Seasonal Settlement in the Medieval and Early Modern Countryside [1 ed.] 9789464270112, 9789464270099

Citation preview

RURALIA XIII

For the first time seasonality is placed at the centre of the study of rural settlement. Using a Europe-wide approach, it provides a primer of examples, of techniques and of ideas for the identification and understanding of seasonal settlement. As such, it marks an important new step in the interpretation of the use of the countryside by historic communities linked to the annual passage of the year. The particular studies are introduced by an opening essay which draws wider conclusions about the study of seasonal settlement, followed by 31 papers by authors from all parts of Europe and beyond. By its very nature ephemeral, seasonal settlement in the medieval and early modern periods is less well researched than permanent settlement. It is often presumed that seasonal settlement is the result of transhumance, but it was only one facet of seasonal settlement. It was also necessitated by other forms of economic activity, such as fishing, charcoalburning, or iron-smelting, including settlements of pastoralists such as nomads, drovers, herders as well as labourers’ huts within the farming context. The season a settlement was occupied varied from one activity to another and from one place to another summer is good for grazing in many mountainous areas, but winter proved best for some industrial processes. While upland and mountainous settlements built of stone are easily recognised, those that use wood and more perishable materials are less obvious. Despite this, the settlements of nomadic pastoralists in both tundra and desert or of fishermen in the Baltic region are nonetheless identifiable. Yet for all that definitive recognition of seasonal settlement is rarely possible on archaeological grounds alone. Although material remains can be of particular importance, generally it is the combination of documentary information, ethnography, geographical context and palaeo-environmental data that provide frameworks for interpreting seasonal settlements.

IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

SEASONAL SETTLEMENT

SEASONAL SETTLEMENT

SEASONAL SETTLEMENT

IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE edited by

PIERS DIXON & CLAUDIA THEUNE

ISBN 978-94-6427-009-9 ISBN: 978-94-6427-009-9

RURALIA XIII

9 789464 270099

RURALIA XIII

SEASONAL SETTLEMENT

IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

SEASONAL SETTLEMENT

IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE edited by

PIERS DIXON & CLAUDIA THEUNE

RURALIA XIII

© 2021 Individual authors The authors are responsible for the content of the papers (incl. image credits). Published by Sidestone Press, Leiden www.sidestone.com Imprint: Sidestone Press Academics This book has been peer-reviewed. For more information see www.sidestone.com Lay-out & cover design: Sidestone Press Photograph cover: Castle Campbell, Parish Dollar, Clackmannanshire, Scotland © Claudia Theune ISBN 978-94-6427-009-9 (softcover) ISBN 978-94-6427-010-5 (hardcover) ISBN 978-94-6427-011-2 (PDF e-book) ISSN 2565-8883 The conference and this volume has been sponsored by Historic Environment Scotland; the University of Stirling; the University of Vienna; the Historic Rural Settlement Trust and the Aurelius Charitable Trust.

Contents

Foreword9

Piers Dixon and Claudia Theune

SECTION ONE: SEASONAL SETTLEMENT AND MEDIEVAL  ENVIRONMENT

13

Seasonal settlement in the medieval and early modern  countryside: introduction

15

Too much environment and not enough history: the opportunities  and challenges in researching medieval seasonal settlement in Atlantic Europe

23

SECTION TWO: SEASONAL SETTLEMENT IN SOUTHERN EUROPE

33

Piers Dixon

Richard Oram

Archaeological research on seasonal settlements in the Iberian  Peninsula – an overview

35

Catarina Tente and Margarita Fernández Mier

Early medieval seasonal settlement and vertical transhumance  in an agricultural landscape in Ainet, East Tyrol, Austria

45

A multidisciplinary approach to the relationship between seasonal  settlements and multiple uses: case studies from southern Europe (15th-21st centuries)

57

Transhumance in medieval Serbia – examples from the Pešter  Plateau and northwestern slopes of the Prokletije Mountains

69

Archaeology of the commons: seasonal settlements in the  Cantabrian Mountains

81

Plows, herds, and chafurdões. Vernacular architecture and land  use in modern Castelo de Vide (Alto Alentejo, Portugal)

93

Elisabeth Waldhart and Harald Stadler

Anna Maria Stagno

Uglješa Vojvodić

Margarita Fernández Mier and Pablo López Gómez

Fabián Cuesta-Gómez and Sara Prata

From the Roman villa rustica to the early modern farmer’s grange –  105 specific forms of seasonal settlements in eastern Croatia

Pia Šmalcelj Novaković and Anita Rapan Papeša

SECTION THREE: SEASONAL SETTLEMENT IN NORTHERN AND  EASTERN EUROPE

111

Transhumant settlement in medieval Wales: the hafod

113

Imagining and identifying seasonal resource exploitation on the  margins of medieval Ireland

125

Entangled flexibility, adaptability, and seasonality in inland  Scandinavia – the case of agrarian outland use and settlement colonisation

137

Upland habitation at Castle Campbell in the Ochils, Scotland: a  multifunctional historic landscape at Dollar Glen

147

Rhiannon Comeau and Bob Silvester

Eugene Costello

Eva Svensson

Daniel T. Rhodes

Palynological data on vegetation and land use change at a shieling  159 ground on Ben Lawers, central Scottish Highlands, since the 13th century AD

Richard Tipping and Angus McEwen

From seasonal settlement to medieval villages? Early medieval  settlement in the coastal region of Uusimaa, southern Finland

169

Building crannogs in the 9th-12th centuries AD in northern  Scotland: an old tradition in a new landscape

179

This piece of singular bad neighbourhood: the Mamlorn Forest  Dispute, Scotland, c. 1730‑1744

189

Settlements of the Pskov long barrow culture: seasonal,  temporary, or short-lived?

201

SECTION FOUR: INDUSTRY, TRADE AND SEASONAL SETTLEMENT

209

Tuuli Heinonen

Michael J. Stratigos and Gordon Noble

Ian Maclellan

Elena Mikhaylova

Connections between transhumance and whisky distilling in  Highland Scotland

211

Darroch D.M. Bratt

Seasonal iron production in the mountains of Viking Age and  medieval Scandinavia

221

Markets and horse fighting sites in southern Norway – their  socioeconomic significance, origin, and demise (AD 1300‑1800)

229

Seasonality and logistics of the late medieval and early modern  cattle trade in Hungary

243

Kjetil Loftsgarden

Marie Ødegaard

Laszlo Ferenczi

In which part of the year did iron smelting occur in the  Drava valley?

253

Albuen – The king’s great herring market?

263

Research on seasonality and seasonal settlements in the  Czech lands – an overview (High and Late Middle Ages)

273

Ivan Valent, Tajana Sekelj Ivančan and Renata Šoštarić Leif Plith Lauritsen

Tomáš Klír and Martin Janovský

SECTION FIVE: HERDING AND NOMADISM283 Long-term patterns of nomadic and sedentary settlement in the  Crowded Desert of NW Qatar

285

‘We are always coming and going, like migratory birds’.  Diachronic changes in the seasonal settlement of Sámi reindeer herders in the Lake Gilbbesjávri region, northwestern Sápmi,  AD 700‑1950

295

Patterns of seasonal settlement of the forest Sami in Sweden

309

José C. Carvajal López

Oula Seitsonen

Gudrun Norstedt

SECTION SIX: WOODLANDS AND SEASONAL SETTLEMENT321 Dendrochronological research to track shepherds’ summering  in the Pyrenees

323

Mireia Celma Martínez and Elena Muntán Bordas

To browse and mast and meadow glades: new evidence of shieling  337 practice from the Weald of South-East England

Andrew Margetts

Seasonal agro-pastoral and craft-related temporary settlements  in medieval and post-medieval Provence (France)

349

Seasonality, territories, and routes: pannage as a  multi-component practice in medieval and early modern Hungary

361

Sylvain Burri and Aline Durand

Csilla Zatykó

Foreword Piers Dixon* and Claudia Theune**

* Leithen Mill Lodge Leithen Crescent Innerleithen EH44 6JL United Kingdom [email protected] ** Department of Prehistory and Historical Archaelogy Franz-Klein Gasse 1 1190 Vienna, Austria

This volume assembles contributions from the 13th International Conference of the Ruralia association, which took place in Stirling, Scotland, in September 2019. The association’s biennial conferences always focus on a specific theme, and this time it was ‘Seasonal Settlement in the Medieval and Early Modern Countryside’. Geographically, medieval and early modern seasonal settlements are to be found in a variety of often-extreme locations, highland, lowland, or even coastal, from the Arctic to the deserts of the Middle East. They were occupied by people who were driven by economic pressures to take advantage of new pastures and land, changing fishing grounds, new sources of iron and other minerals, and sometimes illicit activities in the longue durée from the early medieval period to post-medieval times (c. AD 500 to c. 1750). The establishment of seasonally dependent temporary settlements in inhospitable places, or even in relatively distant parts of a farm that provided temporary accommodation for the workforce of the farm at critical times of the year, was the subject of the conference. To discuss these themes, some 50 colleagues from 22 countries across Europe followed the call for papers, submitted an abstract, and presented the results of their research in lectures and posters during the conference. The conference opened with a substantial introduction session in which broad environmental and regional perspectives were considered; the 38 contributions were mainly based on studies within national boundaries. In several panels, seasonal settlements in various regions of Europe, in particular Southern, Eastern, and Northern Europe, were presented. A further focus was on seasonal settlements on the coast, in woodlands, and on craft, industry, and trade. Both analogies and very different approaches could be discussed in this way for Europe and beyond. We are very pleased that we can present 31 submitted papers from the conference in this volume. Following the structure of the conference, the six sections start with an introduction to the topic and an elaborate paper that contextualises the environmental and climatic background with socioeconomic continuities and discontinuities and how they are reflected in written and archaeological records. Sixteen papers present case studies from the three regions of Europe mentioned above and the other sections focus on seasonal settlement and industry and trade, herding and nomadism, and woodlands. Interested colleagues will get a very good overview of the variations, opportunities, capabilities, challenges, and solutions of seasonal settlement in very different, often marginal, areas. In the tradition of Ruralia conferences, attendees have the opportunity to visit archaeological sites in the host region that speak to the principal interests of the association and to the conference topic. We were expertly led around Stirling Castle by Peter Yeoman and Nicki Scott; Castle Campbell by Daniel Rhodes and Nicki Scott; Ben Lawers settlements by In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 9-12.

9

Fig. 1. Piers Dixon talking to Sam Harrison, Director of The Shieling Project, Strathfarrar, over a picnic lunch on our way to visit the shielings (© Claudia Theune).

Fig. 2. Daniel Rhodes of the National Trust for Scotland expounding on the possible seasonal settlement at Castle Campbell, which can be seen in the background (© Claudia Theune).

Richard Tipping, Piers Dixon, and John Atkinson; and the Crannog Centre on Loch Tay by Michael Stratigos, Tony Brown, and Mike Benson. The post-conference excursions took us to the Highlands of Scotland, visiting the Highland Folk Museum in Newtonmore, Speyside, introduced by Liz English; The Shieling Project, Strathfarrar, led by Sam Harrison; and a deserted township and illicit distillery site in Strathconon, led by Meryl Marshall. It is a pleasure to express our cordial gratitude to those who have supported the conference and its 10

publication. First and foremost, Historic Environment Scotland provided financial support for the conference and support in kind at the conference, namely Kirsty Owen, Kevin Grant, and Nicki Scott, and also allowed the conference participants to tour Stirling Castle after normal closing time with the help of its staff, which was greatly appreciated. We were welcomed most graciously at the opening reception and dinner held at Stirling Castle by the local member of the Scottish Parliament, Bruce Crawford; Alasdair Gammack, Dean of the Guildry

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 3. The reconstructed crannog on Loch Tay at the Crannog Centre before it was destroyed by fire (© Leif Lauritsen).

Fig. 4. A reconstructed peasant house at the Highland Folk Museum, Newtonmore (© Leif Lauritsen).

of Stirling; and by Kirsty Owen, on behalf of David Mitchell, Head of Conservation at Historic Environment Scotland, who was unable to attend in person. Our thanks also go to Stirling University, whose Deputy Principal Education and Student Experience, Professor Leigh Sparks, opened proceedings at the first session, welcoming us to the university, and also sponsored a second event at the conference centre, Stirling Court Hotel, in Stirling University, which Professor Richard Oram kindly hosted on behalf of the university. We would particularly like to thank Paula Miller of Stirling Court Hotel for providing such an efficient conference venue and her staff, who resolved all our problems so pleasantly and unfailingly. The conference was roundly entertained by Alistair Gentleman

and his fantastic ceilidh band after the conference dinner, persuading many an uncomprehending soul in the delights of Scottish country dancing. In addition to Historic Environment Scotland, Ruralia wishes to acknowledge the generosity of the sponsors, the Historic Rural Settlement Group and the Universities of Stirling, Aberdeen, The Highlands and Islands, and Vienna. Our heartfelt thanks also go to our publisher Sidestone Press, in particular Karsten Wentink and his team. The RURALIA XIII conference in Stirling, Scotland, was organised by the editors of this volume, Piers Dixon and Claudia Theune. We were ably supported by other members of the organising committee: Mark Gardiner, Niall Brady, Richard Oram, Jane Downes, Elizabeth Ritchie, and Foreword

11

Fig. 5. Piers Dixon informing the delegates about the deserted settlements at Tombreck, Lochtayside, Claudia Theune in the background (© Niall Brady).

Fig. 6. Peter Yeoman explaining the layout of Stirling Palace, Stirling Castle, Mark Gardiner in the foreground (© Niall Brady).

Kirsty Owen. We should also like to thank the help of early careerists, Jennifer Allison of Aberdeen University, Darroch Bratt of the University of the Highlands and Islands, and Theo Dixon of Stirling University, who provided an essential service of manning the desk at reception.

12

Special thanks go to Daniel McNaughton for the proofreading and to Justyna Stanko, University of Vienna, for her help throughout the editing work for this volume.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Section One

Seasonal Settlement and Medieval Environment

Seasonal settlement in the medieval and early modern countryside: introduction Piers Dixon*

* Leithen Mill Lodge Leithen Crescent Innerleithen EH44 6JL United Kingdom [email protected]

Seasonal as opposed to permanent settlement suggests to many ethnographical, archaeological, and historical researchers the practice of transhumance in its various guises. However, not all seasonal settlement was related to seasonal pastoral activities of this type. Temporary settlements could be built at different seasons of the year for all sorts of reasons, many of them unrelated to pastoral activity. The question is what were these settlements like, how are they dated, and what were the economic drivers for creating them. At its core, the issue is to address the difficulties inherent in defining seasonal use archaeologically and in disentangling the landscape palimpsest that is presented to us by the relatively ephemeral remains of seasonal settlements in order to understand not only the type of structures that were built for different seasonal activities, but also the environmental context. For many, models derived from ethnographic studies or historical narratives have been a sufficiently convincing context in which to understand the archaeological remains. However, these frameworks should not be taken as read, nor should it be assumed that practices recorded in the 19th century have always been like that. If archaeology and indeed history teaches anything, it is that things change and society adapts to external change. There has been a natural tendency for archaeological research into medieval settlement and rural life to be concentrated on the main foci of settlement, the everyday habitations and cultivation remains of the rural population – agricultural villages, farms, fields, mills, etc. Previous Ruralia conferences are no exception, and have often focused on the changing patterns of agricultural settlement, which is the meat and drink of rural life, and was after all Ruralia’s starting point in 1995. But this is not entirely the case. Ruralia VII (Klápšte – Sommer 2009), for example, took marginal landscapes as its theme. Some of the chapters (e.g. Christine Rendu et al.; Peter Herring; Ingvild Oye) touched on the type of seasonal settlements and activities that were to be revisited at Ruralia XIII, from a marginal perspective. In the introduction to that volume, Eva Svensson and Mark Gardiner (2009) noted that the utmark of Scandinavia ‘might be used for seasonal grazing’ (22). This author addressed the palimpsest issue in Scotland to examine how hunting and settlement, seasonal and otherwise, were often in competition in their use of marginal areas, such as the uplands (Dixon 2009). That is certainly one aspect of the discussion in this volume too. It touches on the strategies that may be adopted for exploiting land that cannot easily be cultivated on a permanent basis enabling an agricultural farm or village to be established. As Eva Svensson (this volume) has In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 15-22.

15

recognised, the taking in of new land for cultivation need not make a farm, but merely what in Scotland is called outfield, cultivated for a year or two and then converted to pasture. This land may have been grazed previously in the summer and perhaps have been occupied by shieling huts, a process that may lead to an improvement in the soils and make its use for cultivation attractive. Such land might become the location of a farm. Indeed, it is recognised that Shiel(d) and Scale place names of farms in parts of northern England and southern Scotland attest in some instances to such a process of colonisation (Winchester 2012, 133‑134), while in other cases a site may drop out of use again (e.g. Alnhamsheles, Dixon 2014). This process of conversion of one land use to another seems to be a feature of the moorland environments of northern Europe rather than southern Europe, where it is the very altitude of mountain pastures in the Alps and Pyrenees that present biogeographical limitations to the conversion of upland pastures to permanent farms, where alpine and subalpine pastures have a short summer growing season (Rendu et al. 2009; Andres 2018). By its very nature ephemeral, seasonal settlement is less well researched than permanent settlement. There has been a rash of recent work on transhumance, a subject that has its own historiography and been variously researched in many parts of Europe. While for prehistorians it has been model of how upland areas might have been farmed in prehistory (Collis et al. 2016), a recent European Association of Archaeologists conference session in 2015 tackled the issue of historical transhumance, leading to a publication in 2018 (Historical archaeologies of transhumance across Europe, edited by Eugene Costello and Eva Svensson). Costello and Svensson identified that transhumance and pastoralism are closely interrelated and that at one extreme there is nomadism, where no agriculture is practised. However, other variations in pastoralism, such as semi-sedentary and semi-nomadic were also defined, as well as transhumance itself in the sense of ‘people engaged in agriculture in specific ecological zones to use seasonally productive pastures in other areas’ (Barth 1962). However, the focus of the Ruralia XIII conference was upon seasonal settlement to give scope to researchers to discuss how seasonal settlement can be distinguished from permanently agriculturally based settlement and was inclusive of all the variants of pastoralism identified above, as well as other activities. Curiously, few chose to tackle the issue of how seasonal settlement can be distinguished archaeologically from temporary or permanent settlement. Types of evidence

Taking a multidisciplinary approach to the question of seasonal settlement was a common methodology applied with varying degrees of success in the chapters here. 16

Early medieval evidence for transhumance is outlined by Andrew Margetts for the Weald of South-East England, a wooded area of upland that was exploited for its summer grazing, from evidence of the detached portions of lowland parishes in the Weald that are a fossil of a transhumant economy, along with the surviving route ways that led to them. Recently, the accidents of development have revealed the ephemeral archaeological remains of an early medieval enclosure and buildings constructed in timber, the first archaeological data that might support this argument: the documentary and landscape evidence providing the framework for interpretation. Eugene Costello decides to sidestep the retrogressive ethnographic approach of presuming the modern evidence of booleying reflected what it may always have been like. Instead, he focuses on the medieval evidence for seasonal exploitation of the Irish uplands before c. 1600, looking at literary texts and archaeological data, particularly in the Wicklow Mountains, where pollen data suggests a landscape of wood pasture in which cattle grazing and hunting could be carried out. Yet actual evidence of seasonal settlement is hard to come by, due to modern farming and forestry, while those which have been interpreted as seasonal in upland areas under pasture, he argues, are associated with enclosures and cultivation ridges suggesting permanent settlement. However, a more compelling argument for seasonal settlement is presented by the circular and oval huts found without any associated cultivation remains or enclosures in the Mourne Mountains. This compares well with the situation in Scotland, where the argument that such sites represent the huts of people who migrated to the summer pastures is based on their very appearance in upland pastures, supported by place names and documentation (Dixon 2018). Palaeoenvironmental evidence

Environmental evidence for pastoral activity was one of the key questions for the conference and the focus of Richard Tipping’s chapter as an established specialist in the field. He analysed the evidence from a pollen core in a known shieling ground on the slopes of the Ben Lawers mountain range dating from the 13th century to the present. This provided a case study in whether summer grazing of the central Scottish highlands in the medieval period is reflected in the pollen rain from a glacial valley on the north side of Loch Tay. Open-canopy birch woodland predominated in the 13th century, suggesting low-level pastoral activity or wood pasture. However, the presence of barley pollen is dismissed as wild barley and there are no signs of ground disturbance in the other taxa to support cultivation. Increased pastoral activity and woodland reduction is late medieval rather than earlier, intensifying in the 17th century, but Tipping argues that cattle-grazing pressures are the cause rather than climate

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

deterioration, based on the loss of dwarf herbs. This is instructive in view of the prospective influence of climate raised by Richard Oram, since Tipping favours economic drivers. Other chapters made reference to this type of data as the background to the seasonal activities they discuss, including the presence of mainly oak woodland in the Drava valley, Hungary, as supporting environmental evidence for documented pig husbandry in the Middle Ages. In Scandinavia pollen evidence is used to suggest a mixed economy of pasture and small-scale cultivation for outland settlement in the boreal forest of Sweden (Eva Svensson) as well as the Finnish outland (Tuuli Heinonen). Svensson, in particular, argues that the presence of barley and weeds of cultivation suggest a more complex and flexible land use in the outland, based on two key case study sites from the late 1st millennium AD suggesting that sites could vary from outfields to satellite farmsteads or indeed shielings, showing the adaptability of the farming communities in response to both geography and external changes. Similar evidence of pastoral activity to Scandinavia is evident in the Welsh uplands of Preseli (Rhiannon Comeau and Bob Silvester) in the 12th-13th centuries with small-scale cultivation, but by the 17th century the grassland had declined as heath became dominant and shieling activity also declined. A parallel for this type of environmental change may be found on the Scottish borders, where sheep grazing took over from transhumance in the early modern period (Davies – Dixon 2007; Tipping 2010). In the Pyrenees (Tente and Celma), the clearance of woodland based on pollen evidence intensified in the Middle Ages and may be associated with exploitation of summer pastures, but in the Tauern Mountains in the Austrian Alps there was uninterrupted exploitation of the summer pastures from Roman times based on these data, although the archaeology for the early medieval period is absent for several centuries (Elisabeth Waldhart and Harald Stadler). Pollen and archaeological data provide a counter to the documented 15th-century origins of sheep and goat herding in the Estrela Mountains of Portugal, suggesting it went back to the Early Middle Ages. Other forms of environmental data were tapped at La Roche Redonne in Provence to suggest summer/early autumn for the occupation from the last growth rings preserved in charcoal, but without any certainty Sylvain Burri and Aline Durand suggest, as it is at variance with other data. In the Pyrenees, scars on trees were recorded and compared with artefactual evidence in museums to show that sheep collars and other objects were made from the bark of trees by shepherds in the summer pastures in the early modern period (Mireia Celma Martínez and Elena Muntán Bordas), extending in a novel way the range of environmental evidence that may be tapped for seasonal settlement in an area otherwise incapable of supporting settlement of any kind.

Ethnographic evidence

Where transhumant practices have continued into the 20th century, they have often been subject to ethnographic research in many European countries, providing models of these practices and evidence of the types of buildings and artefacts. Ethnographic evidence is still a key type of evidence for transhumant practices in Spain and Portugal, with a well-established understanding of the Spanish Mesta, for example, from these 20thcentury sources (Catarina Tente and Margarita Fernández Mier), but for the medieval period such research is scarce and the archaeological data even less so. This lacuna is, however, beginning to change. Ethnography can provide a model to contrast with archaeological and environmental research. Indeed, ethnography was a part of the research methodology for the unusual use of tree bark in the Pyrenees to make collars for sheep, cows, goats, and horses (Mireia Celma Martínez and Elena Muntán Bordas). The ethnography interpreted by the local museums, it was noted, present the shepherds entirely as men, raising the issue of gender, which is not easily addressed archaeologically! In Sweden, ethnographic research reveals a female bias towards transhumant herding and the milking of cows in summer (Eva Svensson), and similar gender biases in those involved in transhumance has been observed in Ireland, for example, based on ethnographic data (Costello 2018, 102). Elena Mikhaylova uses ethnographic evidence for shifting settlements to provide a model to interpret the ephemeral remains of early medieval settlements in the Russian boreal forest as the result of a similar shifting over time. In Hungary, however, ethnographic evidence provides the only model for the type of habitations used by medieval pig herders in the absence of any medieval structures (Csilla Zatykó). This weakness is a strength in Portugal, where the round buildings used for seasonal accommodation by agricultural labourers in the early modern period are still standing and are themselves viewed as ethnography loci as well as something to conserve for their heritage value. In addition, ethnographic evidence of materials, their acquisition, and use is used to interpret the way in which they were constructed (Fabián Cuesta-Gómez and Sara Prata). On the other hand, Rhiannon Comeau and Bob Silvester, investigating medieval transhumance in Wales, prefer a combined archaeological, documentary, and environmental approach to outmoded ethnography, as does Uglješa Vojvodić with respect to vlach herding in the Prokletije Mountains of Serbia. Structures associated with seasonal occupation

While most contributors depended upon ethnographic or documentary sources for providing the context of the seasonal activity represented by the structures in question, Dixon

17

the form of the structures and the evidence of repeated use suggesting seasonal or temporary use is essentially an archaeological question. Repeated use of the same site was evident in a number of chapters, which could be interpreted in diverse ways. In the Provencal mountains, Sylvain Burri and Aline Durand’s research into the archaeology of woodland crafts and agro-pastoral activities found evidence of multiple phases of activity at the huts excavated at Roque Fadade and Castel Diol over the 13th to the 17th centuries and 14th to the 18th centuries, respectively, but only one site produced a complete floor. Cleaning of the floors, bioturbation, and imprecise dating make it difficult to know what phases are actually represented. At another location, La Font de Mars (Le Castellet), there was evidence for wood-tar production at a settlement of two phases, medieval and post-medieval, with a charcoal platform, but relating one to the other with any certainty was not possible. While documentation records annual or multiyear cycles, the archaeology is less tractable. In the Arabian Desert José Carvajal López found two phases of traces of nomad tents dating to the 14th and the 19th century, respectively, in the Mulayḥah Depression interleaved by fluvial deposits, but argues that the intervening years might not be due to absence, but due to erosion of deposits by the later flood event. This reoccupation, he argues, was no accident, as the nomads returned to the same sites, using the same grammar of spatial order in rows oriented towards the southeast and using the same water sources. Elena Mikhaylova grapples with the interpretation of early medieval settlements in the boreal forests of Russia that preserved few traces of buildings and only occasional pits in the region of the Pskov long barrow culture on the edge of the many lakes of the area. She interprets them as temporary settlements, the clearly separated, i.e. stratified, thin cultural deposits (8‑15 cm) being the result of seasonal or sporadic use based around fishing or metallurgical activities, despite the topographic relationship with nearby funerary sites suggesting permanent communities. Types of seasonal practice

The types of seasonal activities practiced are many and varied. Pastoralists may be nomadic or practice a mix of sedentary and nomadic pastoralism. Semisedentary communities of this kind have been identified in northern Sweden (Gudrun Norstedt) and in the Prokletije Mountains of Serbia (Uglješa Vojvodić). At the latter, vlachs occupied settlements called katun(s) at the base of the mountain pastures which they grazed, and developed an agricultural base in the late medieval period, existing alongside nomadic vlachs, whereas in northern Sweden the forest Sami, it is argued, had a fishing economy, as opposed to the nomadic reindeer 18

herders, from the 16th century, if not before. While Sami settlements in Sweden from c. AD 600 to the 1900s on the shores of lakes and rivers may be identified by oval or rectangular hearths, these may be the settlements of the Sami reindeer herders, especially as documentary sources from the 17th century report that the forest Sami built timber huts rather than the tented structures suggested by the archaeology. Adaption is also evident in the Sámi nomads of Finland (Oula Seitsonen). From the 8th century Sámi settlements can be recognised from rows of rectangular or oval hearths (Stallo sites) with an economy based on a mix of hunting and reindeer herding in a limited geographical range. This changed to a nomadic pastoralism in the late Middle Ages, after the dislocations caused by the Black Death pandemic, and was associated with the domestication of the reindeer herds. This is reflected in the form of their habitations. Later, Sámi nomadic habits were increasingly confined and in the modern period their territories permanently altered by the development of national borders. The analogies with the sedentarisation of pastoralism in Qatar are fascinating. José Carvajal López, working in the Persian Gulf on what he calls the ‘sedentarisation’ of nomadic societies in Qatar in the 7th-8th centuries AD during the early Islamic expansion of the tribes of Arabia, excavated permanent settlements of rows of stone buildings. He suggests that they were later reversed in favour of a return to nomadism in the 9th century that continued until the 19th. The adaptability and agency of the desert nomads of Qatar in relation to external pressures is one of the key findings of his chapter. Short-distance transhumance by communities in summer, as identified earlier, was a common response to the exploitation of upland pastures in both southern and northern Europe. The development of the tripartite Alpwirtschaft in the medieval period was a strategy for exploiting the Alpine terrain (Elisabeth Waldhart and Harald Stadler). The early medieval absence of settlement in the alpine and subalpine zones in the Isel valley of the Eastern Tyrol may indicate abandonment of the pastures that were used in the Roman period and that it is only from the 9th century that there is ‘a resumption of seasonal settlement’. However, Waldhart and Stadler question the assumption that it is indeed a resumption of an established system from Roman times and before, suggesting the intervening wooded terraces above the permanent settlement horizon may have been used for summer grazing before the resumption of use of the alpine pastures in the early medieval period. It was recognised that transhumance as defined above was only one facet of seasonal settlement. It included other forms of economic activity, such as fishing, charcoal burning, tar making, and even iron smelting, to name but a few. Indeed, as it transpired in contributions to

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

the conference, it also encompassed kelp-harvesting settlements on the Western Isles of Scotland (Kevin Grant at the conference), horse-fighting sites in Norway (Marie Ødegaard), whisky-distilling sites in highland Scotland (Darroch Bratt), a fish market in Denmark (Leif Lauritsen), and cattle drovers’ temporary and seasonal settlements in Hungary (Laszlo Ferenczi). All of these leave particular forms of seasonal settlement and archaeology associated with the particular economic practice, both seasonal and temporary, and vary in form according to the environment in which the activity is carried out, whether upland or lowland, coastal or inland, woodland or open pasture, mountain or plain (steppe). The archaeology of these activities may rest as much in features that characterise their particular economic base rather than their domestic space, be it embanked drove roads in Hungary or ironsmelting furnaces in the mountains of Norway. Even so, there were wayside inns on the routes of the former for the cattle drovers and small timber shelters at the iron-working sites of Norway (Laszlo Ferenczi; Kjetil Loftsgarden). Another variant of seasonal settlement is the satellite settlement that is part of the farming cycle, but requires a separate establishment for labourers. These take the form of circular domed stone huts amongst the fields that are discussed by Fabián Cuesta-Gómez and Sara Prata in Portugal, typifying a wider expression of these structures in that country and western Spain in the early modern period. These were built to provide temporary housing for summer workers as agriculture expanded beyond the immediate environs of nucleated settlements. In eastern Croatia satellite settlements are described that were occupied by members of the farming community, which appear to be similar to the tradition of Roman villae rustica, but are known as stan granges in the early modern period, for the management of livestock, poultry, bees and orchards (Pia Šmalcelj Novaković and Anita Rapan Papeša). In Sweden, Eva Svensson argues that some of the ephemeral settlements that grew up in the outland were satellite farmsteads – an extended part of the farm (see below). The relationship to the permanent settlement is an important one; the seasonal settlement is an essential part of the farm. Seasonality

Most seasonal settlement is focused on spring-summer, especially where it is related to summer pastures. That some settlements were occupied at times of the year other than the summer season was one unexpected findings of the research. Autumn was the best time for droving of fattened cattle from the Hungarian plains to markets in Austria and further west (Laszlo Ferenczi), while the winter season is suggested by Kjetil Loftsgarden for ironproduction sites in the mountains of Norway and Sweden

where the charcoal and ore was prepared in the spring and summer, despite evidence for shelter only being found in a small numbers of cases and being small in scale. In the Highlands of Scotland, illicit distilling was carried out in the winter season, too, when the barley used to make the whisky was available and agriculture did not require much input of resource (Darroch Bratt), just as in the season for iron working in Norway. Like distilling sites in Scotland, iron-working sites in the mountains of Norway and Sweden occur in the same areas as shielings, but were not occupied at the same time. In contrast, iron working in the Sava plain in Croatia was most probably conducted in the summer on the basis of an alder leaf impression in the clay from a furnace, but also to avoid the wet seasons in the spring and autumn when the bog ore would be inaccessible due to flooding. In Norway, late summer appears to have been the time for horse markets called skeids that included competitive activities. They often occur in the same areas as shieling sites and originate as early as the first millennium AD (Marie Ødegaard), but appear to have left little archaeological trace, being mainly attested through place names and from runic inscriptions. Sylvain Burri and Aline Durand’s fascinating analysis of the exploitation of the woodlands in the Provencal mountains illustrates the complexity of the woodland crafts (bark harvesting to dyeing, tanning, woodcutting, charcoal burning, pasture of sheep, and even cultivation) and the intractability of attributing a season to the occupation of a hut or cabin from the archaeology, while suggesting that the various documented activities were least likely to be carried out in the summer. Social groups

For the most part, it is the poorer social groups that are represented as the occupiers of seasonal settlement (e.g. Anna Maria Stagno; Margarita Fernández Mier and Pablo López Gómez; Darroch Bratt; Elisabeth Waldhart and Harald Stadler and Waldhart), but this was not the story behind the crannog settlements that Michael Stratigos and Gordon Noble describe, where it is the elite who are building and staying in them during the hunting season and using these watery sites to impress their social peers. Something of the same elite influence is at work at Castle Campbell set high on the hillside of the Ochils, imposing park enclosures on the surrounding landscape that may have curtailed the pastoral activity represented by the turfwalled shieling huts in the late medieval period (Daniel Rhodes). If there is a theme here, it is the way in which the landowning elite imposed its influence on the use of upland pastures. This was the point made by Ian Maclellan in his chapter on the royal hunting forest of Mamlorne, in highland Perthshire, during the 17th and early 18th centuries. The royal forester there wished to protect the Dixon

19

deer stock in the face of the competing demands of the landowner to exploit the same land for grazing, leading to a dispute that went to the House of Lords in Westminster to settle. Archaeologically, shielings provide evidence for short-distance transhumance in the forest to make use of the summer pastures, but at the same time the landowner was letting out grazing to commercial graziers, putting further pressure on the deer stocks and the practice of summer shielings by the local communities. Landlordism is also evident in the Serbian context (Uglješa Vojvodić), where the role of the Crown, monasteries, and landlords in the management of mountain pastures is critical to understanding the archaeology, where research to date has been limited. Monastic charters of the 13th and 14th centuries defined the rights and duties of vlach cattle herders in the mountains as including military service, giving them an important status in society that differentiated them from farmers. Drivers of change

External drivers are one of the factors raised in a number of chapters. The practice of pannage in Hungary may or may not originate in the medieval period, but its long-distance practice outlined by Csilla Zatykó may have arisen in the context of the lack of availability of local resources for pig breeders to exploit, due to the growth of permanent settlement and the demands of developing urban markets. That this may have led to the development of a particular breed of pig is a sign of the importance of pannage in the medieval economy, but archaeological evidence is scant and analysis of stable isotopes of pig bones is yet to be employed to determine whether this confirms the movement of sheep suggested by the documentation. The development of the long-distance cattle trade in late medieval and early modern Hungary is another indication of the growing demands of urban markets in central Europe in the early modern period (Laszlo Ferenczi), one that is paralleled by the development of cattle droving from the highlands of Scotland to feed the growing demands of the London market. In Hungary, as in Scotland (Lowdon 2014, 205‑212; Adamson 2014, 317‑318), this led to inns being established alongside the drovers’ roads. Ian Maclellan’s chapter on the forest of Mamlorne in highland Perthshire in the 17th century illustrates the effect of these grazing pressures on short-distance transhumance with the destruction of shieling huts, as the local inhabitants’ customary grazing was squeezed between maintenance of the deer and the introduction of cattle graziers. The same pressures may be the background for the enclosure of land at Castle Campbell in the 16th century, ending the use of the shielings (Daniel Rhodes) and leading to the exploitation of these hills for cattle as well as the designed landscape of the elite. 20

Climate change

Richard Oram in his keynote speech outlined the scientific evidence of climate deterioration along with the Black Death pandemic and panzootic events in the 14th century and suggested that these environmental factors appear in ways that are not always apparent to historians and archaeologists. Since we know these changes and events took place, what were the consequences for late medieval society? How far are these evident in the research presented here? Climate, for example, is mentioned in 18 chapters, mostly only in passing as a possible influence. Oula Seitsonen, however, tackles this more directly. She suggests that the Medieval Climate Anomaly may have provided a situation alongside the demand for furs that stimulated the change in Sámi settlements with an increase from the 8th century in Stallo sites in the high mountains and the sites with rows of hearths at Lake Gilbbesjávri and on the River Leahttáseatnu, for instance. In the late medieval period Stallo sites fall out of use and there are changes in Sámi religious practices, which are attributed variously to the pandemic, the Little Ice Age, and forced Christianisation. New styles of hearth are used called bearpmet, oriented towards the entrance, or round hearths, both related to the development of nomadic pastoralism and reindeer herding as opposed to a huntergather economy. Richard Tipping, by contrast, attributes the attrition on woodland and herbivorous species to the intensification of grazing rather than climate. But perhaps the two are interrelated, the growing cattle economy of the Highlands a reaction to the poorer climate’s limiting cultivation as well as meeting the demands of growing markets in central Scotland and further afield in London in the 17th century, following the Union of the Crowns. Leif Lauritsen showed how herring fishing in the shores of Denmark in late summer led to markets being established on Øresund in the Middle Ages, dominated by the Hanseatic League. Control of the markets became a matter of dispute between the king of Denmark and the Hanseatic League in the 14th century, which led to war and the defeat of the king. In the aftermath, small markets were established such as that on a natural spit of land called Albuen, Lolland, covered with grassy banks marking the site of over 130 huts and enclosures providing stalls for the market traders. Excavated examples had a post at each end and wattle walls marked by smaller postholes along the earthen banks. A series of floor levels were identified, which might indicate seasonal occupation. The fish market continued until the herring disappeared at the end of the 15th century. The collapse in the herring fisheries and the warfare with the Hanse may be a symptom of pressure on fish stocks due to climatic changes. A climatic link was proposed some years ago, with cooler water reducing the spawning of herring and thus the stocks in Scania in the 14th and 15th centuries (Hoffman 2005).

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Jurisdictional control

Another strand in the interpretation of pastorally based seasonal settlement is that the jurisdictional context of the pastures is important for understanding the nature of the archaeological remains. Margarita Fernández Mier and Pablo López Gómez’s account of the use of the mountain pastures in the Cantabrian Mountains of northwest Spain is a fascinating study of short-distance transhumance that benefits from the pastures’ continued use up until the 20th century in providing documentation of the management of the commons. Their study is of two different areas that are both now owned by a neighbourhood council, having been acquired from landowners in the 19th and 20th centuries. The common pastures of the village of Taxa included hay meadows (marked by stones) to which each villager had sole rights but once harvested were turned over to communal grazing as well as pastures in collective use shared with neighbouring villages. The Taxa summer cabins were stone huts of similar size (2 by 3 m) linked to a plot in exclusive ownership, scattered throughout the pastures. It appears these originate in the 18th century, due to population pressure and ownership changes that led to the division of the pastures into individual holdings. At Montoubu the common pastures were not divided and the branas were grouped in three clusters. Individual villagers built a variety of shelters for men and also for animals and animal products, including small huts used as dairies. The importance of who controlled the common pastures also comes into relief in Anna Maria Stagno’s study comparing Basque commons with those in the Ligurian Apennines. This shows how the archaeology reflects the way in which common pastures were managed to produce two contrasting patterns of seasonal settlement. The commons in Liguria were protected from drovers by walled tracks called creste. Between the 19th and the 20th centuries, the commons of Monte Becco were divided between the different inhabitants designed to make the upland pastures more productive, replacing shared access to the commons based on established rights with individual holdings and intensification of the economic base. In the process it destroyed the cultural knowledge of the practices followed by communal graziers of the upland pastures. However, the analysis of the Basque country pastoral activity suggests a rather different trajectory. Here there are low- and high-level pastures belonging to individual villages, which had ancient woodland in the former and wood pastures in the latter. Neither pastures had seasonal settlements, only enclosures and sheds. In the late 19th and 20th centuries the wood pastures were replaced by coppiced woodland used for charcoal production. Beyond these pastures exclusive to the village were common pastures that multiple villages grazed and that

have a high concentration of seasonal settlements, with small nuclei each with its place name that are still used. Conclusions

If one question was answered emphatically, it was that definitive recognition of seasonal settlement is not so far possible on archaeological grounds alone. Generally, a combination of documentary information, ethnography, geographical context, and palaeoenvironmental data provides a framework for making interpretations of seasonal settlement. Ethnographic evidence is never a good substitute for archaeological evidence from the medieval period, but can be a useful foil on which to test the evidence in much the same way as documentary and environmental data. There is more that may be done scientifically to determine mobility of both humans and domesticated animals, such as isotope analysis, while thin sections of floor deposits of cabins and huts may yet yield information on repeated use in certain circumstances (Kupiec – Milek 2018). Richard Oram’s call to think more carefully about what the evidence means in relation to climate change in the increasing scientific evidence for its impact should be considered. However, causal relationships are often difficult to establish when other factors, such as pandemics and its after effects, are thrown into the equation. This is particularly true in drawing conclusions about surviving cultivation remains in upland areas, as opposed to sandblown inundations that are more obviously the result of dramatic climatic events. The limited number of archaeological excavations that were presented in any detail was striking and may be a reason for the lack of archaeological evidence for seasonal use. In Scotland the number of sites that might be seasonal settlements that have been excavated is still very small, as it is in Ireland, Wales, and indeed England. However, the amount of archaeological work in Scandinavia on Sami sites and the important evidence that these provide shows its potential. Eugene Costello’s point about the destructive effects of modern agriculture and forestry on locations where seasonal settlements may once have been is likely to be a factor in many parts of Europe. The ephemeral nature of seasonal settlement in areas that use perishable material exclusively will continue to limit archaeological preservation and research. Yet even tented structures can leave a hearth, if little else. As our period is medieval in the broadest sense, there is a wealth of documentation to be investigated that may yet reveal details about seasonal activities and especially settlement. Anywhere that buildings are constructed can meet with a reaction from the powers of those who control the pastures, whether this is because of conflict Dixon

21

of use or the wish to tax or charge rent for the privilege. So, understanding the jurisdictional regimes that apply to pastures is important, since they provide limits to the agency of farmers, graziers, charcoal burners, mineral exploiters, and others. There is also an inevitable clash between the contradictory evidence of one discipline against the other, but from this better research and interpretation should result. Although palaeoenvironmental data is a key resource for understanding pastoral and woodland land use, it can provide interpretive conundrums for archaeologists, such as the absence of archaeological evidence in the Alpine pastures in Austria for any settlement in the early medieval period, despite the pollen evidence suggesting no change since the Roman period. Similar conundrums have occurred in Scotland for the early medieval period (Tipping 2010, 190). The value of this type of data has been demonstrated in Sweden by Eva Svensson and in Wales by Rhiannon Comeau and Bob Silvester, showing that what we have interpreted as seasonal settlements were also engaged in agriculture. The separation in our models between permanently occupied and seasonally or temporarily settled sites is not so clear-cut. References

Adamson, D.B. 2014: Commercialisation, change and continuity: an archaeological study of rural commercial practice in the Scottish Highlands. PhD thesis presented to University of Glasgow. Andres, B. 2018: Alpine settlement remains in the Bernese Alps (Switzerland) in medieval and modern times: the visibility of alpine summer farming activities in the archaeological record, in Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe, 155‑170. Barth, F. 1962: Nomadism in the mountain and plateau areas of south west Asia. The Problems of the Arid Zone, UNESCO Arid Zone Research 18, 341‑355. Collis, J.R. – Pearce, M. – Nicolis, F. (eds.) 2016: Summer farms: seasonal exploitation of the uplands from prehistory to the present, Sheffield Archaeological Monographs 16. J.R. Collis Publications. Sheffield. Costello, E. 2018: Morphology of transhumant settlements in postmedieval South Connemara: a case study in adaption, in Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe, 93‑108. Costello, E. – Svensson, E. (eds.) 2018: Historical archaeologies of transhumance across Europe, EAA. Routledge. 22

Davies, A. – Dixon, P. 2007: Reading the pastoral landscape: palynological and historical evidence for the impacts of long-term grazing on Wether Hill, Ingram, Northumberland, Landscape History 29, 35‑45. Dixon, P. 2009: Hunting, summer grazing and settlement: competing land use in the uplands of Scotland, in Klápšte, J. – Sommer, P. (eds.), Medieval rural settlement in marginal landscapes, Ruralia VII. Brepols, 27‑46. Dixon, P. 2014: Survey and excavations at Alnhamsheles deserted medieval village, on the Rowhope Burn, Alnham Moor, Northumberland, Archaeologia Aeliana 5th Series 43, 169‑220. Gardiner, M. – Svensson, E. 2009: Introduction: marginality in the pre-industrial European countryside, in Klápšte, J. – Sommer, P. (eds.), Medieval rural settlement in marginal landscapes, Ruralia VII. Brepols, 21‑25. Hoffmann, R.C. 2005: A brief history of aquatic resource use in medieval Europe. Helgoland Marine Research 59, 22‑30. Published online at https://hmr.biomedcentral.com/ articles/10.1007/s10152‑004‑0203‑5 (accessed 31. 01. 2021). Klápšte, J. – Sommer, P. (eds.) 2009: Medieval rural settlement in marginal landscapes, Ruralia VII. Brepols. Kupiec, P. – Milek, K. 2018: Ethno-geoarchaeological study of seasonal occupation: Bhiliscleitir, the Isle of Lewis, in Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe, 75‑92. Lowdon, R.E. 2014: To travel by older ways: a historical-cultural geography of droving in Scotland. PhD thesis presented to University of Glasgow. Rendu, C.- Bille, E. – Calastrenc, C. – Campmajo, P. – Crabol, D. 2009: Margins and centres in the shaping of the Pyrenean slopes. Medieval dynamics within the long-term perspective, in Klápšte, J. – Sommer, P. (eds.), Medieval rural settlement in marginal landscapes, Ruralia VII. Brepols, 235‑251. Tipping, R. 2010: Bowmont: an environmental history of the Bowmont valley and the northern Cheviot hills, 10000 BC-AD 2000. Society of Antiquaries of Scotland. Winchester, A. 2012: Seasonal settlement in northern England: shieling place-names revisited, in Turner, S. – Silvester, B. (eds.), Life in medieval landscapes: people and places in the Middle Ages. Windgather Press, 125‑149.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Too much environment and not enough history: the opportunities and challenges in researching medieval seasonal settlement in Atlantic Europe Richard Oram*

Abstract

* Pathfoot Building University of Stirling Stirling FK9 4LA Scotland United Kingdom [email protected]

Over the last 25 years, advances in paleoenvironmental research have revolutionised our understanding of the physical effects of historic climate change around the North Atlantic rim across the eras of the Medieval Climate Anomaly and subsequent Little Ice Age. This revolution has been marked in respect of marginal upland and coastal zones, where landscape-scale palaeoecological research coupled with excavation at abandoned perennial and seasonal settlement sites has provided high-quality and subtly nuanced data to evidence baseline conditions, impacts, and responses. In Scotland, analysis of these data has been framed largely in terms of system sustainability and environmental resilience but, with few notable exceptions, has offered no examination of human agency in shaping responses to climate change or of wider historical contexts for trends evident in the palaeoenvironmental data. Equally, however, too few archaeologists and historians have engaged with the environmental contexts for socioeconomic discontinuities, site abandonment, and resource-related conflict reflected in artefact and ecofact assemblages or the parchment record. Consilience and inter/transdisciplinary approaches to the study of historic seasonal settlement and associated exploitation regimes can provide insights on human ecodynamic processes, avoiding the risk of unconscious determinism through linear, single-discipline analyses and revealing the complex interplay of natural agency and human cultural responses to the opportunities and threats presented by past climate change. Keywords: climate change, land management, upland exploitation, carrying capacity.

In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 23-32.

23

Résumé

Zusammenfassung

Mots-clés : changement climatique, gestion des terres, exploitation des hautes terres, capacité d’absorption.

Schlagwörter: Klimawandel, Landbewirtschaftung, Hochlandnutzung, Tragfähigkeit.

Introduction

bushfires, forest dieback, and collapse in biodiversity and species resilience, discussion is moving towards change at a societal level and focussing on topics such as food production/dietary staples, energy production and consumption, water security and transport, both shortrange local commuting and international long-haul. Each area affects matters of cultural significance, some of profound importance for religious or philosophical

Trop d’environnement et pas assez d’histoire: les opportunités et les défis de la recherche des établissements saisonniers médiévaux en Europe atlantique Au cours des vingt-cinq dernières années, les progrès de la recherche paléoenvironnementale ont révolutionné notre compréhension des effets physiques du changement climatique historique autour du pourtour de l’Atlantique Nord à travers les périodes de l’anomalie climatique médiévale et du «petit âge glaciaire» qui a suivi. Ces chagements sont décrits pour les hautes terres marginales et les zones côtières, où la recherche paléoécologique à l’échelle du paysage couplée à des fouilles sur des sites de peuplement pérennes et saisonniers abandonnés a fourni des données de haute qualité et subtilement nuancées pour prouver les conditions de base, les impacts et les réponses. En Écosse, l’analyse de ces données a été conçue en grande partie en termes de durabilité du système et de résilience environnementale, mais, à quelques exceptions notables près, n’a proposé aucun examen de l’action humaine dans l’élaboration des réponses au changement climatique ou dans contextes historiques plus larges pour aux tendances évidentes des données paléoenvironnementales. De même, trop peu d’archéologues et d’historiens se sont penchés sur les contextes environnementaux pour expliquer les discontinuités socio-économiques, l’abandon de sites et les conflits liés aux ressources reflétés dans les assemblages d’artefacts et d’écofacts ou les registres de parchemins. La cohérence et les approches interdisciplinaires et transdisciplinaires de l’étude des établissements saisonniers historiques et des régimes d’exploitation associés peuvent fournir des informations sur les processus écodynamiques humains, en évitant le risque de déterminisme inconscient généré par des analyses linéaires et disciplinaires uniques et en révélant l’interaction complexe de l’action naturelle et des réponses culturelles humaines aux opportunités et aux menaces présentées par les changements climatiques passés.

Amongst high-level responses to the current ‘Climate Emergency’ is a growing discourse within environmental activist and scientific communities about what practices and behaviours humanity globally will have to stop or change to survive. With the 2010s now confirmed instrumentally as the hottest decade on record (Beament 2020) and against a backdrop of extreme weather events, 24

Zu viel Umwelt und zu wenig Geschichte: die Chancen und Herausforderungen bei der Erforschung der mittelalterlichen saisonalen Besiedlung im westlichen Europa In den letzten 25 Jahren haben Fortschritte in der Paläoumweltforschung unser Verständnis bzgl. der Auswirkungen des historischen Klimawandels am europäischen Nordatlantikrand in den Epochen der mittelalterlichen Klimaanomalie und der anschließenden „kleinen Eiszeit“ revolutioniert. Diese Entwicklung wurde in Bezug auf marginale Hochland- und Küstengebiete beschrieben, in denen paläoökologische Untersuchungen in der Region in Verbindung mit Ausgrabungen an verlassenen permanenten und saisonalen Siedlungsstandorten qualitativ hochwertige und subtil nuancierte Daten lieferten, um die Ausgangsbedingungen, Auswirkungen und Folgen nachzuweisen. In Schottland wurde die Analyse dieser Daten weitgehend im Hinblick auf die Nachhaltigkeit des Systems und die Widerstandsfähigkeit der Umwelt durchgeführt. Aber mit wenigen bemerkenswerten Ausnahmen wurde keine Untersuchung der menschlichen Handlungsfähigkeit bei der Gestaltung von Reaktionen auf den Klimawandel oder eines breiteren historischen Kontextes auf Trends in der Paläoumwelt durchgeführt. Gleichermaßen haben sich jedoch zu wenige Archäologinnen, Archäologen, Historikerinnen und Historiker mit den Umweltkon­ texten für sozioökonomische Diskontinuitäten, Standort­ verlassenheit und ressourcenbezogene Konflikte befasst, die sich in Artefakt- und Ökofakt-Assemblagen oder in schriftlichen Aufzeichnungen widerspiegeln. Konsilienz und inter- / transdisziplinäre Ansätze zur Untersuchung der historischen saisonalen Besiedlung und der damit verbundenen Ausbeutungsregime können Einblicke in die ökodynamischen Prozesse des Menschen geben, das Risiko eines unbewussten Determinismus durch lineare Analysen einzelner Disziplinen vermeiden und das komplexe Zusammenspiel von natürlicher Handlungsfähigkeit und menschlichen kulturellen Reaktionen auf die Chancen und Risiken des vergangenen Klimawandels aufzeigen.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

reasons, but all touching on themes of identity at every level from the personal to the pan-societal. Global reporting, 24-hour worldwide news, multiple social media channels, and the rise of open-access research has increased public awareness of climate change, the debate over its causes, and the evidence of its impacts. Human – and animal – experience of climate change is visible and its effects on the environment in which we subsist are patent. Now, more than ever before, environmental history is offering insight on how climate change will affect modern culture and society, through the record of how our ancestors adapted and changed in the face of historic climate transitions. We are awash with readily accessible data on environmental impacts and can model outcomes, but most public debate is couched in terms of weather events. Where the effects of climate change on human populations are discussed, it is sea-level rise, salinizing of water sources, or desertification in distant and usually underdeveloped countries that is highlighted. Rarely does the discourse turn to likely impacts on communities closer to home. Climate change and historical discourse

Climate change research in the high Arctic, commencing in the later 20th century, has amassed abundant data for past episodes of climatic amelioration or deterioration. In respect of changes since the 9th century, it has had the informative side effect of delivering detailed evidence for the impact of such episodes on the Norse Western Colonies along the North Atlantic arc from the Hebrides to Newfoundland. In contrast to the increasingly sophisticated interdisciplinary study of the impacts of medieval climate change on Greenlandic and Icelandic populations, however, there has been limited interdisciplinary exploration of the medieval Scottish experience. Engagement by historians of medieval rural culture and society in the northern British Isles – and Scotland especially – with analysis of environmental factors as motors for long-term and large-scale social, economic, and cultural change has been limited and, too often, quite superficial. There are important exceptions, mainly sharply focused local studies (e.g. Brown et al. 1998; Tipping 1998, 2004; Dixon 2007), but wider analyses over the long durée are lacking. Yet, this position is despite the daily reminders received by Scotland’s people of the presence and impact of the dynamic weather that arises from its location at the north end of an island placed between the northwestern European landmass and the Atlantic Ocean. Awareness of the major climatic episodes of the last millennium is reflected in generalised statements about Scotland’s people maximising opportunities presented by early medieval climate optima (e.g. Atkinson 2016, 77), but more often it is the ‘Golden Age’ of the Medieval Climate Anomaly (MCA) that is alluded to. We are told that they

pushed cultivation and pasture to the altitudinal and marginal maximum during these ‘good times’, but there is minimal reflection on what that expansion meant socially or culturally, or how climate change was experienced by those living at these maxima when the good times ended. What increased or reduced opportunity meant in terms of social organisation and cohesion, cultural adaptability, community resilience, resource management, subsistence levels, production of dietary staples, or levels of health and nutrition rarely features in the broad discussion. Cyclically, such limited engagement has disincentivised the adoption of the interdisciplinary methodologies that often are instruments for collection of environmental data from both documentary and archaeological sources, perpetuating a disjunction between historical and palaeoenvironmental research. Where environmental historical research has been undertaken into Scottish lived experiences, however, it has shown the benefits of integrating environmental data to deliver a richer image of the opportunities and challenges presented by the rises and declines in climatic conditions. Such data from diverse written, climate-proxy, and instrumentally recorded sources illuminate the analysis of population responses in Scotland to the optimum climatic conditions of the MCA of the mid-11th to mid-13th centuries and the following era of deterioration into the ‘little ice age’, with its oscillation between sustained periods of profound climatic variability and relative stability against a generally downward trajectory in annual mean temperatures and upward trend in storminess from the mid-13th to the mid-19th century, culminating in the modern era of rising temperatures globally and greater weather volatility. The present author and colleagues have applied these techniques to examination of social, economic, and political outcomes from the MCA’s decline from the 1250s to the 15th century (Oram 2014a, 2014b, 2014c; Oram – Adderley 2008, 2010). Those studies, however, offered headline overviews that argue for the recontextualising of the well-rehearsed narratives of political reconfiguration and socioeconomic realignment that have characterised Scottish historical discourse. The effects of climate change on key elements of the socioeconomic landscape, such as seasonal uplandexploitation regimes, require more focussed research at a whole system’s level. Scottish historiography has traditionally taken political events, especially the death of kings, as transitional points in the nation’s development. King Alexander III’s death in 1286 is the arch-exemplar, viewed as the end of the kingdom’s medieval ‘Golden Age’. The 60-year struggle for an independent national identity that followed his demise dominates all narratives, despite increasing awareness of the mythologising of which it is part. It was for long portrayed as precursor to an era of Oram

25

weak monarchy, ‘over-mighty’ nobles, and widespread social disturbance, epitomised by plundering raids launched from the Highlands against lowland districts, and noble factionalism and feuding. That portrayal, too, is now recognised as largely an historiographical fiction, which cast the centralising tendencies of the Crown and the resistance to that of the magnate class in terms of a struggle between opposites of stability and chaos. Political narratives thus have dominated discourse in late medieval Scottish history, and the context for these trends has been expressed almost exclusively in terms of social disintegration and rising levels of violence, legal and illegal, consequent upon a failure of political leadership. Whilst the detail of that narrative has been serially challenged in recent years, the primary focus remains on sociopolitical contexts for change, including those visible in the archaeological record; the emphasis has shifted slightly to accommodate economic factors, but the driving themes still centre largely on high-level political conflict. Towards a new environmental history

More recently, as data relating to millennial-scale climatic trends in the northern hemisphere, and especially to the atmospheric and oceanic motors that powered them, have become available, a new historiography has emerged. This new prescription has begun to contextualise the well-recognised social, economic, and political trends in a wider environmental frame. For example, Alasdair Ross viewed the probably 8th- to mid-9th-century establishment in northern Scotland of socioeconomic units known as daibhachean (singular dabhach), organised to provide communities with pro rata access to adequate supplies of arable, lowland and upland grazing, water, building materials and fuel resources, and sustain those communities as the basis for military levies, as occurring in the context of a response by Pictish kings to Viking incursions (Ross 2015, 195‑196). Although he provided a political explanation for the creation of the daibhachean, he understood their organisation in terms of human economic need and responsiveness to climatic opportunity, environmental constraints, and resource distribution. His methodology eschewed environmental determinism but offered climate-driven environmental change as just one vector, alongside the traditional historiographical media of political and economic disruption, for the introduction of unprecedented levels of opportunity or stress into the subsistence regimes upon which the Scottish peasant population depended and upon which long-established lordship structures were sustained. These opportunities and stresses provide context for cycles of growth and contraction, as argued for the expansion of upland exploitation in the late Pictish period and its decline in the 11th century, re-expansion in 26

the 12th, and collapse in the 14th (Strachan et al. 2019; Oram 2014a). In short, in the benign conditions of the third quarter of the 1st millennium AD and again in the MCA, as it got warmer upland settlement, perennial and seasonal, expanded as population grew and lordship structures crystallised, but when it got colder and wetter in the later 10th century and again post-1250, biomass production contracted, carrying capacity declined, and upland regimes suffered. It was not, however, simply a matter of ‘it got colder and wetter, and first livestock then people died’, for we can see evidence for a wide range of human responses, including increasing violence arising from competition over access to now-scarcer resources, changes to stocking levels or shifts towards livestock and cereal crop types better adapted to harsher conditions, abandonment of the most marginal areas of settlement and agriculture, and the imposition of new organisational frameworks and administrative mechanisms to exploit more rigorously the labour of the peasant population. Responsive to both climatic factors and social pressures, Scotland’s political elites, who had creamed the surplus from upland regimes, looked for new means to obtain the incomes upon which their status was founded, while peasant populations adapted to find a sustainable regime upon which to subsist; the social and political tensions evident in the historical record reflected human responses to these changed environmental circumstances and the impulsion to find a new status quo that delivered greater economic security and stronger social resilience. Focussing just on the MCA to Little Ice Age transition, despite the fragmentary nature of the documentary record for much of the later 13th and 14th centuries, it is possible to construct an alternative broad-brush narrative for the era, summarised here. The Alexandrian ‘Golden Age’ is now better understood both as a cultural construction of later medieval Scottish nationalistic historiography and in the context of the climatic volatility of the declining decades of the MCA. That decline brought episodes of extreme weather – drought alternating with deluge, high summer temperatures counterpoised with extreme winter lows – and a generally downward trend in mean annual temperatures. References to headline extreme events – storms, coastal inundations, spate floods, severe frosts, blizzards, crop failures, and epizootic disease – and to their impacts on the human population, punctuate contemporary accounts. Summer and winter pasture and upland livestock regimes seem to have fared badly. A brief respite in the early 1300s led into the horrors of the 1315‑1327 period: the Great European Famine and a pan-European cattle epizootic in a saturated decade of unprecedented rainfall for the eastern Atlantic rim (Slavin 2010, 2019; Oram 2014a). Scottish record evidence for this period is scant, but the surviving records from adjacent regions of

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

England and Ireland offer proxies that leave little doubt as to the likely impacts on Scotland. Although the 1330s and 1340s saw greater stability in weather patterns, a downward trend in mean temperatures continued into mid-century, with the coldest years coinciding with the first pandemic of the Black Death. A recovery in temperatures through the 1370s led to an economic recovery reflected in records of bumper wool yields, but greater climatic volatility in the last quarter of the century, reaching a peak in the early 1400s, ended that boom and stimulated competition for control of resources. A downward climatic trajectory thereafter became entrenched throughout the second and third quarters of the 15th century and the attendant economic precariousness, coupled with recurrent epidemic and epizootic outbreaks, contributed to widespread and protracted political volatility (Oram 2014c; Oram – Adderley 2008, 2010; Tipping 2004). Climate proxy data

While this narrative can be nuanced by deductive reasoning and extrapolation of primarily qualitative data from the documentary record, the nature of the surviving Scottish records hinders reconstruction of the long runs of interrogable quantitative data necessary to model the impacts of climate change and pathogens at regional or local level. No amount of lamenting the absence of the type of manorial accounts that have enabled English environmental historians to express in human terms the consequences of the climatic slide from the highs of the MCA into the lows of the Little Ice Age can alter the fact that equivalent materials are irrecoverable for Scotland before the 1500s. We can, however, continue to broaden and refine the sequences of climate proxy records to enable modelling at a level of regional specificity far greater than most currently available Scottish scenarios, which are grounded in data obtained from geographically remote areas: Greenland, the Baltic Sea, Urals-Siberia. Such regional-level models invite discussion of the lived experience of peasant populations who strove to subsist in a regime of shifting ‘bad year economics’ (Halstead – O’Shea 2004). Models, however, are only as good as the proxies upon which they are constructed. Until recently, most ‘native’ Scottish sources of proxy data were derived from the Atlantic-facing western margins of the country and lacked the fine-grained resolution necessary for socioeconomic impact modelling of even those regions, let alone of areas further east. Scotland’s topography determines a marked east-west differentiation in temperature and rainfall, with higher rainfall in the mountainous, Atlantic-facing west; an appreciable rain shadow in the east; oceanic amelioration of winter temperatures in the west; and colder winter temperatures further east. Comparison with modern,

instrumentally recorded data for eastern and western mainland weather, however, lends significant weight to models for medieval central and eastern Highland conditions, based on long sequences of northwestern mainland and Atlantic west-coast proxy data. What are the sources of these proxy data? Some, which elsewhere have enabled construction of climate models extending back over two millennia – particularly the annual summer growth-ring sequences from temperatureand moisture-sensitive tree species – are currently unavailable for most of Scotland. The ecology of Scotland’s native woodlands, historic woodland-management practices, and poor architectural or archaeological survival of structural timbers from known native sources have until recently limited the value of tree-ring evidence as a long-sequence climate proxy for summer conditions (for an overview, see Mills – Crone 2012). This position contrasts sharply with the availability of such data in northern and western Ireland. Oak, the main species from which the long growth-ring sequences used to calibrate chronologies in Ireland have been obtained, has been subject in Scotland to such intensive management that few living trees older than c. 550 years remain. A handful of older veterans are present in managed lowland park landscapes, but their geographical isolation from the living populations and preserved timbers upon which English and Irish tree-ring sequences have been established has restricted their effective use. Recent discovery of medieval oak timbers – for example, in the refectory roof of Paisley Abbey (Canmore 2020) and in the submerged oak structure in Hunterston Sands (History Scotland 2014), both in the southwestern lowlands – offers the prospect of a linkage of surviving living trees in the Cadzow and Loch Lomond oakwoods into northern Irish sequences. Too little survives, however, to enable the kind of summer rainfall and temperature models constructed for Ireland to be advanced with confidence from the currently available Scottish evidence. Pine, which has a broader distribution than oak and which was used to a greater extent as a constructional timber throughout upland Highland areas, has been advanced as a potential proxy-data source with a wider Scottish range than oak (Rydval et al. 2017; Mills – Crone 2012; Wilson et al. 2011; Mills 2008). Although there are no living Scottish pines older than around 225 years, a focus on surviving structural timbers and on wood recovered from waterlogged or peat-preserved contexts has allowed a long dendrochronological sequence to be constructed for the mountainous northern Cairngorm region in northeastern Scotland, at present spanning some eight centuries. 14C dating and calibration against trends evident in the oak sequences, speleothems (discussed below), and other climate proxies show similar responses affecting the pinewoods, most noticeably at present the Oram

27

impact of atmospheric cooling and increased precipitation in the 14th century. The palaeohydrological records of peat profiles and speleothems from the Tralligil basin in the northwestern mainland offer a further important proxy-data source (Charman et al. 2001). Used as a measure for winter precipitation, these data have revealed synchronous trends with Greenlandic ice-core data across the last two millennia. In respect of medieval climate change, they show a marked increase in rain and/or snowfall peaking in the late 14th century, around 570 years BP. There is still debate over the fine-grained accuracy of these data, but that is balanced by general acceptance that these palaeohydrological records provide an annual resolution climate reconstruction of local rainfall and regional winter atmospheric pressure system patterns in the North Atlantic Oscillation (NAO). Together with our understanding of shifts in oceanic water circulation, in this context the Atlantic Meridional Overturning Current (AMOC), these NAO-related data enable construction of generalised models for winter weather across parts of northern Highland Scotland in the late medieval period. Temperature data is more contested, but using instrumentrecorded modern data for similar NAO patterns it is likely that increasing precipitation was accompanied by decreasing winter temperatures across the 14th and 15th centuries in northwest Highland Scotland, which would accord with historical records of extreme winters, delayed springs, and cooler summers with higher precipitation. Accompanying that decline came a shortened growing season that was accentuated by altitude, affecting traditional short-range transhumance regimes. When taken in combination with research into deepwater renewal in fjord-like west coast lochs like Loch Etive and Loch Sunart, where changes in biotic and isotope indicators for temperature and relative salinity have been used as palaeoclimate proxies, the palaeohydrological evidence becomes more promising yet (Cage – Austin 2010; Stott et al. 2010). Increased or lowered salinity levels in these almost-landlocked inlets are indicative of drought (high salinity) or high precipitation (low salinity) reflective of variation in freshwater runoff from the surrounding land into the lochs. Like the palaeohydrological records, the principal driver for these changes is understood to be the NAO, working in conjunction with the AMOC. At Loch Sunart, a marked lowering of bottom-water temperatures from an MCA high occurred after the middle of the 13th century; temperatures recovered in the first decade of the 14th century before falling through the 1310s; and there were significant annual fluctuations through to the 1350s, when a profound drop was recorded, with a recovery into the first decades of the 15th century ending in a slide to the lowest temperatures recorded since the pre-MCA early 10th century. This record from 28

the southwestern Highland zone correlates closely with the palaeohydrological record from the northwest zone, pointing generally to rising levels of precipitation from the 1250s, interspersed with brief episodes of prolonged drought, coupled with a downward trend in temperatures remarkable for extended periods of extreme coldness in the late 1400s. In summary, the weakening of the NAO after the highs of the MCA brought a general shift into colder, wetter, and – bringing in a third proxy record in the form of optically stimulated luminescence (OSL) dates for major episodes of dune deflation and sand-blow in the Outer Hebrides (Gilbertson et al. 1999; Dawson et al. 2011) – stormier winters and generally cooler and wetter summers. Human responses: practice and legislation

Together, these climate proxies chart long-term shifts in the weather systems reaching Scotland from the North Atlantic. It is unlikely that such shifts had no cumulative impact on resource exploitation regimes founded on traditional environmental knowledge that had been built up across the three centuries of the MCA. For communities in Scotland’s Southern Upland and Highland zones, increased rainfall and cooler temperatures affected everything from fuel supply – dominated in such regions by peat consumption – through to meat and milk yields from cattle and yield ratios of cereal crops, although this may have been offset in part by an increased reliance on more wet- and cold-tolerant oat species (Tipping 1998). A survey of the renders in kind from royal estates recorded in the surviving accounts of the Scottish exchequer reveals a progressive shift towards oat production through the 14th and 15th centuries, continuing into the 16th century (Stuart et al. 1878‑1908), paralleled by increasingly frequent legislative efforts to manage planting and landmanagement practices and the trade in victuals (e.g. Brown et al. 2007‑2020, 1370/2/8, 1401/2/15, 1426/39, 1452/5, 1454/3, 1458/3/28‑9, 1486/3/2, 1540/12/70). Here, Scotland’s royal administration and parliament were seeking to address the perceived causes of declining productivity, food shortages, and the concomitant decline in revenue yields through punitive legislative measures. They viewed the recurrent supply crises purely in human terms as the consequence of poor practice and solvable through human action impelled by legal enforcement. Scotland’s 15th-century legislators, however, were seeking to address the results of poorer growing conditions through instructing landowners and their tenants simply to plant and grow higher volumes of grain crops or increase their livestock numbers. It is in these legislative measures and in the records of efforts by landowners to adopt them that we can see best the interplay of human and environmental factors and detect

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

on the ground the results of such endeavours. The financial accounts of the monastic landlords of the Coupar Angus abbey estates in upper Glen Isla, for example, detail efforts to regulate carrying capacity on seasonal upland pasture and increase the planting of woodland, management of scrub species, and sowing levels on the arable land there (e.g. Rogers 1880, 141‑142, 261‑262). They were, however, struggling in an age of reduced biomass production, with more days per annum failing to reach the critical 6˚C temperature threshold for grass growth. This reduced grass growth affected carrying capacities of summer and winter grazing land and entailed either reductions in livestock-to-acreage ratios (known as soums) or increases in the extent of grazing assigned to a community (Ross 2006). Where such changes were not negotiated locally, conflict over control of, or access to, resources proliferated, escalated, and intensified. Whilst recognising that we lack at present much data from the eastern watershed zones, extrapolation from instrumentally recorded atmospheric and precipitation records in lowland and especially east coast locations suggests that even with the buffering effect of the mountains, eastern agricultural regimes were confronted with only moderately ameliorated challenges and their seasonal upland grazing resources suffered the same impacts as were experienced further north and west. For perennial settlement, the same temperature threshold and increased rainfall affected production of less cold- and wet-tolerant wheat, again potentially offset by a move to oat cultivation, and may have been in part responsible for the altitudinal drop in cultivation evident in the abandoned field systems in the upland districts of southern and southeastern Scotland. Estuarine lands brought into agriculture during the peak of the MCA were abandoned in the face of repeated marine transgression, as evidenced from the districts around the head of the Solway Firth in the southwest and the carselands of the Forth estuary in the east. Lowland districts from Ayr to Sanday and Westray in Orkney, Eldbotle in Lothian, and Forvie in Aberdeenshire experienced repeated episodes of dune deflation and sand-blow, which either inundated or stripped away the cultivated ground (Oram 2014b, 205). In each case, this arose from a disastrous interplay of cultivation of calcareous shell-sand zones, stripping of turves from the dune systems for anthrosol formation in garden plots, animal action, and strengthening of westerly winds attendant on the negative NAO (Oram 2014b, 206‑211). Human and natural agency delivered a veritable ‘perfect storm’. Conclusions

When expressed in direct human experiential terms rather than as a catalogue of symptoms of environmental stress, the impact of long-term climate change on the seasonal

upland regimes upon which much of Highland Scotland’s population subsisted was devastating. Even against the backdrop of plague mortality and still-falling population levels from the middle of the 14th century, the continued downward climatic trend meant that little slack was created in a system that had been poised on a subsistence knife edge since before the end of the MCA. It is in this context that we should see the record of increased resort to legitimate and illegitimate violence in later medieval Scotland. This phenomenon is characterised or, indeed, caricatured by the rise of the militarised Highland kindreds with their culture of conspicuous consumption, predation, ritualised raiding, and mercenary service. It is evident also in the political factionalism of the non-Highland nobility of the 15th century, much of which had competition for resources at its root (Oram 2014c; Oram – Adderley 2008, 2010). Warfare and political instability at national level exacerbated the difficulties, but these were more symptoms of wider socioeconomic dislocation than a cause. In short, people were moving into new relationships of dependency and power, often expressing those relationships in the language of continuity from a near-mythical Golden Age as they negotiated their way through an era of profound environmental transformation. And this is perhaps the greatest significance in closer examination of the Scottish experience, the fact that Scotland as a polity endured despite the multiplicity of insults to its socioeconomic metabolism. For very understandable methodological and opportunistic reasons, much of the palaeoenvironmental and environmental historical research effort of the last three decades has focused on Greenland and Iceland as laboratories for how complex societies were affected by catastrophic climate change. The historical experiences of those lands, it is suggested, provide insight on likely consequences of current climate change in terms of both environmental and societal impacts. Indeed, they do; but those societies failed or survived only through reduction to dependency at the margins of an external provider, Norway. Medieval Scotland, however, proved to be resilient, but with resilience won at significant cost, including successive episodes of profound social reconfiguration in which social dislocation and violence played powerful roles. Yes, Scotland had access to a broader suite of resources and, in comparison to the extremes of climatic deterioration experienced in 14th- and 15th-century Greenland, suffered far less dramatic decline in climate and weather conditions, but what differentiates it from the northern Atlantic territories is the record that survives for responses to the palpably negative changes that it endured. At present, however, our understanding of the motors driving change and how those changes were experienced is limited still to headline statements and mainly related to locations on the geographical margins of the medieval kingdom. Oram

29

This skewed national picture has the positive value of elucidating the mechanisms for the transformations evident in the archaeological and palaeoenvironmental record of regions otherwise lacking in the documentary sources that elsewhere drive the change narrative. Any epiphany in understanding of the social, economic, cultural, and, probably, political reconfigurations occasioned in medieval Scotland by the distressed condition of its rural regimes and especially their seasonal upland constituents, however, is yet to be expressed in human rather than more-abstract environmental terms. The data exist; we now require the awakening of the imagination of most Scottish archaeologists and historians to interpret their meaning and translate them into terms accessible to non-specialist audiences. References

History Scotland 2014: Early medieval coastal site discovered at Firth of Clyde – Scottish Archaeology. History Scotland. Published online at https://www.historyscotland. com/history/early-medieval-coastal-site-discoveredat-firth-of-clyde-scottish/ (accessed 22. 01. 2020). Atkinson, J. 2016: Ben Lawers: an archaeological landscape in time. Results from the Ben Lawers historic landscape project, 1996‑2005. Scottish Archaeological Internet Report 62. Published online at https:// archaeologydataservice.ac.uk/library/browse/issue. x h t m l ? r e c o r d I d = 1 1 3 7 4 9 5 & r e c o r d Ty p e = MonographSeries (accessed 16. 01. 2020). Beament, E. 2020: Climate scientists confirm 2010s was hottest decade recorded, The Independent Online, 15 January 2020. Published online at http://independent.co.uk/ environment/climate-crisis-decade-warmest-scienceweather-el-nino-a9284746.html (accessed 16. 01. 2020). Brown, K. – MacIntosh, G.H. – Mann, A.J. – Ritchie, P.E. – Tanner, R. J. 2007‑2020: The records of the Parliaments of Scotland to 1707. University of St Andrews. St Andrews. Published online at http://www.rps.ac.uk. Brown, T. – Crane, S. – O’Sullivan, D. – Walsh, K. – Young, R. 1998: Marginality, multiple estates and environmental change: the case of Lindisfarne, in: Mills, C. M. – Coles, G. (eds.), Life on the edge. Human settlement and marginality. Oxbow Monograph 100. Oxbow Books. Oxford, 139‑148. Cage, A.G. – Austin, W.E.N. 2010: Marine climate variability during the last millennium: The Loch Sunart record, Scotland, UK, Quaternary Science Reviews 29, 1633‑1647. 30

Canmore 2020: Canmore, Paisley, Place of Paisley. Published online at https://canmore.org.uk/site/43140/paisleyplace-of-paisley (accessed 22. 01. 2020). Charman, D.J. – Caseldine, C. – Baker, A. – Gearey, B. – Hatton, J. – Proctor, C. 2001: Paleohydrological records from peat profiles and speleothems in Sutherland, Northwest Scotland, Quaternary Research 55 (2), 223‑234. Dawson, S. – Dawson, A.G. – Jordan, A.T. 2011: North Atlantic climate change and Late Holocene windstorm activity in the Outer Hebrides, Scotland, in: Griffiths, D. – Ashmore, P. (eds.), Aeolian archaeology: the archaeology of sand landscapes in Scotland, Scottish Archaeological Internet Report 48. Published online at http://sair.org.uk/sair48 (accessed 16. 01. 2020). Dixon, P. 2007: Hunting, summer grazing and settlement: competing land use in the uplands of Scotland, Ruralia VII, 27‑46. Gilbertson, D.D. – Gilbertson, D. – Schwenninger, J. – Kemp, R. – Rhodes, E. 1999: Sand-drift and soil formation along an exposed North Atlantic coastline: 14,000 years of diverse geomorphological climatic and human impacts, Journal of Archaeological Science 26, 439‑469. Halstead, P. – O’Shea, J. (eds.) 2004: Bad year economics: cultural responses to risk and uncertainty. New directions in archaeology, new ed. Cambridge University Press. Cambridge. Mills, C. 2008: Historic pine and dendrochronology in Scotland, Scottish Woodland History Discussion Group: Notes XIII, 9‑14. Mills, C.M. – Crone, A. 2012: Dendrochronological evidence for Scotland’s native timber resources over the last 1000 years, Scottish Forestry 64:1, 18‑33. Oram, R.D. 2014a: ‘The worst disaster suffered by the people of Scotland in recorded history’: climate change, dearth and pathogens in the long 14th century, Proceedings of the Society of Antiquaries of Scotland 144, 223‑244. Oram, R.D. 2014b: From ‘Golden Age’ to depression: land, lordship and environmental change in the medieval Earldom of Orkney, in: Gulløv, H. C. (ed.), Northern worlds – landscapes, interactions and dynamics. Studies in Archaeology & History Vol. 22. Publications from the National Museum. Copenhagen, 203‑214. Oram, R.D. 2014c: Between a rock and a hard place: climate, weather and the rise of the Lordship of the Isles, in: Oram,

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

R. (ed.), The Lordship of the Isles. Brill. Leiden, 40‑61. Oram, R.D. – Adderley, P. 2008: Lordship and environmental change in central Highland Scotland c.1300 to c.1450, Journal of the North Atlantic 1, 74‑84. Oram, R. – Adderley, P. 2010: Lordship, land and environmental change in West Highland and Hebridean Scotland c.1300-c.1450, in: Cavaciocchi, S. (ed.), Economic and biological interactions in pre-Industrial Europe from the 13th to the 18th centuries. University of Florence Press. Florence, 257‑268. Rogers, C. (ed.) 1880: Rental Book of the Cistercian Abbey of CuparAngus. Volume II. Grampian Club. London. Ross, A. 2006: Scottish environmental history and the (mis)use of Soums, Agricultural History Review 54, 213‑138. Ross, A. 2015: Land assessment and lordship in medieval northern Scotland. The Medieval Countryside 14. Brepols. Turnhout. Rydval, M. – Loader, N.J. – Gunnarson, B.E. – Druckenbrod, D.L. – Linderholm, H.W. – Moreton, S.G. – Wood, C.V. – Wilson, R. 2017: Reconstructing 800 years of summer temperatures in Scotland from tree rings, Climate Dynamics 49, 2951‑2974. Slavin, P. 2010: The fifth rider of the Apocalypse: the Great Cattle Plague in England and Wales and its economic consequences, in: Cavaciocchi, S. (ed.), Economic and biological interactions in pre-Industrial Europe from the 13th to the 18th centuries. Florence University Press. Florence, 165‑179.

Slavin, P. 2019: Experiencing famine in fourteenth century Britain. Environmental histories of the North Atlantic world 4. Brepols. Turnhout. Stott, K J. – Austin, W.E.N. – Sayer, M.D.J. – Weidman, C.R. – Cage, A.G. – Wilson, R.J.S. 2010: The potential of Arctica islandica growth records to reconstruct coastal climate in north west Scotland, UK, Quaternary Science Reviews 29, 1602‑1613. Strachan, D. – Sneddon, D. – Tipping, R. 2019: Early medieval settlement in Upland Perthshire: excavations at Lair, Glen Shee 2012‑17. Archaeopress Publishing Ltd. Oxford. Stuart, J. – Burnett, G. – McNeill, G.P. (eds.) 1878‑1908 The Exchequer Rolls of Scotland, 1326‑1600. 23 vols. H.M. General Register House. Edinburgh. Tipping, R. 1998: Cereal cultivation on the Anglo-Scottish border during the ‘Little Ice Age’, in: Mills, C.M. – Coles, G. (eds.), Life on the edge. Human settlement and marginality. Oxbow Monograph 100. Oxbow Books. Oxford, 1‑11. Tipping, R. 2004: Palaeoecology and political history; evaluating driving forces in historic landscape change in southern Scotland, in: Whyte, I.D. – Winchester, A.J.L. (eds.), Society, landscape and environment in Upland Britain. Society for Landscape Studies supplementary series 2. The Society for Landscape Studies, 11‑20. Wilson, R. – Loader, N.J. – Rydval, M. – Patton, H. – Frith, A. – Mills, C.M. – Crone, A. – Edwards, C. – Larsson, L. – Gunnarson, B.E. 2011: Reconstructing Holocene climate from tree rings: the potential for a long chronology from the Scottish Highlands, The Holocene 22:1, 3‑11.

Oram

31

Section Two

Seasonal Settlement in Southern Europe

Archaeological research on seasonal settlements in the Iberian Peninsula – an overview Catarina Tente* and Margarita Fernández Mier**

Abstract

The archaeology of seasonal occupation and its structures is not a subject that has dominated medieval research in southwestern Europe. Some work has been done, mostly linked to multidisciplinary and multi-chronological teams, which have been focused on mountain areas. More recently, projects specifically addressing the study of medieval occupation in mountain areas have been developed and some results have now been published. Other research has studied Islamic contexts related to seasonal fishing. The role played by preventive archaeology must also be stressed, particularly in the characterisation of ethnographic and archaeological structures. Therefore, this paper presents the state of the art of seasonal settlement research in the Iberian Peninsula, approaching the currently available data on seasonal settlement in this part of Europe. It also presents the authors’ perspective on research in the near future. Keywords: Iberian Peninsula, seasonal settlement, ethnography, archaeology. Résumé

* Instituto de Estudos Medievais – Universidade Nova de Lisboa Av. Berna 26C, 1069‑061 Lisboa, Portugal [email protected] ** Universidad de Oviedo, Campus de Humanidades ‘El Milán’ C/ Amparo Pedregal, s/n, 33011 Oviedo, Asturias, Spain [email protected]

Recherche archéologique sur les occupations saisonniers dans la Péninsule Ibérique – un aperçu L’archéologie des occupations saisonnières et ses structures n’est pas un sujet ayant dominé la recherche médiévale dans le sud-ouest de l’Europe. Pourtant, il y a eu des efforts, principalement liés à des équipes multidisciplinaires et multi-chronologiques, qui se sont concentrées dans les zones de montagne. Plus récemment, certains projets spécifiquement consacrés à l’étude de l’occupation médiévale dans les zones de montagne ont été développés et certains résultats ont été mis en lumière. Il y a eu des recherches effectuées dans certains contextes islamiques liés à la pêche saisonnière. Il faut aussi souligner les résultats issus de l’archéologie préventive, qui donnent une nouvelle vue sur notamment la caractérisation des structures ethnographiques et archéologiques. Par conséquent, cet article présente un état de la recherche relative au peuplement saisonnier dans la péninsule ibérique, en s’appuyant sur les données actuellement disponibles dans cette partie de l’Europe. L’objectif des auteurs se focalise également sur la perspective de la recherche future. Mots-clés : Péninsule ibérique, peuplement saisonnier, ethnographie, archéologie. In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 35-44.

35

Zusammenfassung

Archäologische Untersuchungen zu saisonalen Siedlungen auf der Iberischen Halbinsel – ein Überblick Die archäologische Erforschung saisonaler Berufe und ihrer Strukturen ist kein Thema, welches die mittelalterliche Forschung in Südwesteuropa dominiert hat. Es wurden jedoch einige Arbeiten durchgeführt, die sich hauptsächlich auf multidisziplinäre und multichronologische Forschungen beziehen und sich auf Bergregionen konzentrieren. In jüngerer Zeit wurden einige Projekte speziell zur Erforschung der mittelalterlichen Besiedlung in Bergregionen entwickelt und einige Ergebnisse konnten gewonnen werden. In

Introduction

Research on Iberian seasonal settlements has fundamentally been focused on mountainous regions – particularly in the mountain ranges of the centre and north of the peninsula – and thus dominated by topics related to pastoral activities. Therefore, the research domains that have approached seasonal settlement – in which, as we will see below, medieval archaeology has played a minor role – have been fundamentally threefold: the history of transhumance, in particular of the Spanish Mesta; ethnographic studies on built structures and economy; and, in the field of archaeology, livestock management, mainly in high mountain areas. However, these studies have been scarce in the case of medieval archaeology, because greater emphasis has been laid on the prehistoric and Roman periods. Additionally, environmental impact assessments and rescue archaeology programmes have permitted the surveying of large territories, which has enabled the identification of several types of structures of all time periods, some of which were also subjected to excavation. However, work of this kind is by definition devoid of a coherent research framework. In the particular case of Spain, studies on the long-range mobility of herds in the Middle Ages and modern period were naturally dominated by the Mesta. The Mesta was an institution that was in charge of the regular transhumance of sheep and goats in the central areas of the peninsula. Several types of built structures (pens, fences, trails, drinking structures, etc.) can be related with this important economic activity. However, among the countless studies already conducted, only few have used archaeological data (Gerbet 1991; Pascua Echegaray 2007). Ethnographic research has been more uneven. Clearly, the work by the Escuela Vasca de Etnografía, which initiated its research activity on the Basque communities in the 1920s under the direction of J.M. Barandiarán (1927, 1935), stands out, as well as the work on ethnohistory developed by J. Caro Baroja 36

einigen islamischen Kontexten wird im Zusammenhang mit der saisonalen Fischerei geforscht. Hervorzuheben ist auch die Rolle der präventiven Archäologie, insbesondere bei der Charakterisierung ethnografischer und archäologischer Strukturen. In diesem Aufsatz wird daher ein aktueller Stand der saisonalen Siedlungsforschung auf der Iberischen Halbinsel vorgestellt, ausgehend von derzeit verfügbaren Daten zur saisonalen Siedlung in diesem Teil Europas. Ziel ist es auch, die Perspektive der zukünftigen Forschung zu diskutieren. Schlagwörter: Iberische Halbinsel, saisonale Abrechnung, Ethnographie, Archäologie.

(1972), both precedents of the current Basque social anthropology (Homobono 1992). For the Pyrenees, particular note should be given to the work by J. Costa (1888, 1902), who did an important study of customary law that is still up to date. Also noteworthy is the contribution of the Hamburg School, namely through the studies on language and traditional culture carried out by F. Krüger (1936), the research by R. Violant i Simorra (1949) on ethnographic artefacts, and the work by J. Amades i Gelats (1931, 1950‑1956) on Catalonian traditional culture. A wide number of topics were approached, particularly the typical pastoral buildings with vegetal roofing, an ethnographic feature of great importance in the Cantabrian mountains (Graña García – López Álvarez 1990, 2007). Since the 1990s several inventories have been made, owing to the abandonment of rural areas. More recently, attention has also been paid to all kinds of livestock buildings and associated material culture. These are, however, dispersed, small-scale studies that do not correspond to planned inventories, but instead to particular initiatives reflecting an increasing interest in disappearing herding activities (Mateo Pérez – Orduña Portús 2017). Portuguese research is somewhat different regarding the material culture of the rural world. Between the end of the 19th century and the 1970s there was a very important school of ethnography that gathered data on rural communities exhaustively all over the country (Oliveira – Galhano 1992; Oliveira et al. 1969). The collection of objects, and even complete structures, also took place and these can be found today in the Museu Nacional de Etnologia in Lisbon. Most of the work was published, thus providing a unique account of this specific moment in Portuguese history when the abandonment of more-traditional rural lifeways began. The main studies on transhumance in Portugal were carried by the geographer and ethnographer O. Ribeiro, who studied this topic systematically (Ribeiro 1939, 1941,

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

1948). Among many other aspects, he charted traditional transhumance routes, distinguished different mobility ranges, and characterised the composition of flocks and the communities and shepherds involved. He also made detailed descriptions and graphic/photographic records of objects, structures, social features, festivities, and so on. Although his perspective was mostly ethnographic, he was able to give it a historical dimension, using written sources to seek the origins of transhumance and changes through time. His most important research region was the Estrela mountain range, in the centralnorthern region of the country (see below), where the most particular traditional transhumance system in Portugal existed until the 1960s (Ribeiro 1941). The important ethnographic and geographical contributions both in Spain and Portugal mentioned above were not accompanied by comparable archaeological research on the same topics. The traditional marginality of montane territories was one of the obstacles to the development of archaeological research. Moreover, medieval archaeology, in particular, has long focused on monumental remains, such as castles or churches, and has not paid enough attention to coeval agrarian structures and lifeways. Due to the fact that some seasonal structures – some of them with prehistoric origins – were in use until as recently as the 1960s, it was not assumed that they could be studied by archaeology. This preconception constituted another research obstacle. Therefore, most of the archaeological studies in montane environments were aimed at older sites, favouring the study of the earliest permanent settlement and agrarian structures, thus referring mainly to the Neolithic, protohistoric, and Roman times, not the Middle Ages. One exception is the research carried out in the area of Granada, in the Spanish province of Andalusia, on the Islamic occupation of Sierra Nevada, though it was heavily focused on land use strategies and management of water (Ruiz Ruiz – Martín Civantos 2017). In both countries, environmental impact assessments and recue archaeology have been permitted seasonal settlements to be identified. This was the case with Ponta do Castelo (Fig. 2), on the western coast of Algarve (Portugal), where an Islamic seasonal fishing site dated to the 12th century was discovered (Gomes et al. 2001). The complete study of this settlement, however, is not yet published and its associated alcaria (Islamic village), where the community was permanently settled, is presently unknown. Other coeval fishing settlements have also been identified in Portugal in recent years, but the lack of systematic research programmes prevents their adequate characterisation and interpretation. The same is true of several seasonal structures also recognised in environmental impact assessments, such as those that took place in the Suído mountain range, in Galicia (Ballesteros Arias 2004). In most cases only inventories

were made, without a study of their relation with the surrounding landscapes. On seasonal settlements

The map in Fig. 1 shows the distribution of the main seasonal sites excavated or studied in the Iberian Peninsula. Despite the presence of fishing sites, it is clear that the large majority of the sites are located in montane environments. The main references in Spain for the study of seasonal settlements are the research groups working in the Pyrenees. The DEPART project, coordinated by C. Rendu (Toulouse University, France) and E. Gassiot (Universidad Autónoma de Barcelona, Spain), with the collaboration of M. Le Couédic (Université de Pau, France), I. Clemente Conte (Spanish National Research Council), and A. Fortó (Goberment of Andorra, designed a GIS for the inventory of all archaeological heritage sites and the modelling of pastoral territories (Gassiot 2016; Gassiot Ballbé – Pèlachs Mañosa 2017; Le Couédic et al. 2015; Rendu 2016). Intensive archaeological fieldwork on all types of structures has allowed a long-term occupation to be traced in the region, from prehistory to the present, with there being periods of greater impact on the landscape (such as the Neolithic and Roman periods) along with gaps in the occupation sequence (e.g. the Bronze Age). In addition, in the Pyrenees the landscape archaeology research group from ICAD (Institut Catalá d’Arqueología Clásica, Spain) has developed a research project on the high mountain areas of its eastern sectors. Although their main research focus is the Roman occupation, it has been possible to document the exploitation of these areas since Neolithic times, as elsewhere in the mountains (Palet i Martínez et al. 2017) (Fig. 3). These studies have assembled abundant datasets on the medieval era. However, this particular period has not been subject to detailed analyses, because most of the work was carried out by researchers with a greater interest in prehistory. But since there have been some PhD theses on more-recent periods, it is hoped that attention to the medieval occupation will be renewed in the near future (Garcia Casas 2018). In fact, the greatest volume of information in the Pyrenees has been provided by palaeoenvironmental studies. The GEODE research group, directed by Didier Galop, focuses on the long-term co-evolution between environment and societies. These studies are based on an interdisciplinary retro-observation of socio-ecological processes that combines multi-proxy analysis of palaeorecords (pollen, microfossils, stomata, microcharcoals, etc.) with eco-historical studies (written sources, ancient maps, etc.) that documented an intense deforestation process that took place throughout the High Middle Ages associated with the building of livestock structures (Lerigoleur et al. Tente and Fernández Mier

37

Fig. 1. Distribution of rural medieval sites studied archaeologically in Iberia and location of regional areas mentioned in text: white dots – fishing settlements; black dots – agro-pastoral settlements (© Margarita Mier Fernández and Catarina Tente).

2015; Pérez-Díaz et al. 2015). This phenomenon is well attested, both in the central-eastern Pyrenees (Galop 1998; Rendu 2003) and in the western sector of the range (Mazier et al. 2009), although it developed at non-coincident chronological rates (Galop et al. 2013). In the Cantabrian Cordillera (northern Spain), a team is working on the organisation of village territories and communal spaces (Fig. 4). Through the archaeological excavation of habitation sites and areas of agricultural and livestock use, it has been possible to define the main chronological phases underlying the creation of landscapes (Fernández Mier – López Gómez 2021). More recently, within the theoretical and methodological framework of the systematic work carried out in the Alps and the Pyrenees, a new project was started on the human impact on the landscape seen from a long-term perspective. Although it has focused on prehistoric times, this project has been able to identify occupations dating 38

up to the present and therefore also including the Middle Ages (González Álvarez 2019). Interesting research was also carried out at the beginning of the present century in the Sierra de Aralar, in the Basque Country. This was able to observe the uninterrupted occupation of this mountain range since the Neolithic. In this case, the study of several seasonal settlements, together with palaeoenvironmental studies, has already enabled some hypotheses to be put forward about seasonal settlements during the High Middle Ages. Here, several seasonal settlements were occupied in the 5th and 6th centuries. These consisted of cabins that were rebuilt over several generations. During the 7th and 8th centuries, the greater intensity of livestock occupation coincided with the formation of the villages in the lowlands (Moraza – Mujika 2005; Agirre et al. 2008a, 2008b; Mujika et al. 2013). Obviously, both phenomena are closely linked, because they were part of the same

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 2. Fishing settlement of Ponta do Castelo (Aljezur), dated from the 12th- 13th centuries (© Catarina Tente).

Fig. 3. Montane site of Coma de Vaca (Queralbs, Pyrenees), dated from the Roman period (© Palet i Martínez et al. 2017).

Tente and Fernández Mier

39

Fig. 4. Montane site of Andrúas (Asturias, Cantabrian Cordillera), dating to the 12th century (© Pablo López Gómez).

Fig. 5. Montane site of Lameiros Tapados (Vouzela, Caramulo mountain range), datingto the 10th century (© Catarina Tente).

broad economic system based on a close integration of intensive herding and farming in a regime of shortdistance transhumance (Narbarte Hernández 2019). In addition, zooarchaeological datasets show significant changes in animal exploitation techniques. The most striking phenomenon is, without a doubt, the reduction in sizes of the three main domestic species (cattle, sheep, and goat) documented from the 8th century onwards (Grau Sologestoa 2014; Sirignano et al. 2014). However, it is not known if this greater intensity in the use of seasonal settlements corresponds simply to an increase in the use of montane areas, or if there were also changes in the ways pastures and livestock were managed (Fernández Mier – Quirós Castillo 2015; Stagno et al. 2020). In Galicia (northwest Spain) two teams are currently addressing the topic of seasonal settlements. The Arqueología del Paisaje group from the University of 40

Santiago de Compostela – later on integrated in the INCIPIT attached to the CSIC – launched a project for the study of the landscape in its long-term environmental, social, symbolic, and formal dimensions. This project also responded to the need of designing a strategy to address the environmental impact of extensive infrastructures in Galicia, such as gas distribution networks (Criado-Boado 1991). In this context, the investigation of all kinds of elements related to agricultural (e.g. terraces, cultivation fields) and livestock (e.g. delimitation ditches) practices has been undertaken. Also of interest are the structures for livestock associated with transhumant practices that are object of analysis from the ethnographic and archaeological perspective (Ballesteros Arias 2004). More recently, research conducted by a different team in the Serra de Barbanza (Barbeito Pose et al. 2015), a coastal mountain range, allowed the discovery of a few seasonal

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

settlements, which were attributed to the Middle Ages after the radiocarbon dating of two of them. One is Porto Traveso, where a hearth was dated to the 9th century, and the other is Río Barbanza, where a sequence spanning from the Late Bronze Age to the 9th century was discovered. The inventory and excavation of a number of structures showed the continuous use of these sites and territories for millennia, although the relationship that these settlements may have with seasonal habitats has not yet been described in detail. The chronologies in Serra de Barbanza are younger than those in the Pyrenees and the Basque Country, but the Galician research is still in an initial stage. Peneda-Gerês (northern Portugal) is the only montane area in Portugal where short-/medium-distance cattle transhumance is well documented, both historically and ethnographically. This movement implied the existence of two main seasonal sites, the so-called brandas and inverneiras that correspond to summer and winter settlements, respectively. Here it was the whole community who moved across the landscape. Agriculture was carried out in both types of settlements. The most striking aspect of the Peneda-Gerês area is that livestock mainly consisted of cattle, thus sharply contrasting with evidence from the other places in Portugal (Ribeiro 1939; Lima 1996). However, historical sources allow us to go back only as far as the 17th century, and archaeological research has never been carried out on a systematic basis allowing a deeper exploration of older manifestations of this particular type of transhumance in the area. The Estrela mountain range (centre-north Portugal) is surely the best-studied case of transhumance in Portugal. Indeed, this has been a focus of research since the mid-19th century. Here an important set of historical data on sheep and goat herding goes back to the 15th century (Ribeiro 1941) although archaeology and pollen records show that it began much earlier, probably as far back as the Early Middle Ages, if not before (Tente 2010; Carvalho – Tente forthcoming). Available ethnographic evidence and historical sources indicate that sheep/goat transhumance in the Estrela range was carried out by shepherds only, that is, individuals who specialised in the herding tasks. However, recent archaeological research suggests that a regional settlement system similar to that observed at Peneda-Gerês could have existed around the 10th century (Fernández Mier – Tente 2018). The main difference from the more-recent practices, in terms of site structures, is that the medieval sites were fixed in the landscape: the settlements where seasonal huts were rebuilt every summer were surrounded by thick stone walls. Another contrast is that the modern practices were carried out by small teams of shepherds who used mobile, transportable structures (impossible to identify archaeologically) that did not involve the building of permanent, stone structures.

In the last four years a project has been underway in the Caramulo mountain range, also in centralnorthern Portugal, with a research topic aimed at the study of settlement mobility in the area. Although obtained recently and still limited, preliminary results seem to indicate that around the 10th-11th centuries vertical mobility encompassed lowland and mid-altitude mountain (700‑800  m a.s.l.) seasonal sites (Fig.  5). Further excavations are however necessary for a better characterisation of this reality, which is very poorly documented, even ethnographically. The future of research

Clearly, a line of research on vernacular architecture is urgently needed to approach the study of sites/structures that have remained in use until recent times, but are currently suffering accelerated deterioration due to abandonment. With the exception of the above-mentioned project in the Pyrenees, there are no available databases where structures related to livestock practices are systematically collected. This heritage, which was in use until the mid-20th century, is essential for the understanding of human occupation of highland areas in the Iberian Peninsula since late prehistoric times. Such structures demonstrate the importance that the surrounding pastures must have held for local communities. Despite the above limitation, the available records do show us that a major research problem is the archaeological invisibility of most built structures. This is, of course, more evident in some regions than others, and depends heavily on the kind of materials that were in use and also on the processes at play in the formation of each archaeological site. Indeed, the ethnographic record shows that most mobile structures were built with perishable materials – this is particularly the case with transportable huts or cabins – that do not leave an archaeological footprint. Most of the time it is impossible to overcome this limitation, but it is still necessary to bear these facts in mind in order to understand the overall complexity of montane occupations, even when apparently absent. Therefore, future (and present) research must be focused on multidisciplinary, complementary projects specifically designed for the study of seasonal settlements and their correlation with permanent ones. Current research already provides us with a valuable track record of this kind of approach in the Iberian Peninsula, which has proved promising. Another topic for future research is the identification of archaeological markers of seasonality. This approach should be carried out according to a two-step approach: firstly, a judicious, sound definition of units of analysis (i.e. regions) must be constructed, given the acknowledged regional variability observed in Iberia; secondly, systematic comparisons need to be made between different types Tente and Fernández Mier

41

of settlements within units. The mentioned variability may be present at several levels, such as the types of used materials, the different sizes of the habitation structures, and their locations in the landscape, etc. As far as we know in the current state of the research, there is no common model applicable to the whole peninsula. Within seasonality, another research topic that has been under study in recent years will apparently be crucial in the future – this is the archaeological study of commons, a landscape feature that still exists in most of the Iberian mountains and is closely related with seasonal activities. In this research topic, a specifically orientated analysis of the (sometimes) abundant written medieval and postmedieval records should be a priority. These documents, which were usually produced by local entities in charge of the management of commons, contain qualitative information on exploitation strategies that, if crossed with archaeological data, will enable a multifaceted and more complete vision of the large variability of management processes. In some cases, it may even be possible to analyse the decision capacities of the various social actors involved (Fernández Mier – López Gómez 2021; Stagno 2017). On the other hand, it must be stressed that a regressive approach, from ethnographic to archaeological data, is clearly needed. Indeed, ethnographic studies are fundamental in this specific archaeological research, not only to help decipher the archaeological record, but also to plan field surveys and excavations. Conversely, this has to be combined with a progressive, or historical, approach that moves from prehistory towards modern times; as seen in all of the different territories, single sites may have been successively reoccupied for millennia. Finally, palaeoenvironmental studies will play a fundamental role in future research, as they enable the understanding of millennial changes in the landscape. These will contribute to a qualitative analysis of exploitation processes in pasture areas in a diachronic perspective. References

Amades i Gelats, J. 1931: Vocabulari dels pastors, Butlletí de dialectología Catalana 19, 65‑240. Amades i Gelats, J. 1952‑1956: Costumari Català. El curs de l’any. Editorial Salvat. Barcelona. Agirre, J. – Moraza, A. – Mujika, J.A. – Reparaz, X. – Telleria, E. 2008a: Primeros vestigios de un modelo económico de ganadería estacional especializada. Los fondos de cabaña tumulares de Arrubi y Esnaurreta (Aralar), Kobie, Serie Paleoantropología XXVII, 105‑131.

42

Agirre, J. – Moraza, A. – Mujika, J.A. – Reparaz, X. – Telleria, E. 2008b: La transición entre dos modelos de ganadería estacional de montaña. El fondo de cabaña pastoril de Oidui (Sierra de Aralar). Kobie, Serie Paleoantropología XXVII, 163‑190. Ballesteros Arias, P. 2004: Arquitectura tradicional ganadeira na Serra do Suido. A resposta dunha tradición, Cuadernos de Estudios Galegos 17, 9‑48. Barbeito Pose, V.J. – Fábregas Valcarce, R. – Rodriguez Rellán, C. – Blanco Chao, R. – Martín Seijo, M. – Paz Camaño, A. – Fariña Costa, A. – Gordgoso López, L. 2015: Ocupacións domésticas na Serra do Barbanza: Resultados preliminares, Gallaecia 34, 125‑158. Barandiarán, J.M. 1927: Contribución al estudio de los establecimientos humanos y zonas pastoriles del País Vasco, Anuario de Eusko-Folklore VII, 137‑141. Barandiarán, J.M. 1935: Albergues veraniegos. Trashumancia intrapirenaica, Anales del Museo del Pueblo Español 1, 88‑97. Caro Baroja, J. 1972: Los Vascos. Ediciones Akal. Madrid. Carvalho, A. F. – Tente, C. forthcoming: Orlando Ribeiro’s work on the pastoralism of centralnorthern Portugal: guidelines for a research project on ethno-history and archaeology, Historia Agraria. Costa, J. 1888: Introducción a un tratado de política textualmente de los refraneros, romanceros y gestas de la Península. Librería de Fernando Fe. Madrid. Costa, J. 1902: Derecho consuetudinario y economía popular de España. Henrich y Cª. Barcelona. Criado-Boado, F. (dir.) 1991: Arqueología del paisaje, el área Bocelo-Furelos entre los tiempos paleolíticos y medievales (campañas de 1987, 1988 y 1989). A Coruña. Xunta de Galicia. Fernández Mier, M. – López Gómez, P. in press: Archaeology of the commons: Seasonal settlements in the Cantabrian mountains, in: Theune, C. – Dixon, P. (eds.), Ruralia XIII, Seasonal settlement in the medieval and early modern countryside, Sidestone. Leiden. Fernández Mier, M. – Quirós Castillo, J.A. 2015: El aprovechamiento de los espacios comunales en el Nororoeste de la Península Ibérica entre el período romano y el medieval, Il capitale culturale XII, 695‑723. Fernández Mier, M. – Tente, C. 2018: Transhumant herding systems in Iberia, in: Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe. London, 219‑232.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Galop, D. 1998: La forêt, l’homme et le troupeau dans les Pyrénées. 6000 ans d’Histoire de l’environnement entre Garonne et Méditerranée. GEODE. Toulouse. Galop, D. – Ruis, D. – Cugny, C. – Mazier, F. 2013: A history of long-term human-environment interactions in the French Pyrenees inferred from the pollen data, in: Lozny, L.R. (ed.), Continuity and change in cultural adaptation to mountain environments studies in human ecology and adaptation. Springer. New York, 19‑30. Garcia Casas, D. 2018: Arqueologia d’un territori d’alta muntanya del pirineu central. Persones, ramats i prats al llarg de la història al Parc nacional d’aigüestortes i estany de Sant Maurici. Universitat Autonoma de Barcelona. Barcelona. Gassiot Ballbè, E. – Pèlachs Mañosa, A. 2017: La ocupación ganadera de los Pirineos occidentales de Catalunya en época romana e inicios de la Edad Media, Treballs d’Arqueologia 21, 287‑306. Gassiot Ballbè, E. (ed.) 2016: Arqueología del pastoralismo en el Parque Nacional d’Aigüestortes i Estany de Sant Maurici: montañas humanizadas. Organismo Autónomo Parque Nacionales. Madrid. Gerbet, M.C. 1991: L’élevage dans le royaume de Castille sous les Rois Catholiques (1454‑1516). Casa de Velázquez. Madrid. Gomes, R.V. – Teixeira, V. – Miranda, M.J. 2001: Povoado muçulmano na Ponta do Castelo (Aljezur), Notícia Preliminar, Al-Madan 10, 200‑201. González Álvarez, D. 2019: The need to understand the cultural biographies of alpine and subalpine landscapes during Later Prehistory: Upland Archaeology in the Cantabrian Mountains, Cuadernos de Investigación Geográfica 45 (1), 143‑165. Graña García, A. – López Álvarez, J. 1990: Las cubiertas vegetales en Asturias: notas sobre la historia de la casa redonda, in: Sánchez Gómez, L.A. – Cea Gutiérrez, A. – Fernández Montes, M. (eds.), Arquitectura popular en España, 415‑430. Graña García, A. – López Álvarez, J. 2007: Los teitos en Asturias. Un estudio sobre la arquitectura con cubierta vegetal. Red de Museos Etnográficos de Asturias. Gijón. Grau Sologestoa, I. 2014: Livestock management in Spain from Roman to post-medieval times: a biometrical analysis of cattle, sheep/goat and pig, Journal of Archaeological Science 54, 123‑134.

Homobono Martínez, J.I. 1992: Evolución del estado actual de la antropología social en el País Vasco, Anales de la Fundación Joaquín Costa 9, 147‑170. Krüger, F. 1936: Die Hochpyrenäen. Seminar f. roman. Sprachen u. Kultur. Hamburg. Le Couédic, M. – Rendu, C. – García Casas, D. – Gassiot Ballbè, E. – Calastrenc, C. – Clemente Conte, I. – Fortó, A. – Guillot, F. – Nuñes, J. – Pujol, F. – Rey Lanaspa, J. – Contamine, T. – Mazzucco, N. – Obea, L. – Quesada, M. – Rodríguez Antón, D. 2015: Comparer et modéliser les sites, les territoires et les systèmes pastoraux pyrénéens dans la diachronie: présentation et premiers résultats du projet collaboratif DEPART, in Pays pyrénéens et environnement, Congrès de la Société Ramond, Bagnères-de-Bigorre, 12‑14 juin 2015. Fédération historique de Midi-Pyrénées, Paris, 321‑338. Lerigoleur, E. – Mazier, F. – Jegou, L. – Perret, M. – Galop, D. 2015: PALEOPYR: un système d’information pour les données paléoenvironnementales nord-pyrénéennes,  Ingénierie des Systèmes d‘Information 20:3, 63‑87. Lima, A. 1996: Castro Laboreiro. Povoamento e organização de um território serrano. ICN. Braga. Mateo Pérez, M.R. – Orduna Portús, P. 2017: Construcciones de uso ganadero en las Bardenas Reales de Navarra: corrales y barreras, Cuadernos de Etnología y Etnografía de Navarra 91, 139‑186. Mazier, F. – Galop, D. – Gaillard, M.J. – Rendu, C. – Cugny, C. – Legaz, A. – Peyron, O. – Buttler, A. 2009: Multidisciplinary approach to reconstructing local pastoral activities: an example from the Pyrenean Mountains (Pays Basque). The Holocene 19, 171‑188. Moraza, A. – Mujika, J.A. 2005: Establecimientos de habitación al aire libre. Los fondos de cabaña de morfología tumular: características, proceso de formación y cronología, Veleia 22, 77‑110. Mujika Alustiza, J.A. – Agirre García, J. – Edeso Fito, J.M. – Lopetegi Galarraga, A. – Pérez Díaz, S. – Ruiz Alonso, M. – Tarriño Vinagre, A. – Yusta Arnal, I. 2013: La continuidad de la actividad pastoril durante la época romana en la zona de Argarbi (Sierra de Aralar, Gipuzkoa). Kobie, Serie Paleoantropología 32, 217‑258. Narbarte Hernández, J., 2019: Late medieval and modern settlement dynamics in three Atlantic Basque villages: an approach on the rural landscape, in: Grau-Sologestoa, I. – Quirós Castillo, J.A. (eds.), Paisaje y prácticas sociales: arqueología agraria en el País Vasco. Archaeopress. Oxford, 103‑122.

Tente and Fernández Mier

43

Oliveira, E.V. – Galhano, F. 1992: Arquitectura tradicional portuguesa. D. Quixote. Lisbon. Oliveira, E.V. – Galhano, F. – Pereira, B. 1969: Construções primitivas em Portugal. Centro de Estudos de Etnologia. Lisbon. Palet i Martínez, J.M. – García i Molsosa, A. – Orengo Romeu, H.A. – Polonio Alamino, T. 2017: Els espais altimontans pirenaics orientals a l’Antiguitat: 10 anys d’estudis en arqueologia del paisatge del GIAP-ICAC, Treballs d’Arqueologia 21, 77‑97. Pascua Echegaray, P. 2007: Las otras comunidades: pastores y ganaderos en la Castilla medieval, in: Rodríguez, A. (ed.), El lugar del campesino. En torno a la obra de Reyna Pastor. Universidad de Valencia. Valencia, 209‑237. Pérez-Díaz, S. – López-Saez, J.A. – Galop D. 2015: Vegetation dynamics and human activity in the Western Pyrenean region during the Holocene, Quaternary International 364, 65‑77. Rendu, C. 2003: La montagne d’Enveig. Una estive pirènnéene dans la longue durée. Perpignan. Rendu, C. (ed.) 2016: Estives d’Ossau; 7000 ans de pastoralisme dans les Pyrénées. Le Pas d’oiseau éditions. Toulouse. Ribeiro, O. 1939: Brandas e inverneiras em Castro Laboreiro, Revista da Faculdade de Ciências VI:1‑2, 297‑301. Ribeiro, O. 1941: Contribuição para o estudo do pastoreio na Serra da Estrêla, Revista da Faculdade de Letras de Lisboa IV:1‑2, 213‑303. Ribeiro, O. 1948: Notícia do pastoreio na Serra do Montemuro. Miscelânea de Estudos à Memória de Cláudio Basto. Porto, 333‑339.

44

Ribeiro, O. – Santos, M.A.P. 1951 Montanhas pastoris de Portugal. Tentativa de representação cartográfica. Compte Rendu du XVIe Congrès International de Géographie. Travaux de la Section IV, vol. III. Union Géographique International. Lisbon, 59‑69. Ruiz Ruiz, J.F. – Martín Civantos, J.M. 2017: La gestión comunitaria del agua en la cara norte de Sierra Nevada: acción colectiva y saberes etnoecológicos en los sistemas de riego de origen andalusí E-rph, Revista Electrónica de Patrimonio Histórico 20, 76‑103. Sirignano, C. – Grau Sologestoa, I. – Ricci, P. – GarcíaCollado, M.I. – Altieri, M. – Quirós Castillo, J.A. – Lubritto, C. 2014: Animal husbandry during Early and High Middle Ages in the Basque Country, Quaternary International 346, 138‑148. Stagno, A. 2017: Archeologia e storia di uno spazio precario. Le colture temporanee tra pratiche collettive e appropriazione privata (Paesi Baschi,  XV-XX secolo), Quaderni storici 2, 499‑536. Stagno, A. – Tejerizo-García, C. – Echazarreta Gallego, A. – Santeramo, R. – Portillo, M. – Pescini, V. – Hernández Beloqui, B. 2020: De montes comunes y sociedades campesinas. Los resultados del proyecto ARCHIMEDE en el País Vasco. In: Grau Sologestoa, I. – Quirós Castillo, J.A. (eds.), Arqueología de la edad moderna en el País Vasco y su entorno. Access Archaeology. Oxford, 165‑181. Tente, C. 2010: Arqueologia Medieval Cristã no Alto Mondego. Ocupação e exploração do território nos séculos V a XI. PhD thesis presented to FCSH-UNL, Lisbon. Violant i Simorra, R. 1949: El Pirineo español, Plus Ultra. Madrid.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Early medieval seasonal settlement and vertical transhumance in an agricultural landscape in Ainet, East Tyrol, Austria Elisabeth Waldhart* and Harald Stadler**

Abstract

In Alpine regions, seasonal settlement is closely related to vertical transhumance and various forms of Alpine pasture management. In addition, ores and other resources such as rock crystal, medicinal plants, and antlers can also be collected during a temporary stay in the high mountains. During the course of archaeological surveys and several excavation projects in the East Tyrolean Isel Valley, structures and finds with a wide temporal distribution were discovered. However, there was a striking gap in the finds material for the Early Middle Ages from the 5th to the 11th century. For other high Alpine sites in the Eastern Alps, a decline in settlement structures, especially from the 5th to the 7th century, is evident. Based on an examination of the landscape, individual stray finds, field names, and the historical development of the area under investigation, this paper will discuss preliminary hypotheses about seasonal use in these areas and the shifting settlements and patterns of use. Further investigations ought to focus on the forested areas at medium altitudes. Since seasonal pasture farming cannot be considered as isolated from permanent settlements and agriculture, changes in this context must also be included. The possible drivers of these shifts (climate, politics, or migration) are still to be determined. Keywords: Alpine farming, Eastern Alps, landscape archaeology, Early Middle Ages, Slavic period, vertical transhumance, field-name research.

* Schießstandgasse 2e/12 6020 Innsbruck, Austria [email protected] ** Institut für Archäologien Langer Weg 11 6020 Innsbruck, Austria [email protected]

Résumé

Implantation saisonnière et transhumance verticale au début du Moyen-Âge dans un paysage agricole à Ainet, Tyrol oriental, Autriche Dans les régions alpines, les migrations saisonnières vont de pair avec une transhumance verticale et différentes formes d’économie alpestre. L’exploitation d’autres ressources, comme les minerais, plantes médicinales où des ramures, peut également être faite dans le cadre d’un séjour temporaire en haute montagne. Au cours d’un relevé et des fouilles archéologiques dans la vallée de l’Isel, des structures et des objets datant d’une large In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 45-56.

45

période ont été découverts, avec une absence remarquable d’éléments du haut Moyen-Âge, soit entre le Ve et le XIe siècle. Dans d’autres sites archéologiques des Alpes orientales, on remarque un déclin respectivement une interruption des structures d’implantation entre le Ve et le VIIe siècle. L’examen des paysages, des découvertes isolées reconnues, de la toponymie et des moments clés historiques liés au site étudié peut inciter aux premières réflexions sur l’utilisation saisonnière et le déplacement des structures d’habitation et d’exploitation. La surface forestière aux altitudes moyennes doit être prise en compte pour la suite des recherches étant donné que l’économie pacagère saisonnière ne peut être considérée comme totalement séparée des implantations permanentes et de l’agriculture, des changements dans ce contexte (climat, politique, migration) doivent être envisagés pour le travail de recherche. Mots-clés : économie alpestre, Alpes orientales, archéologie du paysage, Haut Moyen-Âge, transhumance vertical, époque slave, recherche par nom de domaine / toponymie. Zusammenfassung

Saisonale Siedlung und vertikale Transhumanz im Frühmittelalter in einer landwirtschaftlich geprägten Landschaft in Ainet, Osttirol, Österreich In alpinen Regionen ist saisonales Siedeln eng verknüpft mit vertikaler Transhumanz und

Introduction

Seasonal settlement in Alpine regions is embedded in practices of vertical transhumance, meaning high Alpine regions are exploited seasonally (for definitions, see Costello – Svensson 2018, 3‑6). This can happen due to different types of summer pasturing, but can also result from other agricultural uses of high-altitude meadows, such as haymaking, or mining activities and the extraction of other raw materials. During a small-scale survey project in the East Tyrolean Isel Valley, structures and finds as well as historical sources with a wide diachronic range were detected, but with a gap in the Early Middle Ages. This raises questions of the (dis)continuity of this economic practice during the early medieval period between the 5th and 13th century. In order to investigate seasonal settlement in this period, the very limited number of stray finds and data from the 9th and 10th centuries are brought into context with place names, historical sources, and settlement structures.

46

verschiedenen Formen von Almwirtschaft. Neben Erzen kann auch die Nutzung von anderen Ressourcen wie Bergkristall, Heilpflanzen oder Geweihstangen während eines zeitlich begrenzten Aufenthaltes im Hochgebirge erfolgen. Im Rahmen eines Survey- und Grabungsprojektes im Osttiroler Iseltal wurden Strukturen und Funde mit einer weiten diachronen Streuung entdeckt, wobei ein Hiatus im Fundmaterial für das frühe Mittelalter vom 5. bis zum 11. Jahrhundert auffällig ist. Analog zu weiteren bekannten hochalpinen Fundplätzen in den Ostalpen zeigt sich insbesondere für das 5. bis 7. Jahrhundert ein Rückgang, beziehungsweise ein Umbruch in den Siedlungsstrukturen. Anhand einer Landschaftsanalyse, Fundtopographie, Flurnamenerhebung und historischer Quellen im Untersuchungsgebiet können nun erste Überlegungen zur saisonalen Nutzung und sich verschiebenden Siedlungsstrukturen angestellt werden. In den Fokus zukünftiger Untersuchungen rückt das bewaldete Gebiet in den unteren und mittleren Höhenlagen, da saisonale Hochlagenwirtschaft nicht losgelöst von den permanenten Siedlungen und der Agrarwirtschaft im Tal betrachtet werden kann. Die möglichen Auslöser dieser Verschiebungen (Klima, Politik, Migration) sind noch zu diskutieren. Schlagwörter: Alpine Landwirtschaft, Ostalpen, Landschaftsarchäologie, Frühmittelalter, Slawen, vertikale Transhumanz, Ortsnamenforschung.

Early medieval vertical transhumance in the Eastern Alps

To put it simply, two main strategies of pastoralism are considered to have taken place in the Alps. One focused on flocks of sheep, often driven over longer distances, and involved the production of wool and meat. This type of transhumance often took place on the fringes of the Alps. The other, known as Almwirtschaft, Alpwirtschaft, or malghe, was embedded in a system of Alpine agriculture in which high Alpine pastures were linked in various ways with permanently settled farms on the valley floors. In this system, the emphasis was on the production of dairy products and meat, and the livestock consisted of cattle, sheep, and goats in various combinations (Costello – Svensson 2018, 3‑6; Reitmaier 2017, 8‑10). Archaeologically speaking, each of these usage systems of high Alpine pastures is considered to have left different traces. Almwirtschaft is thought to be more easily detected, due to the less ephemeral structures involved (Carrer 2016).

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

The exploitation of high Alpine areas is known to have taken place in Roman times and Late Antiquity, but developed variably from region to region during early medieval times according to their topography and political development. In the Dachstein area southeast of the Isel Valley, the archaeological traces vanish in the 5th and 6th centuries, reappearing again in the 7th. In the Steiner/Kammniker Alps and in the Karawanken in the northeast, on the other hand, Alpine farming flourished from the 4th to the 6th century and seems to have ended with the beginning of Slavonic rule (Fig. 1). At Gastein – south of the Hohe Tauern – pollen profiles indicate an uninterrupted exploitation of summer pastures. This might be linked to a supply network for miners (Winckler 2012, 272‑278). On the Dachstein plateau, small finds and the completely excavated groundplan of a Roman-period hut suggest a peak in summer pasturing in the 3rd century (Hebert 1996; Mandl 1996, 54‑61; for a critique, see Gleirscher 2006). After this peak, the material decreases. For the early medieval period, the best-researched structure is a hut measuring 5.5 m by 9 m at Steiniggrube in the eastern part of the Dachstein region. Two radiocarbon dates, a potsherd, an arrowhead, and a fragment of a knife suggest seasonal settlement from the 7th century onwards. The rectangular stone layout is interpreted as being the base for a blockhouse. Another hut with a rectangular floorplan was documented above the tree line, and charcoal from a small test pit has been radiocarbon dated to the 11th century AD. Both structures are situated at around

1,650 m a.s.l. and are therefore in a comparatively low altitude for Alpine farming. Place names rooted in a Slavic language indicate human activity in the upland pastures. The 14C dates sampled from small test pits increase in number from the 10th century onwards, coinciding with an improvement in climatic conditions. This is seen as the starting point of a medieval summer pasturing economy (Mandl 1996, 62‑72; Mandl 2002, 145‑146; Kral 1994). In the Tennengebirge mountain range, a small test pit sample from inside a roughly four-sided structure, indicated by stones protruding through the surface, has produced 14C dates ranging from AD 660‑730 and AD 740‑770. These dates match the mention of summer pastures and a seasonal settlement on the northern fringe of the mountain range in a register from the 8th century (Brandner 2014). The Late Antique settlement pattern in the Steiner/ Kammniker Alps changes from the Roman period to Late Antiquity (4th to mid-5th centuries), with a shift towards hilltop settlements at around 1,000 m a.s.l. The transition to the early medieval period is marked by the cessation of these permanent settlements and a lower population density. This is linked with the end of most of the detected seasonal settlement sites, which disappear at the same time. The early medieval remains are far less numerous than those of former periods (Horvat 2002, 128‑130; Horvat 2019, 7). In the Julian Alps, the pattern of seasonal settlement changed at the end of the 6th and the beginning of the 7th century, with sites being abandoned and other

Fig. 1. Map showing the mountain ridges named in the text (© Elisabeth Waldhart).

Waldhart and Stadler

47

structures being built. The number of known sites remains almost the same from Late Antiquity to the early medieval period, with 11 detected sites dating to the former and 12 detected sites dating to the latter. These locations are situated in a narrow altitude range of only 1,475 to 1,670 m a.s.l., compared to previous sites, which ranged from 1,400 to 2,000 m a.s.l. Three buildings with substantial drystone foundations from the Julian Alps indicate uninterrupted habitation from the 6th to the 9th century. New buildings used the same drystone technique for foundations as previous buildings (Horvat 2019, 8‑9). Apart from summer pasturing, other seasonal activities in areas of extensively farmed land on the mountainsides have been studied or can be taken into consideration, such as mineral extraction on the one hand and the exploitation of mountain flora on the other. Potstone, a soft stone that can be worked easily, was quarried and worked into rough shapes in the southwestern part of Tyrol (Fig. 1). First 14 C dates from charcoal-rich layers containing roughly worked potstone vessels and sherds of these pots indicate an early medieval horizon dating from the 7th century. Though many quarries are known, less can be said about the accommodations of the miners (Bachnetzer 2014). Mining of various ores was carried out at high altitude throughout medieval times and into the early modern age (Theune 2019). The collecting of medicinal herbs was described by Oswald von Wolkenstein around 1600 and, from more-recent times, the collection of Gelber Enzian (Gentiana lutea) roots as an ingredient for hard liquor is known, while the diverse and rich blossoms of Alpine flowers could also have been used in beekeeping (Wopfner 1997, 374‑375). Historical and economic development

The region known today as East Tyrol was once part of the Roman province of Noricum Mediterraneum, with the Municipium Aguntum as its centre. In the 5th and 6th centuries, it was within the sphere of influence of Byzantium and the power blocs of the Francs, Bavarians, and Langobards. At the end of the 6th century, new agents – people subsumed under the term ‘Slavs’ – emerged from the east. The Langobard chronographer Paulus Diaconus reports a battle between the Bavarians and the Slavs near Aguntum in AD 610. To this day, however, the exact location has not been identified and the victims of this conflict have not been archaeologically documented. The encounter between the Slavic immigrants and the native population has been variously discussed in research. Up to the 1950s, the Migration Period was pictured as one of invasion and devastation. Today, influenced by research on place names, the view has shifted to one of a more peaceful encounter, as far as this seems plausible 48

under circumstances of land grabbing. In the 7th and 8th centuries, the Lienz Basin and Isel Valley became part of an independent Slavic principality named ‘Carantanien’. It is assumed that its centre of power was at Zollfeld (Karnburg?) or at St. Peter near Spittal (Teurnia), in modern-day Carinthia. It remains unclear how the rulers managed to establish this formation, which was made up of newcomers and local residents and which was, moreover, the oldest known political structure in the Slavic world. As late as the 11th century and, in peripheral areas, up to the 13th, documents still refer to a Slavic-speaking population in East Tyrol (Wiesflecker 1949‑1952, 37‑40). The Roman and Late Antiquity settlement structure around the cities of Aguntum and Lavant is relatively unknown, and little is known about houses from the 7th century onwards (Auer – Sperger 2018). This makes it hard to determine traditions, continuities, breaks, and changes in the area of interest. In the second half of the 8th century, the principality of Carantanien became part of the Franconian Kingdom and was thus subject to new influences from Western Europe and a new growth of Christianity, which was strongly connected to power relations (Štih 2014, 264‑280). However, some churches in East Tyrol do show continuity from Late Antiquity to the Middle Ages and provide evidence for ongoing Christian worship (Gleirscher 2018, 241‑246). In the 9th century, the last Slavic sovereign was replaced by a Franconian count, which resulted in a big shift in administration and law, although fragments of older laws seem to have remained valid into the 11th century (Štih 2014, 279‑282). The Franconian system encompassed bigger farms, owned by a caretaker and farmed by unfree people, and smaller farms cultivated by men who had to pay tax to and perform work duties at the bigger farms (Gleirscher 2018, 296‑299). Archaeological traces from this time are predominantly graves, with only rare traces of settlements (Fig. 2). From the 10th to the 13th century, new settlers took new land into cultivation. Starting in the 12th century, permanent farms at higher altitudes with a focus on livestock breeding – so-called Schwaighöfe – were established. In the 13th and 14th centuries, more attention turned to these upland areas, which were difficult to access and economically less attractive (Štih 2014, 292‑293; Sendele 1963, 23‑26). Some of these farms were in existence for only 50 to 100 years and then disappeared from historical sources (Sendele 1963, 124). These transformations between seasonal and permanent settlements, which were possible in both ways for multiple reasons, show the difficulties of interpreting archaeological structures. During these periods of transformation, local populations must have felt the impact of the changing rule in terms of administration, organisation of religion, settlement structures, and trade. For seasonal settlement,

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 2. Map of known settlements, graves, and small finds from the 9th and 10th centuries in East Tyrol. The red box marks the boundaries of the map in Fig. 4 (© Elisabeth Waldhart; data from: Land Tirol tiris 2020).

the most important aspects related to changes in trade. Since Almwirtschaft is a means of enhancing agricultural production, changes in the wider economy are likely to have had consequences for this system of production (Štih 2014, 257‑264). The archaeological landscape above Ainet

The potential for archaeological work to expand our knowledge on seasonal settlement in high mountain environments is being tested by a project in the Schobergruppe mountain range. The focus of this paper is on an Alpine site high above the community of Ainet, where the majority of the material in this project has been gained through survey work (Figs. 3 and 4). The high Alpine landscape above the village of Alkus, Ainet, East Tyrol has been investigated since 2006, with surveys and excavations taking place from 2007 to 2018. While the excavations focused on prehistoric to Roman sites (Waldhart – Stadler 2019; Stadler – Waldhart 2020), the survey projects gave insight into a broader timeframe, ranging from the early Bronze Age through the Roman period to the 20th century (Klocker 2013; Weishäupl 2013; Klocker 2017).

The area under investigation is part of a landscape characterised by the deeply cut Isel Valley. From the bottom, steep hillsides rise upwards, interrupted by smaller terraces that are used in different ways – the lower ones are permanently populated, while on the upper ones seasonal farming is practised. At an altitude level between 2,000 and 2,800 m a.s.l., peneplains form flat areas that are dominated by individual peaks. While the community of Ainet is located on an alluvial cone, the villages of Gwabl and Alkus, as well as some stray farms, are found on small terraces. The high plateau of Potschepol, which is, together with Lake Alkus, the highest area under investigation, is situated at approximately 2,080 m a.s.l., 100 m above the recent treeline. A further 150 m upwards, Lake Alkus is embedded in a peneplain, surrounded by mountain peaks reaching up to 3,000 m a.s.l. A second lake, Gutenbrunn, is situated to the east. A little river, the Taber Bach, runs from Lake Alkus across the Potschepol and further down into the valley, where it flows past Ainet. The landscape in this area has been formed by the people living, working, walking in, and talking about it. The valley floor and the lower terraces are now free of Waldhart and Stadler

49

Fig. 3. Alkus, Potschepol: view from the north of the Potschepol plateau, where archaeological investigation of seasonal settlement in a high mountainous environment has been ongoing since 2006 (© Department of Archaeologies, University of Innsbruck).

woods. Many of the flatter areas are used for permanent settlements and intensively farmed patches of land. The steeper parts of the hillsides and the area above 1,500 m a.s.l. are covered with woods, interspersed with patches of cleared land, where summer farms can be found today. Around those clearances, which offer good pastures with sheltered areas and species-rich fauna, lies open woodland consisting of larch trees (Larix). This kind of woodland has been shaped by humans, maybe through fire clearance, and can be found in various Alpine areas (Oeggl 2009, 85‑86). The time of origin of the clearances in this area cannot be determined. However, a single 14C date was taken from a 5‑7 cm thick carbonaceous layer found beneath the Potschepol at 2080 m a.s.l. This layer can be read as the result of a forest clearance in a wider area and a charcoal sample taken from it is dated to AD 970‑1170 (95,4%, VERA-4972) – quite an imprecise date, considering the old wood effect and a plateau in the 14C calibration curve in the Early Middle Ages. A species identification showed that the charcoal sample was of stone pine (Pinus cembra), a type of tree that reaches the uppermost fringe of the treeline in this area. Future projects might provide more information about this. Above the modern-day, managed summer farms, 34 undated, partly dilapidated drystone structures can be found, amongst them walls forming pasture boundaries, huts, pens, and smaller shelters (Weishäupl 2013). The name Ainet refers to Einöd, which translates as ‘a remote area’. The first mention of the Ainet community can be found in written documents that date from 1277, which refer to a farm above Ainet, which was given to 50

the Dominican nunnery by Count Albert von Görz-Tirol. The Görzer Urbar, a rent roll, mentions a farmyard near Ainet in 1299. On the hillside terraces, the villages of Gwabl, Alkus, and Lassnig can be found; farms in Gwabl and Alkus are mentioned in 1206 in a bill of sale, which states that Bishop Walther of Gurk bought farms from Count Meinhard von Görz and returned them to the peasants as compensation for their military service. In this document, an inhabitant known by the Slavic name of ‘Ilcona’ as well as inhabitants with Germanic names are mentioned (Bergmann 2005, 15‑16). The references to farms in the written sources give clues to permanent settlements and farms in the 13th century in the areas around Ainet and on the lower terraces. Archaeological traces in Alkus might date the origins of the settlement further back. Finds hint at beginnings of human activity in this area in the Roman period (Stadler 2008). In the area where the easternmost houses of Alkus are found, two graves were discovered during building activities in the 1950s (Fig. 4.1). Both are east-west oriented and the poorer-preserved one lay beneath a layer of black, charcoal-rich material. In the betterpreserved grave, a temple ring with rounded ends was found (Fig. 5.1). This ring type is dated to the 10th-11th centuries (Stadler 1993), although the relatively large diameter – 6 cm – might indicate a slightly earlier form (Eichert 2010, 45). It is common in the eastern Alpine area, especially in Austria and Slovenia. The second find, which is a headdress ring, indicates a human presence above today’s settlement fringe (Figs 4.2; 5.2). It was a stray find detected on a terrace that has

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 4. Map of the recent landscape around the community of Ainet, with current settlements and recently used mountain farms in the wooded areas and in cleared patches visible. Above the treeline, the prehistoric to Roman Age structures and the undated structures detected by surveys are marked; 1‑4 are the locations of small finds (see Fig. 5) (© Elisabeth Waldhart, data from: Land Tirol tiris 2020).

recently been used for summer pasturing, with three summer farms nearby. Situated at about 1,590 m a.s.l., it is slightly above the highest farms that were historically known to be permanently settled, and in the lowest part of the seasonally farmed areas. The semi-lunular ring with a basket-shaped body and central extension – the bracket is missing – is cast in bronze. Indentations point to enamel inlays, a typical style of decoration for this type of headdress ring. These ornaments are typical of eastern Alpine regions, especially in Austria and Slovenia. Cast pieces are dated to the late 9th century and, principally, to the 10th (Eichert 2010, 65‑75). Both types of ornament belong to an archaeological group often known under the slightly misleading name ‘Köttlach’, which is linked by its main distribution to the Franconian principality of Carantanien (Eichert 2013). On the high plateau of Potschepol, where seasonal settlement is likely to have occurred due to its favourable topographic conditions, only items that cannot be dated clearly and have a long span of use have been found. A lyre-shaped fire striker with open, rolled-up ends and a

characteristic spur in the middle (Figs. 4:3 and 5:3) has parallels in finds from graves from the early medieval period (Eichert 2010, 138) to the High Middle Ages (Steuer – Ilkjær 1994; Pierobon – Bombonato 2015). Altogether, 31 complete horseshoes as well as 30 fragmented ones have been found on the Potschepol plateau. Most of them can be roughly dated to medieval or post-medieval times and are seen as traces of horse pasturing in this area (Klocker 2013). One specimen with a waved edge, relatively flat heels, three nail holes at each side, one remaining nail, and no toe clip is of particular interest in relation to the question of seasonal settlement above Alkus (Figs. 4:4 and 5:4). As horseshoes are small finds with relatively little chronological accuracy, often having been made in small workshops and without many differentiating characteristics, dating is relatively vague. The veterinary surgeon and archaeologist Urs Imhof has made a chronology of horseshoes for the western part of Switzerland, in which he states that horseshoes with waved edges are the earliest form of the more commonly used horseshoes with nails in Western Europe. In his Waldhart and Stadler

51

Fig. 5. Small finds from the area of Ainet (see Fig. 4): 1) temple ring with rounded ends, KG Alkus; 2) semi-lunular headdress ring, Obermair auf GlorachAlm; 3) fire striker, Potschepol; and 4) horseshoe, Potschepol (© Elisabeth Waldhart).

typologies, he traces dated finds from castles. Horseshoes with waved edges were first crafted in the late 10th/11th centuries and were still in use in the 12th (Imhof 2010). Two horseshoe fragments from graves in Carinthia that cannot be typologically assigned due to poor preservation indicate that horseshoes can also be assumed to have been found in the Eastern Alps from the 10th to the 11th centuries (Eichert 2010, 129). The archaeological traces for seasonal settlement in the area of investigation are very scarce, with only one characteristic stray find and a few other objects that might hint at seasonal settlement. The objects cover a time span from the late 10th to the 11th century and the typologically insensitive horseshoe and a fire striker have been found that can be dated as late as the Middle Ages. A hut and pen structure indicates that the pastures on Potschepol were used in the Roman period (Waldhart – Stadler 2019, 15‑18) with small finds up 52

to the 4th century. After this, a hiatus, mirrored in the material record, takes place. Whether the absence of finds indicates a cessation of seasonal settlement from the 4th century up to medieval times (horseshoes indicate renewed pasture usage in the 12th century), or whether it is the result of changes in land use practices, such as the adoption of other herding strategies that leave fewer traces in the archaeological record, is still to be determined. The semi-lunular headdress ring might indicate a migration of seasonal settlements from the highest areas to lower areas within the (cleared) woodland. This matches findings from other areas in the eastern Alpine region, like the Dachstein and the Julian Alps. Nonetheless, the early medieval Slavic reign and the Slavic-speaking population – who were most probably present up to the 11th century – did leave traces on the landscape. Place names and settlement layouts can persist for long periods of time. The area around Ainet,

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

next to the Virgen Valley, has the highest density of Slavic place names in East Tyrol. Slavonic place names in the area

In the context of this project, the name of the high plateau is significant. The term ‘Potschepol’ derives from the Slavic -pól, meaning a flat and wet piece of land. Also, several mountain peaks nearby have Slavic names. This can be read as an indicator that Slavic speakers populated the area. Places get their names from people such as farmers, foresters, and herders who travel through the landscape and communicate with each other about its features; later on, they are used by people who write down claims of ownership. The first record of the names of pastures can be found in written documents of the 15th century; such records become somewhat more numerous in the 16th century (Rampl 2009). The primary names denominate, among other things, topographical features, the use of an area for agriculture or other economic activity, vegetation patterns, and settlement structures (Rampl 2008). These first written records of place names only capture names that were somehow related to administration, recording ownership, border demarcations, or fiefs. Most names were still being passed down orally until the 18th and 19th centuries, when they were written down in the course of early land registration projects or due to scientific interest in documentation. This leads to an incomplete record of the names in use at a particular time and creates a patchwork pattern of names from different historical periods and times of recording, since different people used different names and, depending on whose version was taken down, this would have affected the record. Viewed in a wider context, place names can tell us about land use and the people who managed the land, as well as provide a history of claiming power on the land (Antenhofer 2014, 105‑109). The place names with Slavic roots are widely distributed and can be found from the valley floor to the high mountains. In the development of the landscape, the names of patches of clearances are of special interest. For example, place names deriving from the Slavic *lazъ refer to treeless areas in the wood, virgin soil, a new field, or pasture. Seven place names with this root can be found in the area of Ainet, Gwabl, and Alkus. Five of them can be located in the current landscape (Fig. 4), while three are only traceable from historical documents. At an elevation of around 1,500 m a.s.l., stray farms are combined under the names Ober- and Unterlassnig. They are spread out across two small terraces on the steep hillsides, with two farms on the upper terrace and three on the lower one. The first written sources that record names for these farms date back to 1385 (Bergmann 2005, 203‑205). A meadow

on the upper part of the colluvial cone where the village Ainet is situated is named Las; a field in Gwabl is named Las Ackerle. Also linked to a recent farmyard is the field Lasín on the terraces in Gwabl. In Alkus, the field Lass and Zelàs (which translates as ‘behind a place called Las’) have the same etymology (Bergmann 2005, 198‑103). On the uppermost fringe of the forest belt, the pasture Lusene also bears a name derived from the same root (Bergmann 2005, 214). The place name ‘Las’ only refers to a clearance to maintain new land for farming. It is not clear, however, whether such clearance happened in connection with the extension of a settlement (which could be postulated for the field named Las), or for the creation of new summer pastures (which is conceivably the case for the modernday wooded pasture Lusene). For farms in Ober- and Unterlassnig, some farmhouses can be traced back to the 16th century, when they first appear in written sources. The masonry hearth in one of the farmhouses is comparable to similar structures from the 18th century; though research on this type of structure is limited, the origin of this type in the 16th century is possible. Whether the houses replaced older farmhouses – which could be possible, because the higher altitudes were settled from the 13th/14th centuries onwards – or whether they originated from a former seasonal settlement cannot be determined. Place names suggest that medieval settlements were established on former summer pasturing sites in the Northern Tyrol, where villages founded in the 12th and 13th centuries have names that denote lower summer pastures (Wopfner 1997, 391‑195). However, no information on early medieval Alpine farming in the area of Ainet can be gained from historical texts, with the first documents on Alpine pastures stemming from the 16th century. These documents regulate ownership of summer pastures and therefore tell us nothing about the settlement situation or about the everyday life of the people involved (Klocker 2013, 13‑19). Vertical transhumance, as well as other forms of exploiting high Alpine resources, is connected to the permanent settlements on the valley floor and to the economic situation in the area. Here, documentary and archaeological sources can help us understand the framework of ongoing processes. Conclusion and outlook

While the archaeological evidence for seasonal settlement and human activities in the area between the Potschepol and Ainet is quite dense from the 1st century BC to the 2th century AD, the last archaeological traces of people staying in the area seasonally date to the 3th and 4th centuries. Then, archaeological records vanish, with no hints at seasonal settlement or human activity in the Waldhart and Stadler

53

Fig. 6. Sketch of the development of permanent and seasonal settlement as suggested in the chapter. From prehistory to the Roman Age, only the lower areas were used for permanent settlement. The detected seasonal structures for summer pasturing are adjacent to the highest pastures above the treeline. From the Early Middle Ages to the Middle Ages, patches of woodland were cleared and used for permanent and seasonal settlement with permanent, occupied farms – Schwaighöfe – being located higher up than today’s settlements. In the 19th century, the highest farms were still inhabited or used as seasonal farms. The various locations were used differently throughout the year and were farmed with different animals (© Elisabeth Waldhart).

upland pastures up to the Middle Ages. Beginning in the 9th and 10th centuries, a small number of objects and place names can be interpreted as the first clues that can be used to research seasonal settlement and agricultural use of the Alpine and subalpine zone in the Middle Ages. Whether the apparent lack of archaeological structures from the 5th to the 13th centuries indicates the complete absence of seasonal settlement in high Alpine areas during this period is, however, difficult to judge. There are archaeological traces of high Alpine settlements that have not been dated yet, such as the drystone structures documented by archaeological surveys (Weishäupl 2013). Also, the large number of place names rooted in a Slavonic language can be read as a clue to the exploitation of the high Alpine pastures in early medieval times. Furthermore, it must be kept in mind that early medieval buildings for seasonal use may not have had any stone foundations and thus no archaeological traces have been preserved. The future analysis of pollen profiles has the potential to clarify these questions. Seasonal settlement in high Alpine sites is connected to the permanent settlements on the valley floors and

54

lower terraces in a farming economy. Changing structures in settlement, population density, and supply networks might have led to these farming techniques becoming unprofitable and too risky, leading to the abandonment of upland pastures in the 5th century. The objects and place names discussed in this chapter coincide with a general increase in the archaeological and historical traces of human presence in East Tyrol, starting in the 9th century (Fig. 6). This could point to a resumption of seasonal settlement practices to increase food production or to meet other needs. The preliminary indications of wood clearance and the single semi-lunular headdress ring found in such a clearance raise further questions about changing patterns of exploitation of summer pastures and seasonal settlement during the early medieval history of the area. While the research activities were focused on the upland pastures above the treeline, answers to questions regarding changing practices of seasonal settlement and mountain farming may be sought in the woodland between the upland pastures, as well as in the settlements in the clearances and open wood pastures.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

References

Antenhofer, C. 2014: Erzählte Welt. Die Emotionalisierung der Landschaft über die Geschichte ihrer Benennungen, in: Leitner, U. (ed.), Berg & Leute. Tirol als Landschaft und Identität. Schriften zur politischen Ästhetik 1. Innsbruck, 97‑131. Auer, M. – Sperger, C. 2018: Das Umland von Aguntum – Eine GIS gestützte Kartierung kaiserzeitlicher und spätantiker Siedlungsplätze, in: Auer, M. – Stadler, H. (eds.), Von Aguntum zum Alkuser See. Zur römischen Geschichte der Siedlungskammer Osttirol. Ager Aguntinus – Historisch-archäologische Forschungen 1. Wiesbaden, 33‑39. Bachnetzer, T. 2014: Frühmittelalterlicher Lavezabbau am Pfitscherjoch in den Zillertaler Alpen, Blei für Silber – Zink für Messing: der Bergbau am Südtiroler Schneeberg, Res montanarum Sonderband 2, 183‑194. Brandner, D. 2014: Das interdisziplinäre Projekt „Pitschenbergalm“ Tennengebirge, Land Salzburg. Erste archäologische Ergebnisse, in: Mandl, F. (ed.), Forschung in den nördlichen Kalkalpen. Forschungsberichte der ANISA 5. Haus im Ennstal, 43‑48. Carrer, F. 2016: The ‘invisible’ shepherd and the ‘visible’ dairyman: ethnoarchaeology of alpine pastoral sites in the Val Fiemme (eastern Italian Alps), in: Collis, J. – Pearce, M. – Nicolis, F. (eds.), Summer farms. Seasonal exploitation of the uplands from prehistory to the present. Sheffield, 97‑107. Costello, E. – Svensson, E. 2018: Transhumant pastoralism in historic landscapes. Beginning a European perspective, in: Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe. Themes in Contemporary Archaeology 6. London – New York, 1‑13. Bergmann, H. 2005: Slawisches im Namengut der Osttiroler Gemeinden Ainet und Schlaiten. Österreichische Namensforschung Beihefte 5. Wien. Eichert, S. 2010: Die frühmittelalterlichen Grabfunde Kärntens. Die materielle Kultur Karantaniens anhand der Grabfunde vom Ende der Spätantike bis ins 11. Jahrhundert. Klagenfurt am Wörthersee. Eichert, S. 2013: Zur Absolutchronologie des Ostalpenraumes im Frühmittelalter unter besonderer Berücksichtigung 14C-datierter Grabinventare, Bericht der Bayrischen Bodendenkmalpflege 54, 419‑427.

Gleirscher, P. 2006: Zum Nachweis römischer Almhütten am Dachsteinplateau und in den Steiner Alpen (Kaminške Alpe), in: Mandl, F. (ed.), Alpen. Festschrift: 25 Jahre ANISA. Haus im Ennstal, 23‑30. Gleirscher, P. 2018: Karantanien. Slawisches Fürstentum und bairische Grafschaft. Klagenfurt/Ljubljana/Wien. Hebert, B. 1996: Ergrabung einer römerzeitlichen Hütte in den Rotböden. (Steiermark, Östliches Dachsteinplateau, Katastralgemeinde Gröbming), in: Mandl, F. – Cerwinka, G. (eds.), Dachstein. Vier Jahrtausende Almen im Hochgebirge. Band 2. Mitteilungen der Anisa 18/2. Gröbming, 200‑231. Horvat, J. 2002: Archäologische Zeugnisse im Slowenischen Alpengebiet, Historie des Alpes 7, 117‑133. Horvat, J. 2019: Inhabiting the mountains of the south-eastern Alps. Slovenia, in: Mandl, F. (ed.), Archäologie und Geschichte. Siedlung und Wirtschaft im Alpinen Raum. Tagungsband. Forschungsberichte der ANISA für das Internet 1. Haus im Ennstal, 4‑29. Imhof, U. 2010: Die Geschichte des Hufbeschlags, Schweizer Archiv für Tierheilkunde 152:1, 21‑29. Klocker, C. 2013: Almurkunden und Hufeisenfunde vom Potschepol. Gemeinde Ainet, Osttirol. Mit einem Beitrag von Burkhard Weishäupl. Nearchos Beiheft 13. Innsbruck. Klocker, C. 2017: Hochgebirgsforschung in Osttirol. Die Metallfunde vom Potschepol, Gem. Ainet. MA thesis presented to Universität Innsbruck. Kral, F. 1994: Zur Frage früher Almwirtschaft am östlichen Dachsteinplateau (Pollenprofil Plankenalm), Mitteilungen der Anisa 15, 37‑43. Mandl, F. 1996: Das östliche Dachsteinplateau. 4000  Jahre Geschichte der hochalpinen Weide- und Almwirtschaft, in: Mandl, F. – Cerwinka, G. (eds.), Dachstein. Vier Jahrtausende Almen im Hochgebirge. Band 1. Mitteilungen der Anisa 18:1. Gröbming, 9‑165. Mandl, F. 2002: Almen im Herzen Österreichs. Dachsteingebirge. Niedere Tauern. Salzkammergut. Mitteilung der ANISA 23. Haus im Ennstal. Oeggl, K. 2009: Prähistorische Landnutzung im mittleren Alpenraum, in: Oeggl, K. – Prast, M. (eds.), Die Waldhart and Stadler

55

Geschichte des Bergbaus in Tirol und seinen angrenzenden Gebieten. Proceedings zum 3. Milestone-Meeting des SFB HiMAT vom 23.26.10.2008 in Silbertal. Innsbruck, 79‑90. Pierobon, F. – Bombonato, G. 2015: Un importante recuperodi reperti dal fianco occidentale del colle di Castel Leuchtenburg, in: Kaufmann, G. (ed.), Archäologie des Überetsch. Archeologia dell’Oltradige, Schriften des Südtiroler Archäologiemuseums 5. Innsbruck, 503‑544. Rampl, G. 2008: Cambiamenti e costanti nei nomi di malghe e di montagne nel Tirolo Orientale. Veränderung und Konstanz bei den Alm- und Bergnamen Osttirols, in: Pascolini, M. (ed.), Le alpi che cambiano. Nuovi abtianti, nuove culture, nuovi paesaggi. Die Alpen im Wandel. Neue Bewohner, neue Kulturen, Neue Landschaften. Udine, 239‑247. Rampl, G. 2009: Mountain and pasture names of East Tyrol as an example of cultural heritage, in: Jordan, P. – Bergmann, H. – Cheetham, C. – Hausner, I. (eds.), Geographical names as a part of the cultural heritage. Wiener Schriften zur Geographie und Kartographie 18. Wien, 99‑105. Reitmaier, T. 2017: Prähistorische Alpwirtschaft: eine archäologische Spurensuche in der Silvretta (CH/A), 2007‑2016, Jahrbuch Archäologie Schweiz 100, 7‑53. Sendele, H. 1963: Die Almwirtschaft von Matrei in Osttirol. In geschichtlicher, rechtlicher und volkskundlicher Betrachtung mit einem Exkurs über die Hausgemeinschaften. Dissertation presented to University of Innsbruck. Stadler, H. 1993: Alte und neue archäologische Entdeckungen in Osttirol, Osttiroler Heimatblätter 61, 7‑8. Stadler, H. 2008: KG Ainet, OG Ainet, VB Lienz, Fundberichte Österreich 47, 590. Stadler, H. – Waldhart, E. 2020: Das Hochgebirge als gefährlicher Raum im Ressourcennetzwerk der Römer. Das Fallbeispiel Alkuser See/Potschepol in Ainet, Osttirol, in: Kasper, M. – Rollinger, R. – Rudigier, A. (eds.), Wirtschaften in den Bergen. Von Bergleuten, Hirten, Bauern, Künstlern, Händlern und Unternehmern. Montafoner Gipfeltreffen 4. Wien – Köln – Weimar, 205-223. Steuer, H. – Ilkjær, J. 1994: Feuerzeug, Reallexikon der Germanischen Altertumskunde 8, 403‑411.

56

Štih, P. 2014: Begegnung, Akkulturation und Integration am Berührungspunkt der romanischen, germanischen und slawischen Welt, in: Härtel, R. (ed.), Akkulturation im Mittelalter. Vorträge und Forschungen, herausgegeben vom Konstanzer Arbeitskreis für mittelalterliche Geschichte 78. Ostfildern, 236‑294. Theune, C. 2019: Climate change and economic development in the Alps during the Middle Ages and the early modern period, in: Brady, N. – Theune, C. (eds.), Settlement change across medieval Europe. Leiden, 435‑444. Waldhart, E. – Stadler, H. 2019: Ora et Labora – Die archäologische Landschaft Alkuser See und Potschepol im Spannungsfeld zwischen Almwirtschaft und Kult, in: Mandl, F. (ed.), Archäologie und Geschichte. Siedlung und Wirtschaft im Alpinen Raum. Tagungsband. Forschungsberichte der ANISA für das Internet 1. Haus im Ennstal, 4‑23. Weishäupl, B. 2013: Anthropogene Strukturen im Gebiet Kunigalm – Potschepol – Alkuser See, in: Klocker, C. (ed.), Almurkunden und Hufeisenfunde vom Potschepol. Gemeinde Ainet, Osttirol. Mit einem Beitrag von Burkhard Weishäupl. Nearchos Beiheft 13. Innsbruck, 134‑159. Wiesflecker, H. 1949‑1952: Die Regesten der Grafen von Görz und Tirol, Pfalzgrafen in Kärnten, Bd. 1: 957‑1271 (= Publikationen des Instituts für Österreichische Geschichtsforschung, 4. Reihe: 1. Abt., Innsbruck. (unter Mitarbeit von Johann Rainer), Die Regesten der Grafen von Görz und Tirol, Herzoge von Kärnten, Bd. 2/1: Die Regesten Meinhards II. (I.) 1271‑1295 (= Publikationen des Instituts für Österreichische Geschichtsforschung, 4. Reihe: 1. Abt, Innsbruck. Winckler, K. 2012: Die Alpen im Frühmittelalter. Die Geschichte eines Raumes in den Jahren 500 bis 800. Wien – Köln – Weimar. Wopfner, H. 1997: Bergbauernbuch. Von Arbeit und Leben des Tiroler Bergbauern. 3. Band: Wirtschaftliches Leben. Aus dem Nachlaß herausgegeben von Nikolaus Grass unter Mitarbeit von Dietrich Thaler. SchlernSchriften 298. Tiroler Wirtschaftsstudien 49. Innsbruck.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

A multidisciplinary approach to the relationship between seasonal settlements and multiple uses: case studies from southern Europe (15th-21st centuries) Anna Maria Stagno*

Abstract

Building upon the increasing recent interest in the archaeology of seasonal occupation, this chapter will focus on settlements associated with short and long transhumance. These sites were mainly located on common land, with some still existing today. Using case studies from the Basque Country and Ligurian Apennines, this chapter will establish interconnections between transformations in the structural characteristics of buildings, practices of environmental resource management, and the social organisation of mountain spaces, together with ownership and access rights to commons. This interdisciplinary methodology – which combines landscape archaeology, historical ecology surveys, environmental archaeology analysis, and documentary research methods – allows for an analysis of transformations of management systems of environmental resources and their connection with changes to seasonal settlements. This approach also reflects on the progressive disappearance of multiple agro-forestry-pastoral activities and changes in the jurisdictional realm of the management of collective spaces within peasant societies. This provides opportunity to reflect on the great complexity of the multiple dimensions related to pastoral practices and on the necessity to retain this complexity in the deciphering and interpretation of archaeological evidence. * Dipartimento di Antichità, Filosofia, Storia Laboratory of Enrivonmental Archaeology and History (DAFIST-DISTAV), Università degli Studi di Genova via Balbi 6 16126 Genova, Italy [email protected]

Keywords: seasonal settlements, common lands, rural archaeology, multidisciplinary approach. Résumé

Une approche multidisciplinaire à la relation entre occupation saisonnier et usages multiples. Etudes de cas dans l’Europe du sud (XVe-XXIe siècles) S’appuyant sur l’intérêt accroissant pour l’archéologie de l’occupation saisonnière, cet article se concentre sur les colonies associés à la transhumance courte et longue. Ces sites étaient principalement situés sur des terres communes et certains In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 57-68.

57

existent encore aujourd’hui. À partir des études de cas du Pays basque et des Apennins liguriens, cette étude établira des interconnexions entre les transformations des caractéristiques structurelles des bâtiments, les pratiques de gestion des ressources environnementales, l’organisation sociale des espaces de montagne, ainsi que les droits de propriété et d’accès aux biens communs. Cette méthodologie interdisciplinaire – qui combine l’archéologie du paysage, les enquêtes d’écologie historique, l’analyse de l’archéologie environnementale et les méthodes de recherche documentaire – permet d’analyser les transformations des systèmes de gestion des ressources environnementales et leur lien avec les changements des établissements saisonniers. Cette approche reflète également la disparition progressive des activités agroforestières-pastorales multiples ainsi que les changements dans le domaine juridictionnel de la gestion des espaces collectifs au sein des sociétés paysannes. Ceci permettra de réfléchir à la grande complexité des dimensions multiples liées aux pratiques pastorales et à la nécessité de conserver cette complexité dans le déchiffrement et l’interprétation des preuves archéologiques. Mots-clés : établissements saisonniers, communaux, archéologie rurale, approche multidisciplinaire. Zusammenfassung

Fallstudien aus Südeuropa (15.-21. Jahrhundert) Aufbauend auf dem in jüngster Zeit zunehmenden Interesse an der Archäologie der saisonalen Besiedlung werde ich mich in diesem Beitrag auf Siedlungen konzentrieren, die mit kurzer und langer Transhumanz verbunden sind. Diese Orte befanden sich hauptsächlich auf kommunalem Land, welches heute noch teilweise existiert. Anhand von Fallstudien aus dem Baskenland und dem ligurischen Apennin werden Zusammenhänge zwischen Veränderungen der strukturellen Merkmale von Gebäuden, Praktiken der Bewirtschaftung von Umweltressourcen, der sozialen Organisation von Bergregionen sowie den Eigentumsund Zugangsrechten zu Gemeingütern erörtert. Dieser interdisziplinäre Ansatz kombiniert Landschaftsarchäologie, historisch-ökologische Erhebungen, umweltarchäologische Analysen sowie dokumentarische Forschungsmethoden. Dadurch wird eine Analyse der Transformationen in den Managementsystemen von Umweltressourcen und ihres Zusammenhangs mit Veränderungen der saisonalen Besiedlung ermöglicht. Diese Herangehensweise reflektiert auch das fortschreitende Verschwinden der vielfältigen agro-forstwirtschaftlich-pastoralen Aktivitäten und Veränderungen im Rechtsbereich der Verwaltung kommunale Räume innerhalb bäuerlicher Gesellschaften. Dies ermöglicht, über die große Komplexität der vielfältigen Dimensionen im Zusammenhang mit den pastoralen Praktiken und über die Notwendigkeit nachzudenken, diese Komplexität bei der Entschlüsselung und Interpretation archäologischer Zeugnisse beizubehalten.

Ein multidisziplinärer Ansatz für die Zusammenhänge zwischen saisonaler Besiedlung und Mehrfachnutzung.

Schlagwörter: Saisonale Siedlungen, Gemeindeland, ländliche Archäologie, multidisziplinärer Ansatz.

Introduction: investigating the inner social dimension of the landscape

Basque Country. The research in Ligurian Apennines is being carried out by the Laboratory of Environmental Archaeology and History (LASA) at the University of Genoa within the framework of a collaboration with the Soprintendenza Archeologia, Beni culturali, Paesaggio della città metropolitana di Genova e delle province di La Spezia, Savona, and Imperia. Coming back to those regions (Fig. 1), I would like to understand better whether and how the different characteristics of seasonal settlements and their related spaces reflect differences in the access rights used by people for their agro-forestry-pastoral activities (i.e. pasture, wood pasture, fields and private plots of land), and in their social meaning. The case studies share a similar settlement pattern, which is characterised by scattered small hamlets that historically owned and used large extensions of common land (ville in the Ligurian Apennines and pueblos or lugares in Álava). These places were involved in both medium- and short-distance transhumant movements. In contrast to

Seasonal settlements are one of the main elements of transhumance systems and investigation of them has the potential to provide better understandings of the differences between types of transhumance practice. The increasing importance of this topic in archaeology is well testified by two recent monograph volumes (Collis et al. 2016; Svensson – Costello 2018), both of which offer a wide overview of this topic at a European level. In my previous research of the Basque mountains and Eastern Ligurian Apennines (Stagno 2016, 2018a), I focused on the differences between seasonal settlements related to short-distance transhumance activities (monticazione, ganadería estante, or transmiterante) and long-distance ones (transumanza, transhumancia). Investigations in the Basque Country were carried out during my Marie Curie Project ARCHIMEDE at the Grupo de Investigación en Patrimonio y Paisajes Culturales at the University of the 58

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

long-transhumance movements, like the Mesta or Dogana delle Pecore di Foggia, which were centrally organised and governed (and covered long distances), these mediumdistance were locally controlled and based on reciprocity agreements between the hamlets and municipalities of the coasts with those that owned the common lands used as summer pastures. These medium-distance transhumances ceased completely in the Ligurian Apennines at the beginning of the twentieth century, but have continued in the Basque Country to some extent until the present. In fact, the main interesting reason for this difference is that while the two areas experienced similar changes, the Basque one is extensively in use until today, while Ligurian one was almost completely abandoned during the last century. Short-distance transhumances referred to the daily movement from hamlets to upland common pastures (of the same village or of neighbouring ones). The plateau of Aizkorri, since at least the late Middle Ages, was used as summer pasture for long-distance transhumant movements from the coastal winter pastures of Bizkaia and Gipuzkoa (Urzainki Mikeleiz 2007, 95‑99; Aragón Ruano 2003), while the areas of the western Ligurian Apennines were involved in transhumance routes that linked coastal areas of Liguria (winter pastures) to the inner Apennines and Po valley (summer pastures). These kinds of longdistance transhumance movement were regulated by rental contracts or reciprocal agreements between communities or local social groups, on which access to grazing resources (both common lands and private pastures) depended. The reconstruction of the changes in these agreements in relation to the changes in the use of resources enables us to understand management strategies and the tensions and conflicts between social groups and individuals concerning collective resources (cf. Stagno 2018).

In both these areas, the profound changes in husbandry practices documented since the 19th century are closely related to the intensification of environmental resource management. From this period, central authorities have instituted significant changes in environmental resource management practices, forbidding multiple or mixed uses and changing the rules governing access rights to the commons and the co-sharing of resources. The effect of this was to dissolve central elements of the social organisation of rural communities (Thompson 1993), consequently weakening them. This topic merits further study, as few archaeological attempts have been made to study the consequences of capitalism in rural southern Europe (Mientjes et al. 2002; Dalglish 2003). Until the end of the 19th century, parcels of mountain areas had multiple uses based on a cyclical rotation that included animal grazing, timber and wood harvesting, and temporary cultivation. Wooded pastures were the main constituent part, even if today they have almost completely disappeared, as is well documented in many parts of Europe (e.g. Grove – Rackham, 2001; Moreno 1990; Rotherham 2013; Agnoletti – Emanueli 2016). These changes were part of the new approach to mountain areas pushed by central states during the 19th century to make them more productive within the framework of the states’ administrative consolidation, as well as through the capillary regulation and the tight control of the exploitation of environmental resources for fiscal purposes as well (Stagno – Tigrino, 2020). The new regulations were also aimed at ending collective and shared use of resources and directing environmental resource exploitation towards uses considered more ‘productive’ and ‘rational’, such as monoculture. The reforms carried out with this aim led to the progressive disappearance of multiple land uses and of long-distance transhumance.

Fig. 1. Location of the sites discussed in this chapter (© A.M. Stagno).

Stagno

59

In the first example, focused on the area between the western Álava plain and the Aizkorri plateau, I will show how, using archival and field sources, it is possible to reconstruct the social organisation of mountain spaces and to characterise and qualify their different uses as well as the key role played by seasonal settlements in the materialisation of the access rights regarding the commons claimed by different social groups. These kinds of changes have the potential to inform us about wider transformation in the jurisdictional environment. In the second example, involving case studies of two locations in the western Ligurian Apennines (Monte Becco, Levá hamlet; Conscenti), I will consider in depth the relationship between jurisdiction and seasonal settlements and discuss the archaeological evidence related to the private appropriation of mountain slopes during the 19th century and the first decades of the 20th. This work will illuminate the social and cultural implications of such changes. Comparison between these case studies will shed new light on the opacity of our sources and allow us to reflect on the difficulty of making generalisations without considering the complexity of agro-forestrypastoral activities and their multiple and complementary dimensions (i.e. in production, management of environmental resources, building and preserving social relationships, and claiming and maintaining jurisdiction). This chapter will also consider changes over time and in different contexts. This demonstrates the fragility of this heritage and the extent to which its conservation depends upon the hidden social and historical dimensions of environmental resource management. In the conclusion, I will discuss the effects that our interpretation and restitution of the past have on the construction of the meaning of rural heritage and our responsibility for the future. Methodology

The research approach used here is closely related to the archaeology of environmental resources, as elaborated in the Laboratory of Environmental Archaeology and History at the University of Genoa over the last 20 years (Moreno et al. 2010). The archaeology of environmental resources merges approaches and methodologies of English historical ecology (EHE) (since Rackham 1976) with the perspective of Italian social microhistory (Grendi 1993). Together with environmental history and EHE research, the archaeology of environmental resources focuses particularly on the reconstruction of past agro-forestry pastoral practices and their associated landscapes. On the one hand, through present-day observation of vegetation and archaeobotanical studies, it aims at studying and understanding the historical processes of biodiversification and the environmental effects of the abandonment of rural activities, as well as developing the applicative meaning of 60

the research (Maggi et al. 2002; Cevasco 2007; Moneta – Parola 2014). On the other hand, as I will exemplify here, the rural archaeology researches are more focused on the investigation of the relationship between resource management practices, their jurisdictional dimension (in particular with the study of commons), and their links to the trajectories of the local social groups and population dynamics (Stagno 2018b). In this second approach, the role of possession practices in the social relationships to land and access rights (Grendi 1986) is a key element, because from it stems the double nature of resource management practices as both ‘technical facts’ and ‘legal facts’ (Raggio 2001). It is for this reason that archaeology can investigate the evidence left by practices, trying to decipher their different values in the proof of possession. The study of practices of environmental resource management as possession practices opens up the possibility of bringing to light their meaning as expressions of long-term social relationships: these practices that resulted from the dialogue between individuals, communities, and central states, and how they shaped and changed landscapes. Thus, the investigation of practices is a means of deciphering not only changes in the landscape, but also the transformation of social dialectics through time, thus helping to shed light on unclear processes such as the slow process of the marginalisation of mountain areas that began in the 18th century. In the research presented in this chapter I will refer to the archival research carried out on the 16th-century conflicts around access rights to common lands and on rural statutes in the Basque Country, while for the Ligurian Apennines, the study of the 19th Register of properties and of the process of the sale of common lands was already available (Croce et al. 1992). Archaeological methods include investigations of archaeological and historical ecology through present-day vegetation observations, surveys, and small test pits (1 x 0.70 m) carried out in the Basque Country and combined with archaeobotanical (pollen, NPPs, phytoliths, charcoal). Investigation was particularly focused on evidence of the use of common lands and of those elements that allow for the construction of complex and multiple systems of environmental resource management. Socialising mountain areas: overlapping access rights, seasonality, and spaces

The hamlets studied in the Basque Country (Zuazo, Luzuriaga, Narbaiza, and Zalduondo, with territories located between 400 and 1000 m a.s.l.) have been under the jurisdiction of the new town of Salvatierra since 1332. Salvatierra was created (on the hamlet of Agurain) in 1256 by King Alfonso X of Castile. Alfonso X tried to ensure that towns and councils were loyal to the

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

monarchy with the aim of limiting the power of the local aristocracy. The new jurisdiction of Salvatierra was extended to include already existent hamlets with their own jurisdictions. This superimposition generated a complex situation with many consequent conflicts between Salvatierra and its hamlets (Santos Salazar 2011). Those conflicts, together with other conflicts among the same hamlets, are one of the main elements that enable a reconstruction of the organisation of local environmental resource management system between the late 15th and the 19th century. Another key source was the rural statute (Ordenanzas de Luzuriaga), recorded for the first time in 1542 and updated at various times since (1577, 1691) up through 1826 (see Stagno 2019 for a full description). The analysis of such documentation served as the basis for the reconstruction of the changes in the sharing of resources and the organisation of common lands (Fig. 2). Commons were articulated in spaces located at different distances from the permanent hamlets; these were montes bajos (low commons), montes altos (mountain commons), and Parzonería (commons shared between hamlets of the same jurisdiction or of different ones). These spaces also differed in terms of the organisation of their access rights: the montes bajos were usually owned and used by single hamlets (but with the possibility for Salvatierra to use them). Montes altos were, like montes bajos, owned by a single hamlet, but some agreements allowed the daily grazing of herds from neighbouring hamlets (the herds were not allowed to overnight in pastures). Inside montes altos spaces existed that were removed from common use and rent for long periods. These were called seles, and they were used (through precise agreement and contract) by foreign shepherds for the nightly accommodation of their herds. Finally, there were areas shared between different hamlets of the same jurisdiction (e.g. Montes de Udala and Montes de Alzania), and areas shared between hamlets of different jurisdictions (Parzonería, related to the Montes de Urbía and Oltza, which corresponds to the Aizkorri plateau and now are defined as Parzonería general de Gipuzkoa y Álava). Each of these spaces was used in different ways according to seasonality and types of resident livestock. For example, cattle grazed in Parzonería during summer, while herds of pigs were kept there during winter. In montes altos and in Parzoneria seasonal settlements (chozas, cabañas) were built or restored yearly. The restoration of routes used by herds was compulsory labour for the inhabitants of Luzuriaga. No written reference to the presence of those kinds of structures in montes bajos is provided. Fieldwork has identified differences in the uses of spaces. No evidence of sheds or other structures were recorded in montes bajos. So, it is possible to assume that here herds grazed during the day and returned nightly to sheds in the settlements. At the same time, surveys

allowed us to qualify those areas, as characterised by the presence of the evidence, as ancient wooded pastures of oak (in many cases fenced like in dehesas and sotos) inside present oak-beech mixed woodland. It is interesting to note that the right of use of these spaces was only reserved for the neighbours of the hamlet; Salvatierra was excluded from the co-sharing. This was explicated due to their artificiality, testified by the presence of fences and embankment: only the heirs (successors?) of those who built those spaces had the right to use them). In those spaces, temporary cultivations were allowed and practiced. Montes altos (between 700 and 1000 m a.s.l.) were also characterised by wooded pastures; there, evidence of temporary cultivation was absent. The presence of fences/ enclosures with sheds without evidence for domestic use suggested their use inside short-distance transhumance systems, in which herds were left in high-altitude pastures throughout the grazing period, while shepherds came back at night to their homes in the settlement (locally named transterminantia; a detailed description of the evidence is provided in Stagno et al. 2020; a first reconstruction in Stagno 2018a). The six structures identified date back to between the 16th and the 18th century and their distance from the hamlet is less than 5 km (Stagno et al. 2020). In both those areas, during the late 19th and the 20th centuries the wooded pastures were replaced by coppiced woodlands for charcoal production, similar to what happened in other European mountain areas during the rationalisation of rural activities (Stagno et al. 2018). The spaces of Parzonería in Aizkorri plateau are mainly characterised by pastures with little wood pasture, and a very high concentration of seasonal settlements with evidence of domestic spaces and articulated in small well-defined nuclei, like small hamlets, and with different place names. This spatial organisation suggests that the local social groups who inhabited them during the summers arrived here from different part of the Basque coast, as is testified by the information collected from oral sources in Urbía, where many ‘hamlets’ are still in use. A different situation was documented in Malla (at the western margin of Parzonería, and a sel in a list of the 1439 but not in the later documentation), where the remains of another seasonal settlement was identified. All of the site is constituted by structures visible at the ground level (Fig. 3), with the exception of a big fence abandoned just few years ago. The comparison with previously studied sites in Urbía (Gandiaga et al. 1989; Ugalde et al. 1992‑1993) and previously verified sites in other southern European mountainous areas and the relationship between the structures’ states of preservation and their chronologies (Rendu et al. 2016; Le Couedic 2010; Gassiot Balbé – García Casas 2014) enabled the selection of representative structures to verify the chronology of the site and to start to characterise their Stagno

61

Fig. 2. Articulation of the investigated areas in the Basque country and locations of the identified evidence (© Stagno et al. 2020, 168, fig. 3.3.2, modified).

Fig. 3. Topographic relief of the Malla site (© by C. Tejerizo, after Stagno et al. 2020).

62

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

precise functionality and their changes. The small test pits and archaeobotanical analyses (Stagno et al. 2020) carried out in 3 of the 13 identified structures demonstrated that, during their life, these structures were first used, in the 12th century, as domestic spaces and then converted into sheds (inside a sort of life cycle, similarly to what happens at present, when old domestic structures are converted into sheds and then into enclosures). Their domestic use (testified by cooking pottery fragments) in combination with evidence (from the archaeobotanical analysis of two areas sampled with shovel test pits) of agricultural uses suggest that the area was seasonally used for summer pastures within long-distance transhumance networks, a use which ended in the late 15th century, when the structures were completely abandoned and all the archaeological sequences show a period of abandonment. Since the 17th century, two of the earlier structures have been reused, having been enclosed by new fences, but there is no evidence of domestic use that might relate to the use of the area for short-distance transhumance (ganadería estante). This shift could be interpreted as in line with the use of those spaces by the local collectives of Oñati (Arantzazu), with the re-appropriation of their access rights to the area, after the dismission of the sel. It is interesting to highlight that the affirmation of the local use corresponds to the appearance of wooded pastures and of temporary cultivation and hay production. These types of evidence disappear during the last phase of use (after the 19th century), when the structures were completely abandoned. This further change indicates the single use of the area as pasture, which continues today, but without any connection to other uses, as in the previous period. One is better than many: the productivisation of mountain areas in the Ligurian Apennines

The sites located in the Ligurian Apennines that are discussed in this chapter are Monte Becco (between 600 and 900 m a.s.l., Levá, Canepa) in the municipality of Sori, Monte Lungo (500 m a.s.l., Terisso) and Campo Martino (300 m a.s.l., Graveglia) in the municipality of Nè. The first is situated near the coast, while the latter two are located inland; however, their historical trajectories are very similar. These sites reflect two different moments in the long process of transformation of the uses and nature of common land during the 19th and the first decades of the 20th century (cf. also Stagno – Tigrino 2020). Before the 19th century, the area of Monte Becco was part of the commonland of Levá, a small hamlet of the parish of Santa Maria of Canepa (Sori). This area was used by transhumance routes marked by the presence of mule tracks and walls called créste. Créste are documented at least since the 16th century and were built to stop herds

from straying into private plots or commons reserved for local herds (Moreno 1990). Between the 19th and the 20th centuries those commons were divided between the different inhabitants of Levá, and today the property is highly fragmented (Croce et al. 1992). The archaeological evidence includes the mule track with créste and an impressive number of structures (c. 150) concentrated over less than two hectares (Fig. 4). These buildings are of different types and forms: 20 small structures with a quadrangular plan with 2 openings (a door and a small window on the opposite side) and maximum height of 2 m, which, according to oral sources, were small storehouses (locally named caselle); 10 terraces; 15 filtering walls in the small watersheds of the seasonal stream (ria in local dialect); 50 so-called maxere or maghé (heaps of stones with rectangular, circular, or semi-circular arrangements); and other small walls along the slopes to mitigate against soil erosion. Just one casella shows evidence of two phases of use (the first possibly in the late 19th century and the second dated up the second half of the 20th). The study is ongoing, but it is possible to identify at least three types of caselle, differentiated by the use of timber in the construction, which possibly could correspond to different chronologies. The ceramic material collected during surveys dates between the beginning of (tache noir, terraglia near; Cameirana 1970) and the late 19th century (terraglia gialla, pentoline provenzali; Milanese 1997). Structural elements within the buildings also support this chronology (e.g. external wooden door lintels are not documented in western Liguria before the second half of the 19th  century). From the late 19th century until the 1970s, a teleferica (rope line or small-sized cable car, locally called strafia), located 500 m below the site served to bring hay down to the haybarns of Levá. The starting and the ending points of the rope line, which are made of concrete, are still preserved. All this evidence confirms a well-known general process that during the 19th century the local environmental management system was reorganised towards a more intensive use of the fodder resources of the mountain. This spatial organisation suggests the extensive exploitation for pasture and for hay production. Although this general process has been already investigated (Moreno 1990), this systematic archaeological analysis enables a better qualification and characterisation of the process and brings to light new questions. The topographical location of the caselle suggests a relationship with the articulation of the property or an internal division of the pastures. In many cases, the maxere are aligned to define geometrical spaces, suggesting a topographical organisation of the pastures and probably the existence of a constant negotiation around the use of resources. In addition, the precision of their construction and their shape suggest that some of them were also used to create artificial platforms along the slope. One possible hypothesis Stagno

63

Fig. 4. The Monte Becco slopes in the 2020 with examples of structures: a) casella and b) maxera (© A.M. Stagno).

is that they were used for the packaging of hay once it had been cut and before its transport downslope. In addition, the extreme regularity and precision of their construction suggests that the accumulation of stones had more than one function. However, the current interpretation of maxere is that they were stone heaps created from clean grazed areas. This is also the interpretation provided by the last users of these pastures 40 years ago when the first geographical-historical investigation was carried out in the area (Moreno 1970, 1990). The systematic analysis of all the structures suggests that this could be a simplified understanding of the aims and function for which they were built. This research is ongoing, and it is difficult to propose a coherent interpretation of these structures. This shows that our knowledge of the recent past presents the same problems for our understanding as moreancient periods. In addition, even if the construction of these structures took place in a period of privatisation of common lands, and so of new jurisdictional and possession relationships between the inhabitants of Levá, their articulation suggests that they were created within a common strategy – the result of a collective reorganisation of husbandry resources carried out by the inhabitants of Levá, and not as individual initiatives. Monte Lungo (500 m a.s.l., Terisso) and Campo Martino (300 m a.s.l., Graveglia) in the municipality of Nè show evidence for the intensification of husbandry in the mountains slightly inland from the Ligurian coast. These two areas share many similarities: they are both terraced and have well-defined slopes with a building and a small chapel at their margins. In both cases the terraced areas are open

64

areas within chestnut groves, with evidence of terraces and tree coppicing that suggests in the past they were wooded meadow pastures (Fig. 5). The buildings themselves are sheds with haybarns above, while the chapels provide dating evidence for the two building episodes: in 1918 at Monte Lungo and 1927 at Campo Martino. The inscriptions on the facades of the chapels (‘San Vincenzo Martire, nato a Saragozza e martirizzato in Spagna nell’anno 290, Cappella fatta costruire da Raffo Benedetta l’anno 1913’ and ‘Fratelli Podestá 1923’, respectively) also suggest that these terraces were built by local families, and the important presence of haybarns above the new sheds suggests the role played by hay production. Like Case Becco, this is new evidence of the intensification of husbandry activities, even during the first decades of the 20th century. In this case, the small area occupied by the terraced areas, the presence of a chapel with the name of the family who built it, and their precise boundaries, all suggest a private initiative very different from the one documented in Levá less than a century before. It is interesting to note that both these processes occurred in a period of depopulation of their respective municipalities (Tuttitalia 2013a, 2013b), and as already documented in the earlier case study, the intensification of the agro-forestry-pastoral activities does not seem to be related to any demographic increase (Stagno 2018b). Instead, this is a local response to changes in the socioeconomic infrastructure, such as the lack of importance of the road network, disappearing transhumance practices, and the progressive prohibition of the seasonal work when the population of these areas was drastically decreasing.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 5. Terraced slope at Campo Martino (UT 5, 300 m a.s.l., Graveglia) (© A.M. Stagno).

Conclusions

In this chapter I have considered the different natures of documentary, archaeological, and oral sources to underline the different perspectives they open up to the past. In the Basque country, the complexity of the social organisation related to different husbandry activities and the seasonal and topographical specialisation, dependent on the different types of herds, that has emerged through the study of the rural statutes of Luzuriaga is not as clear in the archaeological record. Nevertheless, archaeological and historical ecological observations have made it possible to characterise and qualify spaces and reveal the intimate relationship between local short-distance transhumance and wooded meadow pastures, and to identify the close connection between uses and access rights to resources. In the case of Case Becco in the Ligurian Apennines, the different types of structures identified through archaeological analysis do not fit with the interpretation derived from present and past oral sources. This difference suggests that a loss of jurisdictional and social meaning of the agro-forestry-pastoral practices occurred in the 20th century (probably accompanied by the disappearance of any collective dimension in the use of those spaces) that contributed to the suppression of the memory of the multiple functions of those structures. In one sense,

the simplification of the landscape during the 19th to 20th century, with the disappearance of shared use of resources and multiple uses, corresponds to a simplification of the meaning of the objects and therefore an erosion of memory. The end of the transmission of the knowledge related to the multiple meanings and functions of maxere well exemplifies one of the various ways through which the local naturalistic knowledge has been destroyed over the last two centuries (about this erosion as a consequences of the changes in the control and management of environmental resources, see Moreno 1990). In both cases, the transformations documented for the 19th and the 20th centuries, with a simplification of uses in a monocultural sense, testify to the process promoted by all the European central states within the framework of their consolidation and also through the capillary regulation and control of the exploitation of natural resources. The supposed ‘rationalisation’ of agroforestry-pastoral practices during the 19th century was driven by an ideology that aimed to reduce the collective dimension of agricultural activities (Grossi 1977; Congost i Colomer – Lana Berasáin 2007) and the use of different practices (e.g. temporary cultivations, wood pastures, etc.) that were considered irrational and unproductive. The sale of common lands represents the macroscopic effect of this process. However, most of these changes did not correspond with the privatisation of common goods, but resulted from the individualisation of the management of common resources, removing the social values embedded within these practices. Reforms of the 19th and the first decades of the 20th century aimed to make mountain areas productive. On the one hand, they removed any jurisdictional element of environmental resource management practices, and consequently their social and historical dimension resulted in negotiation around the use of resources (Stagno 2019). On the other hand, this process resulted in a homogeneity of the uses and forms of exploitation and an under-appreciation of local naturalistic knowledge and expertise. In this sense, the deep changes in the husbandry system promoted since the 19th century negated the need for the transmission of knowledge between generations within local communities (Stagno – Tigrino 2020). In this way, a less-visible, but no less-important, process of folklorisation of rural societies was promoted and took hold along with the ideological ruralisation promoted by the Fascist regime (Armiero 2011; Torre 2011). This approach to rural societies still has a deep influence on contemporary considerations of rural areas. Often, they are promoted as ‘natural’ and ‘atemporal’ spaces (the narrative of the wilderness), or for their aesthetic values, without any consideration of the historical dimension or the values that focussed observation of the landscape could generate. Stagno

65

The investigation of the archaeological evidence of these practices, and of their abandonment, brings to light the environmental and cultural dimensions of these processes. From a methodological stance, this is of crucial interest, because we see and interpret what we are used to seeing. Therefore, the erosion of memory, in all its multifaceted dimensions, pushes us to pay more attention to what we usually do not see or consider. In addition, the lack of understanding surrounding the multiple meaning of the objects, like the maxere at Monte Becco, and of the memory of their function makes them somehow less meaningful and important and therefore superfluous and easier to abandon. This is crucial in the investigation of heritage, because it pushes us to reflect on what heritage and its meaning are. We work with memory, and we engage with the construction of memory around our natural and built environments. If memory has no function, is its perpetuation useful or not? Acknowledgements

Research in the Basque country discussed in this paper has received funding in the framework of the Marie Curie IEF project ARCHIMEDE (G.A. 630095). I wish to thanks the anonymous referees for their important suggestions and Claudia Theune and Piers Dixon for their openness and help to improve this paper. References

Agnoletti, M. – Emanueli. 2016: Biocultural diversity in Europe. Springer. London/ New York. Aragón Ruano, Á. 2003: Gestión, uso y aprovechamiento de comunales y parzonerías en la Edad Moderna: el ejemplo de Entzia y Urbía, in: Pastor Díaz de Garayo, E. (ed.), Sortaldeko Lautada historian zehar: gaurko tresnez baliatuz, joandako denborak argitu. La Llanada oriental a través de la historia: claves desde el presente para comprender nuestro pasado. Diputación Foral de Álava. Vitoria-Gasteiz, 81‑91. Armiero, M. 2011: A rugged nation. Mountains and the making of modern Italy. Cambridge. The White Horse Press. Cameirana, A. 1970: La terraglia nera ad Albisola all’inizio dell’Ottocento, in: Atti del  III Convegno Internazionale della Ceramica. Centro Ligure per la Storia della Ceramica. Albisola, 61‑114. Cevasco, R. 2007: Memoria verde. Nuovi spazi per la geografia. Diabasis. Reggio Emilia.

66

Collis, J. – Pearce, M. – Nicolis, F. 2016: Summer farms: Seasonal exploitation of the uplands from prehistory to the present. J.R. Collis Publications. Sheffield. Congost i Colomer, R. – Lana Berasáin, M. 2007: Campos cerrados, debates abiertos. Análisis histórico de la propiedad en Europa (siglos  XVI-XIX). Universidad Pública de Navarra. Pamplona. Croce, G.F. – Montanari, C. – Moreno, D. 1992: Antiche praterie appenniniche, in: Maggi, R. (ed.), Archeologia preventiva lungo il percorso di un metanodotto. Quaderni della Soprintendenza Archeologica della Liguria 4, 159‑176. Dalglish, C. 2003: Rural society in the Age of Reason: an archaeology of the emergence of modern life in the southern Scottish Highlands. Plenum/Kluwer. New York. Gandiaga, B. – Ugalde, T. – Urteaga Artigas, M.M. 1989: Prospecciones arqueológicas en Urbía: yacimientos catalogados en las campañas de 1988 y 1989, Kobie, Serie Paleoantropología XVII, 123‑166. Gassiot Balbé, E. – García Casas, D. 2014: Histories d’ovelles i pastures. Arqueoiogia deis darrers segies de ramaderiaa l’alta muntanya, L’arqueologia del món modern i contemporani 78, 452‑470. Grendi, E. 1986: La pratica dei confini: Mioglia contro Sassello, 1715‑1745. Quaderni Storici 63, 811‑846. Grendi, E. 1993: Storia di una storia locale: l’esperienza ligure: 1792‑1992. Marsilio. Venezia. Grossi, P. 1977: Un altro modo di possedere: l’emersione di forme alternative di proprietà alla coscienza giuridica postunitaria. Giuffrè. Milano. Grove, A.T. – Rackham, O. 2001: The nature of Mediterranean Europe. An ecological history. Yale University Press. New Haven. Le Couedic, M. 2010: Les pratiques pastorales d’altitude dans une perspective ethnoarchéologique. Cabanes, troupeaux et territoires pastoraux pyrénéens de la préhistoire à nos jours. PhD thesis presented to Université François Rabelais de Tours. Maggi, R. – Montanari, C. – Moreno, D. 2002: Atti del Seminario Internazionale ‘L’approccio storico ambientale al patrimonio rurale delle aree protette’, Archeologia Postmedievale 6, 9‑214. Mientjes, A. – Pluciennik, M. – Giannitrapani, E. 2002: Archaeologies of recent rural Sicily and Sardinia: a comparative approach, Journal of Mediterranean Archaeology 15, 138‑166. Milanese, M. 1997: La ceramica postmedievale in Toscana. Documenti archeologici su produzione e consumo. All’Insegna del Giglio. Firenze.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Moneta, V. – Parola, C. 2014: Oltre la rinaturalizzacione. Studi di ecologia storica per la riqualificazione dei paesaggi rurali. Oltre Edizioni. Sestri Levante. Moreno, D. 1970: Per una storia della montagna ligure. Note sul paesaggio della Montagna di Fascia, Miscellanea Storica Ligure 2, 71‑135. Moreno, D. 1990: Dal documento al terreno. Storia e archeologia dei sistemi agro-silvo-pastorali. Il Mulino. Bologna. Moreno, D. – Montanari, C. – Stagno, A.M. – Molinari, C. 2010: A plea for a (new) environmental archaeology: the use of the geographical historical microanalytical approach in mountain areas of NW Italy, in: Tzortzis, S. – Delestre, X. (eds.), Archéologie de la montagne européenne, Errance. Condé-sur-Noireau, 75‑83. Rackham, O. 1976: Trees and woodland in the British landscape, J.M. Dent & Sons Ltd. London. Raggio, O. 2001: Immagini e verità. Pratiche sociali, fatti giuridici e tecniche cartografiche, Quaderni storici 108, 843‑876. Rendu, C. – Calastrenc, C. – Le Couédic, M. – Berdoy, A. 2016: Estives d’Ossau. 7000 ans de pastoralisme dans les Pyrénées. Le Pas d’Oiseau éditions. Toulouse. Rotherham, I.D. (ed.) 2013: Cultural severance and the environment. The ending of traditional and customary practice on commons and landscapes managed in common. Springer. Dordrecht. Stagno, A.M. 2016: Seasonal settlements and husbandry resources in Ligurian Apennines (17‑20 c.), in: Collis, J. – Pearce, M. – Nicolis, F. (eds.), Summer farms: seasonal exploitation of the uplands from prehistory to the present. J.R. Collis Publications. Sheffield, 67‑88. Stagno, A.M. 2018a: Short- and long-distance transhumant systems and common lands in post-classical archaeology. Case studies from southern Europe, in: Svensson, E. – Costello, E. (eds.), Historical archaeologies of transhumance across Europe. Routledge. London, 171‑186. Stagno, A.M. 2018b: Gli spazi dell’archeologia rurale. Risorse ambientali e insediamenti nell’Appennino ligure tra XV e XXI secolo. All’Insegna del Giglio. Firenze. Stagno, A.M. 2019: Investigating rural change. Legal access rights and changing lifestyles in rural mountain communities

(Ligurian Apennines, Italy, 16th-21st centuries), World Archaeology 51:2, 311‑327. Stagno, A.M. – Parola, C. – Beltrametti, G. 2018: Le charbonnage dans l’Apennin ligure (Italie): sites, pratiques, ressources (XIXème-XXème siècle), in: Paradis Grenouillet, S. – Burri, S. – Rouaud, R. (eds), Charbonnage, charbonniers, charbonnières. Presses universitaires d’Aix-en-Marseille. Aix-en-Marseille, 157‑172. Stagno, A.M. – Tejerizo García, C. – Echazarreta Gallego, A. – Santeramo, R. – Portillo, M. – Pescini, V. – Hernández Beloqui, B. 2020: De montes comunes y sociedades campesinas. Los resultados del proyecto ARCHIMEDE en el País Vasco, in: Grau Sologestoa, I. – Quirós Castillo, J.A. (eds.), Arqueología de la edad moderna en el País Vasco y su entorno. Archaeopress. Oxford, 165‑181. Stagno, A.M. – Tigrino, V. 2020: Borderline landscapes. Ligurian hillsides and shores between environmental history and archaeology (eighteenth to twenty-first centuries). Annali dell’Istituto storico italo-germanico in Trento, 46, 20‑54. Svensson, E. – Costello, E. 2018: Historical archaeologies of transhumance across Europe. Routledge. London. Thompson, E.P. 1993: Customs in common. Studies in traditional popular culture. The New Press. New York. Torre, A. 2011: Luoghi. La produzione di località in età moderna e contemporanea. Donzelli. Roma. Tuttitalia 2013a: Ne Statistiche della popolazione 1861‑2013. Published online at https://www.tuttitalia.it/liguria/28-ne/statistiche/ censimenti-popolazione/ (accessed 25. 01. 2020). Tuttitalia 2013b: Sori, Statistiche della popolazione 1861‑2013. Published online at https://www.tuttitalia.it/liguria/49-sori/statistiche/ censimenti-popolazione/ (accessed 25. 01. 2020). Ugalde, T. – Urteaga Artigas, M.M. – Gandiaga, B. 1992‑1993: Prospecciones arqueológicas en Urbía: yacimientos catalogados en las campañas de 1990 y 1991, Kobie, Serie Paleoantropología XX, 57‑85. Urzainki Mikeleiz, M.A. 2007: Parzonerías y parques naturales: comunidades de montes en Gipuzkoa: las Parzonerías. Universidad de Deusto. Donostia-San Sebastián.

Stagno

67

Transhumance in medieval Serbia Examples from the Pešter Plateau and northwestern slopes of the Prokletije Mountains

Uglješa Vojvodić*

Abstract

* Institute of Archaeology Knez Mihailova 35/IV 11000 Belgrade Serbia [email protected]

Extensive animal husbandry was one of the main economic branches in the medieval Serbian state. Its development was enhanced by the favourable geographic and climate characteristics of the region. An important role was also played by the availability of common pastures to all social groups up to the mid-14th century. These circumstances influenced the development of transhumant movements of livestock breeders – vlachs (as a social category) – from summer mountain pastures and winter abodes in river valleys and the Adriatic coast, where a favourable Mediterranean climate was predominant. Certain changes occurred in the 13th and first half of the 14th century, when Serbian rulers removed high-altitude pastures (planine) from the body of common lands collectively owned by villages and incorporated them into the monastery estates of their endowments. In addition to high-altitude pastures (planine), the monasteries were also given groups of vlachs, who thereby became tied to a certain territory. This was certainly a contributing factor in the founding of semi-sedentary livestock-breeding settlements at the foot of the designated pastures. A part of the population remained in these settlements over the summer and gradually took up agriculture. Although the question of vlachs and vlach settlements (katuni) has been thoroughly examined in Yugoslavian historiography and ethnography, unfortunately no relevant efforts have been made so far to discover the remains of winter and summer settlements used for animal husbandry and subject them to archaeological research. Over the last two years, smaller-scale archaeological surveys in the area of the Pešter and Sjenica Plateaus and the northwestern slopes of the Prokletije Mountains have been done. This research has shown in the example of the Pešter village of Gračane that archaeological methods can be used to determine the cattle-herding character of a medieval rural settlement. On the other hand, the survey of the slopes of Prokletije Mountains has supplemented previous knowledge of the organisation of summer pastures at high altitudes. Specifically, it has discovered the existence of clear boundaries between summer settlements as well as the parcelisation of space within a single katun between the constituent families. Keywords: medieval Serbia, common lands, semi-sedentary settlements, transhumance, vlachs, katun, sedentarisation.

In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 69-80.

69

Résumé

La transhumance sur le territoire de la Serbie médiévale – deux exemples : les hautes plaines du plateau de Pešter et la partie nord-ouest de la chaine des montagnes de Prokletije L’élevage extensive présentait une des activités commerciales fondamentales du système politique de la Serbie médiévale. Les spécificités géographiques et climatiques favorables de ce territoire influençaient son développement comme la disponibilité des pâturages communs à toutes les catégories sociales de la société jusqu’à la moitié du XIVe siècle. Ces circonstances exerçaient une influence importante sur le développement des déplacements saisonniers des éleveurs – valaques (comme catégorie sociale) – entre les pâturages montagneux d’été et les pâturages hivernaux dans les bassins fluviaux et la côte Adriatique, avec un climat méditerranéen favorable. Un changement important a eu lieu au XIIIe et au début du XIVe siècle quand les seigneurs serbes ont exclu les pâturages de haute montagne de la propriété commune, les offrant en cadeaux aux monastères sous leur patronat. A la même occasion, ils les offraient également à des groupes des valaques, les rattachant ainsi, autrefois nomades, à un terrain bien précis. Cela influençait sans doute la formation des communautés semi-sédentaires d’élevage (katuni) aux pieds des montagnes offertes. Une partie de la population valaque restait dans ces communautés pendant l’été et commençait avec la cultivation de la terre. Bien que la question des valaques et de leurs communautés d’élevage a été bien traité en détail dans l’historiographie et l’ethnographie yougoslave, les recherches archéologiques afin de retrouver les sites précises d’élevage hivernales et estivales font encore défauts. Or, pendant les deux dernières années, des prospections archéologiques pédestres ont été effectuées sur les champs de Sjenica et de Pešter ainsi que sur les versants nord-ouest de la chaine des montagnes de Prokletije. Ces prospections montrent, à l’exemple du village Gračane de Pešter, que cette méthode archéologique nous permet de déterminer les caractéristiques d’élevage d’un site rural médiéval. En plus, la prospection pédestre aux versants de Prokletije nous a enrichi des connaissances sur l’organisation des pâturages d’été à une grande altitude. On y a su déterminer l’existence des frontières très strictes entre les différentes communautés (katuni), et même une répartition exacte du terrain entre les familles appartenant à une communauté (katun). Mots-clés : Serbie médiévale, biens communaux, semisédentaires communautés, transhumance, valaques, katun, sédentarisation. Zusammenfassung

Transhumanz im mittelalterlichen Serbien – Beispiele aus der Pešter (Peschter) Hochebene und aus dem nordwestlichen Gebiet des Prokletije-Gebirges 70

Extensive Viehzucht gehörte zur wirtschaftlichen Grundlage im serbischen mittelalterlichen Staatswesen. Die Entwicklung wurde durch günstige geografische und klimatische Bedingungen in der Region gefördert. Eine bedeutende Rolle spielte auch der Zugang zu den gemeinsamen Weideflächen für alle sozialen Schichten der Gesellschaft im Zeitraum bis zur Mitte des 14.  Jahrhunderts. Diese Umstände führten zur Entwicklung von transhumanen Bewegungen der Viehzüchter – Walachen (als soziale Kategorie) – zwischen den Weideflächen im Gebirge während des Sommers und dem Winterlager in den Flussniederungen sowie dem adriatischen Küstengebiet, wo ein überwiegend günstiges mediterranes Klima herrschte. Zu gewissen Veränderungen kam es im Laufe des 13. und in der ersten Hälfte des 14. Jahrhunderts, als die Weideflächen im Hochgebirge von den serbischen Herrschern aus dem Gemeinbesitz ausgenommen und den im Klosterbesitz befindlichen Stiftungen geschenkt wurden. Zusammen mit den Regionen im Hochgebirge bekamen die Klöster Gruppen von Walachen zugeteilt, die man auf diese Weise an ein bestimmtes Gebiet gebunden hatte. Solches führte jedenfalls zu halbsesshaften sog. Katunen (katuni) als Gemeindeformen (comune) oder Gemeinschaften (compagnia) für Viehzucht in den vorgeschriebenen Gebirgsfußgegenden. Ein Teil der walachischen Bevölkerung verbrachte den Sommer in diesen Siedlungen und begann schrittweise Land zu bearbeiten. Trotz der en Detail erfassten Fragestellungen zu den Walachen und den walachischen Katunen in der jugoslawischen Geschichtsschreibung und Ethnografie, wurden bis dato keine bedeutenden Anstrengungen unternommen, um Reste der Viehzuchtsiedlungen in der Winter- und Sommerzeit aufzufinden und archäologisch zu untersuchen. In den vergangenen zwei Jahren wurden archäologische Surveys kleineren Umfangs auf dem Gebiet der Ebene um Sjenica und der Peschter Hochebene sowie der nordwestlichen Ausläufer des Prokletije-Gebirges unternommen. Diese Untersuchungen zeigen am Beispiel des Dorfes Gračane (Gratschane) auf der Peschter Hochebene, dass es archäologisch möglich ist, den durch Viehzucht geprägten Charakter einer ländlichen Siedlung im Mittelalter zu belegen. Andererseits ermöglichte die Begehung der Prokletije-Ausläufer eine Ergänzung der bestehenden Kenntnisse zur Organisation der im Hochgebirge befindlichen Weideflächen im Sommer. Es wurde nämlich zweifelsfrei festgestellt, dass unter den Katunen klare Grenzen herrschten, dass aber auch der Raum innerhalb eines Katuns unter den Familien, aus denen es gebildet wurde, aufgeteilt war. Schlagwörter: Serbischen mittelalterlichen Staatswesen, gemeinsamen Weideflächen, halbsesshafte Siedlungen, Transhumanz, Walachen, Katun, Sedentarisierung

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Introduction

Based on recent research, seasonal movements related to production, distribution, trade, etc. were carried out all across Europe and other parts of the world (Costello – Svensson 2018). Certain differences are observed based on the regional geographical and climate distinctions. In order to deal with the mining, cattle breeding, fishing, beekeeping, etc., transhumant movements were significant for the economy of medieval Serbia. Despite that fact, transhumant movements and temporary settlements associated with seasonal economic activities were outside the range of past archaeological research. Therefore, this chapter provides an overview of research conducted by historians and ethnologists, and presents an attempt to include previous and newly obtained results of archaeological surveys and excavations in the debate. Since the scale and frequency of all seasonal movements in medieval times is beyond the scope of this paper, the emphasis will be on the extensive cattle breeding and seasonal movements of sheepherders in the area of medieval Serbia. Previous research

Transhumance and the question of the professional herders – vlachs (vlasi in Serbian) – have been the main topic of several academic conferences and papers written by Yugoslavian historians and ethnographers. Although earlier scholars did outline the method and main course of research (Jiriček 1879; Novaković 1965; Dedijer 1914, 1916), a significant upturn was achieved during the multidisciplinary symposia held in Sarajevo in 1961 and 1973 (Sarajevo 1963; Sarajevo 1983) and in Belgrade in 1975 (Belgrade 1976). Significant recent research has also been conducted by historian Zef Mirdita (2004, 2009). Animal husbandry and vlachs in the medieval period and during Ottoman rule have recently been the topic of a local journal (Braničevski glasnik 2010). This previous research on transhumance was largely undertaken by historians. Based on comparative analyses of data provided by written sources preserved from medieval and Ottoman times, these studies were focused on the interpretations of legal provisions and prohibitions for vlachs and their rights and duties to the ruler. Though providing many new insights into the life of the vlachs and cattle breeding during the Middle Ages, this scholarship unfortunately remained purely paper based: field research was neglected, and no attempt was made to locate medieval seasonal settlements. The reduced scope of archaeological research of medieval rural communities and their settlements in Serbia also affected the investigation of these ‘marginal’ locations and poorly preserved remains of seasonal settlements. That was the result of unrepresentative and chronologically unspecific artifacts discovered in few excavated rural

houses. Another contributing factor to this situation were difficulties in the chronological identification of the remains of seasonal settlements, due to the chronological insensitivity of simple residential structures and the lack of movable archaeological evidence in them. More recent small-scale surveys, jointly undertaken by experts from the Institute of Archaeology in Belgrade and the Museum of Novi Pazar, represent some of the pioneering attempts to include archaeologists in the research on seasonal animal husbandry in this area. They will be presented in the following text, along with an overview of the results of research undertaken so far. Much significant ethnographic research was carried out in the 20th century. A particularly relevant part of this research was carried out in the regions where the tradition of transhumant pastoralism still survives. The most relevant are Milisav Lutovac’s written reports about 19th- and 20th-century livestock herding in the area of the Pešter and Sjenica Plateaus and the northwestern parts of the Prokletije Mountains (Lutovac 1933, 1962, 1973, 1977). Vlachs in medieval Serbia

The people predominantly engaged in animal husbandry in the territory of the medieval Serbia were named vlachs. Even though the debate about medieval vlachs in the Balkans is extensive and as such lies beyond the scope of this paper, it is of particular importance to note that the term vlach was also used for the Romanised inhabitants of the Balkans (Cvetković 2012, 20, 30; Maksimović 2017, 402). Some scholars argue that the origin of these nomad cattlemen is to be found in Roman and probably prehistoric times (Mirdita 2009; Gušić 1962). An examination of the modern calendar of herdsmen from the Prokletije Mountains conducted by Branimir Gušić showed that the traces of pagan rituals can be recognised in the customs of modern sheepherders (Gušić 1962). In view of that, it is important to highlight the significance of a gradual influx of a Slavic element into vlach groups during the Middle Ages (Mirdita 2009, 50; Kursar 2013, 117). The name vlach referred to all herdsmen over the course of the Middle Ages, regardless of their origin, as Stojan Novaković, one of the greatest Serbian historiographers emphasised (Novaković 1965, 29). This claim is generally accepted and has been verified based on the data contained in available written sources, mostly from preserved monastic charters from the 13th and 14th centuries, as well as other medieval documents that defined the rights and duties of vlachs. In other words, the term vlach was used during the Late Middle Ages to define a social category that, within the medieval Serbian society, meant cattlemen Vojvodić

71

Fig. 1. Map of the examined area (© Uglješa Vojvodić).

who lived a nomadic or semi-nomadic life (Maksimović 2017, 402). Their settlements, usually located at higher altitudes, were called katuns (katuni). During the 14th century, vlachs started to establish their katuns in the parish among agricultural villages, which will be discussed later. According to medieval written sources, in addition to their lifeways, they differed from farmers in terms of the taxes they paid and the duties they had towards their feudal lords (Mihaljčić 2006, 41‑42, 103‑104, 128, 131, 190‑192; Mišić 2010). Most of their duties were related to military service, the transfer of goods, and professional pasturing, and these stemmed from the existing internal social stratification of the vlach population (Mišić 2010). As we know from written sources, the vlachs had similar obligations to 72

the Byzantine Empire in the 10th and 11th centuries (Cvetković 2012, 30‑32). Positive preconditions for cattle breeding

The development of livestock breeding in the territory of medieval Serbia was facilitated by favourable geographical and climate conditions in the Balkans. A significant factor in that process was also the accessibility of common pastures to all social strata of the Serbian medieval society (Blagojević 1966, 4; Gušić 1976; Ršumović 1976). Based on the historical data, these common pastures existed before the middle of 14th century, positioned within the parish or on the surrounding mountains (Novaković 1965; Blagojević 1966, 5). As previous research has pointed out,

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

during the medieval period two different terms were used for grazing land. Pastures above 1000 m were called planine and those at lower altitudes were called pašnjaci (Jiriček 1952, 163; Blagojević 1966, 8‑15). Pastures, especially those in the mountains (planine), could be bestowed by the ruler upon his monastery endowments or upon noble lords (Dečanske hrisovulje 1976, 56). It seems that Serbian rulers excluded those pastures from common lands before donations. That process started at the turn of the 13th century, when the founder of the ruling Serbian dynasty Stefan Nemanja, with his son and successor Stefan the First-Crowned, donated two mountain pastures to the Hilandar Monastery (Blagojević 1966). During the 14th century, this practice became common. Following the example of their ancestors, Serbian rulers donated at least ten mountain pastures to newly founded monasteries (Blagojević 1966; Maksimović 2017, 403). The peak of this process is reflected in a stipulation recorded in the chrysobull of Dečani, which declared all planine property of the king (Blagojević 1966, 35‑36). After the said provision was enacted, common pastures at high altitudes could be used with the payment of a special toll – the so-called travnina. The possibility of accessing pastures was regulated in a similar way in other parts of medieval Europe (Stagno 2018). Livestock types

In higher-altitude zones, the vlachs were mainly engaged in sheep and goat herding. Based on the earlier ethnographic research in the area around the Prokletije Mountains, the breeds of sheep were moved seasonally (Lutovac 1933). In other words, herded sheep flocks differed between the areas of the Metohija valley, Lim valley, around Lake Plav and the Gusinje valley and Lake Skadar and the Adriatic coast (Fig. 1). For example, livestock, mostly sheep and goats, which arrived on the Adriatic coast from other areas quickly fell ill and died, since the breeds herded in these areas were not used to the type of grazing land and climate prevalent in the Adriatic (Lutovac 1933). At lower altitudes cows and cattle were more common, as were pigs in the zones of oak forest (Trifunoski 1963, 32‑33; Katić 1978). Based on historical sources and preserved toponyms, in the area of the Metohija Valley the breeding of buffalos was especially widespread (Lutovac 1933, 36‑37; Dečanske hrisovulje 1976). The obligations and duties of the vlachs, which included military service, transport of goods and salt, also affected the breeding of horses and mules. As is well known, the livestock bred by vlachs, previously bestowed by the Serbian rulers upon their monasteries, were prepared and served for the monastic community. As shown by recent archaeological studies on the animal bones discovered in the trash pits at the Studenica Monastery, the age structure of the animals

indicates that the young were less frequently used in the diet. Such a trend is especially representative of cattle, which were, according to the aforementioned results, primarily bred for milk, hauling, and ploughing (Marković 2015, 406‑407). Archaeozoological studies have also pointed out that all species of domestic animals were of a slightly smaller stature than in Roman times, which was also the case in other parts of Europe (Marković 2015). Transhumance types in medieval Serbia

On the basis of available data, several variants of the shepherds’ seasonal movements can be identified. In the earlier periods, up to the second half of the 14th century, shepherds under obligation to the Serbian rulers mostly lived as nomads, ascended with their flocks to altitudes between 1000 and 1600 m and stayed there in katuns, and then during the autumn they descended to the parish or remote coastal areas with a Mediterranean climate. That form of animal husbandry, practised in the area of medieval Serbia and the Balkans, could be defined as Braudel’s ‘inverse transhumance’, which implies movements of people based with their flocks in the high mountains down to overwinter at more-clement altitudes (Costello – Svensson 2018, 4). This model of transhumance practised in that way is also attested by information from historical sources, which report that the stockbreeders paid for keeping their livestock on the Adriatic coast and on the shores of Lake Skadar during the winter (Đurđev 1960, 13, 16; Pavković 1976). In other words, they did not have permanent settlements at a lower altitude. As in other parts of Europe, it is not possible to make a clear distinction between transhumance and shorter seasonal movements in medieval Serbia. According to 14th-century written sources, Serbian rulers bestowed high mountain pastures (planine) and vlach groups on their endowments (Maksimović 2017, 403). These vlachs had the duty of guarding and feeding the monastic herds. Written records also indicate that in that period the settlements of the vlachs – katuns, with clearly defined boundaries – were situated in the parish alongside agricultural villages (Miljković 2010, 19‑20). One part of the vlach population gradually began to engage in agriculture and hay harvesting, while the other, bigger part continued to make their seasonal movements. They still covered significant distances on their way to summer pastures but, before winter, they would return to the same village. This type of movements could be characterised as middlerange transhumance (Fernandez Mier – Tente 2018). Also, the sedentarisation of nomadic groups made tax collecting easier for the ruling class. The shepherds’ seasonal movements continued in the second half of 14th century and later on, though on a smaller scale. Vojvodić

73

In their transhumant or middle-range seasonal movements, the herdsmen drove their herds through sedentary villages. This is indicated by a provision in Emperor Dušan’s Code (1349), where village residents are required to provide a passage for livestock in their farmland (Blagojević 1966, 21‑23; Bubalo 2010, 90‑91, 176‑178). Similar pasture zones, up to 300 m wide, where ploughing for agriculture was not allowed and that served for the passage of livestock, are known in Italy – tratturo – and in Sicily – trazzera (Antonijević 1976, 61‑62). In addition to the aforementioned provision, another one in Emperor Dušan’s Code regulated the movement through sedentary villages: only one family from a given vlach group was allowed to lead their herds through those settlements at a time (Novaković 1965, 40; Bubalo 2010, 90‑91, 176‑178). In this way, an attempt was made to protect common pastures at lower altitudes from overgrazing. The existence of certain laws that controlled transhumant movement through agricultural rural areas suggests that there was a significant number of these movements during the 14th century. Beside transhumant movements, which could last for weeks, there were also shorter, everyday movements practiced by herders from agricultural villages. Those became common mostly after the pastures at high altitudes (planine) were excluded from common lands and attached to the large monastic estates during the 13th and first half of the 14th century. Restricted short-distance movements to the common pastures at lower altitudes certainly caused a shift in the economy of the parish farming villages in the form of herd reduction and an increase in hay-producing meadows inside village boundaries. Written records from the 14th century testify that meadows started to become common between ploughed fields in that period (Blagojević 1966, 52, 58). Location of herders’ settlements – katuns

The herdsmen’s summer settlements with pastures, huts, and dairies were called katuns (Đurđev 1963, 144‑145). Later studies of medieval historical sources, mainly the founding charters of monasteries, showed that under suitable circumstances katuns existed in the lower regions inside the parish, alongside farming villages (Ivanović 1955, 401‑402). Such cases are especially common in areas that were not suitable for agriculture, like hills and karst fields with a favourable climate. Based in a parish, these katuns had borders like other rural settlements. That was the case on the property of the St. Archangels Monastery recorded in King Milutin’s Charter from AD 1316, where the shepherds’ settlements were placed in the ‘flat part of the estate’ (Ivanović 1955). The locations of mountain katuns were primarily influenced by the presence of a drinking water source, 74

which was necessary for the life of the herd. Small plateaus below mountain peaks, moraine deposits, or undulating terrain were mainly chosen, with care being taken to ensure that the slope was protected from wind and had a sufficient amount of sunlight (Trifunoski 1963, 30). The katuns were made up of several families gathered around a chief or a leader. They erected sheepfolds and houses made of light material with a stone substructure (Filipović 1963b, 45). Short distances between these structures led to the formation of a kind of hamlet. The gathering of a few families was influenced by a number of factors such as kinship, personal insecurity, and other interests (Kojić 1958, 77; Trifunoski 1963, 31). Based on data provided by written records, katuns had 30 families on average. Previous studies did not provide significant data on the organisation of summer grazing lands, except for those ethnographic studies conducted on 20th-century seasonal settlements at high altitudes. The aforementioned small-scale surveys conducted in previous years indicate that summer settlements on the mountains had their boundaries. The natural contours in the relief were used as boundary objects, such as the sharp-crested serrate ridge (arête) in the example of the northwestern parts of the Prokletije Mountains, while the katuns were located in opposing valleys – cirques (Fig. 2a). However, there is also evidence of the existence of parcelisation within summer pastures. Drywall structures served as pasture fences with the purpose of dividing grazing space between families (Fig. 2b). Similar drywall constructions were built with the same purpose in other parts of Europe (Andres 2018). Territorialisation of medieval katuns

As previously emphasised, the favourable geographic and climate conditions of the Balkans, certain toll privileges compared to the farming population, as well as the rise of medieval Serbia, all had a huge influence on the development of transhumance in the 13th and 14th centuries. As we learn from written sources, the expansion of existing katuns and the emergence of new ones started from the middle of the 14th century (Kovačević 1963). The beginning of the transition from transhumance to the sedentary way of life started very soon after that expansion. This suggests that previously mentioned stipulation recorded in the chrysobull of Dečani about the exclusion of high-mountain pastures from common lands in favour of the vlachs’ being bestowed upon the monasteries might be the result of process of competition for common lands between farmers and cattle breeders (Hardin 1968). Known as territorialisation in Serbian medieval studies, this process began from the end of the 14th and middle of the 15th century (Trifunoski 1963, 36). Directed by the ruler to certain mountains where they could feed their

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 2. a) View of the katun remains and its borders – a sharp mountain ridge, and b) detail of previous photo showing the drywall borders within the katun (© Uglješa Vojvodić).

flocks, vlach shepherds started to overwinter in a nearby parish more often. That was especially the case in areas away from the sea, where grazing land could be found late into the autumn and the snow would quickly melt in the spring. In addition to the aforementioned regulations by which herdsmen were tied to specific mountains, the gradual disappearance of nomadic routes influenced by the political circumstances in the Adriatic coast and the arrival of Ottomans to the Balkans had an impact on the process of territorialisation. The Ottoman conquest caused the disappearance of previously established socioeconomical relationships, despite the attempt to preserve some of them. With the disappearance of large monasteries and other feudal estates, the bearers of animal husbandry in medieval Serbia, the obligations

of vlach groups to them also disappeared (Blagojević 1966, 92‑93). Although we agree with the statement that vlach groups and their genesis should be observed without the chronological dividing line of the Ottoman conquest (Miljković 2010, 8), an observation of the changes that occurred under Ottoman rule goes beyond the scope of this paper and exceeds its chronological framework. Nevertheless, for the study of the process of territorialisation of the medieval katuns in Serbia it is important to mention Ottoman cadastral censuses, which were created immediately after the conquest of these areas from AD 1455 to 1459. Particularly significant is the census of Isa-beg Isaković land property, in which two groups of vlachs are mentioned, one registered without a village, and the other registered Vojvodić

75

Fig. 3. a) Remains of the village of Gračane (© M. Popović 1993, Fig. 2), and b) position of the village of Gračane on a relief map (© Uglješa Vojvodić).

Fig. 4. Angular stones of the barn discovered during the field survey in 2017 (© Uglješa Vojvodić).

with villages and houses (Đurđev 1963, 148‑149). These data certainly show that semi-sedentary and nomadic groups of vlachs existed at the same time in medieval Serbia before the political changes. The medieval village of Gračani

Although the research on medieval rural settlements has so far been small in scope and overshadowed by the exploration of medieval fortifications, monastery complexes, and churches, in the early 1990s preliminary surveys of the village of Gračane near the Sopoćani Monastery were conducted (Fig. 3). The settlement remains are located at the very edge of the Pešter 76

Plateau, on the sunny hillside away from arable land suitable for agriculture. About 20 houses, a watering place, and the remains of one larger building on the central plateau, which is believed to have been a church, were recorded in this area (Popović 1993, 8‑11). House 2 and the church were the only structures subjected to archaeological investigation. Based on the portable archaeological material, house 2 was dated to the first half of the 15th century (Popović 1993, 11). The church was built in the period after the Ottoman conquest in the second half of the 15th century and was in ruins by the middle of the 16th century. Other archaeological findings of the excavations testify that the settlement lasted until the 17th century (Popović 1993). During

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 5. a) Orthophoto of the barn (© Uglješa Vojvodić), and b) Hypothetical reconstruction (© M. Drobnjaković).

a survey of the same area carried out in 2017, a few angular stones were also registered that are most likely to be the remains of a livestock barn (Figs 4, 5). Such structures, which were made of lightweight materials, were used in this area and in other parts of Serbia until recently (Petrović 1984, 84‑85; Findrik 1976). The aforementioned dating of the medieval remains of the village of Gračane is chronologically consistent with the process of transformation from transhumance to the territorialisation of shepherd communities and emergence of new vlach settlements before the Ottoman conquest (1455). The new examined data related to the Gračane settlement suggests that the positions of boundaries between the parish and karst fields, suitable

for grazing for almost the entire year, could have been the preferred positions of permanent herder settlements newly formed during the 14th and 15th century. The founding of settlements at locations like this enabled the herders to transition to short-range and mid-range seasonal movements. Conclusion

In view of the facts stated above, the examination of the question of vlachs as a social category within the medieval Serbian state was mainly based on data provided by available written sources. On the other hand, the number of field studies devoted to the remains of vlach Vojvodić

77

settlements and activities, especially archaeological ones, are disproportionately small. Still, the results of recent pioneering research introduced in this chapter are encouraging and indicate the fruitful potential from which the need for further exploration of this topic arises. Therefore, future archaeological research should be focused on non-destructive field surveys followed by the creation of a GIS database containing the details of locations and appearance of the remains of these seasonal settlements in the high mountains. The selection of sites suitable for archaeological excavations should be made over time, as well as adoption of modern archaeological methodology suitable with the requirements for such sites. References

Andres, B. 2018: Alpine settlement remains in the Bernese Alps (Switzerland) in medieval and modern times: the visibility of alpine summer farming activities in the archaeological record, in: Castello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe. Routledge. New York, 155‑170. Antonijević, D. 1976: Основна обележја номадских сточарских кретања у светлу обичајног права и позитивног законодавства, in: Belgrade 1976, 57‑66.

Belgrade 1976:

Дечанске хрисовуље, Ивић, П.  – Грковић, М. (ури.). Нови Сад.

Dedijer, J. 1914:

Сточарске зоне у планинама Динарске системе, Гласник Географског друштва, св. 3/4. Београд,

39‑57. Dedijer, J. 1916: La transhumance dans les pays dinariques, Annales de Géographie, tom. XXV, 347‑365. Đurđev, B. 1960: Novi podaci o najstarijoj istoriji brdskih plemena, Istorijski zapisi, god. XII, knj. XII, sv. 1, Istorijsko društvo Crne Gore, Titograd 1960, 3‑20. Đurđev, B. 1963: Територијализација катунске организације до краја XV века, у: Sarajevo 1963, 143‑169.

Fernandez Mier, M. – Tente, C. 2018: Transhumant herding system in Iberia, in: Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe. Routledge. New York, 219‑232. Filipović, М. 1963b: Структура и организација средњовековних катуна, у: Sarajevo 1963, 45‑112.

Findrik, R. 1976:

О архитектури сточарског стана, у: Belgrade 1976,

Чубриловић, В. (ур.), Одредбе позитивног законодавства и обичајног права о сезонским кретањима сточара у југоисточној Европи кроз векове, Зборник радова са међународног научног скупа одржаног 6. и 7. новембра 1975. године у Београду, Посебна издања Балканолошког института 4, Балканолошки институт САНУ. Београд.

335‑354. Gušić, B. 1976: Ekološki uslovi na prostoru transhumantnog stočarenja na Balkanu, у: Belgrade 1976, 143‑158. Hardin, G. 1968: The tragedy of the commons, Science 162, 1243‑1248. Ivanović, R. 1955:

Планине и пашњаци у средњовековној Србији,

Катунска насеља на манастирским властелинствима, Историјски часопис 16‑17,

Blagojević, M. 1966:

2‑3, 3‑95. Braničevski glasnik 2010: Браничевски гласник 7, Пожаревац. Bubalo, Đ. 2010: Историјски гласник

Душанов законик (ур. Ђ. Бубало). Завод за уџбенике. Београд.

Cvetković, M. 2012:

Укључивање Словена и Влаха у систем византијске провинцијске организације на југу Балкана до XI века.Сличности и разлике, Зборник радова Византолошког института 49, 19‑41.

Costello, E. – Svensson, E. 2018: Transhumant pastoralism in historic landscapes: beginning a European perspective, in: Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe. Routledge. New York, 1‑13.

78

Dečanske hrisovulje 1976:

399‑409. Jiriček, К. 1879: Die Wlachen und Maurowlachen in den Denkmälern von Ragusa, Sitzungsberichte der k. böhm. Gesellschaft der Wissenschaft, Prag. Jiriček, K. 1952: Историја Срба, превод: Ј. Радоњић, Београд. Katić, R. 1978: Сточарство средњовековне Србије, Београд. Kojić, B. 1958: Сеоска архитектура и руризам. Грађевинска књига. Београд.

Kursar, V. 2013: Being an Ottoman Vlach: on Vlach identity(ies), role and status in western parts of the Ottoman Balkans (15th-18th centuries), OTAM 34, 115‑161.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Lutovac, M. 1933:

Novaković, S. 1965:

Сточарство на североисточним Проклетијама, Посебна издања Географског друштва 14, Београд.

Lutovac, M. 1962: Iz stočarskog života u nekim našim krajevima, Zbornik za narodni život i običaje 40, 321‑329. Lutovac, M. 1973:

Породичне задруге на Пештеру у старој Рашкој, Гласник Етнографског института САНУ XIX-XX

(1970‑1971), 1‑8. Lutovac, M. 1977:

Сточарство и сточарски живот на ПештерскоСјеничкој висоравни, Зборник радова Географског института „Јован Цвијић“ 29, 157‑170.

Maksimović, M. 2017:

Власи у склопу регалних права српских владара, у: Мишић, С. (ур.), Српска краљевства у средњем веку, Зборник радова са међународног научног скупа одржаног од 15. до 17. септембра 2017. године у Краљеву, у част обележавања 800 година од крунисања Стефана Немањића (Првовенчаног). Краљево, 401‑408.

Marković, N. 2015:

Исхрана у манастиру Студеници: археозоолошка истраживања, у: Марко Поповић, Манастир Студеница, Београд, 395‑405.

Mihaljčić, R. 2006:

Закони у старим српским исправама: правни прописи, преводи, уводни текстови и објашњења, Београд: САНУ.

Miljković, E. 2010:

Власи у домаћој историографији (1960‑2010), Браничевски гласник 7, 5‑22.

Mirdita, Z. 2004: Vlasi u historiografiji, Zagreb. Mirdita, Z. 2009: Vlasi: starobalkanski narod, Zagreb. Mišić, S. 2010: Законске одредбе о Власима Немањића, Браничевски гласник

Село (ур. С.Ћирковић), Српска књижевна задруга. Београд.

Pavković, N. 1976:

Обичајно-правни уговори и производни односи у сточарству, in: Belgrade 1976, 133‑142.

Petrović, P.Ž. 1984:

Рашка. Антропогеографска проучавања. Етнографски институт САНУ. Београд.

Popović, M. 1993:

Околина манастира Сопоћани. Остаци средњовековних насеља, некропола и цркви, Новопазарски зборник 17, Нови Пазар, 7‑24.

Ršumović, R. 1976:

Географски услови и њихова улога у еволуцији сточарских кретања у Југославији, у: Belgrade

1976, 159‑176. Stagno, A.M. 2018: Short- and long-distance transhumant systems and the commons in post-classical archaeology: case studies from southern Europe, in: Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe, New York, 171‑186. Sarajevo 1963: Филиповић, М. (ур.), Симпозијум о средњовековном катуну одржан 24. и 25. новембра 1961. године, Посебна издања, књ. II, Одељење историјско – филолошких наука, књ. 1, Научно друштво СР БиХ, Сарајево.

Sarajevo 1983:

Филиповић, Н. (ур.),

Simposijum Vlasi u XV i XVI vijeku održan u Sarajevu 13‑16. novembra 1973. godine, Radovi ANU BIH LXXIII, Odeljenje društvenih nauka knjiga 22, Sarajevo, 73‑177. Trifunoski, J. 1963: Географске карактеристике средњовековних влашких катуна, у: Sarajevo 1963, 19‑39.

у

повељама

7, 36‑48.

Vojvodić

79

Archaeology of the commons: seasonal settlements in the Cantabrian Mountains Margarita Fernández Mier* and Pablo López Gómez**

Abstract

The Cantabrian Mountains are an outstanding setting for the study of the history of the commons. Large pasture areas used collectively by the community are under the management of different administrative bodies, from the villages to the state, including management boards and municipalities. The material culture associated with the pasture areas has rarely been studied by archaeology. We present the initial results of a research project carried out in two high-altitude areas using a multidisciplinary approach. We give a particular emphasis to the analysis of the written sources produced by the aforementioned entities – which enlighten us on the complex ownership structure – and to the archaeological information that allows us to delve deeper into the antiquity of the commons and forms of its usage. Keywords: commons lands, Cantabrian Mountains, multidisciplinarity, local community. Résumé

* University of Oviedo Campus de Humanidades “El Milán” C/ Amparo Pedregal s/n 33011 Oviedo, Asturias, Spain [email protected] ** University of León Campus de Vegazana s/n 24071 León, Spain [email protected]

Archéologie des communautés : établissements saisonniers dans les montagnes cantabriques La cordillère cantabrique est un observatoire privilégié pour l’étude de l’histoire des communautés. Différentes entités administratives, depuis les villages jusqu’à l’État, en passant par les Conseils d’administration ou municipalités, gèrent actuellement de vastes zones de pâturage utilisées collectivement. On y associe une culture matérielle qui a à peine été étudiée archéologiquement. Dans ce travail, nous présentons les premiers résultats du projet de recherche que nous sommes en train de réaliser dans deux régions montagneuses avec une approche multidisciplinaire. Nous accordons une importance particulière à l’analyse de la documentation écrite produite par les entités susmentionnées, illustrant les formes complexes de propriétés et à l‘information archéologique qui nous permet d’approfondir son antiquité et les modes d‘utilisation  Mots-clés : terres communes, Montagnes Cantabriques, multidisciplinarité, communauté locale.

In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 81-92.

81

Zusammenfassung

Archäologie des Gemeingutes: Saisonale Siedlungen im Kantabrischen Gebirge Das Kantabrische Gebirge ist ein hervorragendes Fallbeispiel für das Studium der Geschichte der kommunalen Selbstverwaltung. Verschiedene Verwaltungseinheiten, von den Dörfern bis zum Staat, einschließlich Verwaltungsräte bzw. Gemeinden, bewirtschafteten große Flächen gemeinsam genutzten Weidelands. Sie sind mit einer materiellen Kultur verbunden, die archäologisch kaum erforscht ist. In diesem Beitrag stellen wir die ersten

Introduction

The extensive bibliography regarding the study of medieval settlement in Europe reveals the complex organisation of the production spaces of medieval villages (Williamson 2012; Rippon 2008), the multifunctionality of their components (Fernández Mier 2010), the importance of collective practices in the performance of daily activities, and the relevant role played by areas of communal use (Oosthuizen 2013). Such complexity has been studied both through the archival records and archaeology, and its historiography has gradually expanded the themes and fields of study from living spaces to ones of production, including areas of communal use, mainly forest areas and pastures. The study of medieval Europe settlement has focused in recent years on deserted villages (Hamerow 2002, 2012; Quirós Castillo 2009). The data provided by consulting archaeology and research projects are of empirical nature and can be used to develop models of medieval settlement networks. However, a fair amount of small- and mediumsized present-day settlements are medieval in origin. A number of issues related to archaeological practice have hindered the study of these settlements and conditioned the interpretation of medieval settlement networks based on data from settlements either abandoned for some reason in the medieval or post-medieval period or that otherwise have become modern towns of large size. This type of settlement has not received much archaeological attention; the subject has been addressed in a recent stateof-the-art study that reviews this archaeological topic in different European countries (Fernández Fernández – Fernández Mier 2019). The research carried out by our team over the last decade on the Cantabrian Mountains (in the northwest Iberian Peninsula) was part of a project that used two currently inhabited villages in a mid-mountain area of Asturias as a laboratory for methodological experimentation. The objective was to understand the origin of the settlement network and the process of the transformation of the area. In order to achieve this objective, some archaeological work was arranged 82

Ergebnisse des Forschungsprojekts vor, das wir in zwei altmontanen Gebieten mit einem multidisziplinären Ansatz durchführen. Besondere Bedeutung messen wir der Analyse der schriftlichen Dokumentation bei, welche von den oben genannten Körperschaften erstellt wurde und die komplexen Eigentumsformen veranschaulicht, sowie den archäologischen Daten, die es uns ermöglichen, tiefer die Ursprünge und die Formen der Ausbeutung zu untersuchen. Schlagwörter: Gemeingut, Kantabrisches Gebirge, Multidisziplinarität, lokale Gemeinschaft.

to study the territory using a holistic approach, from living areas to pastures, including farmlands (Fernández Fernández – Fernández Mier 2019; Fernández Fernández 2017). In mountain villages, an important part of the inhabitants’ activity has been connected with livestock since prehistory and still is in the present. A series of known settlements used seasonally for livestock present a wide typology. Some of them are currently in use; others were in use until the mid-20th century. These areas are marked by collective forms of management, already well documented in the post-medieval period and even as far back as the early Middle Ages, as seen in other areas of the northern portion of the Iberian Peninsula, such as the Pyrenees and the Basque Country (Rendu 2003; Fernández Mier – Quirós Castillo 2015; Gassiot – Pèlachs 2017; Palet et al. 2019). Therefore, we believe that the study of seasonal settlements is always connected with that of stable settlements. We try to understand how both realities complement each other, taking into account the different scales of action that allow us to understand how they work. The different types of settlements and constructions found in communal areas belong to different chronologies, but they also have multiple functions in relation to a wide variety of land management, such as the types of multi-scale territorial transhumance. It should be borne in mind that, although cattle raising is the most important activity in these areas, agriculture has been present, at least since the 16th century, and especially since the 19th century. Furthermore, the commons are characterised by various forms of land ownership and usage rights on the part of diverse social groups who held a wide variety of rights, all of which complicate their study and the decoding of their meaning. This is why our research on seasonal use areas is related to the study of commons management. In order to understand its meaning, we must relate them to the forms of ownership, management, and exploitation of the large pasture areas in which they are located. The complexity

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

of ownership structures overlapping communal pastures and the customary rights of local communities has to do with the location, typology, and function of seasonal settlements. The different strategies adopted by the peasantry to preserve their rights vis-à-vis monasteries, secular lords, or the state and the internal tensions derived from inequalities within communities help us to understand the meaning of the brañas and the majadas. Ethnographic and anthropological studies are the basic tools here in a methodological sense. The continued existence of traditional land management systems until very recently provides a great volume of information about ownership structure and usage by the different social actors. This is the starting point to understand the forms of exploitation in all their complexity, and the conflicts generated by their control. It is necessary to confirm verify the moral economy of local communities (Scott 1976) and their forms of resistance (Scott 1985) at the present time, even if they are only marginal realities now, and to identify the parallels found in written sources and archaeological data. This view of resilient local communities able to maintain or recover ownership over spaces they have managed thanks to customary practices since the Middle Ages (Thompson 1991) provides a complex conceptualisation of meaning, uses, and chronology of seasonal settlements and their relationship with different social groups. The numerous studies addressing the legislation that governs commons in the last two decades (e.g. De Moor et al. 2002; De Keyzer 2018) have rarely questioned how the people of those communities perceive their usage rights of the commons, both collectively and individually or the diverse governance forms employed by the peasantry after different historical problems (Montesinos – Campanera 2017). It is necessary to reflect on concepts of ownership, not only those imposed after liberal reforms, but also those still in the imagination of the peasantry linked to customary uses that have resulted in many conflicts over the management of the commons and can be decoded with a contextualised reading, both of the legislation and of the abundance of lawsuits at certain historical moments (Congost 2002). It is necessary to keep in mind the friction that may exist between the law, the agrarian practices, and the legal rights depicted in written sources on the one hand, and tradition – that is, the management modes implemented by the peasantry – on the other. These are based on day-to-day practices and an intensive experience of the environment, which yields detailed empirical knowledge (geographical, climatic, and soil type conditions), important for informed decision making about exploitation and management, and also imbued with a symbolic significance that fills this place with the identity and past of the group. The combination

of both creates the ‘local landscape’ as such, a construct that can only be understood within the community. The last generations of livestock farmers who used these seasonal settlements called brañas or majadas in the Cantabrian Mountains hoard a detailed knowledge of the local area that is about to disappear concerning toponymy, the locations of old abandoned buildings, the signs that delimit the pastures belonging to the local communities, the pantries (ocheras) for cooling of the milk, the caves where cheese was made, the small hermitages that reflect the interests of the church in these areas, and the terrace systems for cultivation. All of these elements are increasingly difficult to find through archaeological survey due to the advance of the wilderness and the lack of use. While collective memory has held onto them up to the present, they will soon disappear due to the current process of abandonment of rural areas; they are the mandatory starting point to study the use of mountain areas in some parts of southern Europe. Long-term studies are a second methodological aspect to consider. Many of these livestock seasonal establishments are early medieval in origin, as seen in the Pyrenees and the Basque Country (Rendu 2003; Mujika Alustiza et al. 2013; Fernández Mier – Quirós Castillo 2015; Gassiot – Pèlachs 2017; Palet et al. 2019), used more or less frequently at different times in the past. But along with this consolidation process of seasonal settlements, there are equally interesting changes in subsequent periods related to specific historical moments of relevant ruptures. The commons have been key for the economy of the peasant communities since the Middle Ages, but some changes arising at certain times due to diverse historical factors that are reflected in the management of the commons – the territorial reorganisation associated with the establishment of a monastic community (Fernández Mier 2020), the consolidation of municipalities from the 13th century (Monsalvo Antón 2001), the strengthening of local communities from the 16th century (Rubio Pérez 1993), and the transformation that starts in the 18th century with the advent of liberal thought (Lana Berasaín 2018) – are reflected in written sources and the material records and they must be studied based on case studies analysed micro-territorially. Seasonal settlements in the Cantabrian Mountains: case studies

The Cantabrian Mountains or Cordillera Cantábrica are a mountain range running parallel to the Cantabrian coast in northern Spain. Its highest peaks reach over 2600 m. A particular feature of this range is that, at some points, it runs very close to the Cantabrian Sea, with its northern slopes forming the coastline. The Cantabrian Mountains Fernández Mier and López Gómez

83

Fig. 1. Map of the northwest of the Iberian Peninsula with archaeological sites researched from the agrarian archaeology perspective (© Pablo López Gómez).

are bordered on the south by the central plateau of the Iberian Peninsula, so they form both a geographical limit and a communicating area between different regions of northwest Iberia. We can make an initial classification of the seasonal settlements of the Cantabrian Mountains associated with the different types of transhumance recorded from the late Middle Ages up to the present. Each type of transhumance has a different scale and all of them are practiced by different social groups who clash over the management of the commons: 1) the valley transhumance, which developed within the territory of the villages; 2) the medium-distance transhumance between the coast and the high mountains, associated with marginalised social groups; and 3) the long-haul transhumance, La Mesta, composed of sheep herds in the hands of large owners who have led them since the 13th century from the south to the north of the Iberian Peninsula (Fernández Mier – Tente 2018). We are currently focusing our investigation on the settlements associated with the valley transhumance that developed within the territories assigned to the villages. The traditional model of land usage is marked by the annual cycle of use of the different agroecological niches: the cattle graze in the meadows closest to the living areas during winter, and when spring arrives, they gradually climb to use the upper pastures during summer, where different buildings (brañas, mayaos, cabanos) are built for livestock care and sheltering (González Álvarez et al. 2016). The highland areas are coveted by more than one village and the usual conflicts over the limits of usage of 84

these areas arise, sometimes resulting in the defining of areas of joint use by several villages (facerías) (Fig. 1). The Cueiru range

The Cueiru range is a meeting point/boundary between the municipalities of Teverga, Somiedu, Miranda, and Grau, featuring wide plateau highlands crossed by the Camín de La Mesa, a well-known historical route linking the northern Iberian Peninsula with the central plateau. The large number of archaeological sites dating from prehistoric times found here has to do with cattle drives, which were being reused as communication paths by the time this area was incorporated into the Roman Empire (Mañana Vázquez 2011). In the first part of the research we focused on the study of pasture areas of two villages the properties of which come together in this place; the towns of Taxa in the municipality of Teverga and Montoubu in the municipality of Miranda. Taxa is a village located on the eastern slope of the Cueiru range at an altitude of 1000 m. The village manages large pasture areas. Its constant lawsuits with neighbouring villages and council governments provide invaluable data to understand the ownership structure and usage rights of this area, dating back to the 16th century. The pastures are currently owned by the Neighbourhood Council, a governance institution with greater legal status than a village. The local community of Taxa has maintained the right of communal use of grazing land from the late Middle Ages to the present: from the 16th

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 2. View of the Veiga’l Prau (Taxa, Asturias) (© Pablo López Gómez).

to the 19th century the community had the right of use, despite the lordly control exercised over the pastures. In the 19th century, grazing lands became dependent on the state after the Spanish liberal confiscations; from that moment on, the local community acquired their property through various purchases from the state, becoming the owner: the pasture areas belong to the community collectively and are managed according to rules established by neighbours – a common pool institution, according to the terminology of Ostrom (1990). We have prospected two large pasture areas belonging to Taxa: Veiga de Cueiru and Veiga’l Prau (Fig. 2). Both are currently under semi-collective management: each neighbour has one or more plots in each of the meadows. The plots are separated by landmarks (mojones), small stones stuck in the ground that lay out the perimeter of the plots. The usage of the pasture was mixed: the plots were managed by each neighbour for the production of hay. But once the hay was mowed and stored at the beginning of the summer, the whole pasture – all the plots – was subjected to the derrota regime: the herds, regardless of size, of all neighbours grazed together throughout the area. Beyond the parcelled meadows, there are wide areas of lower-quality pastures, which can be used by all the neighbours. Nevertheless, a certain schedule must be

followed and observed by all of them. In these areas are located the facerías of forest and pasture on the border between two villages. Owned by one of the villages, the facerías can be used and exploited by the people from the other under certain conditions: cattle can cross the boundaries between villages freely, provided the cattle are gathered up for the night; or cattle have to spend the night in the territories or brañas of each locality. In addition to these stipulations, they also have to observe the rules of boundaries that regulate the pastures. Archaeological survey of this large territory has recorded 129 abandoned or half-ruined cabins. All of them share the same typology: a rectangular floor of small dimensions (2 x 3 m), built with a drystone system, made of limestone and mountain quartzite and covered with a gabled roof of curved tiles. We have identified in the field the quarries that supplied the stone and the remains of a teyera or furnace to make the tiles (Fig. 3). The cabins were used to store hay, keep young cattle, and house the shepherds if necessary. Each of these constructions is linked to a plot and was owned and used exclusively by the plot owner. The cabins can be found scattered throughout the wide pastures. They were in use until the 20th century and their abandonment was seemingly due to the disbandment of the rural environment by the Fernández Mier and López Gómez

85

Fig. 3. View of the ruins of the furnace for making the tiles (Taxa, Asturias) (© Belén Iglesias Martínez).

middle of the last century and the construction of access roads from the villages, which made it possible to reach the pastures in mechanical vehicles, reducing the travel time from the village, so that farmers did not need to spend the night in the high-altitude pastures. The large pastures managed by Taxa are an element of great cohesion for the community. Its inhabitants have litigated with their neighbours for centuries, and they have established within the community usage rules for their territories where the collective management of commons is inextricably linked to private property and private rights, something necessary to understand that collective management. The community wrote ordinances organizing commons usage and, in the 18th century, laid out the commons’ territory by adding visible landmarks and engraving marks (crosses) in areas of greater geographic uncertainty to avoid conflicts with neighbouring villages. These marks remain in the collective memory of the inhabitants and are still recognisable today. Indeed, in 2018 they were used to win a lawsuit brought by the village against the government of the Principality of Asturias for the control of these areas. The rules for pasture usage are defined by the village community based on a semi-collective usage system, according to which the large pasture areas are divided to plots assigned to each of the 86

neighbours for the production of hay, but are later subject to communal use for grazing during the summer months. The ownership concept that has arisen is one of private management within joint administrative frameworks obeyed by the entire community. It is a good example of an institution that has managed to protect the community’s common resources, demarcating the community of users and excluding outsiders and favouring the use of those resources by all the neighbours; we can say that it is a successful common pool institution (De Keyzer 2018). No archaeological work has been carried out in any constructions associated with this process, but the records seem to indicate that there was an important transformation in the 18th century, probably associated with an increase in the livestock herd pressure on the land. At this time new management and ownership configurations appeared under the influence of the individualistic liberalism that led to significant changes in the forms of collective management and in turn on many occasions to the strengthening of the common pool institutions. The other village, Montoubu (Miranda), is located on the western slope of the same range at an altitude of 700 m. Its pastures and forest areas were purchased by the Neighbourhood Council from the state over the course of

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

several transactions during the 19th and 20th centuries. Just like Taxa, Montoubu possesses ample pastures located on the western slope of the Cueiro range. We have focused the first archaeological work on the seasonal settlement of Los Fuexos, located in a large communal area and used collectively by all the inhabitants of the village – although there are some private properties at the edge of this pasture area. The braña is located on a small terracing near three springs of water. A cabin is assigned to each inhabitant of the village to shelter their cattle and the person who has to travel daily during summer to attend to it. The cabins, all located in the same area, are arranged in three groups. There are also specific buildings, bel.lares, intended to house exclusively the youngest animals. There are yards linked to the cabins and bel.lares, some private ones attached to the cabins and other collective ones in the middle of them. Other buildings for cooling milk milked at night, ol.leras, are recurrent in these areas. The survey led to the identification of 30 cabins, 5 yards, and 3 ol.leras. The typology and sizes of the cabins vary widely from small circular structures (2 m in diameter) to square and rectangular constructions ranging in length from 2 m to 20 m. Their state of preservation also varies: 4 cabins were in optimum condition and the rest in different stages of ruin. The variety in typology and states of preservation has to do with the extensive span of use of this seasonal settlement. We have carried out archaeological work on 2 structures, C23 and C18, of this seasonal settlement, although at present the results are of little use. Cabin C23 is located in the highest part of the braña, a place called Pimpano, in a small flat area on a mid-slope. It is close to the old road that goes up from the braña to the Campa de Cueiru, where the boundaries of three municipalities came together, forming a meeting point for all the neighbours of the surrounding villages, especially during the fairs that were held there at the end of the summer; an old small inn and a hermitage were located here. The archaeological excavation covered 9 m2, an area of cobbled ground with small edges of less than 20 cm directly arranged over the rock bed. No archaeological remains were recovered, because the area is mostly worn by erosive agents; being a very steep slope, the effect of rain and snow has been so damaging that the construction has been mostly wiped out. Some locals still remember more-visible remains of the cabin and another attachment, which were swept away in the 1950s by a heavy snowfall that remains in the collective memory of the region for its destructive effects. The second excavated structure, C18 (Fig. 4), was chosen for its advanced state of decay and because local memory no longer remembers its owner’s identity, which might indicate it is far older than others used until recently and still linked to different houses of the village. The archaeological excavation covered a total of 20 m2 and,

Fig. 4. General view of the archaeological excavation of cabin C18 (© Pablo López Gómez).

as in C23, the complex post-depositional processes that affected the building were documented. Located on a small terrace occupied by scree comprised of an accumulation of rocks detached from the top of the hillside, C18 was built using the largest rocks, like many other constructions of this braña. It is surrounded by other cabins, yards, and bel. lares, some of which are located at a slightly higher level. The latest level of use was covered by the collapse of the side walls and a large amount of materials eroded from the slope, including medieval and modern pottery, abundant remains of fauna, and metallic pieces coming from constructions located above the excavated cabin. Beneath the levels of hillwash and collapse the remains of a hearth were found under the shelter of a great rock used as a foundation, including remains of animal bone (bovids) and coals of the last fire dated around the 16th century, the estimated time that the building ceased to be in use. The grazing areas of Taxa and Montoubu are adjacent and are administered by the neighbourhood councils. The processes of their acquisition were similar; however, the usage and ownership of these large areas is very different, due to the diverse perceptions of property and management of each community. While in Taxa most of the grazing areas are divided up and distributed among neighbours, with the buildings for livestock scattered over a wide area, in Montoubu communal use predominates and the cabins are concentrated in a very closely defined space. This is why we feel it is necessary to elaborate on the case studies, avoiding false generalisations out of the commons legislation. Fernández Mier and López Gómez

87

Fig. 5. Map of Buxana’s braña (Aramo Range, Asturias) (© Pablo López Gómez).

The Aramo Range. The mountain pass of Andrúas

The mountain passes of Andrúas are an important area of highland pastures located on the western slopes of the L’Aramo range at an altitude of between 1000 and 1500 m. They are currently part of Public Utility Mounts No. 260 ‘Mountain Pass of Andrúas, La Guariza and Siblo Valley’, belonging to the municipality of Quirós. This area is shared jointly (facería) by several localities of the municipalities of Quirós, Proaza, and Santu Adrianu. Although the property owner is the municipality of Quirós, the livestock usage rights jointly benefit the inhabitants of the local communities of Bermiego, Tene, Aciera, Pirueño, Pedroveya, and La Rebollá in the municipality of Quirós; Serandi, Samartin, and Villamexin in Proaza; and Villanueva in Santu Adrianu (see Fig. 1). A 1960 lawsuit provides rich information on the usage rights of these pastures, with data that go back to the 16th century, enough to understand the long-term management problems better. A written source from the late 16th century provides relevant information we can use to ascertain the continuous attempts to build enclosures and constructions in communal areas, a process that can be seen not only since the 18th century, with the establishment of liberal ideas, but as far back as the late 88

Middle Ages. Indeed, in this document Alvaro Pérez de Tene is forced to return an area in Andrúas where he had built a house and enclosed some land for his use. In the 17th century, a lawsuit filed for the usage of these pastures collects the ordinances and regulations for joint usage by the different villages during the summer months (from May to September), as well as their rights to build cabins and pens (bel.lares), and the constructions’ characteristics: they have to be built with the drystone system and without locks on the doors. In this case it is possible to identify a more inclusive type of communally managed resource, in which the property is linked with one community that conferred access on other communities in a geographical area under strict rules of use. Two areas with livestock structures have been identified in the field survey, the braña of Buxana (Fig. 5), and the braña of Andruxas, although modern sources allude to 5 other areas with constructions yet to be found. In Buxana, there is currently only 1 building standing, the communal cabin of Villanueva, a quadrangular structure covered by a corbel arch and a green roof, the building typology of which seemingly dates back to the 18th century. Four other ruined square-plan buildings may be seen, as well as the foundation of a rectangular structure. There is also a large yard built over rocks and terrain features.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

There are three separate areas in the braña of Andruxas. In the first, we can see 2 nearly ruined buildings that are currently identified as the cabins of the villages of Villamexin and Aciera. The remains of 6 other totally ruined cabins can be found in the same area. Following a small stream, we find 3 other circular structures each with an approximate diameter of 1.5 m. To the north, on the other side of the stream, there is another concentration of 5 structures, all of them ruined, 2 of circular layout and 3 square. Due to the diversity of types and sizes and the absence of material remains on the surface, the functionality or chronology cannot be determined. Conclusions

Written documentation, mainly lawsuits between different social groups, and some ethnographic information can be used to define a complex framework of ownership structure and usage of the seasonal settlements of these small villages from the 16th century to the present. After stating the problem and setting out the hypothesis, it is necessary to go further with the archaeological excavations in order to better understand the usage of the seasonal settlements in the Middle Ages. We must deal with this research by taking into account the extensive information from later centuries. We have seen how different forms of transhumance are linked to different types of brañas; but if we analyse small village pasture usage from the smaller territories’ perspective, we find they are diverse, not only in their uses but also in their ownership, linked to different types of brañas, which can be identified by the arrangement of constructions and their relationship with the landscape and the territory, but otherwise similar from a construction point of view. We must go more deeply into the chronology of these cattle-raising constructions and into the archaeobiological study of them, which are just now being undertaken in our area of study. In other areas of the Cantabrian Mountains and in the Pyrenees, there is some evidence of such constructions having been used since the early Middle Ages and of significant pressure on the forest to create grazing areas throughout the Middle Ages (Rendu 2003; Fernández Mier – Quirós Castillo 2015; Gassiot – Pèlachs 2017; Palet et al. 2019). Therefore, we need to carry out lengthy studies to better understand the complexity of their formation process. In addition, the matter of mountain agriculture should be addressed through the study of terraced systems. Farming terraces have been documented in grazing areas of the Cantabrian Mountains, the Pyrenees, and several areas in Italy (Stagno 2016, 2019; Giovannetti 2004), which shows the multifunctionality of the territory (Fernández Mier 2010). From the 16th century on, the written documentation mentions the ploughing up of farmland in

grazing areas, often at the direction of lay or ecclesiastical lords; the pasture of the communities was thereby reduced, forcing the inhabitants to try to defend themselves from such pressure. The creation of farmland in the 16th century is related to the lords’ interest in increasing cereal production for the market. This process intensified in the 19th century, when potatoes were introduced. This food item grew well in mountain areas and soon became very important for the population. In addition, its cultivation was supplemented with the exploitation of the pasture land by cattle raising (García Fernández 1988). The potato crop is the reason for the large number of terrace systems across the Cantabrian Mountains, which, however, we have not documented archaeologically in any of the three micro-territories that we have studied. We have only found a few oral references to terraces in the Andrúas area and have to locate them. This process favours the privatisation of the commons, so that it must be common practice to have the rights and limits for the creation of farmland in grazing areas recorded in the bylaws. One of the most interesting aspects is the complexity of the forms of ownership, the relation between individual and collective property, and how material culture needs to be analysed together with the written documentation in order to understand these processes. The three case studies we have presented show great variability in ownership form and management. First, in the Aramo range we find an inclusive managerial regime: a grazing area belongs to a small village, but several nearby villages have grazing rights to it, giving rise to the drafting of some complex ordinances that regulate its use. In the bylaws to these ordinances, some brañas with cabins are documented, which are used by all the inhabitants of the villages (one cabin for each village). Second, in Braña Los Fuexos there is an exclusive regime of pasture: only the village members have access and they use this space collectively, with the brañas, dating back to the 17th century, having cabins that belong to each of the domestic units of the village. Finally, in the Cueiro range pasture fields are fully parcelled and used semi-collectively, with cabins within each plot of land. We have seen how the management of the common areas defines different types of communities (a small village or a group of them) with the common interest of defending the commons they manage. This gives them some cohesion and identity in relation to other communities or individuals who operate at a supracommunal level; the rural council is the guarantor of the defense and management of the commons. Nevertheless, this does not exclude internal tension or conflicting interests as a result of the contradictions between the maintenance of common land against advancing private ownership; the former type of ownership makes sense within the community and must not be interpreted considering the current private ownership concept, but Fernández Mier and López Gómez

89

within the framework of other cohesive elements, such as families, which help us to understand the complexity of ownership in the pre-modern period and decode the superimposition of rights on certain goods. Finally, we would like to highlight the way these communities, mainly the small villages and rural municipal districts, have been able to reinvent themselves, adapt, and find tools that have enabled them to, at least in the Cantabrian Mountains, recover throughout the 20th century some lands they had been naturally exploiting since time immemorial. The managerial organs have changed in accordance with the political changes, but they have been very resilient, which has enabled these communities to preserve their collective rights in the grazing areas and forests. References

Congost, R. 2002: Comunales sin historia. La Catalunya de los masos o los problemas de una historia sin comunales. In De Dios, S. – Infante, J. – Robledo, R. – Torijano, E. (coords.), Historia de la propiedad en España. Bienes comunales, pasado y presente. Madrid, 291‑328. De Keyzer, M. 2018: Inclusive commons and the sustainability of peasant communities in the medieval Low Countries. Routledge. New York. De Moor, M. – Shaww-Taylor, L. – Warde, P. 2002: The management of common land in north west Europe, c. 1500‑1850. Brepols. Thurnhout. Fernández Fernández, J. 2017: Arqueología de una aldea medieval y su espacio agrario: Villanueva de Santo Adriano (Asturias, noroeste de la península Ibérica), Historia Agraria: Revista de Agricultura e Historia, 79‑107. Fernández Fernández, J. – Fernández Mier, M. (eds.) 2019: The archaeology of medieval villages currently inhabited in Europe. Archeopress. Oxford. Fernández Mier, M. 2010: Campos de cultivo en la Cordillera Cantábrica. La agricultura en zonas de montaña, in: Kirchner, H. (ed.), Por una arqueología agraria. Perspectivas de investigación sobre los espacios de cultivo en las sociedades medievales hispánicas. BAR International Series. Archaeopress. Oxford, 41‑59. Fernández Mier, M. 2020: La construcción de la territorialidad altomedieval. Entre la documentación escrita y la arqueológica: la montaña centroccidental asturiana, in: Martín Viso, I. (ed.), La construcción de la territorialidad en la Alta Edad Media: poderes centrales y sociedades 90

locales edad media: poderes centrales y sociedades locales. Salamanca. University of Salamanca, 223‑242. Fernández Mier, M. – Catarina Tente, C. 2018: Transhumant herding systems in Iberia, in: Svensson, E. – Costello, E. (eds.), Historical archaeologies of transhumance across Europe. Routledge. London, 219‑232. Fernández Mier, M. – Quirós Castillo, J.A. 2015: El aprovechamiento de los espacios comunales en el NW de la Península ibérica entre la romanidad y la Edad Media, Studies on the Value of Cultural Heritage 8, 695‑723. García Fernández, 1988: Sociedad y organización del espacio tradicional en Asturias. Gijón. Silverio Cañada. Gassiot Ballbé, E. – Pèlachs Mañosa 2017: La ocupación ganadera de los Pirineos occidentales de Catalunya en época romana e inicios de la Edad Media, Treballs d’Arqueologia 21, 287‑306. Giovannetti, L. 2004: Archeologia e storia della Montagna della Garfagnana e delle sue risorse. Il caso di Gorfigliano nel più ampio contesto apuano e appenninico, in: Quirós Castillo, J.A. (ed.), Archeologia e storia di un castello apuano. Gorfigliano dal Medioevo. La pratica della pastorizia e gli alpeggi della Garfagnana (versante appenninico) alla luce delle fonti statutarie, orali e dell’archeologia postmedievale. Archeologia Postmedievale 9. All’Insegna del Giglio. Florence, 53‑68. González Álvarez, D. – Fernández Mier, M. – López Gómez, P. 2016: An archaeological approach to the ‘brañas’: summer farms in the pastures of the Cantabrian Mountains (north of Spain), in: Collis, J. – Pearce, M. – Nicolis, F. (eds.), Summer farms. Seasonal exploitation of the uplands from prehistory to the present. Sheffield Archaeological Monographs16. Equinox Publishing Ltd; J.R Collis Publications. Sheffield, 203‑219. Hamerow, H. 2002: Early medieval settlements. The archaeology of rural communities in North-West Europe, AD 400‑900. Oxford University Press. Oxford. Hamerow, H. 2012: Rural settlements and society in Anglo-Saxon England. Oxford University Press. Oxford. Lana Berasain, J.M. 2018: Mejorar la suerte del proletariado. Cuestión social y repartos comunales (Navarra, 1868), Ayer 112, 99‑127. Mañana Vázquez, G. 2011: El Camín Real de la Mesa. Caja de Ahorros de Asturias. Oviedo.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Monsalvo Antón, J.M. 2001: Usurpaciones de comunales: conflicto social y disputa legal en Ávila y su tierra durante la baja edad media, Historia Agraria 24, 89‑122. Montesinos, L. – Campanera, M. 2017: Formas de vida, usos y apropiación de recursos. Propuesta para el estudio de los comunes contemporáneos, Revista de Antropología Social 26:2, 193‑216. Mujika Alustiza, J.A. – Agirre García, J. – Edeso Fito, J.M. – Lopetegi Galarraga, A. -Pérez Díaz, S. – Ruiz Alonso, M. – Tarriño Vinagre, A. – Yusta Arnal, I. 2013: La continuidad de la actividad pastoril durante la época romana en la zona de Argarbi (Sierra de Aralar, Gipuzkoa), Kobie. Serie Paleoantropologia 32, 217‑258. Oosthuizen, S. 2013: Beyond hierarchy: The archaeology of collective governance, World Archaeology 45:5, 714‑729. Ostrom, E. 1990: Governing the commons: The evolution of institutions for collective action. Cambridge University Press. Cambridge. Palet, J.M. – Olmos, P. – Garcia, A. – Polonio, T. – Orengo, H.A. 2019: Occupation et anthropisation des espaces de haute montagne dans les vallées de Nuria et de Coma de Vaca (Gerona, Espagne): résultats des recherches archéologiques et patrimoniales, in: Costamagno, S. – Deschamps, M. – Milcent, P.Y. – Pétillon, J. M. – Renard, C. – Valdeyron, N. (eds.), La conquête de la montagne: des premiéres occupations humaines á l’anthropisation du milieu. Éditions du Comité des travaux historiques et scientifiques. Actes des congrès nationaux des sociétés historiques et scientifiques. Published online at http://books.openedition.org/ cths/7007 (accessed 16. 02. 2020). Quirós Castillo, J.A. 2009: The archaeology of medieval villages in Europa. University of the Basque Country. Bilbao.

Rendu, C. 2003: La montagne d’Enveig. Une estive pyrénéenne dans la longue durée. Trabucaire. Le Canet en Roussillon. Rippon S.J. 2008: Beyond the medieval village: the diversification of landscape character in southern Britain. Medieval History and Archaeology. Oxford University Press. Oxford. Rubio Pérez, L.M. 1993: Estructuras agrarias y modelos organizativos de las comunidades campesinas leonesas durante la Edad Moderna, Mélanges de la Casa de Velázquez 29:2, 253‑274. Scott, J.C. 1976: The moral economy of the peasant. Rebellion and subsistence in Southeast Asia. University of Yale. New Haven. Scott, J.C. 1985: Weapons of the weak. Everyday forms of peasant resistance. Yale University Press. New Haven. Stagno, A.M. 2016: Seasonal settlements and husbandry resources in Ligurian Apennines (17‑20 c.), in: Collis, J. – Pearce, M. – Nicolis, F. (eds.), Summer farms. Seasonal exploitation of the uplands from prehistory to the present. Sheffield Archaeological Monographs 16. Equinox Publishing Ltd; J.R Collis Publications. Sheffield, 67‑88. Stagno, A.M. 2019: Investigating rural change. Legal access rights and changing lifestyles in rural mountain communities (Ligurian Apennines, Italy, 16th-21st centuries), World Archaeology, 51:2, 311‑327. Thompson, E.P. 1991: Customs in common. Penguin Press. London.  Williamson, T. 2012: Environment, society and landscape in early medieval England. The Boydell Press. Woodbridge.

Fernández Mier and López Gómez

91

Plows, herds, and chafurdões. Vernacular architecture and land use in modern Castelo de Vide (Alto Alentejo, Portugal) Fabián Cuesta-Gómez* and Sara Prata**

Abstract

The chafurdões are one of the most characteristic ethnographic buildings in the Alto Alentejo countryside. These are circular drystone structures topped by corbelled domes. Written sources suggest that at least some of them were built from the 17th century onwards. Their usages in the peasant landscapes were numerous: as a shelter for shepherds and farmers, storage facilities for tools and produce, and, less frequently, to keep livestock. Nevertheless, their location, deep within cultivated fields, suggests that they were not permanent housing structures, but rather periodically occupied during certain moments, in which it was necessary to be closer to the fields. Nowadays, many of these structures are still standing, kept by farmers mostly for storage purposes. Their presence in the landscape constitutes an important example of rural cultural heritage that should be further examined and preserved. This paper will consider the distribution of these structures in the countryside, review the techniques used to build them, discuss their seasonal use during agricultural practices of the modern period, and reflect upon their meaning and preservation today, from a cultural heritage perspective. Keywords: vernacular architecture, modern period, corbelled dome, Castelo de Vide (Portugal). * Instituto de Estudos Medievais FCSH, Universidade Nova de Lisboa Avenida de Berna, 26 C, 1069‑061 Lisboa, Portugal [email protected] ** Funding: 2020.01697.CEECIND Instituto de Estudos Medievais FCSH, Universidade Nova de Lisboa Avenida de Berna, 26 C 1069-061 Lisboa, Portugal [email protected]

Résumé

Charrues, troupeaux et chafurdões. Architecture vernaculaire et occupation du sol à Castelo de Vide Moderne (Alto Alentejo, Portugal) Les chafurdões figurent parmi les édifices traditionnels les plus caractéristiques de la campagne de l’Alto Alentejo. Ce sont des structures en pierre sèche avec des dispositions circulaires, surmontées de coupoles en encorbellement. Des sources écrites suggèrent que certains d’entre eux remontent au XVIIe siècle. Leurs fonctions dans les campagnes étaient nombreux : abri pour les bergers et les agriculteurs, entrepôts pour les outils et les produits et, moins fréquemment, pour garder le bétail. Néanmoins, leur localisation, au plus profond des champs cultivés, suggère qu’il ne s’agissait pas de structures d’habitation permanentes mais plutôt temporaires lorsqu’il fallait se rapprocher des champs. De nos In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 93-104.

93

jours, bon nombre de ces structures sont toujours debout, conservées par les agriculteurs principalement à des fins de stockage. Leur présence dans le paysage constitue un exemple important de patrimoine culturel rural qui devrait être examiné et préservé davantage. Nous examinerons la distribution de ces structures à la campagne, passerons en revue leurs techniques de construction, discuterons de leur utilisation saisonnière dans les pratiques agricoles de l’époque moderne et réfléchirons à leur signification et à leur préservation aujourd’hui, dans une perspective de patrimoine culturel. Mots-clés : architecture vernaculaire, époque moderne, coupole en encorbellement, Castelo de Vide (Portugal). Zusammenfassung

Pflüge, Herden und Chafurdões. Volksarchitektur und Landnutzung im modernen Castelo de Vide (Alto Alentejo, Portugal) Die Chafurdões sind eines der charakteristischsten ethnografischen Bauwerke im Alto Alentejo. Hierbei handelt es sich um Trockensteinstrukturen mit kreisförmigen Grundrissen, die von Kragkuppeln gekrönt

Introduction

Vernacular architecture has recently been gaining traction in European research in the context of the protection and analysis of historical rural landscapes (Mileto et al. 2015). With regard to the Portuguese landscape, new approaches are still scarce and most of the available information stems from the extensive ethnographic surveys of the 1960s (Oliveira et al. 1994), although there is recent comparative work from the Spanish countryside. This paper focuses on the chafurdões of the territory of Castelo de Vide (Alentejo). These circular drystone structures, topped by massive corbelled domes, are the most characteristic ethnographic buildings in the countryside. Their location, deep within the rural areas, suggests that they were not for permanent occupation, but were periodically occupied on those occasions in which it was necessary to be closer to the fields. The authors are currently carrying out a research project promoted by the Municipality of Castelo de Vide to revise the local archaeological inventory. The work carried out so far has included extensive surface field survey and a critical overview of previous archaeological documents. The drawings and dimensional data that will be referred to in this paper were chiefly gathered during the 1990s by the archaeology department of the municipality (Seção de Arqueologia Câmara Municipal Castelo de Vide [SACMCV]). 94

werden. Aus schriftlichen Quellen geht hervor, dass zumindest einige davon im 17. Jahrhundert und danach erbaut wurden. Es gab zahlreiche Verwendungszwecke in den bäuerlichen Landschaften: als Unterschlupf für Hirten und Bauern, als Lager für Werkzeuge und Erzeugnisse und seltener für die Viehhaltung. Trotzdem lässt ihre Lage tief in den bebauten Feldern darauf schließen, dass es sich nicht um permanente Wohnstrukturen handelte, sondern um periodische Besetzungen in bestimmten Momenten, in denen es notwendig war, näher an den Feldern zu sein. Heutzutage stehen noch viele dieser Strukturen, die von den Landwirten hauptsächlich zu Lagerzwecken aufbewahrt werden. Ihre Präsenz in der Landschaft ist ein wichtiges Beispiel für das kulturelle Erbe des ländlichen Raums, das weiter untersucht und erhalten werden sollte. Wir werden die Verteilung dieser Strukturen auf dem Land betrachten, ihre Bautechniken überprüfen, ihre saisonale Nutzung während der landwirtschaftlichen Praktiken aus der modernen Zeit diskutieren und ihre heutige Bedeutung und Erhaltung aus der Perspektive des kulturellen Erbes reflektieren. Schlagwörter: Volksarchitektur, Neuzeit, Corbelled Dome, Castelo de Vide (Portugal).

Although the territory of Castelo de Vide is our study area, this kind of drystone hut is a much broader phenomenon, having been in use in several areas of the Iberian Peninsula and the Mediterranean area until recently. In some regions, especially in the north of Portugal, there was a clear connection to seasonal pastoralism such as transhumance (Barroso et al. 2018), whereas in others their uses might have been more diverse, but in all documented cases they were only used seasonally. For the purpose of a comparative analysis, only the charfurdões documented in the neighbouring regions of Alto Alentejo and Spanish Extremadura will be considered, because of the spatial proximity of the two regions and the coherence of them both as landscape features and in relation to land use patterns. There are several other types of drystone structures and buildings preserved in the countryside: field walls, terraces, stone hut bases, watermills, cattle pens, and some corbelleddome buildings that are still standing, such as small bread ovens or medium-sized pigpens. These are all relevant examples of vernacular architecture from the modern period that survive in today’s landscape and should consequently be analysed together. For now, as a starting point, the focus of this research is on the chafurdões, since these are the most noteworthy architecturally and comprehensively documented to date. It is also our notion that their primary function was to be used by people as temporary

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 1. left: location of Castelo de Vide in the Alto Alentejo region (Portugal); right: distribution of the 59 chafurdões in the territory of Castelo de Vide (© Fabián Cuesta-Gómez and Sara Prata).

accommodation. Understanding the location of chafurdões in relation to permanent settlements, the seasonality of their use, and their connection to different types of land use models is key to characterising the transformation in agricultural practices of the modern period. The village and the countryside

Castelo de Vide is a municipality located in the Portuguese region of Alto Alentejo (Fig. 1). Its territory covers an area of 265 km2, which includes the northwestern hilltops of the São Mamede mountain range but is mostly composed by a gentle peneplain crossed by seasonal streams and marked by rocky outcrops. The Alto Alentejo region preserves abundant archaeological evidence in its rural areas, mainly from the Neolithic, the Roman era, and the early medieval period, that reflects different models of dispersed settlement occupation. From the 12th century onwards a strong nucleated settlement pattern started to emerge, organised around castles and fortified boroughs, controlled directly by the Portuguese monarchy or assigned to noble families and religious orders. During the late medieval period these medieval settlements grew into villages and continued to expand during the

modern period, their castles and walls transformed into bastion fortifications. Like Castelo de Vide, most of the urban areas in Alto Alentejo occupied today saw this development sequence. However, evidence of what can be described as rural settlements is very scarce during the high and late medieval periods and it seems likely that most farming activities were carried out in the areas surrounding the fortified urban centres. Only in the second half of the 16th century is there evidence of structures, suggesting that the outermost areas of the territory were starting to be farmed regularly. From the 17th century onwards drystone walls were built for the enclosure of the fields, a reflection of land use and ownership transformations caused by the many changes of the modern period, such as economic growth, diversification of crops and agricultural processes, and population increase (Videira 2008, 178). In Castelo de Vide the average size of the rural enclosures (under 10 ha) suggests small-scale properties scattered around the urban areas. Written sources indicate that most of the population was focused in this territory’s two main urban areas, Castelo de Vide and Póvoa e Meadas, both of which were granted charters during the early 16th century. Farmsteads and Cuesta-Gómez and Prata

95

agricultural estates, in the sense of places where the rural population both lived and worked in the surrounding fields, here seem to be much later phenomenona, possibly dating from the 19th century. The building of the charfurdões marks the first point in the modern period when a gradual increase in agricultural activity can be seen. The construction was probably carried out not by full-time farmers, but rather as an extensive secondary activity of people who lived mainly in the urban areas and were periodically required to support different kinds of agricultural and cattle-raising activities. In this context, chafurdões were used as short-term accommodation for these local farmers, smallholders, or even for seasonal farm workers coming from the northern provinces of the country to work in the harvest. Information about early modern agricultural activities in this region is still scarce. Overall, the average depth of soil in the territory of Castelo de Vide is very low, with vast areas of exposed outcrops, and a highly acidic soil pH, resulting in soils with reduced ecological value, according to current standards. Nowadays polyculture is frequent in a continuous-tilling regime, simultaneously with grazing land and small-scale cereal production. Some parcels still maintain old cork oak forests (Quercus suber), chestnut (Castanea sativa), and traditional olive groves (Olea europea L.). While today the most common practice is extensive livestock production, mainly cows for meat consumption, not so long ago flocks of sheep and goats ‒ mainly for wool and milk for cheese production but also for meat ‒ and herds of pigs for fresh and cured meat were the norm. Although further research is necessary, considering the natural characteristic of this area, it seems feasible to suggest that agricultural practices in the Modern Era were based on a diverse exploitation strategy that combined traditional livestock (sheep and goat) husbandry in parallel with small-scale agriculture of cereals, fruit trees, cork oaks, and vineyards. This land-use model is still visible today in some of the small farmsteads located around the village. Spatial distribution

The information on the Castelo de Vide chafurdões comes from the data included in its original archaeological inventory (Rodrigues 1975) and, fundamentally, from the surveys carried out by the SACMCV during the 1980s and 1990s. These works focused on three of the four municipal freguesias (civil parishes). The territory of Póvoa e Meadas, located in the north, was an autonomous municipality up until 1836, and the land ownership regime, with larger properties and more land suitable for crops, was quite different from that of Castelo de Vide. These socioeconomic differences and the lower incidence of archaeological surveys in the territory explain the absence of chafurdões in the area 96

Fig. 2. Front view, section, and upper view of the chafurdão in Vale de Cales (© J. Magusto, SACMCV).

of ​​Póvoa e Meadas, so our analysis will mainly focus on examples in Castelo de Vide. The relationship between the countryside and the village was direct and intense right from its foundation in the first half of the 13th century. Regardless of the land ownership regimen, there were no other places in the vicinity that attracted and concentrated the population. The proximity to the border with Spain, a constant menace until the beginning of the 19th century; the limited productivity of its poor soils; and the thrust towards industrial textile production made Castelo de Vide the local centre for trade, ideas, and people. As far as farming was concerned, most parcels would be worked directly from the village, meaning periodical trips to the fields whenever necessary.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 3. Top: map of walking isochrones (60 and 120 minutes) from the village of Castelo de Vide; left: scatter plot graph with height and diameter values distribution; right: histogram with the representation of the internal area (m2) of the chafurdões (© Fabián Cuesta-Gómez and Sara Prata).

Cuesta-Gómez and Prata

97

A cost path analysis of the time taken to travel from the village to the fields was carried out, which seems to indicate that many of the chafurdões are between the 60- and 120-minute isochrones (Fig. 3). This reinforces the idea that these structures were temporary dwellings: places for occasional shelter, storage of produce, and keeping animals. Their relative proximity to the village, where permanent homes were situated, and their character reinforce this notion. Although there is no existing study of local property changes, the spatial distribution of the chafurdões in this territory seems to be linked to small- and medium-sized properties. Our working hypothesis is that this municipality’s rural subdivision has not undergone major changes in its configuration during the last three centuries, regardless of whether the property was dispersed or numerous single-family plots are reflected in the landscape. Building techniques and formal features

Chafurdões are a classic example of vernacular architecture. They are built with local materials – irregular stones (granite and/or slate) and compacted earth – and feature a rather straightforward, if demanding, building technique. They were built by overlapping layers of stones in successive rings, the top being closed with a corbelled dome (Fig. 5). Lastly, the dome was covered with earth and gravel to guarantee that the outer stones remained protected and in place. Thus, the steadiness of the chafurdão was achieved by its dead weight, the vertical forces combined with the horizontal as a result of the interlocking of the stones (Rovero – Tonietti 2014). No lime mortar was used during the building process, but sometimes the walls were coated with packed earth. Thick walls and the earth used for covering the dome also ensured thermal insulation. The masonry units used were small and irregular, very similar to the ones used in the enclosure walls, and most likely resulted from quarrying granite outcrops and fieldclearing actions. The exception is the entrance, which is usually framed by large dressed stones (Fig. 4). A total of 59 chafurdões were considered for this analysis. In terms of dimensional data, the maximum height, diameter (orthogonal measurements), wall thickness, and entrance orientation were recorded. For 28 of the analysed structures (47.5%) all parameters were recorded, for an additional 20 structures (33.9%) at least 2 of the measurements were recorded, and for the remaining 11 only 1 data factor or none were recorded. This uneven record is justified by the fact that most chafurdões present some degree of decay: either the domes or the walls have partially collapsed, which no longer allows for rigorous measurement. Finally, 9 (15.3 %) are destroyed, and the data were obtained from previous surveys (Rodrigues 1975; SACMCV 1987‑1995). 98

As a result, a series of parameters that portrays the morphology of these structures has been established. From a total of 45 cases, it can be determined that the average height of these constructions is 3.51 m (SD = 0.64 m), with a maximum interior diameter of 4.93 m (SD = 0.56 m, for 52 cases) and an internal area of 19.35 m2, although in this case the variability is significantly greater (SD = 4.24 m2) with examples ranging from as little as 9 m2 to as much as 27.34 m2 (mode = c. 22 m2). However, it does not appear that the variations in size imply differences in the primary use of the buildings studied. Finally, of the 38 cases in which the entrance’s orientation is recorded, over half (21) face east, and if those that face northeast and southeast are included, this totals 82% of documented cases. This circumstance reflects the need for protection against the most frequent winds in the region, as well as to optimise exposure to the sun in the morning. The interior structure of the chafurdões is quite similar, with vertical walls that begin to curve inwards from the base of the dome. Niches, between 1 and 6 square-shaped spaces delimited by slabs that were used as built-in shelves for storing domestic utensils or food, are frequently found on the inside walls. Less frequently these openings go through the wall, functioning as windows. Ventilators are also frequent as small triangular- or square-shaped holes located at the base of the dome, or, less frequently, at ground level to facilitate debris discharge (Fig. 4). Outer structures added to the chafurdão are also common, notably semicircular walls and squared buildings with thatched roofs, which would be used as additional spaces to keep animals or for storage of wood, hay, building materials, and farming tools. Of the 59 documented cases, 32 had at least one attached structure, such as a cattle pen or a tiled-roof rectangular building (Fig. 2). In some cases, the charfurdões were eventually included in farmstead building complexes, evidence of the later process when the first permanent settlements developed in the countryside. Unfortunately, there are very few elements to date precisely when the chafurdões were built and first used. Although the possibility of excavating a test pit inside one of these structures with the aim of identifying pottery remains or other dating elements was considered, most of the chafurdões were built directly over the granite bedrock. In the few cases that they preserve deposits, these are completely sterile or rather the result of recently added soil layers for cattle keeping. The granite inner-door frame of the chafurdão from Tapada do Couto presents the peculiarity of an engraved name Bartholome[u?] and the date 1733 (Fig. 4). The inscription was made using cursive characters, similar to those found in civil documents from the 18th century. This could be the name of the builder, or, most likely, the owner and the date when the chafurdão was built

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 4. Chafurdão from Tapada do Couto, which presents the peculiarity of an engraved name, ‘Bartholome’[u?], and a date ‘1733’ on the doorway frame (right); left: inner view of Maria Cecilia’s structure, where a niche and an air vent can be seen (© Fabián Cuesta-Gómez and Sara Prata).

or rebuilt. It is rather frequent to find dates such as this on stone buildings, but this is the only known case for a chafurdão of Castelo de Vide. Although these types of buildings are very common in different parts of the Iberian Peninsula (Vegas et al. 2009), the functionality and morphology of the chafurdões of Castelo de Vide find their closest reflection in similar structures in the neighbouring territories of Marvão (Oliveira et al. 2007) and in the Spanish provinces of Salamanca, Cáceres, and Badajoz. While in the Spanish Extremadura the variability of the materials used and the size of these auxiliary agricultural structures seems to be greater (Cruz 2010; Galindo – Muñoz 2004; Blanca 2004; Lorenzana 2017), the drystone hut with corbelled dome and marked agrarian polyfunctionality is frequently

present. In fact, the presence of Portuguese workers who specialised in their construction was common in these border territories (Martín Galindo 2006, 848‑849). Lasting structures for seasonal use?

The building of a chafurdão was a substantial material investment, and the chafurdão’s ability to last was certainly a significant feature. The review of ethnographic studies carried out in the neighbouring Spanish region of Extremadura indicates that once the stone blocks were obtained ‒ the most physically demanding part of the work ‒ the construction of the drystone hut could be completed by a small group of workers in just a few days (Martín Galindo 2006, 852). Experience and the Cuesta-Gómez and Prata

99

Fig. 5. Top: Alto da Cumeada’s chafurdão, one of the largest examples of this form of vernacular architecture preserved in Castelo de Vide; bottom: detail of the corbelled dome, seen from below (© Fabián Cuesta-Gómez and Sara Prata).

coordination of work and materials, as well as knowledge of stone construction, were needed to raise the chafurdão from a simple circle drawn on the ground to delimit its foundations (Mena 2003). It was common for specialised crews of quarrymen to travel around the region to work where required, both in the construction of new huts and in the maintenance of others. There is evidence for numerous different uses for these structures in the peasant landscape: as seasonal shelters for shepherds and land workers, storage facilities for tools and produce, and, less frequently, for keeping livestock. During the ethnographic surveys carried out in the 1960s (Oliveira et al. 1994, 157‑163), at least some of the chafurdões from Castelo de Vide were still in use, but mainly for storage purposes, and it was clear they had already lost their original dwelling function. It is worth noticing that their name originates from chafurda, a term 100

given to pigpens, which might also refer to poorly kept living accommodations. In relation to this, it is also worth considering another type of ethnographic structure: the choça. Choças are drystone circular structures with a similar layout to chafurdões, but instead of corbelled domes they have conical thatched roofs made with oak beams and branches of broom (Cytisus sp.). They have been identified in places such as Cabeçudos (Marvão) as permanent household structures (Oliveira et al. 1994, 130‑135). Even though the thatched roof would have to be replaced periodically, it offered some breathability and a more comfortable living environment, while the inside of a chafurdão retains a lot of moisture. For its part, the main advantage of a chafurdão would be its durability and the possibility of reusing construction materials. In this sense, it would be possible to invest in the construction

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

of one of these buildings, next to a plot of land, and to ensure that it would survive for a considerable period. If repairs were needed at some point, the materials would be easily available, since, unlike organic components that required total replacement, the stone elements can be continually reused. It remains unclear whether chafurdões were built with a single purpose in mind rather than a multifunctional structure meant to support a wide range of agricultural activities. While there seems to be no correlation between the different documented layouts and known uses, it would appear that they were mainly built to be used as temporary accommodation for land workers, as suggested by the fact that there are no chafurdões around the urban areas and their spatial distribution in relation to the villages. While smaller types of drystone corbelled-domed structures were built specifically to keep sows and raise piglets, it must be considered that at least some of the chafurdões were built to keep livestock, as reflected by the ground-level openings for the discharge of liquid waste and stone mangers. These structures could be used for periodic stabling and to stockpile manure to be used as fertiliser. As farming practices progressed, there was an increased demand for permanent rural labour and farmstead type settlements from the late 19th century onwards. Some of these farms ‒ called montes ‒ developed in areas where chafurdões had previously been built, suggesting a transition from seasonal land use and thus seasonal occupations to progressively more intense agricultural activities. Charfurdões became storehouses and barns, as some of them still are today. Chafurdões from a cultural heritage perspective

Chafurdões were the result of a traditional agricultural system that is no longer in use. They survived due to their structural robustness, while their uses were continuously adapted to the new needs of the population of the region in each era. Only by the second half of the 20th century did they become obsolete, when the region was profoundly transformed and the countryside progressively abandoned. The increase of land dedicated to permanent pasture throughout the year and the incorporation of motor vehicles and metal and electrified fences made the need for people in the fields more sporadic, as well as less demanding and time-consuming. Agricultural activity and livestock husbandry together no longer formed a generalised way of life: for a few it became a professionalised sector, and for others it was consigned to a temporary occupation or a hobby in small family plots. Between both worlds, the presence of the chafurdão became less relevant. While some of these structures remained as storehouses or barns, especially

those that were closer to or integrated into contemporary houses or farms, most accentuated their degradation. The memory of their use is being eroded, as well as the know-how needed to build and repair them. Currently there are 59 known chafurdões in the territory of Castelo de Vide with different degrees of preservation. A logical step for these structures might be to regard them as part of the local heritage and ensure specific programmes for their preservation. Only a few of them are integrated in walking trails that are maintained by the municipality for this purpose. The others are primarily located on private lands and the interest and effort to maintain them resides solely with the will of the landowners. The Alto Alentejo region has been steadily growing as a tourist destination. Although projected as ‘rural’, most tourist activity is organised around historic urban centres and the countryside is essentially being left out. This area preserves an impressive number of archaeological sites, but only a few are open to the public. Museums and collections are equally scarce. As a result, there are no effective links between these elements of the past and the current villages of Alto Alentejo, whether to inhabitants or to visitors. In this context, chafurdões and other examples of ethnographic buildings have an added difficulty: they are no longer important in today’s agricultural practices, whilst not old enough to be considered as archaeological, and thus lack the legal protection inherent to other buildings and sites. For this reason, specific lines of research and protection must be promoted. Fortunately, some important steps have already been taken, such as the addition of ‘Art of dry stone walling, knowledge and techniques’ (Croatia, Cyprus, France, Greece, Italy, Slovenia, Spain, and Switzerland) to the Representative List of the Intangible Cultural Heritage of Humanity by UNESCO in 2018. From an Alto Alentejo perspective, and focusing specifically on the chafurdões, it might be advantageous to establish work groups integrating both the Portuguese and Spanish inland regions where these stone structures are common. Both territories have regional and municipal archives where documents from the modern era are kept, which include testaments, deeds, and other similar records. While it seems unlikely that these might include direct references to the building or use of chafurdões, they could be mentioned as part of plot descriptions, especially in documents dealing with land ownership. Documentary research is certainly an area that deserves further attention. Perhaps specific programs for vernacular architecture could be promoted, based on multidisciplinary approaches that would include architecture, ethnography, archaeology, history, and geography. A project of this nature would also need to involve traditional drystone masons and engage directly with the last rural workers who used the chafurdões. Thus, preserving the physical Cuesta-Gómez and Prata

101

structure and the memories of the last charfurdões could also be an important tool to engage local communities with cultural heritage. Conclusion

Chafurdões were mainly built to be used as temporary accommodations for land workers while they were carrying out seasonal agricultural tasks in the 18th century, if not before. Some of the chafurdões were built to keep livestock, as is evident from the presence of mangers. Over time their use diversified, with them becoming storehouses and barns as more permanent settlement and agriculture developed from the late 19th century. Chafurdões are not the only example of vernacular architecture buildings in the Alto Alentejo region, but they are certainly the most demanding technologically. Corbelled domes are an impressive example of technical execution and a vivid testimony to the knowledge, skills, and way of life of people of the recent past. Their conservation should indisputably be connected to the revival of the rural hinterland of Portugal. However, this is unfortunately still a poorly defined issue on political agendas and has no clear solution in sight. As in other similar areas of the Mediterranean, depopulation and aging inhabitants handicap the continuity of rural territories, which suffer the dichotomy of abandonment versus intensive agriculture practices and energy use (solar panels, opencast mines, ponds, dams etc.). With the increase of these landscape altering land uses perhaps it is not realistic to preserve all these vernacular buildings that are abandoned in the contemporary countryside. But efforts must be made to ensure their adequate survey, recording, and analysis before more information is lost. Ethnographic heritage must be examined collectively and from different points of view. In this paper, chafurdões have been analysed from an archaeological and historical perspective, considering their seasonal use in the context of contemporary agricultural practices. The understanding of their locations in relation to the village of Castelo de Vide have helped to shed light on a poorly understood period and to characterise the progressive changes undergone in the countryside during the modern period. Ultimately, relevant insights can only be gathered by considering ethnographic heritage such as chafurdões as an historical source from which to collect useful information about past communities. Appreciation of this point might in turn help develop effective tools for vernacular heritage management and preservation, as well as raise awareness among professionals, cultural managers, and local communities.

102

References

Barroso, C.E. et al. 2018: The construction of the transhumance territory of the Gerês-Xurés: vernacular heritage identification, analysis and characterization, in: Villegas, L. et al. (eds.), Rehabend 2018, Euro-American Congress on Construction Pathology, Rehabilitation Technology and Heritage Management. Universidad de Cantabria /UNICAN, 274‑283. Blanca Pecero, J.A. 2004: Chozos en el término municipal de Fuente del Maestre, in: Martín Galindo, J.L. – Orovengua, J.M. (eds.), Arquitectura tradicional en la provincia de Badajoz. Diputación de Badajoz. Badajoz, 383‑408. Cruz Sánchez, P.J. 2010: Ensayo de tipología de las construcciones secundarias en piedra seca en las Arribes del Duero salmantinas, Estudios del Patrimonio Cultural 4, 5‑24. Galindo Mena, J. – Muñoz Castro, Mª.D. 2004: Bujardas en el término de Llerena: manifestaciones de la arquitectura popular, in: Martín Galindo, J.L. – Orovengua, J.M. (eds.), Arquitectura tradicional en la provincia de Badajoz. Diputación de Badajoz. Badajoz, 409‑444. Lorenzana de la Fuente, F. 2017: Algunos ejemplos de arquitectura vernácula de uso agrario en la Sierra de Gata extremeña, Trócola 5, 75‑97. Published online at http://iesvp.educa.aragon. es/TROCOLA/REVISTA5.pdf (accessed 30. 01. 2020). Martín Galindo, J.L. 2006: Los chozos extremeños referente histórico y recurso socio-cultural para el futuro, Revista de Estudios Extremeños 62:2, 839‑890. Mena Cabezas, I. 2003: Arquitectura tradicional de chozos de piedra en Palomero, Alcántara: Revista del Seminario de Estudios Cacereños 58, 47‑65. Mileto, C. – Vegas, F. – García Soriano, L. – Cristini, V. (eds.) 2015: Vernacular architecture: towards a sustainable future. CRC Press/Balkema. Leiden. Oliveira, E.V. – Galhano, F. – Pereira, B. 1994 [1969]: Construções primitivas em Portugal. Ed. Dom Quixote. Lisboa. Oliveira, J. – Pereira, S. – Parreira, J. 2007: Nova carta arqueológica do Concelho de Marvão, Ibn Maruan 14. Ed. Colibri/Câmara Municipal de Marvão. Lisboa. Rodrigues, M.C. 1975: Carta arqueológica do Concelho de Castelo de Vide. Junta distrital de Portalegre. Lisboa.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Rovero, L. – Tonietti, U. 2014: A modified corbelling theory for domes with horizontal layers, Construction and Building Materials 50, 50‑61. SACMCV, 1987‑1995: Relatórios dos Levantamentos Arqueológicos do Concelho de Castelo de Vide. Unpublished.

Vegas, F. – Mileto, C. – Cristini, V. 2009: Corbelled dome architecture in Spain and Portugal, in: Lotti, G. – Mecca, S. (eds.), Earthen domes and habitats. Villages of Northern Syria. Edizioni ETS. Pisa, 81‑89. Videira, C. 2008 [1908]: Memoria histórica da muito notável villa de Castello de Vide. Colibri/CIDEHUS. Lisboa.

Cuesta-Gómez and Prata

103

From the Roman villa rustica to the early modern farmer’s grange – specific forms of seasonal settlements in eastern Croatia Pia Šmalcelj Novaković* and Anita Rapan Papeša**

Abstract

This paper provides information about specific forms of seasonal settlements in eastern Croatia throughout the centuries. We can observe a pattern from the Roman period up to modern times that reflects a long-lasting tradition of separated estates. Keywords: villa rustica, early modern grange, organisation of space, question of continuity, special way of seasonal living, Syrmia. Résumé

De la villa rustique romaineau grange du début de l’époque moderne – un moyen spécifique de peuplements saisonniers dans l’est de la Croatie Cet article appporte des informations sur la spécificité des établissements saisonniers de l’est de la Croatie au cours des siècles. Nous pouvons observer un modèle particulier de peuplement a été observé de l’époque romaine aux temps modernes qui reflète dans une longue tradition de domaines séparés. Mots-clés : villa rustica, grange moderne, organisation de l’espace, continuité d’occupation, mode de vie particulier, Syrmia. * Institute of Archaeology Ljudevita Gaja 32 10000 Zagreb, Croatia [email protected] ** Municipal Museum Vinkovci Trg bana Josipa Šokčevića 16 32100 Vinkovci, Croatia [email protected]

Zusammenfassung

Von der römischen Villa Rustica zum frühneuzeitlichen Gutshof – eine spezifische Art der saisonalen Besiedlung in Ostkroatien In diesem Artikel wird die besondere Art der saisonalen Besiedlung in Ostkroatien im Laufe der Jahrhunderte diskutiert. Wir können ein Muster von der Römerzeit bis zur Neuzeit beobachten, dass die langjährige Tradition abgegrenzter Ländereien widerspiegelt. Schlagwörter: Villa Rustica, frühneuzeitlicher Gutshof, Raumorganisation, Frage der Kontinuität, besondere Art des saisonalen Lebens, Syrmien.

In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 105-110.

105

Introduction

East Croatia, the southernmost part of the Pannonian plain, is an area of lowlands with one of Europe’s most fertile soil types that has provided ideal conditions for permanent settlements without the need for ‘classical’ seasonal settlement (e.g. transhumant animal husbandry) (Sić 1975, 5‑7). Primarily because of the state of research, with new survey being hampered particularly by the fact that we are dealing with a region severely affected by war (minefields on potential and registered sites), we cannot prove the existence of structures, either archaeologically nor historically, that are usually interpreted by archaeologists as seasonal settlements. On the other hand, from prehistory onward, but especially from the Roman period we can observe a specific form of seasonal activity and organisation of space in the area. The easternmost part of Croatia belongs to an area called Syrmia. It is a geo-strategically separated area, now divided between East Croatia and West Serbia, surrounded by the rivers Sava, Danube, and Bosut. In the Roman period, it was a part of the province of Pannonia Secunda with Sirmium (Sremska Mitrovica, Serbia), Mursa (Osijek, Croatia) and Cibalae (Vinkovci, Croatia) as important centres. This area has emerged as an extremely interesting (micro-) region for the study of the

continuity, or rather, the survival of the Roman tradition in later periods. For instance, we have strong indications of continuity of a Late Antique jewellery tradition in the early medieval communities of the area around Cibalae (Dimitrijević 1957; Šmalcelj Novaković 2020). Syrmia stands out as a geo-strategically very important area, particularly in the early Middle Ages. Control over the (micro-) region meant controlling the Roman Pannonian routes/routes in Pannonia, but also the routes to the European part of the former eastern Roman empire, Byzantium (Filipec 2015, 23). The overlap of early medieval burial sites with Roman limes (military) fortifications and settlements is one of the most obvious arguments in support of the conclusion that Roman organisation of space in Syrmia remained crucial much longer than the Roman government and political system (Filipec 2015, 23, 28; Fig. 6). Research on the continuity or survival of Roman settlement organisation in the region has so far focused mainly on comparing existing results of research on the Roman and early medieval horizons, of which the limes fortifications and the area of Roman Cibalae (today Vinkovci) on the one hand, and the burial sites of the Avar Age on the other, are the most extensively explored sites (see, for example: Rapan Papeša 2011, 2012; Sekelj

Fig. 1. Published sites in Vukovarska-Srijem county defined as highly probable villae rusticae: (1) Jarmina-Kamenica; (2) Vinkovci-Marica; (3) Zalužje-rasadnik; (4) Lipovac-Podlučje, Kloštar; (5) Berak-Gradina; (6) Cerić-Plandište (sources: 1, 2, 3, 4, and 6 – © The database of Antique archaeological sites of the Republic of Croatia [Jarmina-Kamenica, Vinkovci-Marica, Zalužje-rasadnik, Lipovac-Podlučje, Kloštar, CerićPlandište]; 5 – © Registry of Cultural Heritage, Ministry of Culture [Berak-Gradina]; accessed date: 08. 06. 2020; made with Google Earth, version 7.0.2.8415, by P. Šmalcelj Novaković).

106

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Ivančan 2001; The database of Antique archaeological sites of the Republic of Croatia: site Zalužje). So, in this geo-strategically important and very fertile region we can presume some kind of continuity of Roman organisation of space into the early medieval period, but the question remains whether we find any traces of the influence of Roman organisation of space in later periods, albeit indirect? Syrmia

Recently, as a result of intensive field survey and the study of historical sources and cartographic records, the possible remains of villae rusticae in the region, as well as of late medieval settlements and early modern granges, have been identified (Fig. 1). It is noted that late medieval settlements often appear precisely in the positions of supposed villae or other forms of Roman settlements (Petković 2006, 36, 94, 240). One should not forget the fact that the Romans changed the Pannonian Plain, not only in terms of communication (roads and paths), but also in terms of improving natural conditions, which significantly increased the economic (agricultural) utilisation of the area, and certainly also the quality of life (for more on this, see Iskra-Janošić 2001, 25‑29). In general, these parts of the Pannonian Plain were very desirable, as the density of recorded sites in the early medieval period suggests. It is also to be assumed that locally in Syrmia, the remains of Roman architecture (in the east of Croatia, we are speaking mainly of brick/ stone buildings (Roksandić 2019) was a kind of sign or confirmation to the early medieval population that these were good locations at which to settle. We have no reason to doubt that it remained so in the later periods. Field survey reports from the region reveal that one of the main methods of locating villae rusticae is by their architectural remains (stone, Roman brick) – so they were certainly visible in the more distant past. (It should be mentioned that one of the limes forts was still visible in the 18th or 19th century, as well as the remains of the Roman town of Cibalae; see Balen-Letunić – RadmanLivaja 2008, 422, 426; Brunšmid 1902; Iskra-Janošić 2001, 9). As the economic activity of the majority of the population (agriculture) has remained unchanged in this region almost from prehistoric times to the present day, and we can conclude that in the post-Roman period any traces of the prehistoric organisation of space in the region were no longer visible, and because we know that at least traces of Roman infrastructure were still visible in the early medieval period, we can cautiously presume that the Roman organisation of space determined the natural and, partially, the social conditions of settlement of Syrmia even in later periods. Of course, there is no doubt that Romans themselves took note of the

organisation of space and communications built by prehistoric communities when they occupied Syrmia (Gračanin 2010, 9‑10). Roman spatial organisation

Along with cities (municipalities, colonies), villages (pagus), and roads with stations (mutatio), an important part of the economy was based on villae. The Roman villa rustica was often a hub of a large agricultural estate. The villa rustica would thus serve both as a (usually part-time) residence of the landowner and his family (and retainers), and also as a farm management centre. It would often comprise separate buildings to accommodate farm labourers and sheds and barns for animals and crops (Cambi 2002, 83). Although Roman villae rusticae cannot be considered as a form of classic seasonal settlement, in the explicitly flatlands of Eastern Croatia they did usher in a tradition of a specific way of life on a separate estate. The very idea of a majur, a grange, a temporary residence created for economic purposes (agriculture) and to enjoy the natural benefits of the space, was originally the ideal setting of the villa rustica as well. If we look at the example of a typical Syrmia grange, it is easy to see the similarity in the (architectural) layout of the space with the standard layout of the villa rustica, especially if we rely on the nearest investigated example: the late Antique-period villa rustica Livade kod Ćuprije (Fig. 2; Jovičić – Redžić, 2012). Early medieval period

The organisation of space in the Roman period was wellmanaged and adhered to general rules for the management of infrastructure; by comparison, the early medieval period (7th to 9th centuries) looks like a step backwards, with parallels closer to prehistory than to the Roman period. A further problem is that early medieval settlements are still poorly known to researchers, the first ones having been excavated in the last 10‑15 years (Sekelj Ivančan 2010; Sekelj Ivančanet et al. 2017). Because there is no published detailed research of an early medieval settlement in Syrmia, we have used the geographically closest example of an early medieval settlement at the Kaznica-Rutak site (7th to 9th centuries according to material typology, 8th century according to 14C results), as its geographic characteristics are similar to those in Syrmia; it reveals that early medieval communities were well aware of their Roman heritage. The furnaces and hearths in the settlement were built of Roman bricks and tegulae most likely from the villa rustica site at Selci Đakovački-Brstine (about 3.5 km away) (Šmalcelj Novaković – Hršak 2017). Similar observations can be made for Syrmia as well, based on field surveys and comparison with neighbouring geographical and historical lands (Bačka and Banat in Serbia; see Bugarski 2008).

Šmalcelj Novaković and Rapan Papeša

107

High and late medieval period

Settlements in the High and Late Middle Ages (10th to 16th century) are divided into urban centres, markets, villages, and satellite settlements. The latter consist of several houses separated from their home settlement, comprising no more than five residential buildings. Historical sources have noted the existence of dozens of such smaller, satellite villages that oriented to a larger settlement with the church (Petković 2006). We also know that some of the settlements, or more precisely, their inhabitants, were specialised in terms of economic activity: for instance, the etymology of the village Tovarnik, first mentioned in the 14th century, points to the work of raising cattle; while the name of its nearest northern neighbour, Lovas, points to inhabitants focussed on horse breeding (Petković 2012, 163‑165). It is assumed that a wider family community lived in such settlements and cultivated land in close proximity. Confirmation of this is found in the Ottoman population censuses, where the householders generally have the same names and are only occasionally recorded as immigrants. Today, the names of these villages are recognised through toponyms on maps and through finds from field survey. Historical sources, especially during the Habsburg-Ottoman wars at the end of the 17th and the beginning of the 18th century, record refugees living in forests, that is, perennial settlements of the refugee population who moved into dense oak

forests fleeing the horrors of war and organised their lives there until the return to their original households was possible (Gligorević 2006). We believe this type of seasonal settlement was also used in earlier times, especially if we take into consideration the military techniques of the Ottoman occupation (seasonal war activities – the akindžija phenomenon (Živković 2016, 53). We still do not have any concrete archaeological evidence of these settlements. On the other hand, we have a clear example of a settlement from the 14th century located in immediate vicinity of a Roman villa rustica at the site of RetkovciStaro selo (Rapan Papeša 2010, 333). Early modern period

After liberation from the Ottoman occupation, the Habsburg Empire organised a special form of administration, the Military Frontier (18th – 19th century), which included the policy of village enlargement (the so-called ušoravanje) and the organisation of streets in today’s plan (Sić 1975, 7). An old family community was called a peasant cooperative, and apart from their house in the village, they also had a separate seasonal abode, a grange (hrv. stan), where they lived mainly in the summer months. The Slavonian grange is an economic-residential entity in a field, somewhere next to a forest, in the form of a farmhouse surrounded by other outbuildings. The members of the cooperative were

Fig. 2. Comparison of the layout of a modern farmer’s grange (Sikirevci, © Gligorević, Lj. 2006, Iz tradicijskog života, Vinkovci 2006) and that of a Late Antique-period villa rustica (Livade kod Ćuprije, © Jovičić, M. – Redžić, S. 2012).

108

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

exchanging at the grange, taking care of large and small livestock, poultry, bees, orchards, and so on (Knežević 1990). In current society, this tradition of living has almost died out, and some granges have been given a new purpose as tourist housing. Often such granges have become, in effect, a kind of modern villae rusticae, where agricultural production is maintained primarily for the needs and enjoyment of visitors, whose primary reason for visiting is the enjoyment and full experience of the region in which they reside. One of the sites in the wider Vinkovci area, Takšića stan-Kunjevci, better known as Zalužje, is a good example of continuity of settlement in the same place (Fig.1). The area of Zalužje corresponds to that of a medieval village, Zalwsa/Zalwsya/Zalujiye (in 15th- and 16th-century records), and in its southern part there are granges (stanovi) that include the aforementioned ‘Takšićastan’. There is a presumption of a villa rustica in the vicinity, indicated by finds of Roman graves and Roman gold coins at the same site (Petković 2006, 240‑241; Rapan Papeša 2010, 346).

economic activity has remained the same throughout history, the settlement sites overlap and we know that the major reorganisation of space was done in Roman times and was actually followed in later periods also. There remains a bigger question: can we really determine what is and is not continuity? Can we really trace the continuity of tradition? Can we objectively determine where and how corresponding circumstances and contexts of geographical characteristics end and a tradition of settlement begins? For these questions to be answered, further research (primarily, concrete excavations of a number of sites) is needed.

Conclusion

References

If we compare the function and purpose of the Roman villae rusticae in Antiquity and the šokački granges in the early modern period, we can see how closely they resemble each other. These are isolated residential and commercial buildings in which some members of the community stayed occasionally and took care of production facilities essential for their survival. The very idea of the existence of Roman tradition in the early modern period was introduced for the first time in ethnological papers, pointing to similarities between the hair braiding of local maidens and Roman imperial female portraits on coins (Radauš-Ribarić 1982). Only further research can establish whether there are examples of satellite settlements during the Middle Ages with a function similar to that of Roman villae rusticae or early modern granges. Some results of field surveys in Syrmia point out that the choice of places to settle has been very similar in various periods of history, which can be interpreted in light of the geographical characteristics of or continuity in the organisation of space in the region. At the same time, the existence of the stan, a special type of seasonal settlement, which in its organisation greatly corresponds to that of the villa rustica, leads us to the question of whether it is the result of some type of continuity in the organisation of space from the Roman period (considering the continued visibility of Roman infrastructure in the 19th century) or just an superb example of how different communities in different time periods but similar circumstances react similarly? In case of the Syrmia region, where the main

Acknowledgements

This study was supported by the Croatian Science Foundation through the project ‘Life on the Roman Road: communications, trade and identities on Roman roads in Croatia from 1st-8th century CE’ (UIP-05‑2017‑9768).

The database of antique archaeological sites of the Republic of Croatia. Published online at http://baza.iarh.hr/public/ locality/map (accessed 20. 01. 2020). Balen-Letunić, D. – Radman-Livaja, I. 2008: Izvorište arhivskih podataka: izvadci iz putnih bilježnica Josipa Brunšmida o rimskim nalazima s dunavskog limesa, Vjesnik Arheološkog muzeja u Zagrebu, 3.s., XLI, 417‑438. Brunšmid, J. 1902: Colonia Aurelia Cibalae: Vinkovci u staro doba, Vjesnik Hrvatskoga arheološkoga društva N.s. 6, 117‑166. Bugarski, I. 2008: The geomorphological matrix as a starting point for determining the position of Avar-time settlements in Pannonia: the example of the Bačka Region, Archäologisches Korrespondenzblatt 38:3. Mainz. Cambi, N. 2002: Antika, Naklada Ljevak. Zagreb. Dimitrijević, S. 1957: Četiri groba iz slavenske nekropole u Otoku kod Vinkovaca s posebnim osvrtom na naušnice lunulasto-zvjezdolikog tipa, Opuscula Archaeologica 2, Radovi Arheološkog intituta Filozofskog fakulteta Sveučilišta u Zagrebu, 21‑38. Filipec, K. 2015: Donja Panonija od 9. do 11. Stoljeća, Sarajevo, Bosna i Hercegovina: Univerzitet u Sarajevu. Gligorević, Lj. 2006: Iz tradicijskog života. Vinkovci.

Šmalcelj Novaković and Rapan Papeša

109

Gračanin, H. 2010: Rimske prometnice i komunikacije u kasnoantičkoj južnoj Panoniji, Scrinia Slavonica 10, 9‑69. Iskra-Janošić, I. 2001: Urbanizacija Cibala I razvoj keramičarskih središta. Gradski muzej Vinkovci. Vinkovci. Jovičić, M. – Redžić, S. 2011: Late Roman Villa on the site Livade kod Ćuprije – a contribution to the study of Villae Rusticae in the vicinity of Viminacium, Archaeology and Science 7, 369‑386. Knežević, A. 1990: Šokački stanovi. Županja. Petković, D. 2006: Srednjovjekovna naselja sjeverozapadnog dijela vinkovačkog kraja. Gradski muzej Vinkovci. Vinkovci. Petković, D. 2012: Pregled vijesti o Tovarnika u povijsnim izvorima od 13. do 16.stoljeća, Acta Musei Cibalensis 5, 159‑185. Radauš-Ribarić, J. 1982: Tradicijsko djevojačko češljanje u Panonskoj Hrvatskoj, Etnološka tribina 1982, 97‑109. Rapan Papeša, A. 2010: Pregled novorekognosciranih i reambuliranih srednjovjekovnih lokaliteta na području bivše Općine Vinkovci, Vjesnik Arheološkog muzeja u Zagrebu 3.s., XLIII. Arheološki muzej. Zagreb. Rapan Papeša, A. 2011: Lokalitet: Nuštar – avarsko groblje (park dvorca KhuenBelassy), Hrvatski arheološki godišnjak 8, 65‑68. Rapan Papeša, A. 2012: Prvi nalazi s avarskog groblja na položaju Nuštar, katalog izložbe. Gradski muzej Vinkovci. Vinkovci. Roksandić, D. 2019: Građevinski materijali I tehnike u arheologiji na temelju naselja istočne Hrvatske, Kolegij: Položaj naselja I graditeljske tehnologije, inaugural lecture

110

for assistant professor, Faculty of Philosophy and Social Sciences, Department of Archaeology. Zagreb. Sekelj Ivančan, T. 2001: Neki arheološki primjeri zaposjedanja ruševina antičkih urbanih cjelina u sjevernoj Hrvatskoj tijekom srednjeg vijeka, Prilozi Instituta za arheologiju u Zagrebu 18, Instiut za arheologiju, 189‑212. Sekelj Ivančan, T. 2010: Podravina u ranom srednjem vijeku. Rezultati arheoloških istraživanja ranosrednjovjekovnih nalazišta u Torčecu, Institut za arheologiju, Zagreb. Sekelj Ivančan, T. – Tkalčec, T. – Krznar, S. – Belaj, J. (eds.) 2017: Srednjovjekovna naselja u svjetlu arheoloških izvora, Zbornik Instituta za arheologiju/Serta Instituti archaeologici 5, Zagreb. Sić, M. (ed.) 1975: Istočna Hrvatska, Geografija SR Hrvatske knjiga 3. Školska knjiga. Zagreb. Šmalcelj Novaković, P. 2020: What are they doing here? Raceme type earrings with elongated pendant/Traubenohrringe mit ährenförmiger Anordnung at the Late Avar age site Privlaka-Gole njive, Life and death in mediaeval and early modern times 14, 33‑46. Šmalcelj Novaković, P. – Hršak, T. 2017: The beginning of early medieval settlement in Eastern Croatia: a case study of Kaznica-Rutak, Srednjovjekovna naselja u svjetlu arheoloških izvora [Medieval settlements in the light of archaeological sources]. Zbornik Instituta za arheologiju 6. Zagreb. Živković, G. 2016: Renesansno vojno umijeće u Hrvatsko-turskim ratovima. Graduation thesis presented to Diplomski studij Hrvatskog jezika i književnosti i povijesti, Sveučilište J.J. Strossmayera u Osijeku, Osijek.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Section Three

Seasonal Settlement in Northern and Eastern Europe

Transhumant settlement in medieval Wales: the hafod Rhiannon Comeau* and Bob Silvester**

Abstract

The transhumant settlements of Wales were formerly a significant element of its upland landscapes. This paper summarises and reviews research into the hafod (‘summer residence’) and investigates some of the key points that emerge. The presence of summer settlements is attested in medieval documents, place names, and through abundant remains of structures, the last having been examined during extensive field surveys from the early 1990s onwards. These seasonal dwellings have a long history: a variety of evidence indicates the likely presence of transhumant practice in the early medieval period, and although it largely disappears in south Wales by the 16th century, it continued in some other areas until the 19th century. Material evidence of probable seasonally occupied structures from excavations and surveys is summarised. In the absence of any nationallevel synthesis of data from the field surveys, the paper uses selected sites to explore some key questions about summer settlements – problems of identification of seasonal and permanent phases, relationships to permanent settlement, location, and environmental setting. The principal case study, a recently excavated site in the Preseli Hills of southwest Wales, provides a multidisciplinary contextualisation that shows a relationship between longue durée settlement patterns and changing environmental conditions. Keywords: transhumance, summer hafod, seasonal settlement, upland landscape, field survey. Résumé

* 21 Ulleswater Road Southgate London N14 7BL United Kingdom [email protected] ** Braich-uchaf, Cil Road Meifod Powys SY22 6YA United Kingdom [email protected]

L’habitation transhumante dans le Pays de Galles médiéval: le hafod Les installations liées à la transhumance au Pays de Galles ont été jadis un élément essentiel du paysage des hautes terres. Cet article résume et passe en revue le travail réalisé sur le hafod («résidence d’estive») et examine quelques points clés qui ont émergé suite à cette recherche. Les documents et toponymes médiévaux, ainsi que de nombreux vestiges, ces derniers ayant été examinés au cours de prospections depuis le début des années 1990, attestent de la présence des habitations d’estive. Ces habitats saisonniers ont une longue histoire : une multitude de preuves indique sur la potentielle présence des pratiques de la transhumance dans la période du début du Moyen Age, et, si cette pratique disparaît dans le sud du Pays de Galles avant le XVI e siècle, elle continue néanmoins dans d’autres régions jusqu’au XIXe siècle. La preuve matérielle des structures probablement occupées de manière saisonnière, trouvée dans en fouilles et en prospections, est résumée dans l’article. En l’absence d’une synthèse au niveau In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 113-124.

113

national des données des enquêtes de terrain, cet article utilise certains lieux sélectionnés afin d’explorer un nombre de questions clés sur les habitations d’été – des problèmes liés à l’identification des phases saisonnières et permanentes, aux liens à l’habitat permanent, à l’emplacement et au contexte environnemental. Le site de référence, récemment mis au jour dans les monts Preseli dans le sud-ouest du Pays de Galles, fournit une contextualisation multidisciplinaire qui montre un rapport entre les modes d’occupation sur la longue durée et les adaptations aux conditions environnementales. Mots-clés : transhumance, cabane d’estive, habitation saisonnière, paysage de montagne, enquêtes de terrain. Zusammenfassung

Die wanderweidewirtschaftlichen Siedlungen im mittelalterlichen Wales: die hafod Die wanderweidewirtschaftlichen Siedlungen in Wales waren früher ein bedeutendes Element der Hochlandlandschaft. Dieser Artikel beleuchtet und fasst die bisherige Forschung zum Thema Hafod (‚Sommerresidenz‘) zusammen und analysiert bedeutende Aspekte. Das Vorkommen von im Sommer bewohnten saisonalen Siedlungen ist sowohl durch mittelalterliche Dokumente als auch durch

Introduction

The hafod is the term commonly given to the summer houses used for transhumance in medieval and postmedieval Wales. These were once a significant element in the Welsh upland landscape, the equivalent of the north British sheiling and the Irish booley, and generally account for the large numbers of small rectangular structures identified by upland archaeological surveys over the last 30 years. Their site patterning, ecological contexts, and the links between specific sites and permanent settlements are still substantially unexplored, and this paper will consider aspects of these from selected case studies, after briefly assessing the current state of research. The historically documented background

Understanding of Welsh transhumant settlements was initially shaped – as elsewhere in Europe – by ethnographic and place-name studies. The mid-20thcentury ethnographers Iorwerth Peate and R.U. Sayce, together with the topographically focussed place-name studies of Elwyn Davies in the 1970s and 1980s, explored the tradition of summertime relocation of farming households and livestock to the hills that continued in 114

Ortsnamen belegt. Des Weiteren sind Reste von Siedlungsstrukturen bekannt, welche durch eine umfassende Prospektion seit Anfang der 1990er Jahre untersucht wurden. Diese saisonalen Siedlungen haben eine lange Geschichte: eine Vielzahl von Belegen deuten auf wanderweidewirtschaftliche Praktiken im frühen Mittelalter hin. Trotz ihres Verschwindens in Südwales zum Anfang des 16. Jahrhunderts, scheinen sie in einigen Gebieten bis ins 19. Jahrhundert weitergeführt worden zu sein. Eine Zusammenfassung von materiellen Hinterlassenschaften für saisonal genutzte Strukturen durch Ausgrabungen und Prospektionen wird präsentiert. Da es keine nationale Sammlung der Prospektionsdaten gibt, basiert dieser Artikel auf einer Auswahl bedeutender Fundstellen. Es sollen Fragen zur Bestimmung von saisonalen und permanenten Phasen, zur Relation zu permanent bewohnten Siedlungsgebieten und der näheren Umwelt erforscht werden. Die primäre Fallstudie, eine aktuelle Ausgrabung in den Preseli-Hills im Südwesten von Wales, erlaubt eine interdisziplinäre Kontextualisierung des Verhältnisses zwischen Siedlungsmustern in einer longue durée Betrachtung und sich ändernden Umweltbedingungen. Schlagwörter: Transhumanz; Sommersiedlung; saisonale Siedlung; Hochland; Prospektion.

some areas into the 19th and even early 20th centuries (Roberts 2006a, 2006b). These practices were, until the 1960s, assumed to be a legacy of (hypothetical) Celtic semi-nomadic pastoralists, but these preconceptions were questioned from the early 1960s onwards by investigations of early legal sources that connected seasonal pasture with pre-Norman lowland estates (e.g. Jones 1964). In the 1970s, the Llandaff charters of south Wales were found to contain grants of both permanent lowland hamlets with arable infields and of several large upland areas starting as early as the 7th century and using the terminology of late Roman estates (Davies 1978, 35). Summer settlements are not specifically identified, but seasonal pasture is suggested by the comments of Gildas, a 6th-century cleric from the west of Britain, who listed ‘mountains especially suited to varying the pasture for animals’ as a feature of the postRoman landscape (Winterbottom 1978, 17/§ 3.3). It is in this light that the earliest uses of the hafod term can be understood, found today as a farm name across Wales and conventionally taken to mean ‘summer residence’ – haf (‘summer’) + bod (‘dwelling place’). It is cognate with Old Cornish havos, which is first recorded as an 11th-century place name, but refers to a much older

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 1. Wales, with principal locations mentioned in the text: the Preseli sites are shown in Fig. 5. Shading indicates land above 200 m (© Bob Silvester).

practice of seasonal settlement that Peter Herring (2012) has discussed for Bodmin Moor. In Wales, although seasonal dwellings must have been used from the earliest times, the earliest recorded usages of hafod are for areas of summer pasture rather than for the buildings on them: its first appearance is in a charter of 1185 and denotes a large area of upland pasture (‘all the land called Havot Oweyn’) given to the monastery of Strata Marcella (Thomas 1997, 151‑152). Its widespread application to dairies and dwellings on upland pastures emerges clearly in the 16th century, a period when increasing numbers of hafod sites were becoming permanent, year-round farms and when documentary records proliferate after legal reforms (Davies 1980). Two other terms appear at this time – the hafoty in north Wales, literally a ‘hafod house’, and in mid-Wales the lluest, which initially denoted a temporary military encampment in 13th- to 14th-century sources, but by the 16th century was used for small summer houses or dairies (Davies 1980, 9). Dwellings on summer pastures were first specifically mentioned in 12th- and 13th-century legal references, long after their probable emergence in the landscape. The insubstantial hafdy (‘summer house’) of the bondsman or unfree worker, occupied from May to August or September, figured in the earliest surviving versions of

the Welsh laws that claimed a 10th-century formulation and enforced the practice, familiar elsewhere in Britain, of moving livestock in summer from crop-growing infield areas to seasonal pasture (Comeau 2019a, 140, 146‑167; Roberts 2019, 85‑86). Cattle were the principal livestock on summer pastures at this point, with sheep on yearround farms replacing them in later periods. They were accompanied by milkmaids and township herdsmen who are mentioned in Welsh law (Jenkins 1990, 62, 176, 321), but information is otherwise very sketchy. In the 16th century, boys and young men worked year-round as cattle herders in the unenclosed Pembrokeshire landscape (Comeau 2019b, 155), while 18th- and 19th-century accounts talk of whole families occupying small summer dwellings in north and mid Wales (Davies 1984, 88). Post-medieval evidence for the practice shows strong regionalisation. In parts of north and mid Wales, it continued into the early 19th century and, in a much diminished form, into the mid-20th century, with shortterm relocations of farm households for sheep washing and shearing (Peate 1946, 126). In other areas, notably south and east Wales, it had largely disappeared by the end of the medieval period, its decline linked to enclosure of open fields and shared pastures and to the establishment of permanent upland sheep farms. Comeau and Silvester

115

The archaeological background

Medieval and post-medieval records and place names, therefore, give the outlines of seasonal settlement practice. Identifying the physical traces of upland settlements began with scattered excavations from the 1930s onwards at places like Gelligaer in the Glamorgan uplands (Fox 1939). More sites were recorded by sporadic upland fieldwork and aerial photography from the 1970s onwards (e.g. Ward 1991), and concerted programmes of study began in the 1990s with two major survey initiatives that examined both permanent and seasonal sites. Between 1996 and 2004, a programme sponsored by Cadw (the Welsh government historic environment service) considered all known deserted rural settlements of medieval and later date, the majority being in upland regions (Roberts 2006a). The other, the Upland Archaeology Initiative, organised by the Royal Commission on the Ancient and Historical Monuments of Wales, ran for nearly 30 years from 1987 and aimed to identify all archaeology in the open uplands of Wales above 250 m OD, an area comprising around 40% of the Welsh landscape (Browne – Hughes 2003; Silvester 2011; Leighton 2012; Olding 2016). Output for both has taken the form of regionally focussed publications. The buildings recorded have been essentially simple rectangular structures, often around 4 m to 7 m in length, with (usually) stone foundations that carried walls of turf, wattle or rubble stone (Fig. 2). They are commonly called long huts, to distinguish them from the stonewalled multi-celled longhouses of medieval and later date. Similar structures are found elsewhere in Britain (e.g.

Dixon 2018). On sloping ground the Welsh structures use purpose-built platforms levelled into the hillside. These are common across Wales, sometimes with no visible stone remains, suggesting entirely timber- or turf-built structures (Roberts 2006b, 174; Silvester – Kissock 2012, 153). Ancillary features where they exist consist of the occasional enclosure that might have acted as a fold; perhaps a sunken coldstore for milk, cheese and butter; and sometimes (but rarely) an adjacent field. Some areas display distinct locational patterns, for instance on the Radnorshire commons in central Wales, where seasonal sites frequently occupy platforms spaced along the edges of the numerous commons, above and overlooking enclosed land in the valleys some two or three kilometres away; and further to the south, on the steep-sided Brecon Beacon mountains some deep V-sided valleys are devoid of settlement, while others have numerous rectangular structures terraced into the valley sides (Silvester 2006). Spatial relationships between the structures also vary. There are occasional groups, and some are spaced at intervals along the contour, or around a favoured grazing ground as with the platforms collectively known as Hafoty Arllen-fawr lying above a streamhead in the Cambrian Mountains. Much more commonly (in the experience of one of the authors) single or perhaps paired buildings appear in isolation. Definitive understanding of the incidence of these patterns awaits systematic analysis of the field survey data. Identification of the seasonal occupation of settlements is at best only indicative, being based on a combination of characteristics: the simplicity of building structures; the

Fig. 2. A solitary hafod site in the Duhonw valley on Mynydd Epynt in Breconshire (© Bob Silvester).

116

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

general absence of other features, such as field systems and ancillary buildings; their presence at particular locations, such as immediately above valley floors; their viewsheds and altitude; and their preference for some geographical areas rather than others. Place names and records help only rarely, the former sometimes showing how occupation patterns can fluctuate between seasonal and permanent over the life of a site (Silvester 2000, 55). Attributing date ranges to sites is challenging, some excavations having produced no dating evidence at all (e.g. Kissock – Johnstone 2007). Material culture survival is limited not only by the pre-Norman aceramic and nonmonetised economy and by the acid soils of the uplands, but also by a more general material poverty, first illustrated by Aileen Fox’s work on Gelligaer Common nearly a hundred years ago (Fox 1939). Only a handful of artefacts and radiocarbon dates firmly support medieval occupation (cf. Leighton 2012, 127‑128). The earliest dates are 11th to 12th century and come from a hearth sample at Carn Goedog in southwest Wales, discussed below. Elsewhere in the region a radiocarbon date from beneath a pillow mound (artificial rabbit warren) suggested a medieval origin for the adjacent long huts at Bryn Cysegrfan in Ceredigion (Austin 1988). In Snowdonia in northwest Wales, 11thto 14th-century dates were provided by wood charcoal in a floor deposit at Ynys Ettws (cal AD 1050‑1390 at 95.4% confidence: Beta-127671, 780±70 BP), where occupation continued intermittently to the 17th century (Smith – Thompson 2006, 114). From Hafod Nant y Criafolen in northeast Wales there is a single 15th- to 17th-century radiocarbon date (cal AD 1435‑1670 at 93.2% confidence: HAR-1435, 330+70 BP) with broadly contemporary pottery and artefacts (Allen 1979, 47; see below; all radiocarbon dates calibrated using OxCal v4.3http://c14.arch.ox.ac.uk/). Research issues

Several issues emerge from this brief review of existing work (see also Silvester – Kissock 2012). Firstly, there has been very little national-level synthesis of available data for seasonal settlements, despite the vast amount of field data collected in the uplands during the 1990s, with overviews restricted to limited geographical localities. Second, and equally fundamental, is the problem of distinguishing seasonal from permanent settlement. Excavation, usually the ultimate arbiter, does not offer an indisputable truth, with some of the classic excavations revealing the contrasting viewpoints of the excavators and later commentators: Aileen Fox’s vision of permanent settlement on Gelligaer Common was questioned by Peate (1946, 128) and more recently Locock (2006, 52); Beili Bedw in Radnorshire, seen as a year-round complex of platform sites in the excavation report (Courtney 1991),

had its permanency questioned owing to the ‘paucity of finds’ (Ward 1997, 106), while conversely Hafod y Nant Criafolen in north Wales has been claimed as permanent because of its range of artefacts (Briggs 1985, 304). Thirdly, there is generally a poor understanding of variations in the location and patterning of settlements. There has been little consideration of how and why sites are located in response to environmental or cultural exigencies and how this may affect the pattern of occupation, seasonal or permanent, of a site. Understanding these from site-specific archaeological material alone is difficult, with the clearest picture provided by the extended programme of excavation at the permanently occupied settlement of Graeanog in northwest Wales (Kelly 1982) or by additional palaeoenvironmental studies, as in the study that follows. This examines the spatial, social, and environmental contexts of a recently excavated site, which is characterised using a multidisciplinary dataset – archaeological and historical records, place names, soils, and pollen records. Case study: Carn Goedog seasonal settlement in the Preseli Hills (Schlee et al. 2018).

This 11th- or 12th-century site is currently the earliest firmly dated seasonal settlement in Wales. It lies in the Preseli Hills (Mynydd Preseli) of southwest Wales, which extend some 20 km east-west, rising at their highest point to 536 m OD. Their open moors encompass much common land and are grazed by sheep and ponies, soils being shallow, podzolic, and often seasonally waterlogged. The site’s name, Carn Goedog, is taken from an adjacent rock outcrop or carn that is first recorded on 19th-century maps and hints at processes of landscape change: the second element, coed(i)og (‘wooded’), suggests the former presence of trees, perhaps wood pasture. There are nine long huts here, aligned along the 240 m contour at 2 m to 15 m intervals, close to a transmontane track and overlooking the rough pasture of the lower hillside (Fig. 3A-J). Nearby is another group of five unexcavated and undated sub-circular structures (Fig. 3I-N). One of the long huts, House C, was excavated in 2015 by Mike Parker Pearson and his team, who were investigating the outcrop as one of the prehistoric sources of the Stonehenge bluestones. The structure measured 6 x 4 m externally after excavation, with one rectangular end cut into the hillside, and the other end (apsidal like the Gelligaer houses), projecting out over the hillside on a slight platform (Fig. 4). It had a centrally placed entrance on its long side with perhaps another facing it in the opposite site, a central hearth, and low stone wall bases that were probably the footings for turf walls, like Scottish shielings. A radiocarbon date of cal AD 1030‑1200 at 95.4% confidence (SUERC-68382; 917±34 BP) was obtained from a hazel twig in the hearth, Comeau and Silvester

117

Fig. 3. Long huts and sub-circular structures at Carn Goedog (© Irene de Luis).

and pottery of late 12th- to early 16th-century type and a spindle whorl were found outside the entrance. Mid-11th to early 13th-century occupation is considered likely (Schlee et al. 2018, 250‑252). No coldstores or enclosures around the long huts were identified and their relationship with relict field walls below them is uncertain. The simplicity of the excavated structure and its upland location suggest a seasonal settlement, and indeed the overall patterning of the group of structures is similar to other sites in south Wales where seasonal usage is a possibility (Locock 2006, 45, 48). The moorland below does, however, contain a range of structures of different types and apparent dates, some with enclosures that suggest more permanent settlement (Dyfed HER, 8403, 9944). How confident, therefore, can we be that this is a seasonal settlement, and if so, why are there permanent-looking settlements nearby? There are no specific references to the site in historical records, though these do indicate a long-term pattern of seasonal pasturage on the north Preseli uplands. They are used today as common grazing by farms on the moorland edge, two kilometres away, under rights derived from the mid-13thcentury Charter of Preseli, which itself confirms earlier rights granted to 12th-century Welsh nobles (Comeau 2019b, 156‑157, 560, 646). This is approximately contemporary with the references to seasonal settlement and transhumant agricultural practice in Welsh law that 118

were outlined earlier, and somewhat predates records of transhumance in north Wales (Hooke 2019). These pasture rights were part of a complex mix of localised and longer-distance links that involved the whole Preseli region. The longer-distance links (first recorded in the 1121 charter of St Dogmaels Abbey) span up to 15 km and link lowland settlements on the northern coastal plain with pastures on the southern Preseli slopes (Comeau 2019a, 140‑142). The scale of exploitation varies, from hamlets to seigneurial herds, and its complexity – with rights involving both Welsh and Anglo-Norman areas – suggests arrangements of pre-Norman provenance. A 13th-century ‘grange’ place name (Grangia, 1291) indicates that some of the southern uplands, held by the north-coast-based St Dogmaels Abbey, were probably used for a yearround vaccary or cattle ranch. To the west of these lies an area where, according to 16th-century records, the Lord of Cemais held summer pasture rights for 120 horses and cattle and 300 sheep (Comeau 2019b, 648). More-localised links are represented by the Charter of Preseli and by place names that, through formations like rhos (‘moor’)/mynydd (‘mountain/common/ unenclosed land’) + community name, encode the rights of communities to specific areas of grazing, usually on nearby hills (Comeau 2019b, 652‑653). The overall impression is of widespread use of these upland

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 4. Carn Goedog, House C: plan and section (© Irene de Luis).

pastures at a range of scales, involving long-standing claims by lowland areas to the hills as a shared seasonal resource, with some limited year-round exploitation. There are other place names indicative of seasonal pasture here that use terms like hafod and menyn (‘butter’). These are largely recorded in the 16th century when records of landholdings become generally available, though ironically this is also when transhumance seems to be disappearing locally, since it is not specifically mentioned in an otherwise fairly comprehensive Elizabethan account of local agricultural practice (Comeau 2019a, 135‑136). Thus it is possible that the isolated moorland farm called Hafod Tydfil, one kilometre along the hillside from Carn Goedog, was already a year-round landholding when it was first noted in 1585. There is also uncertainty about the 16th-century mode of occupation at Summerton on the hillslopes to the south. Now a single farm, it was then

a hamlet of 7 houses, the ownership patterns of which show pairing with landholdings in a lowland township called Winterton, one kilometre below, clearly indicating its genesis as a summer settlement. Suggestions of dairying activities typical of summer settlement are provided by the ‘butter rock’ name (Carn Menyn 1573, now Carn Meini) of an outcrop near Carn Goedog (Comeau 2019b, 649‑653 for all seasonal place-name references). The pollen record is also instructive, and reveals changes over time in the upland Preseli pastures. A general pattern of environmental deterioration was identified in the 1980s by a study of Preseli peat cores (Seymour 1985; Comeau 2019b, 140‑143), similar to (if less closely dated than) the picture revealed by recent work in north Wales (Davies 2019). Pastoralism was established by the Early Bronze Age, with good hilltop grassland maintained by heavy grazing and good drainage until the early medieval Comeau and Silvester

119

Fig. 5. Sites and soils in the Preseli region of southwest Wales: soil data taken from Soil Survey Map of England and Wales 1983 (© Rhiannon Comeau).

period. The north Wales (Snowdonia) pollen record also indicates heath burning from c. 2000 BC to c. AD 1300, a spring practice still used in the Preseli Hills to maintain the quality of seasonal pasture (Davies 2019, 180). An intensification of upland grazing is identifiable both in Preseli and in north Wales in the latter part of the early medieval period (Davies 2019, 192‑193; Comeau 2019b, 142). The north Wales work also shows a clear pastoral upland/arable lowland distinction for the early medieval period, probably related to transhumant agriculture (Davies 2019, 180). In Preseli a less clearly dated and more spatially restricted sampling base prevents such comparisons. Some of the proliferation of settlements on the north Preseli Hills may relate to this period, which is recognised as a time when climatic amelioration encouraged the expansion of permanent mixed farming in upland Scotland (Strachan et al. 2019, 140‑143, 151‑152). Economic (and political) pressures for increased cattle production may also have been at play. Enormous cattle-based tribute payments maintained clientship relations between rulers at this time, like the annual 25,000 oxen promised by a north Wales prince to the Anglo-Saxon King Athelstan in 927 (Davies 1982, 120

114). Cattle, together with weighed silver, were widely used as units of exchange and measures of wealth in the non-coin-using economies of early medieval Wales, Ireland, and Spain, and cows were commonly used for rent payments until the 14th century (Comeau 2019b, 54‑56, 429‑430). This phase of upland pastoral expansion is followed, in the Preseli Hills, by the development of peat and heather moorland, sometimes boggy, and probably associated with wet cool conditions in the 15th to 17th centuries as well as overgrazing. Some settlements on poorer soils may have been particularly vulnerable to these environmental changes. This is suggested by comparing the distribution of archaeological records for deserted settlements with pre-1500 place names, the latter – given the restrictions of historical and archaeological records – providing useful proxies for medieval settlement (Fig. 5). Almost all the place names are located on soils of greater fertility and are frequently close to boundaries with less fertile soils that are often, in modern times, liable to seasonal waterlogging. These variations in fertility and drainage determine whether areas are better suited for arable or for seasonal pasture, with the siting of settlements at soil-zone

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

interfaces providing optimal access. The resulting patterns of common-edge settlement are widely recognised in England (Williamson 2013, 132, 152‑154, 158‑159). By contrast, the archaeologically identified deserted (medieval/post-medieval) rural settlements and long huts can be seen to be largely on poorer, seasonally waterlogged, upland soils where heath and reeds now replace early medieval pasture. This is especially evident on the northern Preseli Hills’ slopes. They are, moreover, often located close to the boundary of different poor soils within the moorland area – specifically on the edge of seasonally waterlogged Hafren soils close to their junction with the wetter Wilcocks 1 soils – hinting that the Hafren soils once presented better opportunities than exist today (Comeau 2019b, 159). Similar patterns can be detected for some enclosed (‘hillfort’) settlements of late prehistoric, Romano-British, and possible early medieval date, the distribution of which overlaps with the medieval placename sites, but extends to higher altitudes in moorland areas where there would have been good grazing up to the early medieval period. Comparisons

Similarities can be identified with evidence elsewhere in Wales, though currently nowhere presents so broad a picture (for Scottish sites see Dixon 2018). Two sites, both in north Wales, can be briefly considered. Like Carn Goedog, neither is known from written records. Hafod Nant y Criafolen on the Denbigh Moors (Allen 1979), excavated in 1973‑1974, is the most extensive and fully published excavation of a hafod site in Wales. Like Carn Goedog it is distinctive for its grouping of houses. Seven rectangular huts with stone foundations, shale walls, and (probably) heather-covered roofs and generally accompanied by small enclosures, were spaced out, 25 m to 90 m apart, beside a tributary stream that cut into the eastern slope of a broad river valley. Finds – largely pottery, metal objects, and spindle whorls from middens outside the huts – were more prolific than at any other Welsh hafod, and gave a 15th- to 16th-century date range that correlates with the excavation’s single radiocarbon date, noted earlier, from a pit under one building. At another site, on Moel Rhiwlug (Silvester 2011, 52) at the head of a boggy valley on the Denbigh Moors, pollen sampling in 2006 rather than excavation provided information. It showed cultivation as well as pastoral activity around the site, though remoteness and isolation argue for seasonal rather than permanent settlement. Nearby was a single unexcavated long hut, 9 x 5 m overall, on a slight platform, with walls no more than 300 mm

high and a large enclosure. Peat here did not develop until the early 12th century, and the pollen shows oats and hemp being grown in the late 12th to 13th century, very possibly taking advantage of soil enriched by animal dung. The surrounding landscape at this point was rich grassland with some trees, and charcoal suggests the use of burning to control vegetation or to prepare for shortterm cultivation. Cultivation appears to have stopped by the 1300s, and though grazing continued, the grassland deteriorated as wetter conditions set in, with heather moorland widespread by the 17th century. Summing up

How these examples integrate with broader patterns of seasonal settlement across Wales is currently somewhat uncertain, given the absence of a systematic national review of survey data, coupled with limited excavation and radiocarbon dating evidence, the shortage of studies of site patterning and of the relationship between permanent and seasonal settlement, and the lack of investigations into lowland seasonal settlements. Some things are clear, though. Seasonal settlement in Wales must be understood against a locally fluctuating combination of factors that include upland environmental deterioration, alterations in agro-pastoral practices, and economic changes. Although the remains of seasonal dwellings are broadly similar across Wales, there is no simple single narrative for their use and patterning. Crop-growing evidence needs cautious interpretation, since this could have taken place within a flexible agricultural system of infield/outfield husbandry where occasional cultivation of seasonal pasture was an established practice, often using land enriched by cattle manure (Comeau 2019a). An opportunistic approach to use of seasonal pasture areas can therefore be expected, with some short-term cultivation as well as periods of more intense exploitation. Some sites might become more permanent, while at others summer settlements follow the abandonment of high-medieval permanently cultivated upland areas (Silvester 2000). Environmental setting and differing modes of exploitation would have made some settlements more vulnerable to change than others. All in all, it is a picture that should have particular resonance for us as we face 21st-century economic and climate change. Acknowledgements

We are grateful to Mike Parker Pearson for permission to use Carn Goedog material.

Comeau and Silvester

121

References:

Allen, D. 1979: Excavations at Hafod y Nant Criafolen, Brenig Valley, Clwyd, 1973‑4, Post-Medieval Archaeology 13, 1‑59. Austin, D. 1988: Excavations and survey at Bryn Cysegrfan, Llanfair Clydogau, Dyfed, 1979, Medieval Archaeology 23, 130‑165. Briggs, S. 1985: Problems of the early agricultural landscape in upland Wales, as illustrated by an example from the Brecon Beacons, in: Spratt, D. – Burgess, C. (eds.), Upland settlement in Britain. BAR British Series 143. British Archaeological Reports. Oxford, 285‑316. Browne, D.M. – Hughes, S. (eds.) 2003: The archaeology of the Welsh uplands. Royal Commission on the Ancient and Historical Monuments of Wales. Aberystwyth. Comeau, R. 2019a: The practice of ‘in rodwallis’: medieval Welsh agriculture in north Pembrokeshire, in: Comeau, R. – Seaman, A. (eds.), Living off the land: agriculture in Wales c. 400 to 1600 AD. Windgather. Oxford, 130‑152. Comeau, R. 2019b: Land, people and power in early medieval Wales: the cantref of Cemais in comparative perspective. Unpublished PhD thesis presented to University College London Institute of Archaeology. Courtney, P. 1991: A native-Welsh mediaeval settlement: excavations at Beili Bedw, St Harmon, Powys, Bulletin Board of Celtic Studies 38, 233‑255. Davies, E. 1980: Hafod, hafoty, and lluest: their distribution, features and purpose, Ceredigion 9:1, 1‑41. Davies, E. 1984: Hafod and lluest: the summering of cattle and upland settlement in Wales, Folk Life 23:1, 76‑96. Davies, T. 2019: Culture, climate, coulter and conflict: pollen studies from medieval Wales, in: Comeau, R. – Seaman, A. (eds.), Living off the land: agriculture in Wales c. 400 to 1600 AD. Windgather. Oxford, 174‑198. Davies, W. 1978: An early Welsh microcosm: studies in the Llandaff Charters. Royal Historical Society. London. Davies, W. 1982: Wales in the Early Middle Ages. Leicester University Press. Leicester. Dixon, P. 2018: What do we really know about transhumance in medieval Scotland? in: Costello, E. – Svensson, E. (eds.), Historical archaeologies of transhumance across Europe. Routledge. Abingdon, 59‑73. 122

Dyfed HER: Dyfed Archaeological Trust Historic Environment Record, published online at http://www. dyfedarchaeology.org.uk/. Fox, A. 1939: Early Welsh homesteads on Gelligaer Common, Glamorgan, Archaeologia Cambrensis 94, 163‑199. Herring, P. 2012: Shadows of ghosts: early medieval transhumants in Cornwall, in: Turner. S. – Silvester, B. (eds.), Life in medieval landscapes: people and places in the Middle Ages. Windgather Press. Oxford, 89‑105. Hooke, D. 2019: Resource management of seasonal pasture: some English/Welsh comparisons, in: Comeau, R. – Seaman, A. (eds.), Living off the land: agriculture in Wales c. 400 to 1600 AD. Windgather Press. Oxford, 37‑56. Jenkins, D. (ed. and trans.) 1990: The Law of Hywel Dda. Gomer Press. Llandysul. Jones, G.R.J. 1964: The distribution of bond settlements in north-west Wales, Welsh History Review 11, 19‑36. Kelly, R. S. 1982: The excavation of a medieval farmstead at Cefn Graeanog, Clynnog, Gwynedd, Bulletin of the Board of Celtic Studies 29, 859‑908. Kissock, J. – Johnstone, R. 2007: Sheephouses and sheepcotes – a study of the postmedieval landscape of Cefn Drum, Gower, Studia Celtica 41, 1‑23. Leighton, D. 2012: The western Brecon Beacons: The archaeology of Mynydd Du and Fforest Fawr. Royal Commission on the Ancient and Historical Monuments of Wales. Aberystwyth. Locock, M. 2006: Deserted rural settlements in south-east Wales, in Roberts, K. (ed.), Lost farmsteads: deserted rural settlements in Wales. Council for British Archaeology. York, 41‑60. Olding, F. 2016: Archaeoleg ucheldir Gwent/The archaeology of upland Gwent. Royal Commission on the Ancient and Historical Monuments of Wales. Aberystwyth. Peate, I.C. 1946: The Welsh house. Brython Press. Liverpool. Roberts, K. (ed.) 2006a: Lost farmsteads: deserted rural settlements in Wales. Council for British Archaeology. York. Roberts, K. 2006b: The deserted rural settlement project; a summing up, in: Roberts, K. (ed.), Lost farmsteads: deserted rural settlements in Wales. Council for British Archaeology. York, 171‑186.

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Roberts, S.E. 2019: Living off the land in medieval Welsh law, in: Comeau, R. – Seaman, A. (eds.), Living off the land: agriculture in Wales c. 400 to 1600 AD. Windgather. Oxford, 78‑92. Schlee, D. – Comeau, R. – Parker Pearson, M. – Welham, K. 2018: Carn Goedog medieval house and settlement, Archaeologia Cambrensis 167, 245‑255. Seymour, W.P. 1985: The environmental history of the Preseli region of South-West Wales over the past 12,000 years. Unpublished PhD thesis presented to Aberystwyth University. Silvester, R.J. 2000: Medieval upland cultivation on the Berwyns in north Wales, Landscape History 22, 47‑60. Silvester, R.J. 2006: Deserted rural settlements in central and north-east Wales, in: Roberts, K. (ed.), Lost farmsteads: deserted rural settlements in Wales. Council for British Archaeology. York, 13‑39. Silvester, R.J. 2011: Mynydd Hiraethog. The Denbigh Moors. Royal Commission on the Ancient and Historical Monuments of Wales. Aberystwyth. Silvester, R.J. – Kissock, J. 2012: Wales: medieval settlements, nucleated and dispersed, permanent and seasonal, in: Christie, N. – Stamper, P. (eds.), Medieval rural settlement, Britain and

Ireland, AD 800‑1600. Windgather Press. Oxford, 151‑171. Smith, G. – Thompson, D. 2006: Results of the project excavations, in: Roberts, K. (ed.), Lost farmsteads: deserted rural settlements in Wales. Council for British Archaeology. York, 113‑132. Strachan, D. – Sneddon, D. – Tipping, R. 2019: Early medieval settlement in upland Perthshire: excavations at Lair, Glen Shee 2012‑17. Archaeopress. Oxford. Thomas G.C.G. (ed.) 1997: The charters of the Abbey of Ystrad Marchell. National Library of Wales. Aberystwyth. Ward, A. 1991: Transhumant or permanent settlement, in: James, H. (ed.), Sir Gar. Studies in Carmarthenshire history. Carmarthenshire Antiquarian Society. Carmarthen, 1‑22. Ward, A. 1997: Transhumance and settlement on the Welsh uplands: a view from the Black Mountain, in: Edwards, N. (ed.), Landscape and settlement in medieval Wales. Oxbow Books. Oxford, 97‑111. Williamson, T. 2013: Environment, society and landscape in early medieval England: time and topography. Boydell Press. Woodbridge. Winterbottom, M.E.T. 1978: Gildas: the ruin of Britain and other works. Phillimore. London.

Comeau and Silvester

123

Imagining and identifying seasonal resource exploitation on the margins of medieval Ireland Eugene Costello*

Abstract

This paper sheds light on the neglected question of how archaeologists might imagine and identify evidence of seasonal land use in medieval Ireland (c. 450‑1600). Up to now, research on seasonal resource exploitation in Ireland’s past has focused on the post-1550 period, and primarily on upland booley sites used in transhumance. Archaeologists have not paid very much attention to earlier seasonality, and when they do there is usually an emphasis on sites rather than landscapes. To understand the role of seasonal land use in the medieval Irish economy, and ascertain where it took place, we need a conceptual framework that takes a site’s physical resource catchment into account as well as its social occupants. A remarkable text from the early second millennium AD provides a starting point. Outlining the story of a rich lowland farmer on a grazing circuit in the Wicklow Mountains, it offers an insight into seasonal upland ‘taskscapes’, including the grazing of cows in woodland pasture, the number of people who travelled, and the association of such activity with hunting. The paper then assesses the archaeological evidence for such activities around Ireland, concluding that it has been quite meagre. In addition to a bias in development-led excavation towards fertile lowland areas, a number of recent land-use trends have impacted the archaeological record of Ireland’s lower upland slopes, i.e. those most likely to be used in medieval times. These trends include 18th- and 19th-century population growth and upland improvement as well as 20th-century afforestation and mechanisation. The paper proposes a nationwide programme of remote sensing to define the country’s remaining sections of unenclosed, unimproved, and unplanted hillsides below 300 m a.s.l., followed by field survey to select structures of potentially medieval form within them, and trial excavation to obtain dates and potentially environmental proxies from them. Keywords: seasonality, long-distance transhumance, hunting, medieval Ireland, house morphology. * Institutionen för arkeologi och antikens kultur Stockholms universitet 106 91 Stockholm Sweden [email protected]

Résumé

Imaginer et identifier l’exploitation saisonnière des ressources aux marge de l’Irlande médiévale Cet article met en lumière la question négligée de la façon dont les archéologues pourraient imaginer et identifier des preuves d’utilisation saisonnière des terres dans l’Irlande médiévale (c. 450‑1600). Jusqu’à présent, la recherche sur l’exploitation saisonnière In: Piers Dixon & Claudia Theune (eds), 2021: Ruralia XIII: Seasonal Settlement in the Medieval and Early Modern Countryside. Sidestone Press, Leiden, pp. 125-136.

125

des ressources dans le passé de l’Irlande s’est concentrée sur la période postérieure à 1550, et principalement sur les sites de hauts plateaux utilisés en transhumance. Les archéologues n’ont pas accordé beaucoup d’attention à la saisonnalité antérieure et lorsqu’ils le font, l’accent est généralement mis sur les sites plutôt que sur les paysages. Pour comprendre le rôle de l’utilisation saisonnière des terres dans l’économie irlandaise médiévale et déterminer où elle a eu lieu, nous avons besoin d’un cadre conceptuel qui tienne compte du bassin de ressources physiques d’un site ainsi que de ses occupants sociaux. Un texte remarquable du début du XXe siècle fournit un point de départ. Décrivant l’histoire d’un riche fermier de plaine sur un circuit de pâturage dans les montagnes de Wicklow, il offre un aperçu des «tâches» saisonnières des hautes terres, y compris le pâturage des vaches dans les pâturages forestiers, le nombre de personnes qui ont voyagé et l’association d’une telle activité avec la chasse. L’article évalue ensuite les preuves archéologiques de ces activités en Irlande, concluant qu’elles ont été assez maigres. En plus d’un biais dans les fouilles axées sur le développement vers les zones de plaine fertiles, un certain nombre de tendances récentes de l’utilisation des terres ont eu un impact sur les archives archéologiques des pentes inférieures des hautes terres de l’Irlande, c’est-à-dire celles les plus susceptibles d’être utilisées à l’époque médiévale. Ces tendances montrent la croissance démographique et l’amélioration des hautes terres aux XVIIIe et XIXe siècles ainsi que le boisement et la mécanisation au XXe siècle. L’article propose un programme national de télédétection pour définir les sections restantes du pays de coteaux non clôturés, non amendés et non culivés en dessous de 300 m d’altitude, suivi par une enquête sur le terrain pour sélectionner des structures de forme potentiellement médiévale en leur sein, et des sondages pour obtenir des datations et des marqueurs environnementaux. Mots-cles : saisonnalité, transhumance sur longue distance, chasse, Irlande médiévale, morphologie de la maison. Zusammenfassung

Saisonale Ressourcennutzung in marginalen Landschaften des mittelalterlichen Irlands In diesem Beitrag wird die bislang wenig erforschte Frage, welche archäologischen Hypothesen und Hinweise es für

Introduction

Archaeological research on seasonal settlement in rural landscapes of medieval and post-medieval Ireland has been relatively sparse and, at least for the post-medieval period, has been dominated by the study of transhumance or 126

die saisonale Landnutzung im mittelalterlichen Irland (ca. 450‑1600) gibt, diskutiert. Bisher konzentrierte sich die Forschung zur saisonalen Exploitation von Ressourcen in der Vergangenheit Irlands auf die Zeit nach 1550 und hauptsächlich auf Standorte im Hochland (sog. Booley-Sites), die für die Transhumanz genutzt wurden. Archäologen haben der älteren Saisonalität nicht viel Aufmerksamkeit geschenkt, und wenn sie dies tun, liegt der Schwerpunkt eher auf Siedlungen als auf Landschaften. Um die Rolle der saisonalen Landnutzung in der mittelalterlichen irischen Wirtschaft zu verstehen und zu lokalisieren, benötigen wir einen konzeptionellen Rahmen, der das Einzugsgebiet der physischen Ressourcen eines Standorts sowie seine Bewohnerinnen und Bewohner berücksichtigt. Ein bemerkenswerter Text aus dem frühen zweiten Jahrtausend n. Chr. bietet einen Ausgangspunkt. Es wird die Geschichte eines reichen Tieflandbauern auf einem Weideareal in den Wicklow Mountains beschrieben, dadurch erhalten wir einen Einblick in die Aufgaben einer saisonale Hochlandnutzung, einschließlich der Beweidung mit Kühen auf Waldweiden, der Anzahl der Reisenden und der Verbindung solcher Aktivitäten mit Jagd. In dem Beitrag werden dann die archäologischen Belege für solche Aktivitäten in ganz Irland diskutiert, im Ergebnis waren solche Aktivitäten sehr selten. Zusätzlich zu Beobachtungen bei Ausgrabungen in fruchtbaren Tieflandgebieten haben eine Reihe von archäologischen Aufzeichnungen zu jüngsten Landnutzungen der unteren Hochlandhänge Irlands neue Erkenntnisse erbracht, also in den Regionen, die im Mittelalter am wahrscheinlichsten genutzt wurden. Diese Trends umfassen das Bevölkerungswachstum und die Verbesserung des Hochlands im 18. und 19. Jahrhundert sowie die Aufforstung und Mechanisierung im 20. Jahrhunderts. Es wird ein landesweites Fernerkundungsprogramm vorgeschlagen, um die verbleibenden Regionen des Landes mit noch nicht kultivierten bzw. bepflanzten Hängen unter 300 m ü. M. zu lokalisieren. Daran anschließen sollte sich eine Felduntersuchung, um Strukturen potenziell mittelalterlicher Form zu erkennen und durch Ausgrabungen archäologische und umwelthistorische Daten zu erhalten. Schlagwörter: Saisonalität, Transhumanz, Jagd, mittelalterliches Irland, Hausmorphologie.

summer pasturing of livestock (e.g. Costello 2017, 2020a). For the medieval period, which in Ireland runs from roughly AD 450 to 1600, there have been nods towards the existence of seasonal ‘booley’ sites by the authors of several archaeological textbooks (Edwards 1990, 34, 53;

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

O’Conor 1998, 96; O’Sullivan et al. 2013, 112). However, there has not been very much consideration given to how and why such sites might have been used in a wider system of transhumance. Historians have, in fairness, discussed the motivations behind medieval transhumance, but only in a general sense and there have been no local examples of how it might have worked (Lucas 1989, 58‑67; Ó Cróinín 1995, 101; Kelly 1997, 43‑45). This is now changing, thanks to the case-study research of Mark Gardiner (2018) in East Ulster and Costello (2020a) in Connemara and Donegal (both partly encouraged by the historical geography of Graham 1954). Nevertheless, the subject of seasonal transhumant settlement in medieval Ireland remains neglected overall. In addition, the role of other forms of land use and resource exploitation in producing seasonal settlement has scarcely been tackled. This chapter therefore takes a broader perspective on summer transhumance in uplands, focusing as much as possible on the pre-1600 period. Firstly, I provide context regarding the difficulty of identifying seasonal occupation in medieval Ireland, due to geographic biases in the recent history of archaeological excavation. I then outline how seemingly opposing processual and postprocessual conceptual approaches to landscape can be combined to produce a more balanced socio-ecological understanding of seasonal land use. This perspective is trialled using a medieval text of extraordinary value relating to the Wicklow Mountains, before going on to discuss the archaeological evidence for seasonality. The chapter highlights the potential pitfalls in identification and proposes a number of mitigating strategies.

period (c.  AD 450‑1100), Ireland now possesses an enormous body of excavated archaeological data (O’Sullivan et al. 2013, 1‑4). This is certainly true, but there remains a difficulty, in that the vast majority of archaeological excavations produced over the last three decades have been connected to development, and many of these excavations were funded by the National Roads Authority in the late 1990s and 2000s (now operating as Transport Infrastructure for Ireland [TII]) (see Stanley et al. 2017). As previous scholars have observed (O’Conor 2001, 331; Gardiner – O’Conor 2017, 133‑134), this has biased the archaeological record of medieval rural settlement towards the east and southeast of the Republic of Ireland, since that is where most of the development, especially the construction of new roads, has taken place. Indeed, with one exception, all motorways have led to Dublin and no motorways were constructed in west Munster, most of Connacht, the north Midlands or west /mid-Ulster (Fig. 1). The TII strategy has been to avoid known archaeological ‘hotspots’ and instead to recommend routes through areas of the landscape where sites and monuments were not as densely distributed and more straightforward to excavate. Although this has led to an exploration of rural settlement beyond the most central elite places, it is debatable whether excavations really have occurred ‘in more peripheral areas’,

Excavation history in Ireland and the bias against ‘marginal’ landscapes

It is very difficult to be confident that the occupation of a site was primarily seasonal – as opposed to temporary or year-round – without an understanding of the wider landscape in which the site is situated and, with that, the kind of activities that its occupants could have been engaged in. Geo-archaeologically, it is possible to identify punctuated occupation at an individual site if wind-blown sand enters it and accumulates when people are absent, that is, by undertaking micromorphological analysis of thin sections prepared from excavated soil monoliths (Kupiec et al. 2016; Kupiec – Milek 2018). However, it is not possible to tell if these layers are the result of seasonal absence or periods of abandonment that lasted several years. Furthermore, wind-blown sand is not likely to be a factor at inland sites in Ireland. There is a more significant obstacle to the identification of seasonality at the site level in Ireland, and that is the lack of excavation at potential sites. Much has been made of the fact that, for the early medieval

Fig. 1. Distribution of known archaeological excavations in Ireland (Data source: Database of Irish Excavation Reports. Accessed online at excavations.ie) (Background map data: © 2021 Google).

Costello

127

and whether they have exposed ‘a cross-section of the later medieval landscape’ (Gardiner – O’Conor 2017, 134‑135; Stanley et al. 2017, 3). At the end of the day, motorways and most other developments have taken place on dry inland areas below 100 m asl, avoiding hills, mountains, the coastline, and boglands, i.e. environments where there might be a higher likelihood of identifying seasonal land use and resource exploitation. The only developments that have occurred in such environments are wind farms, usually in cut-away lowland bog or in upland areas now deemed to be agriculturally marginal; however, the archaeological reports for these are still only being produced and none have been published. Taken as a whole, then, the recent wave of development-led excavation in Ireland has the potential to compound the long-standing bias in research against these ‘marginal’ places. This is important to keep in mind, given the recent upsurge in attempts to build models of human land use and settlement change using very large datasets and series of radiocarbon dates. As things stand, landscapes where seasonal activity is likely to have been common are at best underrepresented and at worst left out altogether, reinforcing the view that they were unimportant. Theorising landscapes of seasonal activity

With the excavation evidence still lacking for now, a contextual and multidisciplinary approach is essential to finding evidence of seasonal resource exploitation. Considering sociocultural, economic, and environmental factors together makes it easier to assess the range of activities that are likely to have been possible in a locality, and these are key to gauging the nature of occupation – be it on a temporary (a few days), seasonal (more than a month), or year-round basis. In this regard, site catchment analysis, which has long been used in North America to determine the productivity of an area in which prehistoric settlement is found, has its merits (VitaFinzi et al. 1970). Its attention to natural resources and carrying capacity is very relevant to a debate on seasonal land use, whether we have in mind the availability of pasture for livestock, the migration of fish, or the ease of accessing particular environments and transporting goods out of them at different times of year. Having said that, the aim of objectively ‘calculating’ the resources of an area means that the approach is inherently biased against less-tangible aspects of past seasonal activity, such as the perception of power and territory, and magicoreligious beliefs about the liminality of certain spaces (as for instance captured in various of the papers contained in Bis-Worch – Theune 2017). What is more, it would be difficult to acknowledge the mutability of seasonal activity within a site-catchment reconstruction, given that each one is primarily a horizontal representation or ‘time-slice’ of the landscape. 128

By contrast, in Tim Ingold’s highly influential concept of ‘taskscape’, the landscape is a socially constructed space imbued with temporality (Ingold 1993). In other words, human activities are seen as ongoing processes that interlock with one another. It also recognises the importance of human perception and non-human agency in shaping these activities over time. Yet even this concept is not ideal as a way of gauging seasonal land use and resource exploitation. While Ingold is keen to reject the false dichotomy between nature and culture, and he acknowledges that things can ‘act back’, his taskscape concept is ultimately more successful at conveying the interlocking nature of human activities than it is at assessing how easy or difficult these activities were in relation to the physical environment. For example, while a taskscape of upland pastures might be very useful at evoking the role of certain landscape features in a herder’s recognition of other communities’ territories, and in remembering not to stray into his/her family’s cultivated fields, it would surely be of less help in imagining how long cattle could be grazed in that upland pasturage during summer or why the crops failed in a given year. The creation and perception of social space is clearly important to the debate and can be focused on if seasonality is clear (see Costello 2017). However, moving the chronological focus back to medieval times, and thinking of seasonal activity in a broader sense, a more explicitly human-environment approach is necessary (Hoffmann 2014; Schreg 2014). Seasonal pastoralism and other activity in medieval uplands: a documentary perspective

The documentary evidence for medieval Irish transhumance is relatively encouraging for a scholar hoping to identify seasonal settlement. While there was no systematic local record-keeping regarding grazing rights and the specific locations of herding sites (usually referred to as a buaile or áirí), Old and Middle Irish texts make reference to the seasonal removal of cattle from the homestead in a way that suggests it was common (see Lucas 1989, 58‑67; Kelly 1997, 40, 43‑45, 427). By far the most informative is the third Life of St Kevin of Glendalough or, more properly, Caoimhghin Glinne Da Lach. It includes a story of how he was ‘found’ in Glendalough in the Wicklow Mountains, having been hiding in this upland ‘wilderness’ as a hermit for seven years (Plummer 1922, 157‑159). Although St Kevin is said to have died in AD 618, and myths and traditions about him were being committed to text from the 7th century all the way up to the early modern period, the Middle Irish text in this particular life would appear to date it to roughly the 11th century AD. The story of how St Kevin was found is of interest to us, because it involves a farmer named Dioma mac

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 2. Map of Ireland showing two medieval Irish kingdoms and other sites and regions mentioned in the text: 1) Glendalough; 2) Eastern Connemara; 3) the Antrim uplands; 4) the Mourne Mountains; 5) the Galtee Mountains; 6) Doonloughan; 7) the Beara Peninsula; 8) South County Kerry (© Eugene Costello).

Fergna, who possessed a herd of one hundred cows and would therefore have been a person of quite high status in a medieval Irish context (for information on social grades, albeit a couple of centuries earlier, see Kelly 1988, 10). Dioma is said to have been from the fertile lowland Kingdom of Mide and was undertaking a cuairt bhuailltechuis (literally, a grazing or booleying circuit) with his cows in the glacial valley of Glendalough in the Wicklow Mountains. With Mide lying some 60 km to the northwest at its closest and as much as 100‑150 km away at its farthest, Dioma would have needed to travel quite a distance to reach this upland area (Fig. 2). At any rate, it so happened that Dioma’s herd of cows was grazing in the very part of Glendalough where St Kevin was hiding, and that these were woodland pastures (translated from original: ‘airghe do bhuaibh Dhioma do bheth ag inghilt ar an c-coill ina raibhe Caoimhghin aga cheilt féin’). Any day the cows grazed in the wood where the saint was hiding, one cow would go to the saint and lick his feet, and upon returning home in the evening with the rest of the cows would give a miraculous amount of milk. This was

eventually investigated by Dioma’s herdsman (buachuill na m-bo), who finds the saint and brings Dioma and his children to the spot. Here Kevin greets them as people ‘on exile’ (ar deoraidheacht) from Mide who have been sent by God to him, and performs a miracle to bring one of Dioma’s children back to life. The Saints’ Lives are hagiographical stories and very little if any historical value should be placed on them as biographies of the people involved. However, the settings in which the stories take place have the potential to reveal some assumptions about non-saintly life around the time these stories were told and written down, particularly the mundane farming practices and landscapes one might be expected to encounter. Viewed as a product of Ireland around the start of the first millennium AD, this life of St Kevin is highly significant for several reasons. Firstly, and least controversially, it is the first recorded instance of the word buailteachas, which is Modern Irish for ‘transhumance’. Then there is the question of the distance travelled by Dioma. At 60 km or more, it would far outstrip anything recorded in post-medieval transhumance in either Ireland or Britain, with echoes only in the so-called ‘creaghting’ or caoruigheacht phenomenon that emerges in the late 16th and 17th centuries. However, creaghts tended to be larger refugeelike wanderings of people and livestock and are most associated with the English conquest and colonisation of Ireland (see Simms 2015; Costello 2020a). While Dioma may have had stops on his way to Glendalough, according to this story he still ultimately had a base in Mide as a brughaidh boi-cheadach (one-hundred cow farmer), and the only other people with him were his children, his muinntir (household or followers) and a herdsman. Having noted all that, the issue of distance is complicated by the shorter earlier versions of the Life of St Kevin, which describe Dioma as a farmer from Laigean who had originally been from Mide (see Byrne 1971, 140, who speculates that he was of the Uí Máil dynasty). Even in this case, though, he would still have to travel at least 20 km to reach Glendalough, and this is greater than most transhumant movements in post-medieval Ireland or Britain. Whether distances for ‘grazing circuits’ in the 11th century really did sometimes reach 60 km or more, as the third Life of St Kevin implies, is uncertain and will continue to be uncertain in the absence of stable isotope analysis of contemporary animal remains from the region. However, the present writer would argue that the hagiographer(s) must have considered it a relatable practice for both himself and a lay audience, since only the saints could be seen to perform miracles. It should also be remembered that 7th- and 8th-century law texts describe it as a general right of cattle to graze freely on mountain pastures (Kelly 1997, 44). If this was still the case by the new millennium, rich lowland Costello

129

farmers matching Dioma’s description would have had an obvious incentive to make these grazing circuits an annual event. The number of people and cattle who are said to have gone to Glendalough has even greater importance for how we envisage medieval booleying activity. While the hagiographer makes no mention of tents, huts, or houses, he clearly imagines a central node in the general activity area, in that the cows are all said to have ‘returned’ each evening for milking. At such a site, perhaps on open ground in the centre of the valley, we can assume that a farmer like Dioma and his children would have had some form of shelter from the elements, along with a wattle fence or earthen bank for keeping the cows in at night. Of course, this brings us to a key point, namely that this was a dairy herd. If Dioma had brought all of his cows with him, and not let them out to client farmers, he would have needed all of his children and others besides to milk them – in a mid-20th-century context, at least 10 adults would be needed for such a task, not to mention churning the milk into butter, and carrying it on the rest of their journey (estimation by author’s older relatives and neighbours). The reference to milking is also important in that it makes a wintertime visit highly unlikely; a grazing circuit of lactating cows would almost certainly have taken place during the summer half of the year, perhaps staying in one valley for a few weeks each. There are parallels for this in Highland Perthshire in Scotland during the 17th and 18th centuries, where some tenant farmers would rotate between shielings over the course of the summer. They stayed no more than three to four weeks at each so as to prevent overgrazing (Bil 1990, 179‑186). If several tens of kilometres were indeed traversed by some rich farmers in Ireland around the 11th century, it is also possible that medieval Irish grazing circuits (cuairteanna bhuailltechuis) were organised into a form of two- or three-step transhumance. This is widely attested in other European countries in recent times. For example, in parts of Norway and the Alps, people would take their cattle up to near pastures in late spring/early summer, before moving on to the highest or most distant pastures in midsummer (Daugstad et al. 2014, 252; Andres 2018, 157). The last information to be gleaned from this Life of St Kevin relates to the wider pastoral landscape in the mountains. Firstly, it is clear that the cows were intentionally grazed in the woodlands, a plausible scenario given that Murphy and Potterton (2010) have pointed to autumn woodland grazing in this region during the later medieval period (c. 1100‑1600). While the practice of grazing cows in woodlands seems to have survived in some upland areas up to the 18th century (see Costello 2020b), it contrasts strongly with most instances of post-medieval transhumance in Ireland, in which cattle grazed on open hillsides and mountains (McDonald 2014; 130

Costello 2020a). In Glendalough itself, the presence of woodland is supported by a pollen record from the Lower Lake. It shows that oak, birch, and alder were much more widespread in the second half of the first millennium and early second millennium than at present, but not totally dominant either, since hazel, grass, and upland herbs were also well represented (Haslett et al. 2006, fig. 3). With the way to St Kevin’s hiding place said to have been very rough and dense (ro-aimhreidh dluith), it is not unreasonable to speculate that woodland covered much of the valley’s steep rocky sides. The story also makes clear that hunting was to be expected in such a landscape. Instead of coming to help St Kevin, one of Dioma’s sons decides that he wants to go off hunting, but divine justice sees his hounds set upon and devour him. It is not stated what kind of prey he had been after; wolves, wild boar, or red deer were all hunted in early medieval Ireland (Kelly 1997, 272‑285), so the hunting could conceivably have taken place in either the woodland or in open heath on higher slopes. This feature of Dioma’s trip is important, as it suggests that seasonal upland grazing in medieval times provided opportunities for other forms of resource use. It could also be said of the post-medieval period, but by then it tended to be peat cutting and illicit distilling that took place near booley sites (Costello 2018). By the 18th and 19th centuries, there were fewer large wild mammals in the Irish landscape and the tenant farmers who practised booleying may in any case have lacked the right to hunt such game (Costello 2020a, ch. 6). However, there are signs that early modern activity in Ireland’s uplands may have retained some elements of what is hinted at in the story of Dioma. In the 15th and 16th centuries, Gaelic Irish elites hunted red deer in parts of southwest Donegal and Connemara (see Lacey 1998, 26; FitzPatrick 2013, 95‑118) where cattle would almost certainly have been grazed as well, at least during the summer. As late as 1699, the English travel writer, John Dunton, could visit a ‘booley or summer habitation’ in the hills of east Connemara that a local magnate of the O’Flahertys was using as a base for hunting deer (MacLysaght 1939, 344). Some Gaelic and Hiberno-Norman lords would also use upland and woodland cabins as hideouts when trying to evade Tudor English forces (see Everett 2014, 42, 46) – one wonders whether they had a prior association with these places through hunting and/or booleying. The archaeological (in)visibility of medieval seasonal settlement in uplands?

The lowest part of Glendalough valley developed into a major monastic centre, dedicated to St Kevin, over the course of the medieval period (Doherty et al. 2011). Thus, it is unlikely that this precise location was being visited

SEASONAL SETTLEMENT IN THE MEDIEVAL AND EARLY MODERN COUNTRYSIDE

Fig. 3. Example of a ‘low’ hillslope in Glencar, County Kerry, ranging in altitude from 80 m to 300 m. Note 19th-century hill farms in the southwest, recent clearances to the northeast, and sections of non-native conifer plantations (Satellite imagery: © 2021 Maxar Technologies).

by large herds of dairy cows and hunting hounds in the 10th, 11th, and 12th centuries AD. However, the story does provide a window into perceptions of pastoralism generally at that time, and into the seasonal usage of uplands especially. The question that remains is whether such activities can actually be traced through archaeology. Before we even begin to discuss potential sites in uplands, the impact of several post-medieval land-use trends on the medieval archaeological record has to be recognised. Between the early 18th and the mid-19th century, the population of Ireland rose from 2 million to 8.5 million, and during this time many landlords began to promote the agricultural improvement of ‘marginal’ lands. As a result, there was a massive expansion of small tenant farms onto lands that would have been ideal for medieval summer grazing, i.e. low-lying blanket bog and the lower slopes of mountain ranges, below c. 400 m asl (Connell 1950; Costello 2015). Following this, in the late 20th century, large sections of these post-medieval farmscapes, and a lot of other hill pasture that had escaped colonisation, were covered by plantations of dense, drain-riven commercial forestry (Ní Dhubháin et al. 2009). Finally, with the

introduction of track machines to the Irish countryside and the incentivisation of land reclamation, many hill farmers have recently been able to clear their uppermost fields of inconvenient ‘lumps and bumps’, definitively erasing many features that had somehow survived both enclosure and afforestation (Fig. 3). Together, these three phenomena have greatly reduced the amount of land where archaeologists might expect to find surface remains of seasonal medieval settlement. This is not to say that remaining rough pasture on the highest slopes and the remotest mountain valleys are devoid of medieval evidence, but simply that it would be less likely to appear there, given that medieval farmers had better grazing options at slightly lower altitudes, i.e. where there was less precipitation, slightly warmer soils, and some woodland grazing. So, at the higher altitudes where ‘upland’ archaeological survey usually now takes place, we should be aware of a possible bias towards later periods of seasonal activity. Having said that, there are still patches of ‘relict’ farmland on Ireland’s lower hillsides, under roughly 300 m asl, albeit the evidence for medieval booleying is not straightforward. The challenge is to differentiate it Costello

131

Fig. 4. A largely intact relict landscape in Cloon West, County Kerry, containing hut sites (red dots), but also a complex field system and probable early medieval cashel ringfort, i.e. a small farmstead. Mapped by Eugene Costello through remote sensing and field survey (© 2021 Maxar Technologies).

from the remains of opportunistic year-round settlement, which could range in date anytime from late prehistory up to the post-medieval period. One of the only well-dated medieval sites discussed in the context of transhumance has been Ballyutoag in Co. Antrim, where Williams (1984) found evidence for late 7th- to 10th-century occupation. Given its location at an altitude of just over 250 m, on what is now a peaty hillside above the limit of post-medieval enclosure, the excavators suggest that this site functioned as a ‘transhumance village’. However, the sheer complexity and extent of the remains – large curvilinear enclosures containing 23 circular and oval houses, plus associated fields containing cultivation ridges – would suggest that this settlement functioned primarily as a year-round settlement. Similar curvilinear settlement enclosures are found on the hill slopes of Killylane and Browndod in Co. Antrim (Williams – Yates 1984). A slightly later candidate for medieval booleying is Tildarg in Co. Antrim, where excavations dated occupation to the period, cal AD 1185‑1375 (Brannon 1984). Although Brannon argues that the site is connected to seasonal transhumant activity because of its altitude of 274 m, the house which he excavated here would, measuring 16 m by 6 m, be extremely large in an Irish context. In any case, the substantial rectangular enclosure around it measuring some 82 m by 53 m is not paralleled at likely summer settlements elsewhere in Ireland (see Gardiner 2012; McDonald 2014; Costello 2015, 2016). All told, the site bears more resemblance to a type of Anglo-Norman defended farmstead known as a ‘moated site’ than to a seasonal livestock enclosure (O’Conor 1998, 88). 132

A number of other broadly later medieval/early modern sites in the uplands of Antrim are also doubtful as seasonally occupied sites. For example, the concentration of 129 sub-rectangular/sub-oval structures at Goodland was once thought to have been used as a seasonal settlement in late medieval and early modern times (Case et al. 1969). However, the large number of dwellings make this doubtful, and Horning (2004) has made the more reasonable suggestion that it was a year-round settlement associated with 16thcentury Scottish immigration. There are even question marks over less-extensive sites like Glenmakeeran, Craigs, and Gortin due to the relatively large size of their dwelling structures (7‑11 m in total length; Williams – Robinson 1983; Williams – Goddard – McCorry 1988; Gardiner 2010). Indeed, discussing later medieval settlement in northeast Ulster generally, Gardiner (2018) has convincingly argued that the evidence for year-round settlement in upland areas has been significantly underestimated. A comparable story is emerging in the mountainous peninsulas of Kerry in southwest Ireland. Many of the circular stone huts or clocháin that tend to be cited as booley huts (Aalen 1964; Monk 1998, 36) are actually found on relatively low hill slopes (