Edinburgh Research Explorer

Unlocking historic landscapes in the Eastern Mediterranean

Citation for published version: Crow, J & Turner, S 2010, 'Unlocking historic landscapes in the Eastern Mediterranean: two pilot studies using Historic Landscape Characterisation', Antiquity, vol. 84, no. 323, pp. 216-229.

Link: Link to publication record in Edinburgh Research Explorer

Document Version: Publisher's PDF, also known as Version of record

Published In: Antiquity

Publisher Rights Statement: Crow, J., & Turner, S. (2010). Unlocking historic landscapes in the Eastern Mediterranean: two pilot studies using Historic Landscape Characterisation. Antiquity, 84(323), 216-229

General rights Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights.

Take down policy The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer content complies with UK legislation. If you believe that the public display of this file breaches copyright please contact [email protected] providing details, and we will remove access to the work immediately and investigate your claim.

Download date: 07. Oct. 2021 Unlocking historic landscapes in the Eastern Mediterranean: two pilot studies using Historic Landscape Characterisation Sam Turner1 &JimCrow2

Historic Landscape Characterisation (HLC) maps landscape with particular reference to its historic character and development. Executed using sources including satellite imagery and aerial photography and presented in a Geographic Information System (GIS), this offers a powerful insight into a landscape story. Here two leading advocates of the approach apply HLC for the first time to historic landscapes in the Eastern Mediterranean.

Keywords: Aegean, medieval and modern, Historic Landscape Characterisation, HLC, GIS, field systems

Introduction Landscapes provide a central concern for many disciplines, and the best ways to understand, value and manage them are hotly debated. Richard Bradley has observed a split in landscape archaeology between economic/functional and social/symbolic approaches (2000), a theme recently taken up by Matthew Johnson in his Ideas of landscape (2007). Such divisions can also be traced in related disciplines including history and geography (Widgren 2004). Scholars and surveyors have created detailed and accurate records of ancient remains, but have often

1 School of Historical Studies, Newcastle University, Newcastle upon Tyne NE1 7RU, England, UK (Email: [email protected]) 2 School of History, Classics and Archaeology, The University of Edinburgh, William Robertson Building, 50 George Square, Edinburgh EH8 9JY, UK (Email: [email protected])

Received: 29 December 2008; Accepted: 11 February 2009; Revised: 26 February 2009

ANTIQUITY 84 (2010): 216–229 http://antiquity.ac.uk/ant/84/ant840216.htm 216 bqiosadta l adcp a oevlea ecie ypol.I uue tmybe may it future, In people. by perceived as value is 1995). some landscape has (Cleere that landscape recognises such landscapes all ELC that management, the cultural and ‘outstanding’, heritage ubiquitous as outstanding places inform recognising particular to defining for than used Rather criteria others UNESCO than the broader as much is definition This D is: 2000; landscape that (CoE explicitly (ELC) states ELC Convention The 2007). Landscape Fairclough by European & ratification Turner and the 2006; signing of the particular countries in policy, 33 international influenced is in this developments in recent thinking Our by by frameworks. landscapes mutual unified present in transforming viewpoints and many of past together bringing of possibility the awareness our the in better-informed In deeper, perspectives. people out varying into different hold their incomprehension to that archaeologies thanks surely ways landscape 2007). diverse is integrative in Fleming theory thing view, see same archaeological the 2007; of see M. much present lessons or anything (Johnson, main past that work the forgotten others’ of have each one to reading But seems from side gained each be that might is traditions 208). post-modern 2000: and (Ingold to future only and not present relevant, the are also that but landscape past, of the meanings understanding the into insights archaeology historicity, particular this provide Through 2000). can (Morris term long the over change of trajectories rhelgclrcr a e ayt hr the share landscape to the of many in led present importance time-depth has the the record of with archaeological appreciation archaeologists’ Along context, everyday as 18-49). culture of material 2007: and multi-dimensionality Forbes the 2004; emphasise (Ingold landscapes that anthropologically-minded perspectives helped 1998). develop has R. landscape archaeologists and Johnson, culture ‘real’ perception: material described between with (‘inherent’ and Johnson division engagement landscapes This no exist Bob ‘perceived’ be (what only people’s should them which there and upon archaeologists landscapes canvasses, many cast neutral for is perception), as gaze ‘explicit’ perceiver’s as landscapes the seeing when contexts than the active as become Rather landscapes with life. landscapes and culture on social material perspectives with for engage rich to develop continued has to it Bender well-placed because is (e.g. archaeology landscapes post-processual past 44), on viewpoints provide al. to approaches phenomenological (Cosgrove writing theoretical overtly and traditional 1988). of reflexive Daniels research openly & empirical the more the geography, towards from cultural studies away of scholars landscape case led the has and representation in negotiated on Particularly constantly emphasis 48). changing, 2004: always (Olwig contested, constituted always culturally but static that were not geographers, stressed are landscapes cultural have landscapes They many that historians. including 119-27). theoretically-minded appreciate post-modernists, 2007: and are archaeologists, (Johnson to post-processual tradition trends or different technological a processes and in economic social Steeped of past reflections simple with just them not link to struggled nufruaersl ftewdnn iegnebtenshlr okn nempirical in working scholars between divergence widening the of result unfortunate An rhelgclsuisi h otpoesa rdto aeatmtdt mlyexplicitly employ to attempted have tradition post-processual the in studies Archaeological 97.Atog hs prahshv trce aiu rtcss(obs20:36- 2007: (Forbes criticisms various attracted have approaches these Although 1997). neato fntrlado ua atr’( factors’ human and/or natural of interaction ‘ ...a na r e a ,a sp e r c e i v e db yp e o p l e ,w h o s ec h a r a c t e ri st h er e s u l to ft h ea c t i o na n d a unr&JmCrow Jim & Turner Sam 217 o 00 ril 1 Article 2000: CoE Annaliste itras ocr o following for concern historians’ ). ejeant-Pons ´ et

Method Unlocking historic landscapes in the Eastern Mediterranean

Figure 1. Map showing the location of the two pilot studies. possible to develop frameworks that bring together not only social and economic approaches, but also those from a whole range of other disciplines, professions and perspectives. Using GIS-based techniques such as Historic Landscape Characterisation (HLC), we may be able to map, analyse, compare and contrast the perceptions of a wide range of people working with the landscapes of the past, present and future (Turner 2006a). This will surely help us to open up debates on the past and future of our landscapes. Although this paper does not attempt to outline how HLC might fulfil this wide-ranging role, it demonstrates how it has helped us to appreciate certain aspects of landscape history, using two case studies in the Eastern Mediterranean (Figure 1). Archaeological field survey over the past 30 years has transformed our understanding of many Mediterranean landscapes. Nevertheless, there remains a tendency to focus on Classical and earlier periods (e.g. Price & Nixon 2005), while medieval and post-medieval landscapes (Byzantine, Venetian, Ottoman, modern) are less well understood (Cherry et al. 1991; Cherry 2003; Vionis 2005). As historical archaeologists, we are particularly interested in how the landscape has developed, and how it was organised at different times in the past. In Britain, HLC is widely appreciated as a useful way to model past landscapes and how they changed over time to take their current forms (Turner 2007). The aim of our research has been to investigate whether similar methods might be used in Mediterranean contexts. In future, our maps – which represent our perceptions of historic landscape character – could form one layer amongst

218 n aste saes h ehdwspoerdi onal(K,adhsbeen has and (UK), Cornwall in pioneered was follows: method as there The developer its areas. by as described development them historical their maps a by is linked and HLC features together Instead, bundles features. that archaeological technique individual generalising map not do HLCs inventories, or Unlike neglected often are historic etc.) particular trees, their lanes, represented. places partially boundaries, give only field to buildings, combine (vernacular that character features ‘ordinary’ The locality. given namp oi a ehr oapeit nidvda oueta ato historic a of part as monument individual line an a appreciate or to record dot hard cannot a inventories be to total such can limited addition, provide In usually it not landscape. is so do data map, they location a Site Nevertheless, on environment. concerned. historic is the sites where of particularly records particular planning, of and management preservation inventories landscape the These research, 2008). for and TAY tools inventories: crucial web-based sophisticated the provide of Monument Turkey overview very an National in provides Scotland’s http://www.heritagegateway.org.uk/gateway/); project Historic are TAY (e.g. (see UK Some Records internet various information. Environment the http://www.rcahms.gov.uk/canmore.html); over associated sites (CANMORE: accessible of of Record are lists range provide they normally a increasingly databases with Archaeological together 2006a). (Turner data landscape pnoe eerhpoet htldt h eeomn fHCa a opeetand present to way a (Fairclough as landscape HLC whole of the development of nature the Heritage historic English to the problem, led analyse this to that response projects In 7-8). research development 1998: sponsored during (Herring ignored UK the often in was planning landscape and cultural the monuments, specific beyond Eastern the in techniques these regions. of other potential many the and explore Mediterranean of to archaeologists yet range 2007), have Gaffney a historians & at Dingwall landscape (e.g. sources and exceptions frameworks relevant few provide from a can With data 2004). HLC, (Rippon incorporate as scales that such histories Britain, landscape in particular, years diachronic In 15 for mapping. last digital diverse the and and over massively developed photographs huge using methods comparing aerial was imagery, of and work satellite combining problems comparative needed including for the such datasets data tools past, that provides of the now meant volumes GIS In maps, large time-consuming. landscapes. paper the like extensive easily sources, of marshal cumbersome study to in-depth possible undertake become to recently only has It Characterisation Landscape Landscape Historic Historic apply and adapt we region. Aegean that the places in time these first the in for is (HLC) Characterisation it and Turkey) (Thrace, of value the managed. on be debate should they facilitate how to and used landscapes (GIS) these Systems Information Geographical in many L rvdsacmlmnaytcnqeta a epda ihteeproblems. these with deal help can that technique complementary a provides HLC L asdfe rmtaiinlacaooia aaae o trn n presenting and storing for databases archaeological traditional from differ maps HLC nte18sadery19s rts rhelgssbcm nraigyaaethat, aware increasingly became archaeologists British 1990s, early and 1980s the In Silivri of hinterland the in and (Greece) Naxos on located are studies case two Our a unr&JmCrow Jim & Turner Sam 219 everything tal. et fhsoi neeti any in interest historic of 1999).

Method Unlocking historic landscapes in the Eastern Mediterranean

‘Closer examination [of the landscape] reveals that particular groupings and patterns of components which recur throughout the county can be seen to have been determined by similar histories. Cornwall’s historic landscape can, therefore, be characterised, mapped and described, using a finite number of categories or types of “historic landscape character”’ (Herring 1998: 11).

A range of HLC ‘types’ is usually classified in advance of mapping. The characteristics each type might be expected to exhibit are identified through archaeological or historical case studies. The researcher therefore needs to understand how patterns in the landscape reflect its historical development, and how physical features in the landscape relate to each other. Like other types of landscape archaeology, HLC mapping is a subjective process of interpretation that is informed by the physical landscape. Since relatively little work has been undertaken on the landscape history of Naxos or Thrace, we used retrogressive analysis of parts of our study areas to help inform the HLC mapping. Retrogressive analysis is a technique for unravelling the physical and chronological relationships between different elements in the historic landscape (e.g. roads, field boundaries; Oosthuizen 2006: 77-9), and this has helped us refine the landscape character types and provide increased chronological definition for our characterisation. A recurrent problem is that some areas may include features from several different eras that contribute strongly to overall character, so that it is unclear which ‘type’ should be mapped (e.g. field systems created in the nineteenth century may contain many features built earlier, such as terraces or field boundaries). Using GIS with an explanatory text provides an adequate solution. GIS systems are more flexible than printed maps, because many pieces of information can be presented in relation to each feature or area. In our project, we linked a database to the GIS which allowed a range of attributes to be recorded for each block of each character type. This allows the user of the data to build up a picture of the historical development of the landscape. Because HLC is a flexible method it can be adapted to suit different places and include a range of differing perspectives. Where landscapes exhibit long-term stability HLCs are often used to map landscape types with ancient origins (Turner 2007: 43-60), but the method can also be applied to analyse more recently-created landscape types (Dingwall & Gaffney 2007). Since the data is held in a GIS, it is easy to add data or change the information linked to each unit. HLCs that have been ‘finished’ should not be regarded as complete: new interpretations or perspectives on the landscape will demand additional data or new characterisations. HLC is not a monolithic approach, and different workers might choose to characterise the same area in different ways in response to different research questions (Turner 2006a). Since landscape histories vary from region to region, different HLC types will be appropriate in different places. In addition, characterisations can be undertaken at any scale and for a range of different uses (e.g. research on various periods’ landscape histories, to inform landscape management or for spatial planning). The nature and intended uses of a characterisation will affect how it looks and what HLC types are chosen for mapping (compare e.g. Foard et al. 2005; Turner 2006b, 2007). The HLC method is therefore very flexible.

220 o e omncaatrtps(o oeo h te ye e u est,Co & Crow website, our see types other 2009). the Turner & choices on Crow our more see behind (for discussion rationale types for the 2008; character Turner describes common section few following The a 2. for and 1 stable Tables model most in to remained shown attempt have can landscapes we which allow, fastest. trace the used to changed we have try sources which and and the land-use, as of far patterns character As landscape earlier of sources. phases IKONOS available earlier the all of from interpretation using evidence an the made on we based Second, imagery. type satellite character characterisations. present-day detailed the more recorded in we result First, could development further user-friendly; imagery database 2008). Turner satellite & the Crow reference see and separately: rectify to described are used we photographs methods air The and half mapping area. western HLC the study for Naxos our Google the undertook via of available we only when was imagery (though high-resolution Earth 2007) (April–July Google and maps military Russian .Hsoi i htgah.Ms forNxspltsuywscvrdb A i photos air RAF by covered was study pilot Naxos our of Most photography. air Historic 3. comprised area study (Thrace) Silivri the for source historic main The maps. Historic European 2. by supplied data satellite multispectral 4m and black-and-white 1m IKONOS were 1. ‘geometry’) each or to ‘polygon’ relating (a data type database. Access the character Microsoft and a landscape using mapping, stored specific the and a recorded 2007), undertake (Turner of to project block 9.1 HLC individual sources. ArcGIS and UK landscapes ESRI’s recent studies’ case chose a Mediterranean We our on suit to modelled adapted were significantly though database and method HLCs Our Mediterranean pilot two Thrace: and Naxos ye htaewdsra nteAga:baddtrae,cnortrae,srih step 1996: Moody straight & terraces, (Rackham contour terraces terrace false terraces, main modern braided six and Aegean: identified fields the terraced have check-dams, in collaborators terraces, widespread his and are Rackham that Oliver types landscape 3). fine-grained and and intricate landscape this 2 historic of (Figures important characterisation an our preparing were in terraces consider so to terraced, type are hillsides steep Naxos’ of Many terraces braided Naxos: h L ye ehv sdt a u ao n iir td rafrti rjc are project this for area study Silivri and Naxos our map to used have we types HLC The the keep to order in HLCs pilot our for types HLC of number the limited deliberately We 000 1:50 twentieth-century including sources digital other to referred we relevant, Where eue he rnia ore oifr u characterisations: our inform to sources principal three used We ht nt(hscleto a ic oe oteRylCmiso nteAncient the on Air Commission University Edinburgh). Royal in Keele the (RCAHMS) the Scotland to of moved by Monuments since Historic supplied has and and collection scanned (this Unit were Photo photographs The photography wartime Silivri). comparable identify to for unable were (we 1943 in sorties is during taken and in (1912-13) War 000 Balkan first 1:25 London. the Library, at after British made produced the survey in map Ottoman available Survey an on Ordnance based British was map a of version a 2006-7). acquired (and Munich LLC, Imaging Space a unr&JmCrow Jim & Turner Sam 221 c 93 This 1943. .

Method Unlocking historic landscapes in the Eastern Mediterranean

Table 1. Naxos HLC types. Enclosures Rough ground

Enclosures (modern) Rough ground (post-medieval/medieval) Enclosures (post-medieval) Rough ground (modern/post-medieval) with Enclosures (post-medieval) Enclosures (post-medieval) based on Fields Rough ground (modern/post-medieval) with (medieval) Enclosures (post-medieval) based on Braided terraces (medieval) Enclosures (post-medieval) with Braided terraces Rough ground (modern/post-medieval) with (medieval) Braided terraces (medieval) Enclosures (post-medieval) with Step terraces – Rough ground (modern/post-medieval) with straight/contour (post-medieval) Terraces [other types] Olives (modern) Woodland (modern/post-medieval) Olives (post-medieval) Outcrop, scree, cliff Horticulture (modern/post-medieval) Sand Settlement Terraces Braided terraces (medieval) Settlement (modern/post-medieval/medieval) Check-dams (medieval/post-medieval/modern) Villas (modern) Step terraces – contour (modern/post-medieval) Recreation (modern) Step terraces – straight (modern/post-medieval) Orchard Terraced fields (modern/post-medieval) False terraces (modern)

Industrial Industrial (modern) Quarry (modern/post-medieval)

140-5; Grove & Rackham 2001: 108). We have adapted this classification as the basis for our terrace HLC types. Here we discuss braided terraces in order to illustrate how our HLC types were classified. Recent debate shows the terraces of different Aegean landscapes have widely varying origins, from prehistory to the twentieth century (Betancourt & Hope Simpson 1992; Brunet 1999; Price & Nixon 2005; Forbes 2007). On Naxos, as on other Cycladic islands, there are very many areas where drystone field boundaries cut across earlier terraces (e.g. Kea: Whitelaw 1991). Virtually all examples of braided terraces on Naxos are overlain by walls that divide the terrace systems into discrete blocks (Figure 3). Individual terraces sometimes abut these walls, but invariably other terraces within the same system underlie them. This shows that braided terrace systems have witnessed long (possibly discontinuous) periods of use with several phases of development. The underlying terraces must antedate the walls, which themselves are no later than the eighteenth or nineteenth centuries in the vast majority of cases (almost none have been newly built since the 1940s). Late-medieval visitors to the Aegean noted the presence of terraces, including areas of desertion (Harfouche 2007: 153). On Naxos, seventeeth-century documentary sources refer to louroi, which may represent terraced subdivisions of large ‘open’ fields called engairies. These engairies occurred across much of the island – probably in every estate and village

222 iue2 ao L,wt eetdlnsaefaue n yatn hrhsaon ah,Nxs KNS1m IKONOS Naxos. Rachi, around churches Byzantine and features landscape selected material with (Includes data. HLC, black-and-white Naxos 2. Figure ti ed og ground Rough (modern/post-medieval/ fields Strip fields Strip types. HLC Silivri 2. Table ilsSettlement Transport Recreation (modern) Fields (modern) Fields–grid (modern/post-medieval/medieval) Fields Fields ils(modern/post-medieval) Fields ils(modern/post-medieval) Fields rhr Industrial Quarry Horticulture Orchard Meadow medieval) (post fields Coaxial medieval) medieval) on ti fields strip oxa ed (post-medieval) fields coaxial ae on based ed (post- fields  c 07 pc mgn L.Alrgt reserved). rights All LLC. Imaging Space 2007, based ae on based a unr&JmCrow Jim & Turner Sam 223 Water forest Thracian ground Rough woodland Other Marsh Village Villas Urban Industrial

Method Unlocking historic landscapes in the Eastern Mediterranean

Figure 3. The valley of Aria, Naxos, showing terraces (green), walls and lanes (red) and streams (blue). IKONOS 1m black-and-white data. (Includes material c 2007, Space Imaging LLC. All rights reserved).

(such fields, including terracing, were widespread in the Aegean: Whitelaw 1991: 410-11; Kasdagli 1999: 99-101; Forbes 2007: 195-9). In a soil erosion study undertaken from 1989-1992, Rainer Lehmann tentatively suggested many Naxos terraces may have been cultivated last between the fourteenth and seventeenth centuries (Lehmann 1993). Lehmann’s general conclusions support the possibility that braided terrace systems could have medieval origins (but cf. Grove & Rackham 2001: 264- 5). Many dated monuments on Naxos appear to stand on terraces, hinting at the antiquity of the latter. Examples include the early Byzantine churches of the Taxiarch and Agios Isidoros in Rachi, where both buildings perch on long terraces constructed along the hillside (Figures 4 and 5). Field survey also hints at the antiquity of Naxiot terrace systems. Around the early Byzantine church of Ag Kyriaki, north-east of Apeiranthos, analysis of fieldwalking data by Athanasios Vionis for the 2nd Ephorate of Byzantine Antiquities suggests that up to 70 per cent of the ancient finds collected belong to the seventh to ninth centuries AD. Curving drystone walls around small fields here only partially and untidily enclose the terraces, which are probably related to the early Byzantine settlement. It seems likely that whatever the original date of Naxos’ braided terrace systems, the vast majority would have existed in or before the seventeeth century. In our HLC database most are interpreted as ‘enclosures (post-medieval) with braided terraces (medieval)’. Though tentative, we have suggested other dates for other terrace types (Crow & Turner 2008). This kind of analysis

224 losu obidu ipemdl fhwtefre adcp fNxsmyhave may Naxos of landscape farmed HLC. the the in how recorded data of the using models Roman periods simple probable historic the a up over of build developed site to the represents us ‘b’ cropmarks) cropmarks. allows and = earthworks (as red fields material lynchets, (Includes strip = data. medieval black-and-white probable (green 1m of fields. IKONOS remains farmstead). coaxial showing Tepesi, post-medieval Fener Eski by of overlain ridge The 4. Figure ed’poal aet hs flnsaerognsto nteerymdr eid In period. example, for modern fields, early coaxial subdividing the by in created been reorganisation clearly landscape have fields of recent Silivri. more phase places, a of ‘coaxial to These north village. date the 8km probably outside Fener, immediately fields’ particularly of 70m, around of village sides In with the smaller 4). around (Figure typically particularly are boundaries fields occur axial Individual they sinuous slightly area but study parallel our roughly long, 2006). including considerable (Crow fields levelled of places, and in are destroyed boundary they completely field and been a area has as fragile study itself acts appear the still monument Wall the of fields Anastasian though the parts the even of many between line in the baulks example, that The for clear antiquity: 1m. range is over can it which to but earthworks, high by impermanent, 0.10m sometimes or than baulks low vegetation, less by weedy from marked or are grass Silivri long around fields with between topped boundaries the Naxos, to contrast In fields coaxial Thrace: h edssesw aecle cail r yie ybok fruhyrectangular roughly of blocks by typified are ‘coaxial’ called have we systems field The c . 0 4m n aeya uha 0mars.Sm are Some across. 200m as much as rarely and 140m, x 100 a unr&JmCrow Jim & Turner Sam 225  c 07 pc mgn L.Alrgt reserved). rights All LLC. Imaging Space 2007,

Method Unlocking historic landscapes in the Eastern Mediterranean

Figure 5. Photograph showing the lynchets visible in Figure 4 to the south of the settlement (hidden in the trees and marked with ‘a’ in Figure 4). Sam Turner, July 2007. the block that runs along the east bank of the Fener Dere. It seems likely that these fields were broken up into orchards or horticultural plots, as suggested by both the Ottoman/OS map and the unusually frequent occurrence of trees in today’s boundaries. In other places, our retrogressive analysis suggests that certain roads depicted on the Ottoman/OS map post-date the coaxial fields, strongly suggesting a post-medieval date at the latest. Elsewhere, on the slopes of the ridge running east from Fener, we identified a probably medieval strip field system preserved both as earthworks and cropmarks that clearly underlies the pattern of coaxial fields (Figures 4 and 5). The detection of varying patterns of field systems around different villages and the recognition of their diverse periods of origin has allowed us to appreciate the time depth in this historic landscape for the first time (Crow & Turner 2009; Figure 6).

HLC: prospect and limitations for research applications For research, HLC allows us to relate broad patterns in the historic landscape to particular classes of monuments, boundaries and other significant cultural features. Because we can trace landscape change, we can also relate changing distributions of monuments to changing patterns of land-use. For example, on Naxos there is a density of Byzantine churches virtually unmatched anywhere in the Mediterranean. By plotting these against our HLC, we can identify how they related to areas exploited by contemporary people in different ways, and how patterns of church foundation varied over time (Figure 2). HLC therefore provides an effective way to bring together aspects of economic/functional and social/symbolic landscapes (Turner 2006b; see Nixon 2006: 7-11; Forbes 2007: 16-18). As made clear above, our interpretations of HLC types rely to a significant degree on analogies from previous historical and archaeological work. In much of the Mediterranean, little attention has been given to dating the origins and development of this ‘landesque

226 adcpsa atclrtms o xml,i svr adfru oko xcl what exactly know to us for hard very particular is to it rise gave example, that For agency times. and particular structure at between relations landscapes undertake the understand to mind. to in important how aims these is with future it the in interpretations, research historical our and improve excavation survey, To appropriate 2007). (Widgren capital’ Turkey. Thrace, in hinterland its and Silivri of HLC 6. Figure oefnaetlpolm r o orlt adcp omt itrclpoess and processes, historical to form landscape relate to how are problems fundamental More a unr&JmCrow Jim & Turner Sam 227

Method Unlocking historic landscapes in the Eastern Mediterranean circumstances gave rise to patterns like the coaxial fields we identified around Fener. Here, though, historical archaeologists are at a particular advantage because they can combine a wide range of relevant approaches, from excavation and survey through historical studies to ethnographies and folklore. Landscape has provided an important medium for negotiating social and economic relationships, and changes in practice are inscribed into its ‘palimpsest’. Using a wide range of sources can help us create rich, contextual understandings of societies and places, and to analyse both the detailed causes and the effects of change in the past (Trigger 2006: 534-6). Using GIS, we believe HLC and related methods could provide unifying spatial frameworks to allow these different perspectives and sources to be brought to bear on particular questions about the past.

Acknowledgements A grant from the UK Arts and Humanities Research Council’s Landscape and Environment Programme funded our research. We are grateful to Athanasios Vionis and David Alderson for their contributions to the project. The data we created are available to download free from the Archaeology Data Service: http://ads.ahds.ac.uk/.

References CROW, J. 2006. Der Anastasische Wall: die letzte Grenze,inG.Klos&A.Nunnerich-Asmus¨ (ed.) BENDER, B., S. HAMILTON &C.TILLEY. 1997. Grenzen des Romischen¨ Imperiums: 181-7. Mainz: Leskernick: stone worlds; alternative narratives; Philipp von Zabern. nested landscapes. Proceedings of the Prehistoric Society 63: 147-78. CROW,J.&S.TURNER. 2008. Unlocking historic landscapes in the eastern Mediterranean. Project BETANCOURT,P.&R.HOPE SIMPSON. 1992. The website. Available at: http://www.shc.ed.ac.uk/ agricultural system of Bronze Age Pseira. Cretan projects/eastmed landscape/ (23 December 2008). Studies 3: 47-54. – 2009. Silivri and the Thracian hinterland of Istanbul: BRADLEY, R. 2000. Mental and material landscapes in an historic landscape. Anatolian Studies 59: 167-81. prehistoric Britain, in D. Hooke (ed.) Landscape: the richest historical record: 1-11. Amesbury: Society DEJEANT´ -PONS, M. 2006. The European Landscape for Landscape Studies. Convention. Landscape Research 31(4): 363-84. DINGWALL,L.&V.GAFFNEY (ed). 2007. Heritage BRUNET, M. 1999. Le paysage agraire de Delos´ dans l’antiquite.´ Journal des Savants 1999 (Janvier–Juin): management at Fort Hood, Texas. Experiments in 1-50. Historic Landscape Characterisation.Oxford: Archaeopress. CHERRY, J. 2003. Archaeology beyond the site: regional survey and its future, in J. Papadopoulos & R. FAIRCLOUGH,G.,G.LAMBRICK &A.MCNAB (ed.). Leventhal (ed.) Theory and practice in 1999. Yesterday’s world, tomorrow’s landscape: The Mediterranean archaeology: Old World and New English Heritage Landscape Project 1992-94. World perspectives: 137-59. Los Angeles (CA): London: English Heritage. Cotsen Institute of Archaeology, UCLA. FLEMING, A. 2007. Don’t bin your boots. Landscapes 8(1): 85-99. CHERRY,J.,J.DAVIS &L.MANTZOURANI (ed.). 1991. Landscape archaeology as long-term history: northern FOARD,G.,D.HALL &T.PARTIDA, 2005. Rockingham Keos in the Cycladic Islands. Los Angeles (CA): Forest, Northamptonshire: the evolution of a University of California, Institute of Archaeology. landscape. Landscapes 6(2): 1-29. CLEERE, H. 1995. Cultural landscapes as world FORBES, H. 2007. Meaning and identity in a Greek heritage. Conservation and Management of landscape: an archaeological ethnography. Archaeological Sites 1: 63-8. Cambridge: Cambridge University Press. COSGROVE,D.&S.DANIELS. 1988. The iconography of GROVE, A. & O. RACKHAM. 2001. The nature of landscape. Cambridge: Cambridge University Press. Mediterranean Europe: an ecological history.New CoE (Council of Europe). 2000. European Landscape Haven (CT)/London: Yale University Press. Convention. Florence: Council of Europe (European HARFOUCHE, R. 2007. Histoire des Paysages Treaty Series No. 176). Available at: M´editerran´eens Terrass´es: Am´enagements et http://conventions.coe.int/Treaty/en/Treaties/Html/ Agriculture (British Archaeological Reports 176.htm. International Series 1634). Oxford: Archaeopress.

228 J 04 utr ntegon:tewrdperceived world the ground: the on Culture 2004. – H M I R K J N L P O O NGOLD OHNSON OHNSON RICE EHMANN ACKHAM ASDAGLI IXON OSTHUIZEN LWIG ERRING ORRIS 315-40. hog h feet. the through skill Routledge. and dwelling livelihood, in Essays Science. niomn n hills and environment Wicherek S. in (ed.) Greece, Island, Naxos on erosion rsnigamto fhsoi adcp character landscape assessment historic of method a presenting Archaeology rhelgclsurvey. and archaeological texts from evidence terraces: agricultural rhelgclDialogues Archaeological methodology. theory, philosophy, landscape: Blackwell. Blackwell. otBznieStudies. Post-Byzantine and Byzantine of Institute Naxos Hellenic seventeenth-century Venice: of study case a Aegean: eeec n adsaig nH aag H. Palang, H. Soov in shaping, land and reference afil:HrfrsieUiest Press. University Hertfordshire Hatfield: fields medieval Cambridgeshire’s of development and hrhsadio tnsi pai,southwestern Sphakia, Crete. in stands icon and churches nvriyPress. University landscape Cretan Academic. a environment in globalising change and persistence landscapes: rural ,S.&L.N .20.‘hsi o adcp’ circulating landscape’: a not is ‘This 2004. K. , .2006. L. , .2000. T. , .2000. I. , l,M nrp&G etn(ed.) Setten G. & Antrop M. ali, ¨ .1998. P. , amln rso ntmeaeplains temperate in erosion land Farm .19.Apoce otepreto of perception the to Approaches 1998. R. , 2007. M. , .1999. A. , ,O.&J.M .19.Traedgaainadsoil and degradation Terrace 1993. R. , xod Oxbow. Oxford: .2006. S. , rr:Crwl onyCouncil. County Cornwall Truro: . 0:665-94. 109: IXON aigalnsaesacred landscape a Making rhelg sclua history cultural as Archaeology h ecpino h environment. the of perception The onalshsoi landscape: historic Cornwall’s acetr Manchester Manchester: . adadmrig eteet nthe in settlements marriage and Land da flandscape of Ideas ora fMtra Culture Material of Journal OODY 05 nin Greek Ancient 2005. . adcpsdcdd h origins The decoded. Landscapes 2-3 mtra:Elsevier Amsterdam: 429-43. : mrcnJunlof Journal American 16.Drrct Kluwer Dordrecht: 41-65. : 1996. . () 54-68. 5(1): h aigo the of making The .Oxford: London: . European . Outlying a unr&JmCrow Jim & Turner Sam .Oxford: 9(3): . . 229 R T W T T 2006b. – W T A.20.T 2008. TAY. V 07 r-ooillneqecptl global a capital: landesque Pre-colonial 2007. – IPPON URNER URNER URNER RIGGER IONIS IDGREN HITELAW Archaeology. countryside the C) nvriyo aiona nttt of Archaeology. Institute California, of University (CA): Islands Cycladic the in northern Keos history: long-term (ed.) as Mantzourani archaeology L. Landscape & in Davis Keos, J. northwest Cherry, in J. use land and settlement rural Society. Devon rural of Press. Exeter of Wessex University and Exeter: Devon Cornwall, medieval early in Research planning. and management research, for archaeology landscape a Characterisation: Palang,H.Soov Archaeology Islands. Cyclades the case-study from a Greece: in archaeology post-medieval tp/tyrjc.r/(0Spebr2008). September (10 http://tayproject.org/ at Available Turkey. of Settlements Archaeological Press. University Cambridge Cambridge: edition. 2nd antCek(A:Lf os Press. Coast Left (CA): heritage Creek Walnut and archaeology in standpoints and (ed.) McAtackney in character Europe,inD.Hicks,G.Fairclough&L. landscape of archaeology the culture: 17.Wlu re C) AltaMira. (CA): Creek Walnut 61-77. J. (ed.) & Martinez-Alier McNeil J. Hornborg, A. in perspective, Academic. a in environment change globalising and persistence landscapes: rural European .20.Dmsi aeilclueand culture material Domestic 2005. A. , .2004. S. , ,S.&G.F 2007. S. , Landscape Historic 2006a. S. , .2006. B. , aigaCrsinlnsae h countryside The landscape. Christian a Making .20.Cnlnsae era? nH. in read?, be landscapes Can 2004. M. , .19.Teehorhelg frecent of ethnoarchaeology The 1991. T. , 14:385-98. 31(4): rieAkooi Yerle”meleri/The Arkeolojik urkiye ¨ 91:172-85. 39(1): ehnigevrnetlhistory environmental Rethinking itrclnsaeaayi:deciphering analysis: landscape Historic xtr eo Archaeological Devon Exeter: . ok oni o British for Council York: . nin onr.Tehsoi character historic The country. Ancient AIRCLOUGH itr facaooia thought archaeological of history A l,M nrpadG etn(ed.) Setten G. and Antrop M. ali, ¨ niinn adcps situations landscapes: Envisioning 5-5 odeh:Kluwer Dordrecht: 455-65. : 0-4 o Angeles Los 403-54. : 07 Common 2007. . Post-Medieval Landscape 120-45. : . : .

Method