Paphos; the Ezousas route comes close to the coastal site of Yeroskipou, some 8 km. to the west of Ancient Paphos (Agapiou, Iacovou & Sarris, 2010;Crewe & Georgiou, 2018, p. 63).

The Port Lagoon and the Paphos Forest Project
Thirteen years since the launching of PULP, a strong consensus has been reached over the foundation of Paphos: it must have been founded as the terminal link in a chain of settlements involved in the procurement and transport of copper to the coast. Hence, half a millennium before the construction of the sanctuary, the exploitation of hinterland resources (primarily, copper but also timber) on behalf of inland communities had led to the colonization of the coastal landscape and the establishment of at least one safe anchorage. River silting and coastal uplift (Zomeni, 2012) may have erased every trace of the site of the original port of Ancient Paphos, but not the collective memory of its existence, which survived to pre-modern times and was recorded by Kyprianos in the Chronological History of the Island of Cyprus (1788, p. 18). Kyprianos states that near the sanctuary there was once a lake large enough to serve as a port; it was thought to have served as the anchorage of the fleet of Agapenor, legendary Greek founder of Paphos (Pausanias 8.5.2). This lakesize lagoon was cut off from the sea and was transformed into an unhealthy marsh, whose stagnant waters caused the people to suffer many illnesses during the harvest season (cf. Iacovou, 2013bIacovou, , p. 286, 2014. The spatial definition of this lagoon is currently as elusive as the precise location of the copper mines that must have been exploited in the 2nd and 1st millennia BC. However, copper procurement from the Paphos catchment has been confirmed for Late Antiquity (cf. Kassianidou, 2013, p. 59), especially following the work of the Paphos Forest project. PULP's Paphos Forest survey was based on the pioneering work of Stos-Gale, Maliotis & Gale (1998). It has so far resulted in the identification and volume estimation of a dozen slag heaps in what would have been in antiquity an industrial landscape on the fringe of the forest (Iacovou, 2014, pp. 170-171). The geological horizon of the upper and lower pillow lavas of the southern foothills of the Troodos share the same altitude with the southern boundary of the thick forest, whose timber could have been used for fuelling the smelting kilns (Socratous, Kassianidou & Pasquale, 2015), and also for shipbuilding (cf. Mitford, 1961, p. 9, no. 17;Michaelides, 1996, p. 140;Raptou, 1996, p. 256).
Copper was primarily an export commodity (cf. Knapp, 2013, p. 308). Textual sources and material evidence leave "little doubt that Cyprus was a significant source of copper in the Levant and the Aegean prior to 1700 BC" (Webb, 2018, p. 84). Established transport routes would have connected the mining landscape of the region of Paphos with the gateway to the sea. The distance between the two via the Dhiarizos river valley is only 25 km. The location of the port lagoon is still debated, and there may have been more than one in the course of the LC and the Early Iron Age. Nonetheless, a protected harbour in a natural lagoon must have been operating for a long time before the construction of the sanctuary. As in the case of the contemporary foundation of Enkomi (ca 1700 BC) on the shores of a natural lagoon on the east coast of Cyprus (cf. Crewe, 2007;Devillers, 2008), the earliest use of the coastal landscape of Ancient Paphos would have taken place in the context of its colonisation by human groups congregating on the site to manage the newly-founded gateway (Keswani, 2012, p. 188). In 2012, the site of Achni (consult Fig. 7) was put forward as a possible early Bronze Age anchorage of Ancient Paphos (Howitt-Marshal, 2012) though no LBA evidence has been recorded either on the nearby coast or along Cha Potami (the easternmost stream in the catchment of Paphos) that could support the candidacy of Achni as a 'proto-harbour' (cf. Knapp, 2018, p. 139). Admittedly, the 120 stone-anchors recorded at a distance of 50m from the current shoreline, and identified as of Middle and LBA types, demand explanation. However, to this day, their location (Howitt-Marshal, 2012, p. 110, fig. 7.5) has not been associated with LBA human activity on land, nor with a natural lagoon that could have served as a safe anchorage. The anchors are littering one of the most inhospitable and treacherous coastal landscapes of the south coasts of Cyprus. Based on the currently available Bronze Age site distribution within the hydrological basin (Agapiou, 2010), the delta of the Ezousa's river by the coastal area of Yeroskipou could have functioned as an alternative, or even competing, gateway to Ancient Paphos for some time during the LBA. Only an ancient harbour palaeo-geographic project that will extend from Cha Potami to Yeroskipou can provide definitive conclusions, as it did in at least four other cases: the port of Bamboula in Kition (Morange et al., 2000), the Pyla natural harbour (Caraher, Moore, Noller, & Pettegrew, 2005), the bay of Enkomi-Salamis (Devillers, 2008), and the port of Hala Sultan Tekke (Devillers, Brown & Morange, 2015).
Turning to the landscape around the plateau of Alonia, where the sanctuary is situated, the data collected and analysed by PULP suggest that it was first used in MCIII/LCI for the establishment of settlement and burial clusters by groups that had settled in what was to become first the gateway and later the polity of Ancient Paphos (Fig. 12). The fact that each group kept its tombs close to their respective residences (Crewe & Georgiou, 2018, p. 65), instead of organising a communal burial ground, suggests that the first occupants came from different MBA settlements of the wider Paphos region. Keswani has observed that this mortuary practise is related to the establishment of new LC gateways, like Enkomi (1996, pp. 221-22, 236). To the best of our current knowledge, both the coastal region of Enkomi and the coastal region of the territory of Paphos were virtually uninhabited in EC and also until the last phase of MC. They were lands that began to receive settlers in MCIII. In the absence of MC regional authorities, the coastal frontiers of Cyprus were not colonised in an organised manner. Instead: [M]embers of different regional communities converged to exploit the opportunities for trade and other economic activities. In the absence of an initial sense of communal identity, the settlers may have dug or built their tombs near their houses to symbolize, through the immediate presence of their ancestors, their local identity and their rights to local residence and resources. (Keswani, 2012, pp. 189-190)  The settlers of Paphos continued to use the family tombs they had dug near their respective houses even after the temenos had been erected ca 1200 BC; the richness of the material evidence dating to LCIIIA suggests that the site clusters were most in evidence in the 12th c. BC, when they co-existed with the sanctuary (Georgiou & Iacovou, in preparation). Assuming that the first settlers would have chosen to live close to the port lagoon, the evidence for their living and burial nuclei could point to its location. When we upload the LC nuclei (Asproyi, Eliomylia, Evreti, Hadjiabdoulla, Kaminia, Mantissa, Marcello and Teratsoudhia) on the geomorphological map of Palaepaphos we observe that they are closely distributed to the north and east of the narrow streambed of Loures (Fig. 13). Loures collects water from the ravines and the deep valleys formed between the plateaus and takes it to the sea (today the water is collected and the streambed is cultivated). The west side of Loures is defined by the sharp slopes of Alonia, the plateau which holds the terrace on which the sanctuary was established (Fig. 14). If the lake-size lagoon that had served as the gateway's first anchorage was situated in the lower south part of Loures then the temenos was not only close by, it also had visual contact with the port.

Unlocking the Significance of the Sanctuary's Location
The location of the LC temenos -in the literature it is often referred to as Sanctuary I (cf. Maier, 2004, p. 39) -should not be considered fortuitous. Given that it is one of only two such sacred monuments in Cyprus, we should turn to its 'sibling', the contemporary Temple 1 of Kition which, together with three more cult buildings, was erected next to the port-lagoon of the urban centre (cf. Webb, 1999, p. 64). According to Gifford's map of coastal changes (1985, fig. 4), a 'marsh or shallow lagoon' existed to the NE of the wall and the towers that protected the temple precinct of Kition (see also, Yon, 2006, pp. 46-47, fig. 24.1). Excavations in Kition have confirmed that copper working was practised in workshops within the precinct of the sanctuary. What is not often mentioned in the case of the Paphian sanctuary is that, back in the 1950s, copper slag was found in the sanctuary in association with LBA pottery (Megaw, 1951, p. 13;Iacovou, 2012, p. 66). Evidently, as Webb has noted, the relationship between cult and maritime trade in LCIIIA was "of at least equal importance to that between cult and metallurgy" (Webb, 1999(Webb, , p. 302, 2014. The intensification of the metal industry in relation to cult in the 12th c. BC, which is also strongly evident in Enkomi, has led Sherratt to suggest that the export trade was managed by the sanctuaries (1998, pp. 300, 304). Was proximity to harbour facilities, "the key factor in the initial location of the sacred area" (Webb, 1999, p.287) in the case of the Paphian sanctuary as it was for Kition? When this relation is confirmed, one will be able to claim that in Paphos as in Kition port and sanctuary functioned as an inseparable production and management unit. The institutionalisation of ritual and the function of LC urban sanctuaries as centres of ritualised authority do not become visible in the archaeological record of Cyprus until after the late second millennium BC Mediterranean-wide crisis, which led to the demise and abandonment of major LC polities (e.g. Kalavassos-Ayios Demetrios, Maroni-Vournes and Alassa-Paliotaverna). These urban centres had administrative complexes with industrial units and significant storage capacity (cf. Fisher, 2014) but urban cult buildings have not been identified in any one of them. Following their abandonment in LCIIC-IIIA (cf. Iacovou, 2014, pp. 662-663) only three coastal centres "witnessed a return to, or stabilisation of, centralised authority in the twelfth century that allowed them to take advantage of new cultural and commercial links" (Webb, 2014, p. 624): Enkomi, Kition and Paphos. The economic and industrial intensification in these three centres appears to have been closely associated with the establishment of urban cult buildings (e.g. the Sanctuaries of the Horned God and the Ingot God in Enkomi: Webb, 1999, pp. 91-113, 119-22), which in Kition and Paphos stood out on account of their monumental and emblematic character. Unknown in Cyprus to that date, the megalithic ritual architecture of the temene of Kition and Paphos "was accompanied by elements exclusively associated with cult, namely stepped capitals, high platforms and horns of consecration" (Webb, 2014, p. 627), which would have played a significant role towards the intensification of ceremonial activity. Sacred ceremonies may have been only one aspect of the sanctuaries' function as industrial and storage centres (Knapp, 2013, p. 372) but they were of vital importance for the legitimisation of regional authority by local elites. In her analysis of ritual architecture in LBA Cyprus, Webb discusses the manipulation of sacred space in the LCIIIA urban sanctuaries and, in the case of Kition and Paphos, points to "a significant investment of wealth, energy and labour [which] was expended in the formal differentiation of elite from non-elite observance" (Webb, 2014, p. 628). The construction of these unprecedented labourdemanding sacred edifices signifies the institutionalisation of authority in the economic regions of Kition and Paphos respectively. Therefore, the establishment of Sanctuary I (in Paphos) and Temple 1 (in Kition), ca 1200 BC, signals the moment when these two Late Cypriot port cities became the capital centres of microstates. "With the emergence of social complexity, religion provides a key source of power, justifying ruling institutions that they mirror […]." (Earle, 2017, p. 14).
It is worth noticing that the sanctuary enjoys an impressive visibility towards the sea and vice versa (Fig. 15). It is not unlikely that the megalithic temenos functioned as a beacon for incoming sea traffic. The connection of the sanctuary with a commercial harbour should explain why Homer knew of Paphos as the site of the temenos of a goddess whom he, first, identifies as Aphrodite (Odyssey 8.363), and as Kypris (Iliad 5.330). It was because of its international harbour that the temenos had acquired visibility beyond the island, and its fame as sacred abode of the Cypriot Goddess reached Homer. Before the Ptolemaic and Roman eras the sanctuary was not an international venue of pilgrimage; nor did it receive offerings from other Mediterranean cultures or even from the rest of Cyprus. "The dearth of foreign imports in [Cypriot] sanctuaries is striking, when compared to the numerous finds from contemporary tombs or palaces" (Papasavvas & Fourrier, 2012, p. 301). This observation is particularly true in the case of the Early Iron Age cemeteries of Ancient Paphos (cf. Raptou, 2014, 2016 and Fig. 24 below). They contain a remarkable number of imported ceramic vessels and finished products made of imported materials (e.g. gold), which reveal the Paphian harbour's maritime activity, especially with the Levant (cf. Bikai, 1983;Bell, 2016).

Urban and Extra-Urban Cypriot Sanctuaries
Researchers working on the LBA and the Iron Age archaeology of Cyprus respectively seem to have reached similar conclusions with respect to the cult sites of Cyprus: ever since the LBA, Cypriot sanctuaries, urban and extra-urban, were founded by dominant authorities, who were in control of regional economic resources. We tend to think of the extra-urban sanctuaries as a novel feature of the Cypro-Archaic period because dozens of them were established all over the island in the context of the consolidation of the Iron Age citystates (cf. Fourrier, 2013). However, Enkomi is credited with the establishment of the sanctuary of Athienou early in the LC period for the same reasons that led to the foundation of extra-urban sanctuaries in the same area by her successor, Salamis, in the Cypro-Archaic period: "large-scale communal participation in ritual performances played a significant role in attempts to build organisational structures in the hinterland." (cf. Webb, 2014, p. 626). These "remarkable organizational instruments", as the extra-urban sanctuaries have been described, gathered local communities around common cults (Fourrier, 2011, p. 130) and served as the main stage for the visible promotion of the rulers' agendas and ideology (Satraki, 2013). Despite occasional visits from further afield, they operated exclusively within their politico-economic regions. An equivalent to the Panhellenic sanctuaries that could have brought together people from the different city-states of the island did not exist in Cyprus (cf. Étienne, 2010, p. 63). No sanctuary had functioned as a pan-Cypriot cult centre until after the Ptolemaic take-over of the island (Papantoniou, 2013b, p. 48).
But "why did the kings of Salamis make dedications to the Apollo of Delphi and Delos and not to the Aphrodite of Paphos?" asks Fourrier, so that she can respond that "[r]elationships between Cypriot kingdoms were relationships between kings; they did not use the sanctuaries as intermediaries" (Papasavvas & Fourrier, 2012, p. 301). Urban sanctuaries in particular had an enhanced symbolic and ceremonial function that promoted the narrative through which the rulers' right to authority was legitimatised: the basileis of Paphos had the right to kingship because they were priests of the wanassa; and they were priests of the goddess because they claimed descend from Kinyras, her legendary high priest. Cypriot sanctuaries, therefore, were state-controlled institutions, and of all the urban sanctuaries of Cyprus, the sanctuary of Ancient Paphos has the longest continuous history as a sacred topos; founded ca 1200 BC by one of the earliest and most successful city-states of Cyprus, it remained inseparable from the authority of the priestkings to the end of the 4th c. BC. When Ptolemy Lagos finally won Cyprus and abolished its independent polities (ca 300 BC), the Ptolemaic administration gave the sanctuary of Paphos a new political role as an island-wide cult centre (Papantoniou, 2013b, pp. 46, 49). It also became the centre of the dynastic cult and, not long after, the seat of the KOINON KYPRION (Papantoniou, 2012, p. 154), which in the Roman era minted the bronze coins of the island with the symbol of the omphalos, the aniconic cult image of the goddess (cf. Maier, 2004, p. 50, fig. 35).

Monumentalizing the Landscape of the Cypriot Goddess
When we launched the project in 2006, we were almost exclusively dependent on a custom-made multidimensional digital platform (with Geographical Information Systems), where cartographic information is combined with archaeological data organised on a relational database. A couple of years later, with the development of three distinct scales of landscape analysis (micro-, medium-and macro-scale analysis) and the unexpected discovery of previously unknown standing monuments, we had to employ advanced documentation and imaging technologies (Global Satellite units, total stations and RGB digital cameras mounted to UAVs), as well as geospatial analyses (viewshed, least-cost path, proximities, site catchment analysis, etc.). In the course of 13 consecutive years of field operations (2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018), the results of our targeted excavations to the north and east of the sanctuary (Fig. 16) have exposed a politically charged landscape, whose built monuments have absorbed some of the most dramatic moments in the history of the urban settlement of Ancient Paphos from the Cypro-Archaic period to almost the end of the Ptolemaic era. Rising to 108-110m. asl Mantissa, Marchello and Hadjiabdoulla share a striking inter-visibility with each other and with the sanctuary on the Alonia terrace, which is at a lower elevation (104 m. asl). Mantissa, the middle terrace, is a heavily eroded plateau where no built structures have survived but it is still very rich in LBA ceramic materials.
The preliminary results of our excavations on the northern plateau of Marchello and our attempt to reinterpret the Cypro-Archaic monument as something other than a part of a city-wall have been reported in a number of publications (Iacovou, 2008(Iacovou, , 2013b(Iacovou, , 2017a. At Marchello there is an east to west stretch of a 165 metre-long wall of a rampart; it includes a dog-leg gate and drafted blocks looking towards the south (Fig. 17). The landscape to the south of the wall has not been investigated and the landscape to the north was thoroughly destroyed in the process of levelling operations. Nevertheless, the results of the excavations, first by the Kouklia Expedition under Mitford and Iliffe and subsequently by the Swiss-German Expedition under Maier (cf. Maier,pp. 35,[59][60][61][62][63][64][65][66][67][68][69][70][71][72][73], leave little doubt that Marchello was the site of a violent confrontation. Dedicatory inscriptions by kings and sculptures bearing royal insignia were purposefully destroyed and buried in a huge ditch to the north of the rampart (cf. Maier, 2004, figs. 55-57). The mixed secular-sacred character of the material suggests that the Marchello plateau may have been the site of a Cypro-Archaic citadel. "The religious function attached to the palaces of the Cypriot kings was certainly fundamental, as it was throughout the ancient Near East." (Hermary, 2013, p. 95). The timing of this act of violence, from which Marchello did not recover its secular-sacred functions, coincides with the initiation of an ambitious building programme on the plateau of Hadjiabdoulla, one km east of the sanctuary (Fig. 18). In our view, an episode of violent conflict, whose outcome seems to have been commemorated with an act reminiscent of a damnatio memoriae, was followed by the implementation of a new place-making strategy on a different site. Apparently, the agency that conceived and carried out this strategy could have been a new dynasty that rose to power ca 500 BC. For the next two hundred years, the Hadjiabdoulla plateau appears to have been the administrative citadel of the Paphian polity.   Besides the palatial edifice (East Complex) partially unearthed by the British Kouklia expedition in the 1950s, but left unpublished (Maier, 2004, p. 74, fig. 58), our excavations along the north side of Hadjiabdoulla revealed a second, contemporary, stone-built edifice (West Complex) preserved to a height of almost 2 m. This complex extends over 65m. along the exterior side of a citadel parapet wall. To this day, we have exposed 290 sq.m., hardly a fourth of the West Complex, which extends down the north slope on stepped terraces supported by retaining walls (Fig. 19). Corridors lead to storage and processing units equipped with an efficient system of drains (Fig. 20). This major investment of the fifth-century kings of Paphos was not residential; the building had well plastered stone walls, sizable roof tiles of a type regularly used on important Classical buildings, mill-stones, hearths, industrial installations (Fig. 21) and plenty of storage vessels. Nikokles, the last sovereign to call himself king of Paphos and priest of the goddess, was also the last to have controlled the economy of the polis of Paphos from this citadel complex. Besides olive oil, he invested in the production of purple dye, and his stores had a wide selection of mostly wine amphorae imported from the Aegean, the Levant and North Africa. Two hundred years after his extermination by Ptolemy Lagos, most sectors of the royal industrial complex continued to function; the latest imported amphorae date to ca 100 BC. Besides the East and the West Complex, a third monument, a fortress, was also being constructed at the same time (ca 500 BC) on the Laona hillock, only 70m to the North of the palace (consult Fig. 18). The excavated east side has two staircases facing each other, apparently giving access to towers (Fig. 22). The walls of the fortress survive to a maximum height of 6m. because sometime in the 3rd c. BC the fortress, already in a state of abandonment, was completely buried under a huge earthen mound (cf. Iacovou, 2017a, pp. 209-210, figs. 20-22;Lorenzon & Iacovou, 2019, fig. 3).    With the establishment of the Cypro-Classical citadel on Hadjiabdoulla and the fortress on Laona the landscape to the east of the sanctuary must have become the prime urban quarter of the capital -but probably not for the first time. Besides the Evreti wells with the rich deposits of LBA elite materials (e.g. scraps of worked ivory) from a settlement context (Rüden et al., 2016), a monumental building seems to have occupied the north side of the plateau of Hadjiabdoulla in the LBA: the drafted ashlar blocks sitting on an equally finely crafted levelling course on the south side of the palace cannot be associated stylistically with the Cypro-Classical period masonry of the building. They are LC ashlar blocks and, since they appear to be in situ, they have to belong to a building of the 13th-12th c. BC. This suggestion is further strengthened by a short-term investigation we carried out directly outside and to the south of the palace. Instead of the usual Iron Age ceramic material, we recovered substantial quantities of LC wares (Georgiou & Iacovou, in preparation). For the first time, the plateau could be identified as an integral part of the LBA urban landscape of Ancient Paphos. The temenos is no longer the only LC built monument of the polity. Evidently, the building was at least partly visible when the decision was taken to build the south wall of the palace on top of the LC ashlar blocks. This was no random act; it was a well calculated political decision. Whoever took it -most likely the founder of a new dynasty -wanted to claim descent from powerful ancestors, who had also monumentalised the landscape of the plateau hundreds of years earlier. From the LC building one would have had a commanding view of the coastline as well as a full view of the temenos precinct. Today, from the ruins of the palace one still has an unimpeded view towards the sea but the sanctuary is completely hidden behind the Mediaeval Manor House of Kouklia (Fig. 23).

From Paphos to Palaepaphos and from Palaepaphos to Kouklia
The urban landscape of Ancient Paphos, as it developed in the course of the first millennium BC -with communal cemeteries organised well beyond the habitation area since the 11th-10th c. BC (Fig. 24) -would not have changed much or suddenly after Ptolemy exterminated the local dynasty at the end of the 4th c. BC. To this, however, there is one monumental exception: the construction, almost certainly in the 3rd c. BC, of a mysterious mound on Laona that completely covered the earlier fortress. Tumuli, especially of this size, are foreign to the material culture of Cyprus (Iacovou, 2017b). This has to be the reason why a man-made structure comprised of 13.700 cubic metres of transported soils was viewed as a natural hillock until the identification of its anthropogenic nature by PULP in 2012 (Karkanas & Goldberg, 2018, p. 219). The emblematic monument is clearly visible from the sanctuary. Unlike the Laona fortress, which was a project of the local dynasties (Lorenzon & Iacovou, 2019, p. 351), the Laona tumulus was almost certainly a project that had been conceived and carried out by an exogenous authority, which in this case had to be the Ptolemies. Although the purpose of its construction has not yet been identified, the Laona tumulus is a monumental introduction to the cultural remodeling of the island's urban (cityscapes), mortuary and ritual landscapes that we associate with Hellenistic Cyprus and Ptolemaic administration (cf. Michaelides & Papantoniou, 2018). PULP's current results suggest that the landscape context of the sanctuary of the Cypriot goddess must have changed even before its transformation into a pan-Cypriot cult centre under the Ptolemies. The port lagoon, in the name of which Ancient Paphos was founded ca. 1700 BC, could not have been functioning as late as the Cypro-Classical period. Because of river silting, but also because the size and type of ships had been changing, it is unlikely that the same anchorage, close to which the sanctuary was established, was still in use in the 4th c. BC. Alternative anchorages, one of which may have been near Yeroskipou, must have been sought before the 4th c. BC, when we have substantial evidence that the last king, Nikokles (he died in 306 BC), established at Nea Paphos the city-state's new official harbour (cf. Młynarczyk, 1990). Nikokles would have also redirected the transport routes from the hinterland so that they could terminate close to the new gateway and the new port facilities. His royal inscriptions, which survive in the monastery of Ayia Moni on a saddle between the rivers Xeros and Ezousa and close to the cupriferous deposits (consult Fig. 10), suggest that he sanctified and protected the new route with a new extra-urban sanctuary (Iacovou, 2012(Iacovou, , p. 65, 2014. This rerouting would have severed the relation of the sanctuary with the settlements along the Dhiarizos river route, which for centuries had been employed for the transportation of resources from the Troodos foothills to Ancient Paphos. Not long after, Nea Paphos became the capital of Cyprus and a major naval station where the Ptolemaic fleet was anchored (Nicolaou, 1976, p. 16;Młynarczyk, 1990, pp. 24, 121;Keen, 2012, pp. 52-53). By then, Palaepaphos was completely cut off from the terrestrial and maritime landscapes that had once been the economic territory of the city-state of Paphos.
In the context of the political and economic unification of the island that was gradually accomplished by the Ptolemaic administration (cf. Bagnall, 1976;Papantoniou, 2012b), Palaepaphos had neither an administrative status nor an economic significance. It had become a sanctuary town (Maier, 2007, p. 17). The sanctuary, on the other hand, continued to function as a ceremonial arena in relation to the political economy not of the city-state of Paphos but of the whole island-state of Cyprus. Inscriptions dedicated in the sanctuary, especially one in honour of naval architect Pyrgoteles, who built for Ptolemy II a thirtyoared and a twenty-oared warship (Nicolaou, 1976, p. 108, no. 74;Młynarczyk, 1990, p. 109), and another in honour of Potamon, a Cypriot who held the high office of antistrategos and director of the copper mines of Cyprus (Mitford, 1961, p. 39, no. 107), suggest that the Ptolemies used the most ancient of the Cypriot cult centres to establish economic and ideological control over the island's resources. Minerals, timber and agricultural produce became state monopolies; they were nationalised (Gordon, 2012, p. 503). Because the colonial regime of the Ptolemies continued the exploitation of the same resources that were previously exploited by the city-states, the territory of Paphos continued to prosper. The main difference was that now the routes from the hinterland were directed towards Nea Paphos. Thus, while the urban structure of Palaepaphos began to shrink, the settlements in the catchment of Paphos experienced a population growth that lasted until the early Roman era (Rupp, 1997). Before the end of the 1st c. BC, when the Romans replaced the Ptolemies as colonial rulers of the island, the urban nuclei on the plateaus beyond the sanctuary of Palaepaphos had been completely abandoned. No concentration of pottery that dates to the 1st c. BC or later has been reported by either the Swiss-German Expedition (Maier, 2007, p. 27, fig. 22) or by PULP. Building activity was concentrated on Alonia, especially to the west of the sanctuary, where houses with Roman mosaics have been found under the houses of the village community of Kouklia (cf. Maier & Karageorghis, 1984, pp. 280-282;Maier, 2004, p. 26). When almost nothing was left standing on the plateaus of Mantissa, Marchello and Hadjiabdoulla of the urban structure of the polis of Ancient Paphos, and the provincial town now called Palaepaphos had shrunk to occupy only the plateau of the sanctuary -from where it catered to the needs of pilgrims -the sanctuary itself under its new colonial identity, became an international centre of pilgrimage and received the patronage of Roman emperors (cf. Maier & Karageorghis, 1984, p. 271).
After 1500 years of use the Palaepaphos shrine "apparently lapsed around the time of the mid 4th century earthquakes, when suburban houses began to be built nearby." (Rautman, 2001, p. 8). The earthquakes provided a handy excuse for the definitive abandonment of the last of the Cypriot sanctuaries, like that of Aphrodite at Amathous and Zeus in Salamis, and their replacement with Christian churches. With the establishment of monumental basilicas, the Cypriot Bishops re-sacralised the main urban towns and gave them a new religious identity. "[A]s protector saints, Tychon [at Amathous] and Epiphanios [in Salamis] replaced Aphrodite and Zeus as bearers of civic -and insular -identity" (Rautman, 2011, p. 11). No such early church was built in Palaepaphos; it was neither a port nor a growing urban centre. Instead, Nea Paphos had a basilica built by its port, appropriately known as Panagia Limeniotissa, and another on the eastern part of the city, known as the Chrysopolitissa, which is one of the largest in Cyprus (cf. Maier & Karageorghis, 1984, pp. 292-293). The new religious ideology, which led to the abandonment of the sanctuary of Aphrodite, also led to the demise of Palaepaphos: from a small urban town that lived off visits by pilgrims, it was gradually transformed into a struggling agropastoral hamlet. From the 13th c., when the Lusignans built the Royal Manor House "as a centre of local administration and as headquarters of the royal official who directed and controlled the profitable sugar-cane plantations and refineries" (Maier, 2004, pp. 28-29), the site acquired a third name: Couvoucle (Kouklia). PULP has shown that the sanctuary's long 'biography' can be divided into three main chapters, which correspond to the three names by which the site to which it belongs was identified from antiquity to the present: Paphos, Palaepaphos and Kouklia. Each of these chapters has to be based on the analysis of the urban structure that expanded or shrunk around the sanctuary and, most importantly, on the study of the settlement pattern in the hydrological basin, the catchment of Paphos. Settlement pattern transformations in the hinterland and along the coastal landscape of the region of Paphos disclose the development of different political economy systems, which correspond to (a) Paphos, the independent polity, (b) Palaepaphos, the Graeco-Roman town and (c) Kouklia the Mediaeval feudal estate (Iacovou, 2014, p. 162). In the course of these three cycles, the sanctuary's identity and its relation to the urban environment underwent major changes: (a) first, it was founded, and maintained for almost a thousand years, by the ruling authorities of the polity of Ancient Paphos in the expanding urban centre that was the city-state's capital; (b) then, the roles were reversed: the shrinking urban centre of Palaepaphos was maintained because of the international aura of the sanctuary, which the colonial rulers of the island had transformed into a pan-Cypriot cult centre; (c) finally, when pilgrims stopped visiting the sacred precinct of Aphrodite, urbanism died out; deprived of its sanctity and its political and ideological significance, the sanctuary was incorporated into the industrial landscape of a feudal estate. The Mediaeval Manor House, as well as sugar mills and refineries for the production of cane sugar, were constructed with building materials readily available in the sanctuary (cf. Maier, 2004, p. 91).
The cane sugar industrial economy was exogenous, short lived (cf. Grivaud, 1998, pp. 344, 409) and environmentally disastrous: it depleted forests, water resources and also what was left of the settlement structure of the Paphos catchment: What really demonstrates the unconviviality of medieval sugar production in Cyprus is its transgression of its environmental limits. The refineries' development, growth and success specifically required them to consume water, fuel and soil at a level which would inevitably bring about their [down]fall. This was exacerbated by the concentration of monoculture and labour on the best land, which both increased the demand for food crops for the workforce and restricted their supply. (Given, 2018, p. 84) By contrast, the island's rise to complexity and the long-term prosperity enjoyed by Cypriot society in the micro-states of the island since the early 2nd millennium BC was linked to an indigenous industrial economy that invested in two of the most abundant raw resources of Cyprus: copper and timber. The metal and shipbuilding industries continued to thrive under the Ptolemies, and even as late as the early Roman era. Thus, in the catchment of Paphos, labour forces and specialised craftsmen continued to support the same regional chaine opératoire, which began on the forested and copper rich slopes of the Troodos and terminated on the coast -until it was dismantled in late antiquity. The impoverishment of the region of Paphos, which led to the abandonment of the hinterland settlement network, was caused by the abandonment of the island's ancient economic system. From the point of view of (Roman and Byzantine) imperial economies, Cyprus' natural resources were of limited significance (cf. Gordon, 2012, p. 287). When these industries were no longer state-managed and state-supported, the whole island's physiognomy was transformed (Iacovou 2014, pp. 168-169). In the isolated and poverty-stricken region of Paphos, even the memory of the ancestors' association with these traditional industries faded away (Iacovou, 2012(Iacovou, , p. 64, 2014. PULP has provided answers -some definitive, others preliminary -but its most important contribution is the identification of serious gaps in the research of the region of Paphos. For example, there is no comprehensive settlement pattern analysis of the region for the 1st millennium BC and the 1st millennium AD, analogous to the one compiled by Georgiou, and digitally upgraded by Agapiou, for the 3rd and 2nd millennia BC (Iacovou, 2014, p. 164). In the process of unlocking the landscape context of the urban sanctuary of the Cypriot Goddess, PULP has shown that no Cypriot sanctuary can be studied in isolation of the social, political and economic landscape to which it belonged. Cult centres, urban and extra-urban, were inextricably linked to the social fabric; they were also extremely sensitive to politico-economic changes, either those taking place close by, within their own micro-region, or as far away as the centre of the empire that ruled over the island.